limnological and paleolimnological …
TRANSCRIPT
LIMNOLOGICAL AND PALEOLIMNOLOGICAL INVESTIGATIONS OF
ENVIRONMENTAL CHANGE IN THREE DISTINCT ECOSYSTEM TYPES,
CANADIAN HIGH ARCTIC
by
BRONWYN ELIZABETH KEATLEY
A thesis submitted to the Department of Biology
in conformity with the requirements for
the degree of Doctor of Philosophy
Queen’s University
Kingston, Ontario, Canada
September 2007
COPYRIGHT © BRONWYN ELIZABETH KEATLEY, 2007
ii
ABSTRACT
The biological remains preserved in the sediments of Arctic lakes and ponds in the
Canadian High Arctic are important indicators of environmental change, especially as long-term
instrumental data are often lacking. Although recent studies have underscored variability amongst
these aquatic ecosystems, data are lacking from several key ecosystems. This thesis addresses
some of these critical knowledge gaps in the Canadian high Arctic, using diatom-based
limnological and paleolimnological techniques.
First, I explore the limnology and diatom ecology along a gradient of bioclimatic zones
on Melville Island in the western High Arctic. Lakes and ponds located in the most lushly
vegetated zone were significantly different from those elsewhere on the island, both in terms of
measured limnological variables and in terms of diatom assemblage composition. Diatom species
distributions from Melville Island can best be explained by differences in pH and related
variables.
Secondly, ponds and lakes located in a High Arctic oasis on northern Ellesmere Island,
recorded significantly higher specific conductivity, nutrients, and dissolved organic carbon than
freshwater bodies from the surrounding polar desert.
In Chapter 5, I provide an examination of long-term environmental change from Melville
Island, a region of the High Arctic for which no paleolimnological data exist. The timing of
diatom shifts in a dated sediment core from a small pond is consistent with the onset of climate
warming in the early 20th century.
Differences in ice cover have often been invoked to explain differences in the timing and
magnitude of diatom shifts in the Arctic, but this hypothesis has not been explicitly tested. In
Chapter 6, I compare two adjacent lakes with similar physical characteristics but different ice
cover regimes from northern Ellesmere Island. I provide strong evidence that extended ice cover
dampens diatom community responses to environmental change.
iii
In the final chapters, I determine that marine-derived nutrients significantly affect the
limnology of ponds on Cape Vera, Devon Island, and are related to the degree of seabird
influence. Although a portion of diatom species distributions can be linked to seabird influence,
the most abundant taxa show little relation to the nutrient gradient. In a paleolimnological
context, diatoms do not appear to provide a robust indicator of seabirds in High Arctic ponds.
iv
CO-AUTHORSHIP This thesis conforms to the Manuscript Format as outline by the School of Graduate Studies and Research. Each chapter has been written in the form of the journal in which it was submitted and contains its own literature cited. Marianne Douglas and John Smol, my thesis co-supervisors, are co-authors on all chapters. All fieldwork was conducted by myself, John Smol and Marianne Douglas, with the assistance of others noted in the acknowledgements section of each chapter. As several of the chapters in this thesis are referred to as “Keatley et al. 2007” or “Keatley et al. in press” etc., each “Keatley et al.” reference is followed by square brackets [ ] enclosing the chapter number. Chapter 2 was co-authored by Marianne Douglas and John Smol. I conducted the statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. This chapter has been published separately. [2] Keatley, B.E., M.S.V. Douglas, and J.P. Smol. 2007. Physical and chemical limnological characteristics of lakes and ponds across environmental gradients on Melville Island, Nunavut/N.W.T., High Arctic Canada. Fundamental and Applied Limnology. 168: 355-376. Chapter 3 was co-authored by Marianne Douglas and John Smol. I conducted the diatom-based lab work, diatom counts, statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. This chapter has been submitted for publication and is currently in review. [3] Keatley, B.E., M.S.V. Douglas, and J.P. Smol. In review. Evaluating the role of environmental and spatial variables on diatom species distributions on Melville Island (Canadian high Arctic). Submitted: 7 June 2007. Chapter 4 was co-authored by Marianne Douglas and John Smol. I conducted the statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. This chapter has been accepted for publication and is currently in press. [4] Keatley, B.E., M.S.V. Douglas, and J.P. Smol. 2007. Limnological characteristics of a high Arctic oasis and comparisons across northern Ellesmere Island. Arctic, in press. Chapter 5 was co-authored by Marianne Douglas and John Smol. I conducted the lab work, diatom counts, statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. This chapter has been published separately. [5] Keatley, B.E., M.S.V. Douglas, and J.P. Smol. 2006. Early-20th century environmental change inferred using sub-fossil diatoms from a small pond on Melville Island, N.W.T., Canadian High Arctic. Hydrobiologia. 533:15-26. Chapter 6 was co-authored by Marianne Douglas and John Smol. I conducted the lab work, diatom counts, statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. This chapter has been accepted for publication and is currently in press.
v
[6] Keatley, B.E., M.S.V. Douglas, and J.P. Smol. Prolonged ice cover dampens diatom community responses to recent climatic change in high Arctic lakes. Arctic Antarctic and Alpine Research, in press. Chapter 7 was co-authored by Marianne Douglas, John Smol, Jules Blais (University of Ottawa), and Mark Mallory (Canadian Wildlife Service). I conducted all the lab work related to diatoms, diatom counts, statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. Jules Blais provided δ15N and elemental data from the surface sediments and Mark Mallory provided seabird ecological information. This chapter is in preparation for publication. [7] Keatley, B.E., M.S.V. Douglas, J. Blais, M. Mallory, and J.P. Smol. In preparation. Impacts of seabird-derived nutrients on water quality and diatom species assemblages from Cape Vera, Devon Island, Canadian High Arctic. Chapter 8 was co-authored by Marianne Douglas, John Smol, Neal Michelutti (Queen’s University), Jules Blais (University of Ottawa), and Mark Mallory (Canadian Wildlife Service). I conducted all the diatom-based lab work, diatom counts, all of the statistical analyses, drafted all figures and tables, and was the primary author of the manuscript. Jules Blais provided the elemental and stable isotope data, and the 210Pb dates. Neal Michelutti provided the sedimentary inferred-chlorophyll a data analysed the 210Pb data. This chapter is in preparation for publication. [8] Keatley, B.E., M.S.V. Douglas, N. Michelutti, J. Blais, M. Mallory, and J.P. Smol. In preparation. Tracking seabirds through time: a multi-proxy paleolimnological study in the Canadian High Arctic.
vi
ACKNOWLEDGEMENTS First and foremost, I would like to thank my co-supervisors, John Smol and Marianne Douglas,
for their support and encouragement over the years. John and Marianne, you are not only
outstanding scientists, but fantastic people. I want to thank you for sharing your insights into
science and life in general, but most of all, thank you both for introducing me to the magical place
that is the Canadian High Arctic
I would also like to thank Brian Cumming, Scott Lamoureux, and Shelley Arnott for providing
feedback as members of my PhD committee, and Jules Blais, Linda Kimpe, Mark Mallory,
Shelley Arnott, and Rene Gregory-Eaves for all their help along the way in various capacities as
collaborators, field assistants, and general scientific discussion partners. All of the past and
present PEARL members have made PEARL a lovely place to work, but I would particularly like
to single out Kat Rühland, Neal Michelutti, and John Glew for their years of scientific discussions
and friendships.
Some of the most amazing friendships I have yet formed have occurred because of my tenure in
graduate school, and I would especially like to acknowledge Anita Holtham, Anne Harris, Carrie
Lyons, Claudia Kraft, Dan Selbie, David Chiasson, Dermot Antoniades, Jon Sweetman, Laura
Lawlor and Roger Bull. Even though “a lot of things can go wrong,” you’ve all been there to
help make sure that, in the end, things usually go right. Extra special thanks go to Angela
Strecker for joining me through all those grad school rites of passage and for being such a
fantastic friend.
I would like to thank my family (Sondra & Evans Keatley, Sarah & Jonathan Sewter, Beth
McFarlane, Franklin Keatley, and Josh & Catherine Keatley,) and extra-grad school friends
(especially Bridget Meigs, Jonathan Hill and Mico Devos) for providing moral support and for
helping me to always keep things in perspective. Finally, I would like to thank my partner,
Alexandre Poulain, for his encouragement, love, and support.
vii
TABLE OF CONTENTS Abstract…………………………………………………………………………………………….ii Co-authorship……………………………………………………………………………………..iv Acknowledgements……………………………………………………………………………..…vi Table of Contents……………………………………………………………………………...…vii List of Tables………………………………………………………………………………...…...xii List of Figures…………………………………………………………………………………...xiv Chapter 1: General Introduction and Literature Review………………………………....……1 Environmental Change in the Arctic …………………………………….……………...…………1 Canadian High Arctic limnology…..……………………………...…………………………….…2 Diatoms as environmental indicators…………………………………………………...…………4 Paleolimnology in the Canadian High Arctic………………………………………………..…….5 Impacts of seabird-derived nutrients on Arctic lakes……………………………..……………….7 Summary of thesis objectives…...……...……………………………………………..……………8 References…………………………………………………...……...………...………………..…10 List of Figures…………………………………………………………………………………….17 Figure…………………………………………………………………...…………………...……18 Chapter 2: Physical and chemical limnological characteristics of lakes and ponds across environmental gradients on Melville Island, Nunavut/N.W.T., High Arctic Canada………19 Abstract……………………………………………………………………………………..…….20 Introduction……………………………………………………………………………………….21 Site description……………………………………………………………………………...…….23 Methods……………………………………………………………………………………..…….25 Results and Discussion……………………………………………………………………..…….28 Conclusions……………………………………………………………………………………….41 Acknowledgements……………………………………………………………………...…..…….44 References…………………………………………………………………………………..…….45 List of Figures…………………………………………..………………….………………….….49 Figures…………………………………………………………………….………………..…….50 Tables……………………………………………………………………………………….…….55 Chapter 3: Evaluating the role of environmental and spatial variables on diatom species distributions on Melville Island (Canadian High Arctic)……………………………….…….65 Abstract……………………………………………………………………………………..…….66 Introduction……………………………………………………………………………………….67 Methods……………………………………………………………………………..............…….69 Results and Discussion……………………………………………………………………..…….73 Summary….……………………………………………………………………………………….81 Acknowledgements……………………………………………………………………...…..…….83 References…………………………………………………………………………………..…….83 List of Figures…………………………………………..………………….…………………..…87
viii
Figures…………………………………………………………………….………………..…….88 Tables…………………………………………………………………………………….…….....94 Chapter 4: Limnological characteristics of a high Arctic oasis and comparisons across northern Ellesmere Island……………………………………………………………………..101 Abstract…………………………………………………………………………………….……102 Introduction……………………………………………………………………………..……….103 Methods………………………………………………………………………………..…..…….105 Results and Discussion……………………………………………………………………..…...109 Summary and Conclusions…………………………………………………………………...….117 Acknowledgements……………………………………………………………………...…….…119 References…………………………………………………………………...……………..……120 List of Figures…………………………………………..………………….……………………124 Figures…………………………………………………………………….…………….....……125 Tables……………………………………………………………………………………………129 Chapter 5: Early-20th century environmental changes inferred using sub-fossil diatoms from a small pond on Melville Island, N.W.T., Canadian high Arctic……………….………...…138 Abstract…………………………………………………………………………………….……139 Introduction……………………………………………………………………………..……….140 Study site……………………………………………………………………………..…..……...142 Materials and Methods………………………………………………………………..…..…….143 Results……………………………………………………………………………..…..…….…..144 Discussion……………………………………………………………………..…………….…..146 Conclusions………………………………………………………………….......................…....151 Acknowledgements……………………………………………………………………...…….…152 References…………………………………………………………………...……………..……152 List of Figures…………………………………………..………………….……………………158 Figures…………………………………………………………………….…………….....……159 Tables……………………………………………………………………………………………164 Chapter 6: Prolonged ice cover dampens diatom community responses to recent climatic change in high Arctic lakes………………………………………………………………...….165 Abstract…………………………………………………………………………………….……166 Introduction……………………………………………………………………………..……….167 Site description…...…………………………………………………………………..…..……...169 Methods………………………………………………………………………………..…..…….170 Results……………………………………………………………………………..…..…….…..172 Discussion……………………………………………………………………..…………….…..173 Conclusions………………………………………………………………….......................…....178 Acknowledgements……………………………………………………………………...…….…179 References…………………………………………………………………...……………..……180 List of Figures…………………………………………..………………….……………………185 Figures…………………………………………………………………….…………….....……186 Tables……………………………………………………………………………………………193 Chapter 7: Impacts of seabird-derived nutrients on water quality and diatom species assemblages from Cape Vera, Devon Island, Canadian High Arctic…………………….…196 Abstract…………………………………………………………………………………….……197 Introduction……………………………………………………………………………..……….198 Site description……...………………………………………………………………..…..……...200
ix
Methods………………………………………………………………………………..…..…….201 Results and Discussion…..………………………………………………………..…..…….…..204 Conclusions………………………………………………………………….......................…....210 Acknowledgements……………………………………………………………………...…….…211 References…………………………………………………………………...……………..……212 List of Figures…………………………………………..………………….……………………218 Figures…………………………………………………………………….…………….....……219 Tables……………………………………………………………………………………………226 Chapter 8: Tracking seabird population dynamics using paleolimnology: A case study from Devon Island, Arctic Canada………………………………………………………………….231 Abstract…………………………………………………………………………………….……232 Introduction……………………………………………………………………………..……….233 Site description…………………………………………………………………...…..…..……...235 Methods………………………………………………………………..…..…………………….236 Results……………………………………………………………………………..…..…….…..238 Discussion……………………………………………………………………..…………….…..245 Acknowledgements……………………………………………………………………...…….…250 References…………………………………………………………………...……………..……250 List of Figures…………………………………………..………………….……………………254 Figures…………………………………………………………………….…………….....……255 Tables……………………………………………………………………………………………276 Chapter 9: General Discussion and Conclusions…………………………………………….277 Modern limnology………………………………………………….…………………..….…….277 Diatom ecology………………………………………………………………………....……….279 Paleolimnology…………………………….…………………………………………....……….280 Future directions…………….……………………….………………………………....……….281 References…………….…………………………………………....……………………………283 Appendices…………………………………………………...…………………………………285 Appendix 1………………………………………………………………………………………286 Detailed description of methods used to collect water samples during field work. Appendix 2………………….………………………………………..………………..…….......287 Spatial variable results from Melville Island diatom calibration set. Appendix 3. ………….…………………….………………………………………..……..…....288 Species response scatterplots relative to specific conductivity for diatoms from Melville Island surface sediments. Appendix 4. ………….…………………….………………………………………..……..…....290 Summary statistics and estimated specific conductivity optima of various weighted averaging models for diatom-inferred specific conductivity from surface sediments of Melville Island. Appendix 5. ………………….……………………………………………………...……..…....292 Raw diatom counts from Melville Island surface sediments.
x
Appendix 6. ………….…………………….………………………………………..……..…....298 Trace metal water chemistry data from northern Ellesmere Island. Appendix 7. ………….…………………….………………………………………..……..…....299 Trace metal water chemistry data from the oasis region northern of Lake Hazen, Ellesmere Island. Appendix 8. ………….…………………….………………………………………..……..…....300 Raw diatom counts from core MVAT, Melville Island. Appendix 9. ……….…………………….………………………………………..……..……....302 Raw diatom counts from Skeleton Lake (EP1), Ellesmere Island. Appendix 10………….…………………….………………………………………..……..…....303 Raw diatom counts from EP2, Ellesmere Island. Appendix 11. ……….…………………….………………………………………..……..…......304 210Pb summary data for Skeleton Lake, Ellesmere Island. Appendix 12………….…………………….………………………………………..……..…....305 Total Pb and Hg from Skeleton Lake, Ellesmere Island. Appendix 13………….…………………….………………………………………..……..…....306 Raw diatom counts from Cape Vera, Devon Island surface sediments. Appendix 14. ……….…………………….………………………………………..……..……..308 Raw diatom counts from CV5 core, Cape Vera, Devon Island. Appendix 15………….…………………….………………………………………..……..…....309 Raw diatom counts from CV6 core, Cape Vera, Devon Island. Appendix 16. ………….………………….………………………………………..……..…......310 Raw diatom counts from CV7 core, Cape Vera, Devon Island. Appendix 17. ……….…………………….………………………………………..……..…......311 Raw diatom counts from CV9 core, Cape Vera, Devon Island. Appendix 18………….…………………….………………………………………..……..…....312 Raw diatom counts from CV9a core, Cape Vera, Devon Island. Appendix 19. ……….…………………….………………………………………..…..…..…....313 Raw diatom counts from CV12 core, Cape Vera, Devon Island. Appendix 20………….…………………….………………………………………..……..…....314 Raw diatom counts from CV13 core, Cape Vera, Devon Island. Appendix 21. ………….………………….………………………………………..……..……..315 Raw diatom counts from CV20 core, Cape Vera, Devon Island. Appendix 22. ……….…………………….………………………………………..…………....316 Raw diatom counts from CV22 core, Cape Vera, Devon Island.
xi
Appendix 23………….…………………….………………………………………..……..…....317 Raw diatom counts from CV24 core, Cape Vera, Devon Island.
xii
LIST OF TABLES Chapter 2 Table 1. …………………………………………………………………….……………………55 The chemical and physical parameters for 46 freshwater sites on Melville Island. Table 2. …………………………………………………………………….……………………61 Selected nutrient ratio parameters for 46 freshwater sites on Melville Island. Table 3. …………………………………………………………………….……………………63 Pearson correlation matrix with Bonferroni-adjusted probabilities. Chapter 3 Table 1. …………………………………………………………………….………………..…..94 Summary of selected limnological characteristics for sites from Melville Island. Table 2. …………………………………………………………………….………………….…95 List of diatom species found in Melville Island surface sediment samples in >1% relative abundance from at least three sites, or >10% relative abundance in at least one site. Table 3. ………..…………………………………………………….………………………..…98 Analysis of similarity (ANOSIM) results for surface sedimentary diatom assemblages from Melville Island. Table 4. ………………………………………………………….…………………………..…..99 List of species contributing the most to the dissimilarity (Diss/SD; average dissimilarity/standard deviation) between Zones 4 and 1 on Melville Island, based on an analysis of similarity percentages (SIMPER). Table 5. ………………………………………………………….…………………………..….100 Summary statistics of various weighted averaging models for pH. Chapter 4 Table 1. ………………………………………………………….…………………………..…129 Summary of selected limnological variables for the northern Ellesmere sites and Lake Hazen. Table 2. ………………………………………………………….…………………………..…133 Summary of selected limnological variables for the oasis sites, with abbreviations as describe in Table 1. Table 3. ………………………………………………………….…………………………...…136 Pearson correlation matrix with Bonferroni-adjusted probabilities. Chapter 5 Table 1. ………………………………………………………….…………………………..…164 Present-day physical and chemical characteristics of pond MV-AT were collected on July 24, 2002. Chapter 6 Table 1. ………………………………………………………….…………………………...…193 Selected limnological characteristics for the three study lakes.
xiii
Table 2. ………………………………………………………….…………………………...…194 Diatom taxa (>3% relative abundance) from the top 2.5 cm intervals from the sediment cores of Skeleton Lake and EP2 and the surface sediment (~2 cm) diatoms from EP3. Table 3. ………………………………………………………….…………………………..….195 Common diatom taxa (>3% relative abundance) found in the sediment cores from Skeleton Lake and EP2. Chapter 7 Table 1. ………………………………………………………….…………………………..….226 Selected limnological data for ponds located near Cape Vera, Devon Island. Table 2. ………………………………………………………….…………………………...…228 Pearson’s correlation matrix with Bonferroni-adjusted probabilities. Significant values are denoted in bold (p<0.01) or in underlined italics (p<0.05). Table 3. ………………………………………………………….…………………………...…229 Summary of dominant diatom species (at least 1% relative abundance in at least 3 sites) found in surface sediments of Cape Vera ponds. Table 4. ………………………………………………………….…………………………..…230 List of dominant diatom species arranged in order of their axis 1 species scores based on a redundancy analysis (RDA) constrained solely to δ15N. Chapter 8 Table 1. ………………………………………………………….…………………………..…276 Summary of limnological data from each pond cored at Cape Vera, Devon Island.
xiv
LIST OF FIGURES Chapter 1 Figure 1. ………………………………………………………….………………………….…18 Regional map of the Canadian Arctic Archipelago with the locations of limnological, diatom ecological, and paleolimnological studies conducted by the Smol and Douglas laboratories prior to this thesis. Chapter 2 Figure 1. ………………………………………………………….…………………………...…50 Regional map of the Canadian Arctic Archipelago with the locations of both Melville Island and previous modern limnological studies to which references are made in the text. The inset map A) indicates the location of the Canadian High Arctic within Canada, and the circle on the main map B) indicates Melville Island, the focus of our study. Figure 2. ………………………………………………………….…………………………....…51 Map of Melville Island, with sites differentiated according to previously defined a) bioclimatic zones (after EDLUND 1994), and b) bedrock geology (after HARRISON 1994). On both maps, numbers indicate the following geographical features: 1) Liddon Gulf, 2) Murray Inlet, 3) Purchase Bay, 4) Leopold Glacier, 5) unnamed ice caps, 6) Bridport Inlet, 7) Sabine Bay, 8) Hecla and Griper Bay. 2a) Bioclimatic zone 1 is the most sparsely vegetated region, while zone 4 has the greatest number and abundance of terrestrial vegetation, including woody shrubs (EDLUND 1994). Figure 3. ………………………………………………………….…………………………....…52 Box plots showing variability of selected environmental variables from islands of the Canadian Arctic Archipelago. Solid lines indicate median values, dashed lines indicate mean values, whiskers represent 10th and 90th percentiles, and dots are 5th and 95th percentiles. Data sources are as follows: Prince Patrick Island (ANTONIADES et al. 2003a), Banks Island (LIM et al. 2005), Victoria Island (MICHELUTTI et al. 2002a), Bathurst Island (LIM et al. 2001), Devon Island (LIM & DOUGLAS 2003), Ellef Ringnes Island (ANTONIADES et al. 2003b), Axel Heiberg Island (MICHELUTTI et al. 2002b), Alert (ANTONIADES et al. 2003a). Figure 4. ………………………………………………………….…………………………....…53 Principal components analysis (PCA) biplot of measured environmental variables (arrows) and sampling sites (symbols). The light lines represent variables that were run passively in the PCA. Sampling sites are differentiated into: 4a) dominant type of bedrock geology (after HARRISON 1994), or 4b) bioclimatic zones (after EDLUND 1994). See caption of Figure 2a and 2b for symbol legends. Figure 5. ………………………………………………………….……………………….......…54 Box plots of selected variables showing sites from bioclimatic zone 4, the mostly lushly vegetated zone on Melville Island (n=16) versus all other bioclimatic zones combined (n=30). Solid lines indicate median values, dashed lines indicate mean values, whiskers represent 10th and 90th percentiles, and dots are 5th and 95th percentiles. Chapter 3 Figure 1. ……………………………………………………….…………………………....…88 Map showing the location of (a) Melville Island in relation to Canada, (b) existing diatom calibration sets in the Canadian Arctic, and (c) the 45 lakes and ponds in this study. The numbers given in (b) correspond to the following studies: 1 = Mould Bay, Prince Patrick Island
xv
(Antoniades et al. 2005); 2 = Isachsen, Ellef Ringnes Island (Antoniades et al. 2003b); 3 = Axel Heiberg Island (Michelutti et al. 2006); 4 = Alert, Ellesmere Island (Antoniades et al. 2005); 5 = Cape Herschel, Ellesmere Island (Douglas and Smol 1993, 1995), 6 = Devon Island (Lim 2004); 7 = Cornwallis Island (Michelutti et al. in press); 8 = Bathurst Island (Lim et al. 2001a, b); 9 = Banks Island (Lim et al. 2007). Figure 2. ……………………………………………………….………………………….…......89 Histograms of the common species found in the surface sediments of 45 Melville Island lakes and ponds. Only species present in at least 1% relative abundance in at least 10 sites are shown, arranged in order of DCA axis 1 species scores. The 45 study sites are also arranged in order of DCA axis 1 sample scores. Measured pH and specific conductivity values are plotted to the far right of the figure. Figure 3. ……………………………………………………….…………………………...….....90 Canonical Correspondence Analysis (CCA) of a) sites and environmental variables, and b) diatom species and environmental variables from Melville Island. See Table 2 for species code numbers. Figure 4. ……………………………………………………….………………………….……...91 Relationship between observed and estimated pH values based on the WAtol inv model (n = 90) for a) bootstrapped pH values (r2
boot = 0.432), and b) bootstrapped pH residuals. Figure 5. ……………………………………………………….………………………….……...92 Species response curves of common species (found in at least 1% relative abundance in at least 10 sites) in relation to pH. Chapter 4 Figure 1. ……………………………………………………….………………………….…....125 Location map of northern Ellesmere Island. Inset a) indicates Ellesmere Island within Canada. Inset b) shows the northern sites around Ellesmere Island. The dashed black line denotes the boundary of Quttinirpaaq National Park and the patterned areas within this boundary represent different climate regions based on Thompson (1994). The black star indicates the location of the oasis sites detailed in inset c). Inset c) details the oasis sites just north of Lake Hazen. Figure 2. ……………………………………………………….………………………….…....126 Plots of total phosphorus unfiltered (TPu) versus total nitrogen (TN) in a) the oasis sites, and b) the northern sites. While there is little relationship between TPu and TN in the northern region, there is a clear positive relationship with TPu and TN in the oasis sites, suggesting that different factors control nutrient cycling within the two regions. Figure 3. ……………………………………………………….………………………….…....127 Biplot of a principal components analysis (PCA) of measured limnological variables for all sites. Oasis sites are represented by filled circles and northern sites are represented by open circles. Lake Hazen is kept separate due to its extremely large size and is represented by a star. Axis 1 most closely represents nutrients and related variables, pH, and conductivity, and explains 52.9% of the variance in the dataset. Axis 2 most closely represents a gradient of metals and explains 16% of the variance in the dataset. The dashed lines represent variables that were run passively in the ordination.
xvi
Figure 4. ……………………………………………………….………………………….…....128 Histograms indicating the change in the values of selected limnological variables in 2003 relative to 1963 for a) pH, b) specific conductivity, c) K, and d) SiO2. Site names with * indicate sites for which identification was approximate. Chapter 5 Figure 1. ……………………………………………………….………………………….…....159 Map showing location of sites discussed in this paper. The oval on the inset map shows location of Canadian High Arctic. Sites are as follows: A) MV-AT, Melville Island; B) Isachsen, Ellef Ringnes Island; C) Alert, Ellesmere Island; D) Tuborg Lake, Ellesmere Island; E) Fosheim Peninsula, Ellesmere Island; F) Devon Ice Cap, Devon Island; G) Cape Hershel, Ellesmere Island; H) Agassiz Ice Cap, Ellesmere Island; I) Melville Island ice caps, Melville Island; J) Meighen Ice Cap, Meighen Island; K) Char Lake, Resolute Bay, Cornwallis Island; L) Mould Bay, Prince Patrick Island; M) Bathurst Island; N) Victoria Island. Figure 2. ……………………………………………………….………………………….…....160 Sedimentation rate for dated sections of the MV-AT core, as calculated based on the Constant Rate of Supply (CRS) method described by Appleby (2001) and Binford (1990). Figure 3. ……………………………………………………….………………………….…....161 Diatom profile showing the common diatom species found in pond MV-AT. Individual species with >5% relative abundance in at least one sample were retained for the profile; “other” is a sum of all other diatoms found in each interval. Dates are based on 210Pb dating using a Constant Rate of Supply model. Percent loss-on-ignition (%LOI 550) is expressed as a % of combustion at 550˚C, and is a proxy for organic matter content of the sediment. Percent carbonates (%LOI 1000) is expressed as a percentage of dry weight combusted at 1000˚C. Zones are based on optimal splitting and broken-stick analysis. While the most marked species change occurs at ~5.25 cm, the shift in diatom assemblage appears to have begun earlier (~5.75 cm). Figure 4. ……………………………………………………….………………………….…....162 Detrended Correspondance Analysis (DCA) of diatom species scores versus depth. The DCA axis 1 species scores are scaled in Standard Deviation (S.D.) units, and provide an estimate of species turnover. Figure 5. ……………………………………………………….………………………….…....163 Mean June-July-August temperature data and annual precipitation data from Mould Bay, Prince Patrick Island (see Figure 1 for location) between 1948-1996 (Meteorological Service of Canada 2004). The smoothed line is a LOWESS curve with a span of 0.35. Chapter 6 Figure 1. ……………………………………………………….………………………….…....186 Map showing the locations of the study site (star) and the other sites (numbered) mentioned in the text: 1. Alert, 2. Hazen Plateau, 3. Agassiz Ice Cap, 4. Fosheim Peninsula, 5. Cape Herschel, 6. Isachsen, Ellef Ringnes Island, 7. Char Lake, Cornwallis Island. Figure 2. ……………………………………………………….………………………….…....187 Topographical map detailing the local topography near Skeleton Lake, EP2 and EP3. Figure 3. ……………………………………………………….………………………….….....188 Ice-off dates for Skeleton Lake (solid bars) and Lake EP2 (hatched bars). The data from the 1960s is from Oliver and Corbett (1966), while the 2003 data is from our own field observations.
xvii
These data are also corroborated by archival air photos that show ice persisting on Skeleton Lake after Lake EP2 has become ice-free. Figure 4. ……………………………………………………….………………………….…....189 Principal components analysis (PCA) biplot based on measured water chemistry variables that are considered to influence diatom assemblages. The arrows represent the measured environmental variables, while the open circles represent 52 lakes and ponds across northern Ellesmere Island. Skeleton Lake is represented by the solid square, EP2 is represented by the open square, and EP3 is represented by the solid circle. The proximity of these three sites represents their highly similar water chemistry. Figure 5. ……………………………………………………….………………………….…....190 5a) Age-depth model for Skeleton Lake based on CRS model (Binford, 1990), and b) Total 210Pb activities as estimated by alpha spectroscopy. The dashed line indicates estimated supported 210Pb. Figure 6. ……………………………………………………….………………………….…....191 6a) Diatom profile of Skeleton Lake showing taxa present in at least at least one interval with a relative abundance of >3%. See Table 2 for synonyms for some of the common taxa. Percent loss-on-ignition (%LOI; an estimate of organic matter) and PCA axis 2 sample scores (PCA2; a summary of change occurring in rare taxa) are presented at the right side of the profile. Figure 6b). Photographs of ice cover on Skeleton Lake and Lake EP2, indicating the physical proximity of the two lakes, Blister Hill, and a nearby pingo for reference (photographs taken 7 July 2003). Figure 6c). Diatom profile of EP2 showing only species that are present in at least 3% relative abundance in at least one interval. See Table 2 for synonyms for some of the common taxa. Percent loss-on-ignition (% LOI), total Pb and total Hg (both expressed per gram organic carbon), and PCA axis two sample scores (PCA2) are presented at the right side of the profile. The rise in total Pb and Hg are interpreted to mark the onset of anthropogenic pollution (mid-19th to early 20th century). Figure 7. ……………………………………………………….………………………….…....192 Schematic diagrams illustrating possible diatom responses to changing ice cover conditions in the two study lakes. Although the length of the ice-free season has likely increased in both lakes, it has yet to reach a critical threshold in Skeleton Lake, resulting in a muted diatom response. In Lake EP2, reduction in ice cover has crossed this critical threshold and resulted in marked diatom changes. Chapter 7 Figure 1. ……………………………………………………….………………………….…....219 Map of study location at Cape Vera, Devon Island. Insets showing location of: a) Devon Island within Canada; b) Cape Vera on Devon Island; and c) sites located throughout Cape Vera. The isoclines are not drawn to scale. The gradient indicates the approximate concentration of northern fulmars along the cliffs at Cape Vera, with the most birds occurring within the southern third of the colony. Figure 2. ……………………………………………………….………………………….…....220 Principal Components Analysis (PCA) biplot of measured environmental variables and Cape Vera sites. The “control sites” (CV11, 16, 17, 18, 22, 23, 24) are denoted with open circles while the rest of the ponds are denoted with solid circles.
xviii
Figure 3. ……………………………………………………….………………………….…....221 Box plots comparing selected measured environmental variables (TPu, TPf, TN, DOC, pH, chla) between Cape Vera ponds (n = 24, this study) and ponds located in the nearby Haughton Crater, Devon Island (n = 22, Lim & Douglas 2003). Figure 4. ……………………………………………………….………………………….…....222 Composite PCA biplot of selected measured environmental variables common to our labs’ previous limnological surveys in the Canadian Arctic Archipelago. References for each study are as follows: Melville Island (Keatley et al. [2]), Mould Bay, Prince Patrick Island (Antoniades et al. 2003a), Banks Island (Lim et al. 2005), Victoria Island (Michelutti et al. 2002a), Isachsen, Ellef Ringnes Island (Antoniades et al. 2003b), Bathurst Island (Lim et al. 2001), Cornwallis Island (Michelutti et al. in press a), Devon Island (Lim & Douglas 2003), Axel Heiberg Island (Michelutti et al. 2002b), northern Ellesmere Island (Keatley et al. [4]), Alert, Ellesmere Island (Antoniades et al. 2003). Figure 5. ……………………………………………………….………………………….…....223 Histogram of dominant diatom species (>1% relative abundance in at least 5 sites) found in ponds from Cape Vera. Both the sites and the species are ordered according to their DCA axis 1 scores. Figure 6. ……………………………………………………….………………………….…....224 Principal components analysis (PCA) biplot of species and sites from the surface sediments of Cape Vera, Devon Island. Figure 7. ……………………………………………………….………………………….…....225 Redundancy analysis (RDA) biplot constrained to the three measured environmental variables that explained significant portions of the diatom species variance (δ15N, specific conductivity, and dissolved organic carbon (DOC)). Biplot 7a) presents the diatom species, and 7b) the sites in relation to the environmental variables. Chapter 8 Figure 1. ……………………………………………………….………………………….…....255 Location of study site. A) Regional map of the Canadian Arctic with inset detailing location of the Canadian Arctic within Canada; B) Enlargement of Devon Island region with star identifying the location of Cape Vera; and C) Location of study ponds at Cape Vera, Devon Island. Figure 2. ……………………………………………………….………………………….…....256 Photograph of a suite of ponds directly below the cliffs at Cape Vera, July 2004. Figures 3 – 11. Plots for each core summarizing overall changes in diatom assemblage composition based on Principal Components Analysis Axis 1 (PCA1) and PCA axis 2 (PCA2) sample scores, changes in diatom assemblage diversity based on Hill’s N2 (N2; Hill 1973), seabird-derived nutrients based on (δ15N (‰)), sedimentary-inferred chlorophyll a concentrations (Chla mg/g dry weight), %N, %C, and C/N ratios. Figure 3. CV5 summary plot. ………….………………………………………..….…....257 Figure 4. CV6 summary plot. ………….………………………………………..….…....258 Figure 5. CV9 summary plot. ………….………………………………………..….…....259 Figure 6. CV9a summary plot. ……….………………………………………..….…......260 Figure 7. CV12 summary plot. ………….……………………………………..…...…....261 Figure 8. CV13 summary plot. ………….…………………………………..….…..........262 Figure 9. CV20 summary plot. ………….…………………………………..….……......263
xix
Figure 10. CV22 summary plot. ………….…………………………………..….…........264 Figure 11. CV24 summary plot. ………….…………………………………..….……....265 Figures 12 – 21. Diatom stratigraphic plots for each core from the Cape Vera ponds. These plots include all species present in at least 3% relative abundance in at least one sample. Figure 12. CV5 diatom stratigraphy. ………….………………………………….….......266 Figure 13. CV6 diatom stratigraphy. ………….…………………………………….…...267 Figure 14. CV7 diatom stratigraphy. ………….…………………………………..……..268 Figure 15. CV9 diatom stratigraphy. ………….…………………………………..…......269 Figure 16. CV9a diatom stratigraphy. ………….………………………………….….....270 Figure 17. CV12 diatom stratigraphy. ………….…………………………………..…....271 Figure 18. CV13 diatom stratigraphy. ………….…………………………………..…....272 Figure 19. CV20 diatom stratigraphy. ………….…………………………………..…....273 Figure 20. CV22 diatom stratigraphy. ………….…………………………………..…....274 Figure 21. CV24 diatom stratigraphy. ………….…………………………………..…....275
CHAPTER 1
GENERAL INTRODUCTION AND LITERATURE REVIEW
ENVIRONMENTAL CHANGE IN THE ARCTIC
The Canadian High Arctic is a region that is especially sensitive to global environmental
changes (ACIA 2004). Due to a number of positive feedback loops (e.g. snow-ice albedo,
permafrost thawing), temperature increases resulting from warming are likely to be amplified in
high-latitude areas. Some of these changes are apparent even over the course of the relatively
short monitoring window. For example, Arctic regions have warmed at a rate that is more than
double the global average over the past century, the extent of Arctic sea ice has diminished by
2.7% per decade since 1978, and the uppermost layers of permafrost have warmed by as much as
3 °C since the 1980s (IPCC 2007). Climate models predict further changes in the future, as
average annual Arctic temperatures are projected to rise by 2 to 4 °C over the next century (ACIA
2005). In terrestrial ecosystems, changes in biogeochemical cycling brought about by warming
are likely to alter carbon source/sink dynamics in Arctic tundra and peatlands (e.g. Mack et al.
2004; Shaver et al. 2006), and these will likely have significant implications for future global
warming. The ramifications of Arctic warming have already resulted in the northward expansion
of several species of fish, birds, and mammals, and wreaked havoc with indigenous peoples’
traditional ways of interacting with their environment (ACIA 2005). As noted by Schindler and
Smol (2006), it will be “hard to overestimate the ecological, social, and economic impacts of such
large-scale ecosystem shifts.”
Unfortunately, the Canadian High Arctic is a difficult area to monitor due to logistical
constraints, and the short duration and poor spatial coverage of the very few instrumental
meteorological records. The Canadian High Arctic is currently serviced by only five weather
stations that have been in place since 1947 (Eureka, Ellesmere Island, and Resolute Bay,
Cornwallis Island), 1948 (Isachsen, Ellef Ringnes Island, and Mould Bay, Prince Patrick Island –
now both automated), and 1950 (Alert, Ellesmere Island). More recently established weather
stations or automatic data collecting devices are also present at Grise Fiord, Ellesmere Island and
Rea Point, Melville Island. Moreover, these meteorological records generally post-date the onset
of warming in many High Arctic regions. Comprehensive historical data for other types of
information critical to understanding the impacts of environmental change on ecosystem
functioning (e.g. physical, chemical, and biological data across regions and ecosystems) are
lacking. Thus, reconstructions that aid in deciphering the natural variability of past
environmental conditions, including information on the interactions of climate, biogeochemistry,
and biota, are key to understanding future environmental changes.
The overarching theme of my thesis is to address some of the critical knowledge gaps
surrounding High Arctic environmental change using limnological and paleolimnological
techniques. Each chapter has been prepared as a separate manuscript and thus each contains
introductory material, literature reviews, and study rationales pertinent to the specific study
objective. In order to keep repetition to a minimum, below I provide some broader perspectives
of High Arctic limnology, the use of diatoms as environmental indicators and paleolimnology in a
High Arctic context.
CANADIAN HIGH ARCTIC LIMNOLOGY
An abundance of lakes and ponds characterize the High Arctic landscape. These
freshwater bodies are often important locations of enhanced biodiversity, as they not only provide
habitat for aquatic organisms, but also perform critical ecosystem services for terrestrial biota.
Arctic lakes and ponds are known to be sensitive to environmental change (Douglas et al. 1994;
Schindler & Smol 2006), and several impacts associated with warming Arctic temperatures on
freshwater systems have already been noted. For example, melting of permafrost has been linked
both to the drainage of thermokarst lakes in subarctic Siberia (Smith et al. 2005) and to increases
in major ion transport from land to water (Kokelj et al. 2005). Elevated temperatures have
resulted in the complete dessication of some High Arctic ponds, previously in existence for
millenia (Smol & Douglas 2007). However, in order to determine the extent of environmental
change, a critical first step is to establish current base-line limnological conditions.
Some of the earliest limnological data for the Canadian High Arctic was based on data
collected during “Operation Hazen”, a Defence Research Board of Canada program based near
Lake Hazen, on northern Ellesmere Island (Oliver & Corbet 1966). However, these important
data fall into the “grey literature” category, and thus have not been widely used. In research
circles, most will recognize the data collected during the International Biological Program (IBP)
from the ultra-oligotrophic Char Lake and the sewage-enriched Meretta Lake, Cornwallis Island,
as some of the earliest widely available data from High Arctic lakes (e.g. Schindler et al. 1974;
Kalff & Welch 1974; Kalff 1975). These two relatively large lakes, however, are somewhat
atypical freshwater systems in the High Arctic. Since that time, other aspects of High Arctic
limnology have been explored, such as the relationships between ultra-violet radiation and
dissolved organic carbon (DOC) (Laurion et al. 1997), the role of lake ice in dissolved organic
matter distributions (Belzile et al. 2002); factors related to primary (Vezina & Vincent 1997;
Markager et al. 1999; Bonilla et al. 2005) and secondary (Rautio & Vincent 2006; Rautio &
Vincent 2007) production, and the geochemistry and physics of complex lake systems (Gibson et
al. 2002; Van Hove et al. 2006).
In a concerted effort to gather baseline environmental data from much of the Canadian
High Arctic, John Smol and Marianne Douglas have led a number of limnological surveys of
lakes and ponds (Fig. 1). Since 1983, they, along with their students, have provided physical and
chemical limnological data from eastern-central Ellesmere Island (Douglas & Smol 1994); Alert,
Ellesmere Island (Antoniades et al. 2003a); Axel Heiberg Island (Michelutti et al. 2002b);
Cornwallis Island (Michelutti et al. in press a); Bathurst Island (Lim et al. 2001a); Devon Island
(Lim & Douglas 2003; Lim 2004); Isachsen, Ellef Ringnes Island (Antoniades et al. 2003b);
Banks Island (Lim et al. 2005); Wynniatt Bay, Victoria Island (Michelutti et al. 2002a); and
Mould Bay, Prince Patrick Island (Antoniades et al. 2003a). Additional limnological surveys
have been conducted on Southampton Island (Mallory et al. 2006), and from 204 lakes and ponds
across the Canadian Arctic (Hamilton et al. 2001). These records indicate that High Arctic lakes
and ponds span large gradients of physical and chemical variables, and underscore the need for a
more complete survey of present-day limnological conditions in regions of the High Arctic that
have not yet been fully explored.
Two of the regions for which limnological data are lacking include the western High
Arctic (currently represented by only two relatively small surveys by Antoniades et al. 2003a and
Lim et al. 2005), and from northern Ellesmere Island (a region that encompasses a warm, High
Arctic oasis). Chapters 2 and 4 of this thesis focus on addressing some of the current gaps in
limnological research by evaluating the variability of ponds and lakes across wide environmental
gradients in the western High Arctic (Chapter 2), and from a High Arctic oasis on northern
Ellesmere Island (Chapter 4).
DIATOMS AS ENVIRONMENTAL INDICATORS
Diatoms, siliceous unicellular algae (class Bacillariophyceae), are particularly useful
environmental indicators because they are ubiquitous, they respond rapidly to changing
conditions, and different species often have distinct optima to given environmental variables
(Douglas & Smol 1999). In complementary studies to those described above for basic
limnological data, the Smol and Douglas labs have provided diatom autecological information
from regions throughout the High Arctic (Fig. 1; Douglas & Smol 1993; Lim et al. 2001b; Lim et
al. 2001c; Lim 2004; Antoniades et al. 2004; Antoniades et al. 2005a; Michelutti et al. 2006; Lim
et al. 2007; Michelutti et al. in press a). Additionally, a diatom biogeographical study at the scale
of the Canadian Arctic Archipelago has provided distributional data for Arctic diatoms (Bouchard
et al. 2004). An historical review of diatom research in High Arctic regions, as well as the first
diatom flora produced for the Canadian High Arctic, has recently been provided by Antoniades et
al. (in press).
Several High Arctic diatom-based calibration sets have generated statistically robust
quantitative relationships between diatom species assemblages and the measured environmental
variables, including pH, specific conductivity, dissolved organic carbon (DOC), and/or nutrients,
that can best explain their distributions. The diatom ecological information gleaned from such
calibration studies is important for interpreting the environmental significance of changes in
diatom communities through time. However, just as limnological data are lacking from the
western High Arctic region, diatom distribtutional data was also limited from the western High
Arctic. Thus, in Chapter 3 I explore the role of large environmental and spatial gradients on
diatom distributions from the western High Arctic.
PALEOLIMNOLOGY IN THE CANADIAN HIGH ARCTIC
In the absence of direct observational or instrumental records, paleoenvironmental
techniques can provide important insights into past environments. Several commonly used
paleoenvironmental approaches, such as dendrochronology, are difficult to apply in High Arctic
regions where woody species are rare (but see Rayback & Henry 2006 for one example). Other
powerful paleoenvironmental techniques, such as ice core analyses, are necessarily limited to the
relatively few locations where ice caps that are suitable for analysis exist (e.g. Ellesmere, Devon,
and Meighen islands).
The abundance of lakes and ponds throughout the Canadian High Arctic represent a
wealth of repositories of past environmental conditions (Wolfe & Smith 2004).
Paleolimnological analyses make use of physical, chemical, and/or biological indicators in lake
sediments to reconstruct historical conditions (Smol 2002). In the Canadian Arctic, for example,
variations in the thickness of sedimentary varve records have been used to infer past climatic
conditions (e.g. Lamoureux & Gilbert 2004; Hambley & Lamoureux 2006), whereas spectrally-
inferred chlorophyll a techniques (Michelutti et al. 2005) have been used to track recent changes
in primary production. A variety of biological indicators have also been successfully employed
in the Canadian Arctic to track environmental changes including fossils of algae (e.g. diatoms and
chrysophytes, Douglas et al. 2004a), and invertebrates (Bennike et al. 2004; Quinlan et al. 2005).
Existing observational data (Serreze et al. 2000; ACIA 2004) and proxy records (e.g.
Paterson et al. 1977; Fisher 1979; O’Brien et al. 1995; Zdanowicz et al. 2000; Grumet et al. 2001;
Kaufman et al. 2004; Smol et al. 2005) indicate that the timing and nature of environmental
changes have not been synchronous across the Arctic. To date, diatom-based paleolimnological
investigations of climatic and environmental change from the High Arctic are largely restricted to
the eastern Canadian Arctic (Douglas et al. 1994; Michelutti et al. 2003; Perren et al. 2003; Wolfe
& Smith 2004; Antoniades et al. 2005b) and to one study from the western High Arctic (Banks
Island, Lim et al. 2007). This regional bias precludes our ability to assess the spatial and
temporal variability of climatic change across the High Arctic. To address this gap and in order to
better compare magnitudes and rates of diatom-inferred environmental change from the western
High Arctic with those from the eastern High Arctic, in Chapter 5 I assess diatom-inferred
environmental change from a small pond on Melville Island.
Furthermore, while diatom-based studies have revealed marked shifts in diatom species
assemblages over the last 150 years, the magnitude and timing of such changes have been
different. Typically, large High Arctic lakes tend to be less sensitive to environmental changes,
and therefore often record a much later onset of diatom assemblage changes and more muted
responses, as compared to High Arctic ponds (e.g. Douglas et al. 1994; Michelutti et al. 2003;
Antoniades et al. 2005b). The persistence of ice cover (Smol 1983; Smol 1988) has been invoked
for more than two decades to explain this phenomenon, yet more conclusive evidence is lacking.
In Chapter 6 I examine the influence of ice cover on diatom community changes through time by
comparing sediment records from two adjacent lakes with differing ice cover regimes.
IMPACTS OF SEABIRD-DERIVED NUTRIENTS ON ARCTIC LAKES
The Canadian Arctic is an important habitat for ~10 million marine birds that may be
under threat from changing environmental conditions (Mallory & Fontaine 2004; Gaston et al.
2005). However, the natural variability of seabird populations is unknown, as even the best
seabird monitoring records extend for less than 30 years. Indicators that could reconstruct seabird
populations, therefore, would represent a highly desirable tool for wildlife managers.
Seabird-derived nutrients (e.g. guano, feathers, eggshells) can provide significant
subsidies to otherwise oligotrophic environments (Polis et al. 2004). While much evidence exists
for seabird-related influences on soils, plants, and secondary consumers in arid, sub-tropical
environments (e.g. Polis & Hurd 1996; Polis et al. 1997; Anderson & Polis 1999; Sánchez-Piñero
& Polis 2000; Polis et al. 2004) and Antarctic regions (e.g. Vincent & Vincent 1982; Lindebloom
1984; Ryan & Watkins 1989; Erskine et al. 1998), little data exist for corresponding Arctic
terrestrial environments (Wainwright et al. 1998) or for freshwater ecosystems (Evenset et al.
2004; Blais et al. 2005). In the Canadian High Arctic, only one study (affiliated with this thesis)
has examined the role of seabird-derived nutrients on freshwater quality (Blais et al. 2005).
The water columns of High Arctic lakes are generally nutrient-poor (Schindler & Smol
2006). Although evidence from Arctic Alaska suggests that phosphorus enrichment results in
phytoplankton responses that are analagous to those of temperate regions (i.e. cyanobacterial
blooms, Hobbie et al. 1999), High Arctic lakes do not appear to respond similarly (Schindler et al.
1974). In High Arctic lakes, these responses may be a result of the synergistic effects of extreme
environmental constraints (Markager et al. 1999) and different factors limiting benthic and
phytoplanktonic production (Bonilla et al. 2005). Nevertheless, paleolimnological studies of the
sewage-enriched Meretta Lake, Cornwallis Island, have previously shown that diatom assemblage
changes can track both the onset (Douglas & Smol 2000) and cessation (Michelutti et al. 2002c)
of nutrient enrichment, although the diatom assemblage responses were markedly different and
muted compared to similar studies in temperate regions (Hall & Smol 1999). Similarly, sewage
input only resulted in diatom species shifts after a significant time lag in Annak Lake, Belcher
Islands, Arctic Canada (Michelutti et al. in press b). In a small Arctic pond, changes in diatom
assemblages, in conjunction with changes in stable nitrogen isotopes were linked to nutrient
enrichment from decaying whale carcasses (Douglas et al. 2004b). Guano input from seabirds is
a far more common source of nutrients to Arctic ponds than either of the two point sources noted
above, and yet the influence of birds on diatom ecology is unknown.
Chapters 7 and 8 examine the effects of seabird-derived nutrient enrichment on the water
quality and diatom species assemblages of several small High Arctic ponds (Chapter 7), and
evaluate the use of diatoms, in conjunction with other paleolimnological proxies (i.e. stable
isotopes of nitrogen δ15N (Robinson 2001), and sedimentary-inferred chlorophyll a (Das et al.
2005; Wolfe et al. 2006)), to track these effects through time as a means to assess changes in the
dynamics of a seabird colony at Cape Vera, Devon Island (Chapter 8).
SUMMARY OF THESIS OBJECTIVES
Recent studies have provided important advances in limnological, diatom autecological,
and paleolimnological research that, together, can provide robust reconstructions of
environmental change in the Canadian High Arctic. However, critical knowledge gaps remain,
including understanding the regional variability of environmental change in different ecosystems
across the Canadian High Arctic. This thesis provides new limnological and paleolimnological
data to address some of these research gaps. Furthermore, this thesis assesses the use of diatoms
as indicators of seabird activity, a novel application of diatom-based paleolimnology, in the
Canadian High Arctic.
Chapters 2 and 4 augment the existing High Arctic database of limnological data by
providing data from Melville Island (western High Arctic), and from northern Ellesmere Island
(including a warm High Arctic oasis). Chapter 3 builds upon Chapter 2 by assessing diatom
distributional patterns across large environmental gradients identified on Melville Island. These
data provide important new insights into limnological variability and diatom distributions across
distinctive ecosystem types in the Canadian High Arctic for which little data had previously
existed.
Chapters 5 and 6 are diatom-based paleolimnological investigations of environmental
change. The objective of Chapter 5 is to assess the nature, magnitude, rate, and direction of
diatom-inferred environmental change in a small pond from Melville Island, western High Arctic,
and to compare these with other areas of the High Arctic. The objective of Chapter 6 is to
examine how ice cover influences diatom assemblage changes through time by comparing two
very similar, small High Arctic lakes that experience differing ice cover regimes.
Chapters 7 and 8 involve assessing the influence of seabird-derived nutrients on small
High Arctic ponds. The objective of Chapter 7 is to determine the impacts of seabird-derived
nutrients on freshwater quality and diatom distributions across a gradient of seabird influence.
The objective of Chapter 8 extends the results of Chapter 7 to an historical context, to determine
whether diatom assemblage shifts are able to track the historical trajectory of a seabird colony at
Cape Vera, Devon Island.
REFERENCES ACIA, 2004. Impacts of a warming Arctic. Cambridge University Press, Cambridge, UK. ACIA, 2005. Arctic climate impact assessment. Cambridge University Press, New York. Anderson, W. B., & G. A. Polis, 1999. Nutrient fluxes from water to land: seabirds affect plant nutrient status on Gulf of California islands. Oecologia 118: 324-332. Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2003a. Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick Island, NWT). Archiv für Hydrobiologie 158: 485-516. Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2003b. The physical and chemical limnology of 24 ponds and one lake from Isachsen, Ellef Ringnes Island, Canadian High Arctic. International Review of Hydrobiology 88: 519-538. Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2004. Diatom species-environment relationships and inference models from Isachsen, Ellef Ringnes Island, Canadian High Arctic. Hydrobiologia 529: 1-18. Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2005a. Benthic diatom autecology and inference model development from the Canadian High Arctic Archipelago. Journal of Phycology 41: 30-45. Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2005b. Quantitative estimates of recent environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. Journal of Paleolimnology 33: 349-360. Antoniades, D., P. B. Hamilton, M. S. V. Douglas & J. P. Smol, in press. Freshwater diatoms from the Canadian High Arctic. Iconographica Diatomologica. Belzile, C., J. A. E. Gibson, & W. F. Vincent, 2002. Colored dissolved organic matter and dissolved organic carbon exclusion from lake ice: Implications for irradiance transmission and carbon cycling. Limnology and Oceanography 47: 1283-1293. Bennike, O., K. P. Brodersen, E. Jeppesen & I. R. Walker, 2004. Aquatic invertebrates and high latitude paleolimnology. In R. Pienitz, M. S. V. Douglas & J. P. Smol (eds), Long-term environmental change in Arctic and Antarctic lakes. Springer, Dordrecht: 159-186. Blais, J. M., L. E. Kimpe, D. McMahon, B. E. Keatley, M. L. Mallory, M. S. V. Douglas, & J. P. Smol, 2005. Arctic seabirds transport marine-derived contaminants. Science 309: 445. Bonilla, S., V. Villeneuve, & W. F. Vincent, 2005. Benthic and planktonic algal communities in a High Arctic Lake: Pigment structure and contrasting responses to nutrient enrichment. Journal of Phycology 41: 1120-1130. Bouchard, G., K. Gajewski, & P. B. Hamilton, 2004. Freshwater diatom biogeography in the Canadian Arctic Archipelago. Journal of Biogeography 31: 1955-1973. Das, B., R. D. Vinebrooke, A. Sanchez-Azofeifa, B. Rivard, & A. P. Wolfe, 2005. Inferring sedimentary chlorophyll concentrations with reflectance spectroscopy: a novel approach to
reconstructing historical changes in the trophic status of mountain lakes. Canadian Journal of Fisheries and Aquatic Sciences 62: 1067-1078. Douglas, M. S. V., & J. P. Smol, 1993. Fresh-Water Diatoms from High Arctic Ponds (Cape Herschel, Ellesmere-Island, NWT). Nova Hedwigia 57: 511-552. Douglas, M. S. V., & J. P. Smol, 1994. Limnology of high arctic ponds (Cape Herschel, Ellesmere Island, N.W.T.). Archiv für Hydrobiologie 131: 410-434. Douglas, M. S. V., & J. P. Smol, 1999. Freshwater diatoms as indicators of environmental change in the High Arctic. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge: 227-244. Douglas, M. S. V., & J. P. Smol, 2000. Eutrophication and recovery in the High Arctic: Meretta lake (Cornwallis Island, Nunavut, Canada) revisited. Hydrobiologia 431: 193-204. Douglas, M. S. V., J. P. Smol, & W. Blake Jr, 1994. Marked post-18th century environmental change in high-arctic ecosystems. Science 266: 416-419. Douglas, M. S. V., P. B. Hamilton, R. Pienitz & J. P. Smol, 2004a. Algal indicators of environmental change in arctic and antarctic lakes and ponds. In R. Pienitz, M. S. V. Douglas & J. P. Smol (eds), Long-term environmental change in Arctic and Antarctic lakes. Springer, Dordrecht: 117-158. Douglas, M. S. V., J. P. Smol, J. M. Savelle, & J. M. Blais, 2004b. Prehistoric Inuit whalers affected arctic freshwater ecosystems. Proceedings of the National Academy of Sciences of the United States of America 101: 1613-1617. Erskine, P. D., D. M. Bergstrom, S. Schmidt, G. R. Stewart, C. E. Tweedie, & J. D. Shaw, 1998. Subantarctic Macquarie Island - a model ecosystem for studying animal-derived nitrogen sources using 15N natural abundance. Oecologia 117: 187-193. Evenset, A., G. N. Christensen, T. Skotvold, E. Fjeld, M. Schlabach, E. Wartena, & D. Gregor, 2004. A comparison of organic contaminants in two high Arctic lake ecosystems, Bjørnøya (Bear Island), Norway. The Science of the Total Environment 318: 125-141. Fisher, D. A., 1979. Comparison of 105 years of oxygen isotope and insoluble impurity profiles from the Devon Island and Camp Century ice cores. Quaternary Research 11: 299-305. Gaston, A. J., H. G. Gilchrist, & M. L. Mallory, 2005. Variation in ice conditions has strong effects on the breeding of marine birds at Prince Leopold Island, Nunavut. Ecography 28: 331-344. Gibson, J. A. E., W. F. Vincent, P. Van Hove, C. Belzile, X. W. Wang, & D. Muir, 2002. Geochemistry of ice-covered, meromictic Lake A in the Canadian High Arctic. Aquatic Geochemistry 8: 97-119. Grumet, N. S., C. P. Wake, P. A. Mayewski, G. A. Zielinski, S. I. Whitlow, R. M. Koerner, D. A. Fisher, & J. M. Woollett, 2001. Variability of sea-ice extent in Baffin Bay over the last millenium. Climatic Change 49: 129-145.
Hall, R. I.& J. P. Smol, 1999. Diatoms as indicators of lake eutrophication. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, New York: 128-168. Hambley, G. W., & S. F. Lamoureux, 2006. Recent summer climate recorded in complex varved sediments, Nicolay Lake, Cornwall Island, Nunavut, Canada. Journal of Paleolimnology 35: 629-640. Hamilton, P. B., K. Gajewski, D. E. Atkinson, & D. R. S. Lean, 2001. Physical and chemical limnology of 204 lakes from the Canadian Arctic Archipelago. Hydrobiologia 457: 133-148. Hobbie, J. E., B. J. Peterson, N. Bettez, L. Deegan, W. J. O'Brien, G.W. Kling, G.W. Kipphut, W. B. Bowden, & A.E. Hershey. 1999. Impact of global change on the biogeochemistry and ecology of an Arctic freshwater system. Polar Research 18: 207-214. IPCC, 2007. Summary for policy-makers. In S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor and H. L. Miller (eds), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK. Kalff, J., 1975. Phytoplankton, phytoplankton growth and biomass cycles in an unpolluted and in a polluted polar lake. Verh. int. Ver. Limnol. 19: 487-495. Kalff, J., & H. E. Welch, 1974. Phytoplankton production in Char Lake, a natural polar lake, and in Meretta Lake, a polluted polar lake, Cornwallis Island, Northwest Territories. Journal of Fisheries Research Board of Canada 31: 621-636. Kaufman, D. S., T. A. Ager, N. J. Anderson, P. M. Anderson, J. T. Andrews, P. J. Bartlein, L. B. Brubaker, L. L. Coats, L. C. Cwynar, M. L. Duvall, A. S. Dyke, M. E. Edwards, W. R. Eisner, K. Gajewski, A. Geirsdóttir, F. S. Hu, A. E. Jennings, M. R. Kaplan, M. W. Kerwin, A. V. Lozhkin, G. M. MacDonald, G. H. Miller, C. J. Mock, W. W. Oswald, B. L. Otto-Bliesner, D. F. Porinchu, K. Rühland, J. P. Smol, E. J. Steig, & B. B. Wolfe, 2004. Holocene thermal maximum in the western Arctic (0-180°W). Quaternary Science Reviews 23: 529-560. Kokelj, S. V., R. E. Jenkins, D. Milburn, C. R. Burn, & N. Snow, 2005. The influence of thermokarst disturbance on the water quality of small upland lakes, Mackenzie Delta Region, Northwest Territories, Canada. Permafrost and Periglacial Processes 16: 343-353. Lamoureux, S. F., & R. Gilbert, 2004. A 750-yr record of autumn snowfall and temperature variability and winter storminess recorded in the varved sediments of Bear Lake, Devon Island, Arctic Canada. Quaternary Research 61: 134-147. Laurion, I., W. F. Vincent, & D. R. S. Lean, 1997. Underwater ultraviolet radiation: Development of spectral models for northern high latitude lakes. Photochemistry and Photobiology 65: 107-114. Lim, D.S.S., 2004. Limnology and diatom paleoecology of diatoms on Banks Island, N.W.T., and Devon Island, Nunavut, Canadian Arctic. University of Toronto. PhD Thesis.
Lim, D. S. S., & M. S. V. Douglas, 2003. Limnological characteristics of 22 lakes and ponds in the Haughton Crater region of Devon Island, Nunavut, Canadian High Arctic. Arctic Antarctic and Alpine Research 35: 509-519. Lim, D. S. S., M. S. V. Douglas, J. P. Smol, & D. A. Lean, 2001a. Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. 86:1-22. Lim, D. S. S., M. S. V. Douglas, & J. P. Smol, 2001b. Diatoms and thier relationship to environmental variables from lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. Hydrobiologia 450: 215-230. Lim, D. S. S., C. Kwan, & M. S. V. Douglas, 2001c. Periphytic diatom assemblages from Bathurst Island, Nunavut, Canadian High Arctic: An examination of community relationships and habitat preferences. Journal of Phycology 37: 379-392. Lim, D. S. S., M. S. V. Douglas, & J. P. Smol, 2005. Limnology of 46 lakes and ponds on Banks Island, NWT, Canadian Arctic Archipelago. Hydrobiologia 545: 11-32. Lim, D. S. S., J. P. Smol, & M. S. V. Douglas, 2007. Diatom assemblages and their relationships to lakewater nitrogen levels and other limnological variables from 36 lakes and ponds on Banks Island, N.W.T., Canadian Arctic. Hydrobiologia DOI 10.1007/s10750-007-0623-8. Lindebloom, H. J., 1984. The nitrogen pathway in a penguin rookery. Ecology 65: 269-277. Mack, M. C., E. A. G. Schuur, M. S. Bret-Harte, G. R. Shaver, & F. S. Chapin, 2004. Ecosystem carbon storage in arctic tundra reduced by long-term nutrient fertilization. Nature 431: 440-443. Mallory, M. L., & A. J. Fontaine, 2004. Key marine habitat sites for migratory birds in Nunavut and the Northwest Territories. Ottawa, Canadian Wildlife Service Occasional Paper No. 109. Mallory, M. L., A. J. Fontaine, P. A. Smith, M. O. W. Robertson, & H. G. Gilchrist, 2006. Water chemistry of ponds on Southampton Island, Nunavut, Canada: Effects of habitat and ornithogenic inputs. Archiv für Hydrobiologie 166: 411-432. Markager, S., W. F. Vincent, & E. P. Y. Tang, 1999. Carbon fixation by phytoplankton in high Arctic lakes: Implications of low temperature for photosynthesis. Limnology and Oceanography 44: 597-607. Michelutti, N., M. S. V. Douglas, D. R. S. Lean, & J. P. Smol, 2002a. Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Island, Arctic Canada. Hydrobiologia 482: 1-13. Michelutti, N., M. S. V. Douglas, D. C. G. Muir, X. Wang, & J. P. Smol, 2002b. Limnological characteristics of 38 lakes and ponds on Axel Heiberg Island, High Arctic Canada. International Review of Hydrobiology 87: 385-399. Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2002c. Tracking recent recovery from eutrophication in a high arctic lake (Meretta Lake, Cornwallis Island, Nunavut, Canada) using fossil diatom assemblages. Journal of Paleolimnology 28: 377-381.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2003. Diatom response to recent climatic change in a high arctic lake (Char Lake, Cornwallis Island, Nunavut). Global and Planetary Change 38: 257-271. Michelutti, N., A. P. Wolfe, R. D. Vinebrooke, B. Rivard, & J. P. Briner, 2005. Recent primary production increases in arctic lakes. Geophysical Research Letters 32: L19715-doi: 10.1029/2005GL023693. Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2006. Ecological characteristics of modern diatom assemblages from Axel Heiberg Island (High Arctic Canada) and their application to paleolimnological inference models. Canadian Journal of Botany 84: 1695-1713. Michelutti, N., M. S. V. Douglas, & J. P. Smol, in press a. Evaluating diatom community composition in the absence of marked limnological gradients in the high Arctic: A surface sediment calibration set from Cornwallis Island (Nunavut, Canada). Polar Biology. Michelutti, N., M. H. Hermanson, J. P. Smol, P. J. Dillon, & M. S. V. Douglas, in press b. Delayed response of diatom assemblage changes to sewage inputs in an Arctic lake. Aquatic Sciences. O'Brien, S. R., P. A. Mayewski, L. D. Meeker, D. A. Meese, M. S. Twickler, & S. I. Whitlow, 1995. Complexity of Holocene climate as reconstructed from a Greenland ice core. Science 270: 1962-1964. Oliver, D. R., & P. S. Corbet, 1966. Aquatic habitats in a high arctic locality: the Hazen Camp study area, Ellesmere Island, N.W.T. Department of National Defence, Ottawa. 115 pp. Paterson, W. S. B., R. M. Koerner, D. A. Fisher, S. Johnsen, H. B. Clausen, W. Dansgaard, P. Bucher, & H. Oeschger, 1977. An oxygen-isotope climate record from the Devon Island ice cap, arctic Canada. Nature 266: 508-511. Perren, B. B., R. S. Bradley, & P. Francus, 2003. Rapid lacustrine response to recent High Arctic warming: A diatom record from Sawtooth Lake, Ellesmere Island, Nunavut. Arctic Antarctic and Alpine Research 35: 271-278. Polis, G. A., & S. D. Hurd, 1996. Linking marine and terrestrial food webs: allochthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. The American Naturalist 147: 396-423. Polis, G. A., W. B. Anderson, & R. D. Holt, 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Reviews of Ecological Systems 28: 289-316. Polis, G. A., M. E. Power & G. R. Huxel, 2004. Food Webs at the Landscape Level. University of Chicago Press, Chicago. Quinlan, R., M. S. V. Douglas, & J. P. Smol, 2005. Food web changes in arctic ecosystems related to climate warming. Global Change Biology 11: 1381-1386. Rautio, M., & W. F. Vincent, 2006. Benthic and pelagic food resources for zooplankton in shallow high-latitude lakes and ponds. Freshwater Biology 51: 1038-1052.
Rautio, M., & W. F. Vincent, 2007. Isotopic analysis of the sources of organic carbon for zooplankton in shallow subarctic and arctic waters. Ecography 30: 77-87. Rayback, S. A., & G. H. R. Henry, 2006. Reconstruction of summer temperature for a Canadian High Arctic site from retrospective analysis of the dwarf shrub, Cassiope tetragona. Arctic Antarctic and Alpine Research 38: 228-238. Robinson, D., 2001. δ15N as an integrator of the nitrogen cycle. Trends in Ecology and Evolution 16: 153-162. Ryan, P. G., & B. P. Watkins, 1989. The influence of physical factors and ornithogenic products on plant and arthropod abundance at an inland nunatak group in Antarctica. Polar Biology 10: 151-160. Sánchez-Piñero, F., & G. A. Polis, 2000. Bottom-up dynamics of allochthonous input: direct and indirect effects of seabirds on islands. Ecology 81: 3117-3132. Schindler, D. W., J. Kalff, H. E. Welch, G. J. Brunskil, H. Kling, & N. Kritsch, 1974. Eutrophication in High Arctic - Meretta Lake, Cornwallis-Island (75 Degrees N Lat). Journal of the Fisheries Research Board of Canada 31: 647-662. Schindler, D. W., & J. P. Smol, 2006. Cumulative effects of climate warming and other human activities on freshwaters of Arctic and subarctic North America. Ambio 35: 160-168. Serreze, M. C., J. E. Walsh, F. S. Chapin III, T. Osterkamp, M. Dyurgerov, V. Romanovsky, W. C. Oechel, J. Morison, T. Zhang, & R. G. Barry, 2000. Observational evidence of recent change in the northern high-latitude environment. Climatic Change 46: 159-207. Shaver, G. R., A. E. Giblin, K. J. Nadelhoffer, K. K. Thieler, M. R. Downs, J. A. Laundre, & E. B. Rastetter, 2006. Carbon turnover in Alaskan tundra soils: effects of organic matter quality, temperature, moisture and fertilizer. Journal of Ecology 94: 740-753. Smith, L. C., Y. Sheng, G. M. MacDonald, & L. D. Hinzman, 2005. Disappearing Arctic lakes. Science 308: 1429. Smol, J. P., 1983. Paleophycology of a high arctic lake Cape Herschel, Ellesmere Island. Canadian Journal of Botany 61: 2195-2204. Smol, J. P., 1988. Paleoclimate proxy data from freshwater arctic diatoms. Verh. int. Ver. Limnol. 23: 837-844. Smol, J. P., 2002. Pollution of lakes and rivers: a paleoenvironmental perspective. Arnold, London. Smol, J. P., & M. S. V. Douglas, 2007. Crossing the final ecological threshold in high Arctic ponds. Proceedings of the National Academy of Sciences of the United States of America 104: 12395-12397. Smol, J. P., A. P. Wolfe, H. J. B. Birks, M. S. V. Douglas, V. J. Jones, A. Korhola, R. Pienitz, K. Ruhland, S. Sorvari, D. Antoniades, S. J. Brooks, M. A. Fallu, M. Hughes, B. E. Keatley, T. E. Laing, N. Michelutti, L. Nazarova, M. Nyman, A. M. Paterson, B. Perren, R. Quinlan, M. Rautio,
E. Saulnier-Talbot, S. Siitoneni, N. Solovieva, & J. Weckstrom, 2005. Climate-driven regime shifts in the biological communities of arctic lakes. Proceedings of the National Academy of Sciences of the United States of America 102: 4397-4402. Van Hove, P., C. Belzile, J. A. E. Gibson, & W. F. Vincent, 2006. Coupled landscape-lake evolution in High Arctic Canada. Canadian Journal of Earth Sciences 43: 533-546. Vezina, S., & W. F. Vincent, 1997. Arctic cyanobacteria and limnological properties of their environment: Bylot Island, Northwest Territories, Canada (73 degrees N, 80 degrees W). Polar Biology 17: 523-534. Vincent, W. F., & C. L. Vincent, 1982. Nutritional state of the plankton in Antarctic coastal lakes and the inshore Ross Sea. Polar Biology 1: 159-165. Wainwright, S. C., J. C. Haney, C. Kerr, A. N. Golovkin, & M. V. Flint, 1998. Utilization of nitrogen derived from seabird guano by terrestrial and marine plants at St. Paul, Pribilof Islands, Bering Sea, Alaska. Marine Biology 131: 63-71. Wolfe, A. P., & I. R. Smith, 2004. Paleolimnology of the Middle and High Canadian Arctic. In R. Pienitz, M. S. V. Douglas & J. P. Smol (eds), Long-term Environmental Change in Arctic and Antarctic Lakes. Kluwer Academic Publishers, Dordrecht. 241-268. Wolfe, A. P., R. D. Vinebrooke, N. Michelutti, B. Rivard, & B. Das, 2006. Experimental calibration of lake-sediment spectral reflectance to chlorophyll a concentrations: methodology and paleolimnological validation. Journal of Paleolimnology 36: 91-100. Zdanowicz, C. M., G. A. Zielinski, C. P. Wake, D. A. Fisher, & R. M. Koerner, 2000. A Holocene record of atmospheric dust deposition on the Penny Ice Cap, Baffin Island, Canada. Quaternary Research 53: 62-69.
Figure captions Figure 1. Regional map of the Canadian Arctic Archipelago with the locations of limnological, diatom ecological, and paleolimnological studies conducted by the Smol and Douglas laboratories prior to this thesis. The numbers correspond to the following studies: 1 = Mould Bay, Prince Patrick Island (Antoniades et al. 2003a, 2005a, 2005b); 2 = Isachsen, Ellef Ringnes Island (Antoniades et al. 2003b, 2004, 2005b); 3 = Axel Heiberg Island (Michelutti et al. 2002b, 2006); 4 = Alert, Ellesmere Island (Antoniades et al. 2003a, 2005b); 5 = Cape Herschel, Ellesmere Island (Douglas & Smol 1993, 1995; Douglas et al. 1994), 6 = Devon Island (Lim et al. 2003; Lim 2004); 7 = Cornwallis Island (Douglas & Smol 2000; Michelutti et al. 2002c, 2003, in press a); 8 = Bathurst Island (Lim et al. 2001a, 2001b, 2001c); 9 = Victoria Island (Michelutti et al. 2002a); 10 = Banks Island (Lim et al. 2005, 2007). The letters correspond to regions studied in this thesis: A) Melville Island (Chapters 2, 3, and 5); B) northern Ellesmere Island (Chapters 4 and 6); and C) Cape Vera, Devon Island (Chapters 7 and 8).
80° 80°
75° 75°
70° 70°
65° 65°120°
120°
100°
100°
80°
80°
60°
60°
0 200 400km
1
3
4
6
5
8
2
9
710
A
B
C
Figure 1.
CHAPTER 2
PHYSICAL AND CHEMICAL LIMNOLOGICAL CHARACTERISTICS OF LAKES AND PONDS ACROSS
ENVIRONMENTAL GRADIENTS ON MELVILLE ISLAND, NUNAVUT/N.W.T., HIGH ARCTIC CANADA
BRONWYN E. KEATLEY, MARIANNE S.V. DOUGLAS, AND JOHN P. SMOL
This chapter has been published separately: Keatley, B.E., M.S.V. Douglas, and J.P. Smol. 2007. Physical and chemical limnological characteristics of lakes and ponds across environmental gradients on Melville Island, Nunavut/N.W.T., High Arctic Canada. Fundamental and Applied Limnology. 168: 355-376.
Abstract
Physical and chemical limnological variables were measured from 40 ponds and 6 lakes
across Melville Island, Nunavut/N.W.T., Canadian high Arctic, an environmentally sensitive
region where very limited limnological data were available. Mean values of most variables were
mid-range when compared to other high Arctic limnological surveys, yet the ranges of most
measured variables were amongst the largest yet encountered in Canadian high Arctic regional
surveys. The first two axes of a Principal Components Analysis explained 55.2% of the variation
in the environmental data. Variables most strongly associated with axis one were pH, dissolved
organic carbon, total dissolved nitrogen, specific conductivity and related variables, while axis
two represented gradients of other nutrients and trace metals. High elevation sites near
permanent ice caps recorded the lowest specific conductivity and Ca2+ values yet reported in high
Arctic systems. High phosphorus values (>20 μg/L) in some of the Melville Island sites are
likely indicative of re-suspended sediments, rather than eutrophic conditions. Total nitrogen to
total phosphorus ratios suggest that ~50% of the sites are P limited, while 33% are N limited,
supporting previous research which suggests N limitation is more commonly encountered in
Arctic than in temperate freshwater ecosystems. Finally, when freshwater sites on Melville
Island were grouped according to predefined bioclimatic zones, only the most lushly vegetated
zone appeared to affect limnological conditions, with these sites having higher mean total
dissolved nitrogen, pH, and specific conductivity.
Introduction
High-latitude regions are known to be sensitive to environmental change. For example,
the region north of 60° latitude has warmed by 0.04°C/year compared to 0.025°/year for lower
latitudes over the last 40 years (ACIA 2004). Data from boreholes in Arctic areas, including
Melville Island, suggest that ground surface temperatures have warmed between 1-2°C since the
late 18th century (MAJOROWICZ et al. 2005), while multiproxy records across the Arctic suggest
that the 20th century experienced the warmest temperatures in about 400 years (OVERPECK et al.
1997). These environmental changes have manifested themselves in a variety of ways, including
declining snow cover, thawing permafrost, melting glaciers and sea ice, as well as a series of
limnological changes (e.g. ROUSE et al. 1997, MAGNUSON et al. 2000, ACIA 2004, ANTONIADES
et al. 2005, HINZMAN et al. 2005, SOLOVIEVA et al. 2005).
Limnological changes are pronounced in high Arctic lakes and ponds which are known to
be highly susceptible to environmental change (e.g. DOUGLAS et al. 1994). Indeed, recent
paleolimnological studies from circumpolar regions have highlighted dramatic shifts in algal and
invertebrate community structure, which are consistent with recent climatic warming (SMOL et al.
2005), and increased biological production (MICHELUTTI et al. 2005). Undoubtedly, a warming
climate is already having, and will continue to have, major repercussions on limnological
conditions in Arctic regions. In order to identify whether lakes and ponds are affected by
environmental change, however, it is necessary to document and understand present-day
conditions. Limnological data are scarce in high Arctic regions, as regular monitoring of lakes
and ponds are often hampered by logistical and financial constraints. Nevertheless, these modern
limnological data are important for general monitoring, and are necessary for assessing the
ecological distributions of biological indicators (e.g. KORHOLA et al. 2005), thereby allowing for
scientifically defensible reconstructions of past environmental conditions using paleolimnological
techniques. For example, these data, together with paleoindicators such as diatoms and
cladocerans preserved in lake and pond sediments, can be used as integrative bioindicators of
environmental quality in remote regions such as the Arctic.
Although many Arctic regions remain unexplored, there has been a marked increase in
Arctic limnological studies over the past 10 to 20 years. In the Canadian Arctic Archipelago,
physical and chemical characteristics have been documented for lakes and ponds on Ellesmere
Island (DOUGLAS & SMOL 1994, HAMILTON et al. 2000, ANTONIADES et al. 2003a), Axel
Heiberg Island (MICHELUTTI et al. 2002b), Banks Island (LIM et al. 2005), Bathurst Island (LIM
et al. 2001), Devon Island (LIM & DOUGLAS 2003), Ellef Ringnes Island (ANTONIADES et al.
2003b), Prince Patrick Island (ANTONIADES et al. 2003a), and Victoria Island (MICHELUTTI et al.
2002a). In addition, limnological data from 204 lakes and ponds across the Arctic have also been
published (HAMILTON et al. 2001). These studies indicate that extensive limnological variability
results from the large diversity in geographical and physical settings of high Arctic lakes and
ponds. For example, freshwater bodies throughout the Arctic range from extremely nutrient poor
(e.g. MICHELUTTI et al. 2002a) to eutrophic (e.g. LIM et al. 2005), from extremely dilute to highly
saline (e.g. MICHELUTTI et al. 2002b), and from very acidic, with pH <4 (MICHELUTTI et al.
2002b), to highly alkaline (pH>9, HAMILTON et al. 2001). These wide-ranging conditions
underscore the need for a more complete survey of present-day limnological conditions in
different regions. Despite this recent work, gaps in limnological knowledge persist in many
regions, especially in the western High Arctic, hampering our ability to confidently recognize and
anticipate the effects of climatic changes.
Melville Island is a large island located in the western-central Canadian High Arctic (Fig.
1). The goals of this study are to provide baseline environmental data on the physical and
chemical characteristics of 40 ponds and 6 lakes from across Melville Island, and to determine
which environmental factors most influence the chemical and physical properties of these lakes
and ponds. Our aim is to place the present-day limnology of Melville Island into context with
respect to limnological studies previously completed on high Arctic islands of Canada. Given
that Melville Island encompasses five bioclimatic zones (EDLUND 1994), we also attempt to
distinguish limnological differences between sites from these pre-defined bioclimatic zones.
In order to achieve the above aims, we have divided the analysis of our data into two
inter-related sections. First, as Melville Island represents a large (i.e. 42 149 km2), diverse, but
previously unstudied reference area for polar limnology, we begin by a descriptive analysis of the
physical and chemical properties of the 46 study sites. In this section, we compare our data to
those collected from other Arctic islands, using identical sampling techniques, thus facilitating
comparisons. Second, using principal components analysis (PCA) and other statistical techniques,
we determine which environmental variables best explain the differences in the measured
limnological characteristics.
Site Description
Melville Island (Fig. 1) is the fourth largest island of the Queen Elizabeth Islands
(consisting of Arctic islands north of the Parry Channel) in High Arctic Canada, and straddles the
border between the territory of Nunavut and the Northwest Territories (N.W.T.). As the only
weather records from Melville Island come from the automated data collector at Rea Point and
are only available since 1975, meteorological data is better estimated from the nearest “long-
term” record from the weather station at Mould Bay, Prince Patrick Island (Fig. 1, 76°13’N,
119°19’W), located ~250 km from the centre of Melville Island, where average February and July
temperatures are -34.0°C and 4.0°C respectively, and the mean annual precipitation is 111.0 mm
(METEOROLOGICAL SERVICE OF CANADA 2002). Sea ice reaches its minimum extent around 10
September and by approximately mid-November sea ice cover is generally 100% around Melville
Island (CANADIAN ICE SERVICE 2002). However, the break-up begins first on the southern
coastal areas (McClintock Channel and Viscount Melville Sound) in late July/August, ultimately
reaching between 1% to 60% ice cover, and continues northeast and northwest. The northern
coastal regions typically retain at least 90% ice cover throughout the year (CANADIAN ICE
SERVICE 2002). Prevailing wind patterns are difficult to ascertain due to a lack of data; however,
the unpredictability of wind patterns in Canadian Arctic regions has been noted as an impact of
climate change (ACIA 2004).
Five bioclimatic zones have been previously described for Melville Island (Fig. 2a),
which are delineated mainly by surficial geology and summer climate (EDLUND 1994).
Bioclimatic zone 0 represents completely unvegetated areas and occurs at high elevations. Zone
1 supports low botanical diversity consisting of vascular plants with less than 5% ground cover.
Zone 2 is dominated by herbaceous species and grasses. Zone 3 is dominated by woody plants
and shrubs (Dryas integrifolia, Salix arctica), with a low diversity of sedges (Carex aquatilis var.
stans and Eriophorum sp.) and emergent and aquatic species including Pleuropogon sabinii and
Ranunculus hyperboreus. Zone 4, the most diverse and abundantly vegetated zone (at least 25%
ground cover), is dominated by the same species as found in Zone 3 with the addition of legumes
(Oxytropis sp. and Astragalus alpinus), Rosacea, Asteraceae, Taraxacum sp., and several sedges
and cotton grasses (EDLUND 1994).
The surficial geology throughout Melville Island is largely composed of weathered
bedrock, which is highly variable within regions. In general, the geology of eastern Melville
Island is comprised of sandstones, mudstones, limestones, shales, and evaporites (BARNETT et al.
1977); the western part of the island is dominated by sandstones and carbonates, and the Dundas
Peninsula contains glacial deposits, including some calcareous tills of southern provenance
(TOZER & THORSTEINSSON 1964, HODGSON 1989). Melville Island is the only island in the
western Arctic that currently retains small permanent ice caps (Fig. 2).
Four main physiographic regions have been described on Melville Island: 1) the southern
half of the Dundas Peninsula is characterized by plateaus on horizontal Palaeozoic rocks; 2) the
western peninsula between Murray Inlet and Purchase Bay is characterized by ridges and plateaus
on Tertiary faults; 3) the central and eastern parts of the island are comprised of ridges and
plateaus on folded Palaeozoic rocks; and 4) the Sabine and Sproule peninsulas in the north parts
of the Melville Island are represented by plateaus developed on Upper Paleozoic to early Tertiary
rocks (TOZER & THORSTEINSSON 1964).
There are no permanent settlements on Melville Island, although stone circles found near
Bridport Inlet and Liddon Gulf suggest occupation by Dorset and Pre-Dorset peoples, and
evidence of hunting camps suggests that the Thule visited Melville Island between ca. 500 and
3000 years ago (CHRISTIE 1994).
Although little is known of the limnology of this large area, the anticipation of an Arctic
pipeline project sparked a number of environmental resource surveys, including some preliminary
aquatic analyses, in the 1970s (reviewed in STEWART & BERNIER 1982). Indian and Northern
Affairs Canada also conducted aquatic sampling of 7 lakes from Melville Island in 1981
(STEWART & BERNIER 1982). These surveys provided only limited water chemistry data for two
lakes and concluded that there are few freshwater fish on Melville Island (STEWART & BERNIER
1982).
Methods
Based on our criteria that lakes are > 2 m deep and do not freeze to the bottom in the
winter, while ponds are < 2 m deep and do freeze completely, our 46 sites comprised 6 lakes and
40 ponds. Due to the danger of weak ice on some sites, we could only estimate depth, and thus
our descriptions are largely restricted to classification of a site as either a lake or a pond. A
subset of 14 ponds was located within a radius of ~5 km of Winter Harbour (Fig. 2) and these
were sampled by foot. The remaining 32 sites were sampled via helicopter. All sites were given
alphabetical names preceded by the prefix MV to designate Melville Island (MVA, MVB, etc.).
The sites were chosen to encompass the widest possible environmental gradients (e.g. size,
proximity to the coast, vegetation, geology). We sampled all the lakes and ponds within a short
time frame between 15-24 July 2002, using identical sampling techniques as other limnological
surveys carried out by our labs (e.g. LIM et al. 2001, MICHELUTTI et al. 2002a, MICHELUTTI et al.
2002b, LIM & DOUGLAS 2003, ANTONIADES et al. 2003a, ANTONIADES et al. 2003b, LIM et al.
2005), thereby facilitating comparisons between our previous studies. Detailed methods regarding
field water chemistry sampling are given in Appendix 1.
Latitude, longitude, and elevation measurements were taken using either a handheld
global positioning unit or the helicopter altimeter. Water samples were collected using either
Nalgene® plastic bottles (major ions and trace metals) or 125 mL glass bottles (for phosphorus
and nutrients) following guidelines outlined in Environment Canada (1994). Temperature (temp),
pH, and specific conductivity (cond) were measured from all sites in the field using handheld
thermometers, calibrated handheld Hanna pHep pH meters, and a YSI model 33 conductivity
meter, respectively. All other analyses were performed at the National Water Research Institute
(NWRI) in Burlington, Ontario (Environment Canada), using protocols described in Environment
Canada (1994).
Water was sampled for major cations including calcium (Ca2+), magnesium (Mg2+),
sodium (Na+), and potassium (K+). The major anions measured included chloride (Cl-), and
sulphate (SO42-). Minor ions measured included barium (Ba), lithium (Li), and strontium (Sr).
Metals measured were aluminum (Al), beryllium (Be), cadmium (Cd), chromium (Cr), cobalt
(Co), copper (Cu), iron (Fe), lead (Pb), manganese (Mn), molybdenum (Mo), nickel (Ni),
vanadium (V), zinc (Zn), and silver (Ag). The nutrients and related variables measured included
dissolved silica (SiO2), total phosphorus unfiltered (TP), total phosphorus filtered (TPF), soluble
reactive phosphorus (SRP), nitrogen fractions (nitrate-nitrogen (NO3-N), nitrate-nitrite-nitrogen
(NO3NO2-N), ammonia-nitrogen (NH3-N), total Kjeldahl nitrogen (filtered, TKN), total dissolved
nitrogen (TdN), and particulate nitrogen (PON)), carbon (dissolved inorganic carbon (DIC),
dissolved organic carbon (DOC), particulate organic carbon (POC)), and chlorophyll a (Chlac,
Chla, corrected and uncorrected for phaeophytin, respectively). Details regarding filter size and
type, sample storage, and sample analysis are given in Appendix 1.
To better understand nutrient trends in these aquatic systems, a number of nutrient ratios
were calculated. For example, ratios of POC:Chla (mg) were assessed to help elucidate whether
carbon sources were predominantly allochthonous (>100) or autochthonous (<100, EPPLEY et al.
1977), and were also used to assess general nutrient deficiency trends (μmol:μg) (HECKY et al.
1993). Total N:Total P ratios were calculated to assess whether sites were more likely to be N or
P limited (DOWNING & MCCAULEY 1992, SAKAMOTO 1966, GUILDFORD & HECKY 2000). We
also calculated molar ratios of POC:PON to examine the degree of N-deficiency in lakes (HECKY
et al. 1993). Finally, P deficiencies were further assessed by examining particulate matter molar
ratios of PON:POP and POC:POP (HECKY et al. 1993).
Statistical analyses
Variables that were below the detection limit in >50% of the sites were removed from
further analyses. Some variables were below the analytical detection limit in <50% of sites and
were replaced with half the detection limit for further statistical analyses to avoid zero values and
to provide an estimate of the true value in those sites. Some values of some variables for a few
sites were unavailable due to broken bottles during shipping; these missing values were estimated
by performing a linear regression between two highly correlated variables and using the equation
of the resulting line to derive the missing value. To assess the normality of distributions,
limnological variables were first examined with exploratory graphing techniques using
CALIBRATE version 1.0 (JUGGINS & TER BRAAK 1992). All variables that were not normally-
distributed were normalized by transforming them using either square-root, log x or log x+1
transformations. The transformed data were subsequently used in the following statistical
analyses. A Pearson correlation with Bonferroni-adjusted probabilities was run using the
program Systat version 9.0 to assess pairs of significantly correlated variables. Principal
components analysis (PCA) was executed using CANOCO 4.0 (TER BRAAK & ŠMILAUER 1998),
to determine the main directions of variation in limnological variables in the dataset. Any
variable whose distribution could not be normalized using the aforementioned transformations, as
well as the spatial measurements of elevation, latitude and longitude, were run passively (e.g.
they were plotted onto the PCA after patterns of variation had been derived from the active
variables and thus did not influence the ordination) in the PCA. Finally, to further investigate
differences amongst lakes and ponds located in different bioclimatic zones, t-tests (assuming
unequal variances) were performed using Systat version 9.0.
Results & Discussion
Descriptive analyses
Physical variables
The Melville Island dataset represents the largest elevational gradient yet sampled in high
Arctic limnological surveys, ranging from ~3 m above sea level (asl) (MVAH) to 655 m asl
(MVAE). Other physical gradients covered on Melville Island included a range of latitude from
74°25.14’N (MVQ) to 76°38.66’N (MVAM), and a range of longitude from 106°03.33’W
(MVAQ) to 114°41.53’W (MVAE). Increasing latitude and elevation typically result in harsher
climatic conditions, and thus are expected to play important roles in water chemistry. Despite the
large latitudinal gradient on Melville Island, the complex topography promotes elevation as a
main driver in many of the water quality parameters in this system.
Water temperature during the short sampling window we used (i.e. 15-24 July 2002)
ranged from 2.5°C to 15.5°C, with a mean temperature of 8.8°C (Table 1). The warmest sites,
not surprisingly, were amongst the smallest ponds we sampled (MVP, MVY, MVX, MVS,
MVAD). Water temperature in small Arctic ponds often closely tracks daily air temperature
(DOUGLAS & SMOL 1994), and all these sites were measured over the course of one relatively
warm day. These shallow sites also had relatively lush catchments with mosses and grasses (MV-
P, MV-S, MV-X), and/or had noticeably dense populations of zooplankton, including copepods,
Daphnia, and Anostraca (fairy shrimp). Conversely, the three coolest sites, MVAP, MVAL and
MVAF, were relatively large lakes that were partially ice-covered at the time of sampling, had
little vegetation in their catchments, and were depauperate in zooplankton.
While we did not measure water clarity directly, the turbid nature of many sites during
windy days was noted. Turbidity affects the re-suspension of sediment particles, and thus likely
influences some of our chemical measurements.
Metals
Similar to other high Arctic surveys (e.g. LIM et al. 2001, MICHELUTTI et al. 2002a,
MICHELUTTI et al. 2002b, LIM & DOUGLAS 2003, ANTONIADES et al. 2003a, ANTONIADES et al.
2003b, LIM et al. 2005), many metals were below detection limit in > 50% of the sites (Ag, Be,
Cd, Co, Cr, Mo, Ni, Pb, and V) and are not discussed further. This resulted in a subset of metals
consisting of Al, Ba, Cu, Fe, Li, Mn, Sr, and Zn.
All metal concentrations, except Al and Fe, examined in the Melville Island dataset were
within ranges for natural Canadian waters (MCNEELY et al. 1979). However, mean
concentrations of Al (0.34 mg/L) and Fe (0.49 mg/L) were higher than most other Arctic sites
(Table 1) with similar means and ranges of Al to Ellef Ringnes Island (ANTONIADES et al.
2003b), and Fe with both Ellef Ringnes and Banks islands (ANTONIADES et al. 2003b, LIM et al.
2005). Al was significantly correlated with both Fe and Mn concentrations on Melville Island
(Table 3) and, not surprisingly, the sites with the highest Al typically had high Fe and Mn values.
These sites included two large lakes, MVAO and MVAL, and two small ponds, MVAH and
MVH, all of which were turbid on the day of sampling. In addition to high Fe and Al values,
these sites also had high TP concentrations (see below). Together, this suggests that the re-
suspension of sediment particles is likely the source of elevated metal concentrations in these
sites.
SiO2
Silica concentrations were also within the ranges reported for natural Canadian surface
waters (MCNEELY et al. 1979) with the largest concentrations occurring in the same sites with
very high levels of Al, Fe, Mn, and TP. This likely represents, once again, the presence of re-
suspended material in these sites.
pH and Specific Conductivity
Sites on Melville Island were circumneutral to alkaline in pH, ranging from 6.9 (MVAF)
to 8.8 (MVB), (mean pH = 7.8). This is a relatively small range of pH, given the large geological
gradients covered in this survey, but it is similar to those found on most high Arctic islands (Fig.
3). The pH changes between sites are likely related mainly to local geology, but also possibly to
longer ice cover (PSENNER & SCHMIDT 1992). The lowest pH (6.9) was encountered in a large,
mostly frozen (~80%) lake located at a high elevation which was underlain by sandstone,
siltstone and shale (MVAF). When ice cover persists, photosynthesis is limited, and the
associated CO2 build up can contribute to declining pH (PSENNER & SCHMIDT 1992). The highest
pH (8.8) was from a small pond (MVB) located on calcareous till in the most diverse bioclimatic
zone on Melville Island, and was inhabited by 4 eider ducks on the day of sampling. The
combination of alkaline surficial material, duck guano and relatively high production may have
elevated the pH of this site relative to others in this region.
Specific conductivity in Arctic lakes and ponds is often inversely related to distance from
the coast. On Melville Island, specific conductivity ranged from 5 μS/cm to 1250 μS/cm, with a
mean of 157 μS/cm. This is similar to the average specific conductivities measured from
Bathurst Island, Banks Island, Cape Herschel, and Prince Patrick Island, but is much lower than
those recorded at Axel Heiberg and Ellef Ringnes islands, as well as that recorded at Alert on
Ellesmere Island (Fig. 3., DOUGLAS & SMOL 1994, LIM et al. 2001, MICHELUTTI et al. 2002b,
ANTONIADES et al. 2003a, ANTONIADES et al. 2003b, LIM et al. 2005). Not surprisingly, the
highest conductivities on Melville Island came from coastal ponds that were influenced by sea
spray; they also had the highest concentrations of Na+ and Cl- (MVAI, MVAB). Interestingly,
some sites on Melville Island recorded the lowest conductivities (5 μS/cm) yet reported from the
High Arctic (MVAP, MVAQ, MVAE, MVAF). The lack of catchment vegetation (which would
accelerate the flow of run-off into the lakes and reduce time for solute concentration in run-off),
the influence of ice cover (which would reduce atmospheric input into the water), the relatively
large volumes of water in these sites, and/or the location at high elevations near permanent ice
caps (which would reduce both the size of the catchment area and the influence of marine-derived
aerosols) all contributed to the very low specific conductivities in these sites.
Major and minor ions
While specific conductivity provides summary information about total ionic composition,
the relative contributions of the major cations and anions allow us to more fully understand the
water chemistry of a given system. Mean concentrations (mg/L) of major cations followed the
pattern Na+>Ca2+>Mg2+>K+, while major anions follow Cl->SO42->DIC
(carbonates/bicarbonates). This is likely due to the large influence of coastal sites with very high
Na+ and Cl- concentrations, as well as the influence of sites with very little Ca2+ (see below). To
remove the influence of these few sites, median ion concentrations were also calculated. Median
concentrations give the following cation pattern of Ca2+>Mg2+>Na+>K+, which is similar to cation
concentration patterns found on Victoria (MICHELUTTI et al. 2002a), Devon (LIM & DOUGLAS
2003), and Bathurst (LIM et al. 2001) islands. The median concentrations of anions follow the
pattern of DIC>Cl->SO42-, which is similar to that described from Alert (ANTONIADES et al.
2003a), but a departure from the more typical anion concentration pattern of DIC>SO42->Cl-
found in many other Arctic sites (e.g. LIM et al. 2001, MICHELUTTI et al. 2002a, MICHELUTTI et
al. 2002b, LIM & DOUGLAS 2003). The elevated concentrations of Cl- relative to SO42-
may be
attributed to the highly irregular shoreline of Melville Island; as such, no site is greater than ~40
km from the coast.
The direct influence of the above coastal proximity, as well as the indirect influence of
elevation, are evident in both Cl- and Na+ concentrations, which were significantly (positively)
correlated with each other, as well as significantly (negatively) correlated with elevation (Table
3). Concentrations of Cl- were relatively high compared to other Arctic islands, with a mean
(37.39 mg/L) and a range similar to those found on Ellef Ringnes Island (ANTONIADES et al.
2003b). Sodium concentrations were close to the average for other Arctic islands (mean 19.1
mg/L, range 0.1 mg/L to 297.0 mg/L). Mean Na+ concentrations on Melville Island were lower
than those reported for Axel Heiberg and Ellef Ringnes islands (MICHELUTTI et al. 2002b,
ANTONIADES et al. 2003b), but higher than Bathhurst and Victoria islands (LIM et al. 2001,
MICHELUTTI et al. 2002a). Two small coastal sites had the highest Cl- and Na+ concentrations,
(MVAB and MVAI), while the two ice-covered, high elevation sites (MVAE and MVAF) had
Na+ concentrations below the detection limit, as well as relatively low Cl- concentrations.
The varied surficial and bedrock geology found on Melville Island is likely responsible
for a range of DIC concentrations that was amongst the largest reported for high Arctic islands,
(Fig. 3, mean 12.3 mg/L, median 8.4 mg/L, maximum 47.8 mg/L, minimum 0.2 mg/L), except for
values documented on Axel Heiberg Island (MICHELUTTI et al. 2002b). For example, calcareous
materials are largely restricted to the Dundas Peninsula and are associated with glacial tills
originating from Victoria Island, rather than from local bedrock (HODGSON 1989). Not
surprisingly, the sites located on this glacial till (MVA through MVM) had the highest DIC
concentrations in our suite of lakes and ponds. The lowest DIC sites were located on the central
and eastern part of the Melville Island (MVAJ, MVAP, and MVAQ), an area underlain mainly by
sandstones, shales, and evaporites (BARNETT et al. 1977).
Surficial geology also plays a role in both Ca2+ and SO42- concentrations. For example,
while the mean concentrations of both SO42- and Ca2+ were similar to previously recorded Arctic
values (SO42-: mean 13.8 mg/L, median 1.8 mg/L; Ca2+: mean 15.8 mg/L and median 7.5 mg/L),
the location of pond MVAM near the base of a sulphate and calcium-rich gypsum evaporite dome
on the Sabine Peninsula resulted in anomalously high concentration of SO42- (357 mg/L) and Ca2+
(179 mg/L). A similar trend was noted on Ellef Ringnes Island in ponds close to gypsum
outcrops (ANTONIADES et al. 2003b). Other relatively high Ca2+ values were found in sites with
high DIC, located on calcareous glacial till (see above).
Interestingly, Melville Island also recorded the lowest Ca2+ and Mg2+ concentrations yet
reported from the High Arctic (Fig. 3; Ca2+ only). Low Ca2+ concentrations were also reported by
Stewart & Bernier (1982) in their survey of 7 Melville Island lakes. Once again, high elevation
and ice cover characterised the sites with the lowest Ca2+ and Mg2+ concentrations (MVAE and
MVAF). These sites also had very low specific conductivities, reflecting the dominant role of
Ca2+ in conductivity measurements for sites located away from coastal areas. Similar low Ca2+
values to sites from Melville Island are reported from SubArctic areas in Alaska (GREGORY-
EAVES et al. 2002), and across treeline sites from the Northwest Territories (RÜHLAND & SMOL
1998).
Potassium concentrations (mean 1.5 mg/L, range 0.1 mg/L to 14.1 mg/L) were similar to
the range recorded on Banks Island (LIM et al. 2005). Lim et al. (2005) found that high
concentrations of K+ on Banks Island were likely attributable to higher terrestrial production
evidenced by low Na:K ratios (MCNEELY et al. 1979), and it is possible that the low Na:K (Table
2) ratios of some of the Melville Island sites are related to relatively high plant production.
However, the sites located in the most lushly vegetated zone did not have significantly lower
Na:K ratios than other sites (see below). The highest K+ concentration was recorded in a small
coastal pond (MVAI); as K+ is significantly correlated with Cl-, the influence of sea spray was
likely a driving force in this high potassium concentration.
Nutrients and related variables
In general, Arctic lakes and ponds are quite oligotrophic relative to temperate sites.
However, lakes and ponds surrounded by relatively lush vegetation would be expected to have
higher nutrients (phosphorus and nitrogen) and related variables (organic carbon) due to increased
input from the catchment. The increased availability of nutrients is expected to translate into
higher primary productivity (as measured by chlorophyll a). Many of our sites followed this
expected pattern, with some notable exceptions.
Many nitrogen species were measured from Melville Island. Ammonia (NH3) and nitrite
(NO2) levels were within ranges reported for other Arctic sites, although many lakes and ponds
situated in the High Arctic are typically below the detection limit for these nitrogen species; this
was the case for nitrate-nitrite (NO3-NO2) values from most sites in our dataset. Total Kjeldahl
nitrogen (TKN) is a measure of ammonia and organic nitrogen; TKN values were similar to those
reported from Bathurst, Victoria and Ellef Ringnes islands (LIM et al. 2001, MICHELUTTI et al.
2002a, ANTONIADES et al. 2003b). In summary, total nitrogen values (TN =
PON+TKN+NO3NO2) were calculated and were within the ranges reported from other high
Arctic surveys (Fig. 3). Total dissolved nitrogen (TdN) values were not directly measured for
other high Arctic limnological surveys. Not surprisingly, the sites with the highest TdN, TN, and
TKN values in our dataset (MVL, MVM) came from small ponds located in a very lush area
around Winter Harbour that was characterized by mosses and grasses. These sites also contained
large populations of Daphnia and Anostraca, suggesting that they are relatively productive. As
expected, the lowest TdN values were found in generally high elevation sites MVAE, MVAF,
and MVAJ that also had low DOC. TdN values were significantly correlated with DOC, DIC,
TKN, Ca, Mg, pH, specific conductivity, and (negatively) to elevation (Table 3).
The only nitrogen fraction that did not fit this pattern was PON, which appears to be
more closely related to particulate matter (POC, TP) in the water column than to other nitrogen
parameters. PON concentrations (mean = 0.051 mg/L, median = 0.041 mg/L) were higher than
all other Arctic sites, although Axel Heiberg Island (MICHELUTTI et al. 2002b) was the only other
high Arctic island with >50% of sites above detection limit. The sites with the highest
concentrations of PON (MVAK: 0.296 mg/L, MVAH: 0.122 mg/L, and MVP: 0.210 mg/L) also
had high POC and, except for MVP, high TP values.
Unlike TN or TdN values, the highest TP values were not found in small ponds located in
lushly vegetated sites, but instead in ponds and lakes that likely experience sediment re-
suspension. On Melville Island, TP values were elevated relative to other high Arctic ponds and
lakes (Fig. 3, mean 21.8 μg/L, median 13.4 μg/L). The range of values, 6.2 μg/L to 135.0 μg/L
(Table 1), was most similar to those reported from Banks Island, the most lushly vegetated Arctic
island yet studied (LIM et al. 2005), but lower than those values reported from Ellef Ringnes
Island (ANTONIADES et al. 2003b). The two sites with the highest TP values (MVAR: 135 μg/L
and MVAO: 123 μg/L) had values of TPF, SRP, and TdN that were below their respective means.
Since MVAO was a large, mostly ice-covered lake that was quite turbid, also had high Fe and Al
values (see above), and was located on phosphorus-rich shale, the high TP value is likely a result
of sediment re-suspension; a similar phenomenon was noted and used to explain very high TP
values from sites located on Ellef Ringnes Island (ANTONIADES et al. 2003b). Sites MVAL (73.0
μg/L) and MVAQ (34.5 μg/L) were also large lakes that had little apparent vegetation in their
catchments, yet had fairly high TPF and SRP values. The lowest TP values were from sites MVV
(6.2 μg/L), a large lake with >70% ice cover containing very little vegetation in its catchment;
MVAM (8.8 μg/L), a pond located right by the gypsum outcropping on the Sabine Peninsula; and
MVI (8 μg/L), a small pond characterized by many mosses and grasses in its catchment.
To avoid problems associated with particulate P, total phosphorus filtered (TPF) may be
used. Unfortunately, TPF values suggest that some of our samples may have been contaminated,
as TPF was much higher than TP in some sites (Table 1, MVAC, MVAH, MVAM, MVAN).
When these suspicious sites are removed, mean and median TPF is 10.0 μg/L and 8.6 μg/L,
respectively. These are the highest average values for TPF yet reported in the High Arctic, but
are similar to values reported for two of the most lushly vegetated high Arctic regions yet studied,
Banks Island (LIM et al. 2005) and Prince Patrick Island (ANTONIADES et al. 2003a). As
expected, sites with high TPF concentrations were typically small ponds in lush, grass-moss
meadows (MVE: 32.2 μg/L, MVF: 24.9 μg/L, MVAA: 22.7 μg/L, MVAK: 25.6 μg/L). Site
MVAL (22.4 μg/L) was unusual in this group as it was a very large, mostly ice-covered lake with
little terrestrial vegetation in its catchment. Interestingly, TPF was uncorrelated with any other
variable in our dataset (Table 3).
Like TPF, soluble reactive phosphorus (SRP) values were uncorrelated with any other
variable used in this dataset and were also relatively high; the mean concentration was the highest
yet reported from high Arctic surveys (mean = 2.8 μg/L). The two sites with the highest SRP
values from Melville Island were MVAJ (13.5 μg/L), a site with very little vegetation and
sediment-covered rocks at the bottom, and MVAM (11.3 μg/L), a pond located near a gypsum
dome.
Dissolved organic carbon (DOC) in lakes and ponds is highly related to the presence and
type of vegetation in the catchment. Due to relatively sparse vegetation cover and very harsh
climate (where water and soils are frozen ~10 months of the year), DOC values from high Arctic
lakes and ponds are often much lower than those present in Subarctic sites, where more extensive
vegetation growth, increased leaf litter, and a longer season for run-off act in concert to increase
DOC. On average, DOC values from Melville Island were in the higher range of values reported
for Canadian Arctic islands (Fig. 3, mean 5.45 mg/L and median 4.65 mg/L). Not surprisingly,
these values were much lower than those reported from Subarctic sites located in Alaska
(GREGORY-EAVES et al. 2000), likely reflecting the more restricted summer growing season in
the High Arctic. Some small ponds on Melville Island had quite high DOC concentrations (MVL,
MVM, MVF, and MVAK); these are likely due to their locations in the most lushly vegetated
regions of Melville Island, which would increase DOC in run-off from the catchment. The lowest
DOC concentrations were found in MVAE and MVAF, two high elevation sites characterized by
persistent ice cover, very low conductivities and low TdN values.
Particulate organic carbon (POC) values were within the range reported for other high
Arctic sites (mean 0.503 mg/L, range 0.162 mg/L to 2.710 mg/L). These values were higher than
those recorded on Victoria Island (MICHELUTTI et al. 2002a), approximately the same as ponds on
Axel Heiberg Island (MICHELUTTI et al. 2002b), but lower than Ellef Ringnes and Banks islands
(ANTONIADES et al. 2003b, LIM et al. 2005). The sites with the highest POC values (MVAK,
MVP, MVAH) also had high DOC and TdN values. POC was correlated with both PON and
DOC (Table 3).
Chlorophyll a
Both chlorophyll a uncorrected (Chla) and corrected (Chlac) for phaeophytin were
measured; however, like most high Arctic studies, Chlac was below the detection limit levels in
nearly all sites. This suggests that much of the algal material in these systems was degraded.
Chla had a mean of 1.41 μg/L, a median value of 0.9 μg/L, and a range between <0.1 μg/L to
10.1 μg/L. The mean and median values are similar to most other high Arctic sites, but the
maximum Chla concentration reported was higher than almost all other measured Chla values
except for Banks Island (Fig. 3, LIM et al. 2005). The site of maximum Chla, MVAK, was a very
small, shallow pond (depth = approximately 15 cm) located in the middle of a grassy moss
meadow, and had moss banks extending into the water. An orange algal crust was present on the
sediment, and the cyanobacterium Nostoc was also prevalent. The lowest Chla values came from
relatively large, ice-covered sites that also had low conductivities and low total dissolved nitrogen
(MVAE, MVAF, MVAG, MVAJ, MVY). Chla was significantly correlated with both POC and
PON, which likely suggests their co-variance as particulate matter in the water column.
Interestingly, Chla was uncorrelated with any other variable in the dataset. Typically
Chla is correlated with TP (DILLON & RIGLER 1974) but a lack of correlation between Chla and
either TN or TP has also been found in most other high Arctic lakes and ponds (MICHELUTTI et
al. 2002a, MICHELUTTI et al. 2002b, ANTONIADES et al. 2003a, ANTONIADES et al. 2003b, LIM et
al. 2005), and suggests that neither N or P are controlling planktonic algal production in these
systems. This may be explained in a number of ways. First, most of the aquatic primary
production in the Arctic is from periphyton, especially in shallow systems. As Antoniades et al.
(2003a) highlighted, our Chla values measure concentrations in the water column, and are not
indicative of periphytic production. Secondly, limitation by factors such as temperature may
affect the responses of primary producers such that Arctic systems may not respond to nutrient
enrichment in the same fashion as temperate lakes (FLANAGAN et al. 2003). Finally, recent
studies have shown that N and P concentrations in interstitial waters associated with
cyanobacterial mats in the High Arctic may be many times higher that those measured in the
pelagic zone (VILLENEUVE et al. 2001), and that these mats are not nutrient limited (BONILLA et
al. 2005). Thus, pelagic nutrient concentrations may not reflect what is functionally available to
periphyton.
Nutrient ratios
There are a number of ways to assess the nutrient status of a lake system, with TP
concentration being the most commonly used. Classification of trophic status based solely on TP
(WETZEL 1983) suggests that 22% of our sites were oligotrophic (<10 μg/L), 67% were
mesotrophic (10-30 μg/L), 6% were eutrophic (30-100 μg/L) and 4% were hypereutrophic (>100
μg/L). However, this TP-based classification may be misleading. In some high Arctic shallow
ponds, high TP values recorded in water samples taken near the shoreline (where we do our
sampling) may reflect the presence of re-suspended sediment particles from high wind and wave
action, and do not necessarily indicate eutrophic systems (e.g. ANTONIADES et al. 2003b). Thus,
other methods of assessing nutrient availability (or deficiency) might be preferable. An
examination of particulate organic carbon (POC) versus Chla, for example, could indicate general
nutrient deprivation (HECKY et al. 1993). On Melville Island, POC:Chla ratios suggest that 96%
are severely nutrient deprived, while the other 4% are moderately nutrient deprived (Table 2). Of
course, some caution must be used when interpreting these ratios, as the input of terrestrial
sources of C, especially in shallow ponds, would inflate our POC values. Indeed, when we
examine the POC:Chla (by weight) ratio, we observed that 96% of the lakes are likely receiving
most of their carbon from allochthonous sources (EPPLEY et al. 1977).
In addition to understanding general nutrient deficiency, it is useful to know whether
systems are generally N or P limited, especially as the input of anthropogenic N becomes of
increasing concern (e.g. VITOUSEK et al. 1997, WOLFE et al. 2001). Sakamoto (1966) suggested
that lakes with TN:TP (μg/L) >17:1 would be P limited, where sites with ratios of <14:1 would
more likely exhibit N-limitation (DOWNING & MCCAULEY 1992). When examined this way, 56%
of sites can be considered P deficient, 35% could be considered N deficient and the remaining 9%
might be limited by either nutrient. Examining TN:TP ratios using molar concentrations indicates
that 47% of our sites are P-deficient, 33% are N-deficient, and 20% could be limited by either
nutrient (GUILDFORD & HECKY 2000, Table 2). Regardless of how we calculate these TN:TP
ratios, they suggest that, while the majority of our sites are P limited, many are N limited.
Further examination of nutrient ratios from particulate matter can provide a better
understanding of nutrient limitation in these systems. For example, POC:PON (molar ratios)
indicate the degree of N-deficiency, where values < 8.3 indicate no N-deficiency, values between
8.3 and 14.6 indicate moderate N deficiency and values > 14.6 suggest severe N deficiency
(HECKY et al. 1993). According to these criteria, all of our sites are moderately N-deficient
except for MVM, which is classified as severely N-deficient (Table 2).
Likewise, P limitation can be examined by calculating molar ratios of PON:POP and
POC:POP (HECKY et al. 1993). We have restricted our calculation of these ratios to 37 sites, due
to suspect TPF values from 9 sites (see above). PON:POP ratios < 23 indicate no P limitation,
while ratios > 23 indicate severe P limitation (HECKY et al. 1993). According to these criteria, 12
sites are P deficient while 25 sites are not P deficient. POC:POP ratios < 133 indicate no P
limitation, while those >133 suggest moderate to severe P limitation (HECKY et al. 1993).
According to POC:POP ratios, 21 sites are P deficient while 16 are not (Table 2).
Taken together, these data suggest that, while all our sites may be considered N deficient,
not all are considered P deficient. These data support previous findings from Alaskan lakes
which suggest that N deficiency might be more common in Arctic lake systems than in temperate
systems (LEVINE & WHALEN 2001). In the Canadian high Arctic, Bathurst Island and Ellef
Ringnes Island were the only two regions studied thus far where the majority of sites are thought
to have been limited by N rather than P, based on TN:TP (by weight) (LIM et al. 2001,
ANTONIADES et al. 2003b). Furthermore, POC:Chla ratios indicate that all our sites are at least
moderately nutrient deprived. These findings corroborate classification of trophic status based on
Chla, but not TP concentrations alone, suggesting that, despite high TP values from some sites on
Melville Island, it is unlikely that any of these systems could be genuinely considered eutrophic.
Statistical analyses
PCA summarizes the main gradients of environmental variables into linear components.
Many of our parameters were highly correlated (Table 3), and thus we selected representative
variables to include in the PCA in an attempt to make the results of the PCA less complex. For
physical and chemical variables from Melville Island, the first two PCA axes explained 55.2% of
the variance in the environmental data. The first axis explained 31.5% of the variance and
represented gradients of pH, specific conductivity and related variables (Ca, Na, K, Cl, Mg, DIC),
TdN, DOC, and elevation (Fig. 4). Axis 2 explained 23.7% of the variance in the environmental
data and was mainly influenced by variables related to nutrients, including Chl a, TP, TPF, and
the metals Mn and SiO2, (Fig. 4).
The location of lakes and ponds on the PCA biplot indicates their relationship to multiple
gradients. Sites did not sort according to any pattern when split into physiographic regions (data
not shown). This is not surprising given that geology, rather than physiography, is known to be
an important factor influencing water chemistry elsewhere in the Arctic. Interestingly, when we
sorted sites based on bedrock geology (Fig. 4a), patterns were also difficult to distinguish,
possibly due to the predominance of sites located on carbonate and sandstone bedrock. When
sites are distinguished based on bioclimatic zones (Fig. 4b), only the sites located in the most
botanically diverse and abundant zone (zone 4) tended to group together in the lower left
quadrant of the ordination biplot. The sites located in bioclimatic zones 3, 2, and 1 are widely
spread throughout the PCA biplot, suggesting that, except for bioclimatic zone 4, variations
amongst these zones are not reflected in differences in water chemistry.
Bioclimatic zone 4 is characterized by shrubs such as Salix arctica and Dryas
integrifolia, several species of sedges and cotton grasses, Rosaceae, Asteraceae, Taraxacum spp.,
legumes (e.g. Oxytropis spp. and Astragalus alpinus), as well as several emergent and aquatic
species (EDLUND 1994). Together, the sites from bioclimatic zone 4 have significantly higher
pH, DOC, DIC, TdN, and TN:TP (Fig. 5, t-test, p < 0.05) than the rest of the sites combined.
When the sites from zone 4 are tested relative to those in individual zones, TdN remains
significantly higher in zone 4 than in all other zones except for zone 3. Interestingly, these sites
are not characterized by higher concentrations of other nutrients and associated variables (TP,
Chla, SRP, TPF, PON). In fact, the other zones combined had significantly higher TP values
than zone 4 (Fig. 5, t-test, p < 0.05).
The higher DIC and pH values may be explained by the location of most of these sites
near Winter Harbour on calcareous glacial till. The abundance of vegetation in this area, which
would contribute terrestrial organic C and would slow the percolation of run-off through the
catchment, is likely to have caused increased DOC in these sites. Both Antoniades et al. (2003a)
and Lim et al. (2005) suggested that the presence of vegetation near some of their sites was likely
responsible for higher DOC, TP, and TN values. On Melville Island, however, both TdN and
DOC are separated from both TP and Chla in the PCA biplot. This is puzzling, and may be
related to both the greater abundance of vegetation as well as the presence of leguminous plants
only in bioclimatic zone 4. Indeed, differences in vegetation type have been known to affect
fluxes of DOC and dissolved organic N in Arctic soils more so than climatic differences (NEFF &
HOOPER 2002). The presence of nitrogen-fixing Alnus near lakes in Alaska has been known to
increase nitrogen concentrations in lakewater due to both the terrestrial decomposition of N-
enriched plant matter as well as through input of N-rich leaves to the lake (GOLDMAN 1960,
DUGDALE & DUGDALE 1961). A similar phenomenon may be occurring in these zone 4 sites (the
only sites in which legumes are present) on Melville Island, albeit on a much smaller scale. Thus,
the higher DOC and TdN concentrations in these sites likely reflects the abundance, and possibly
the type, of vegetation in relatively lush areas.
Conclusions
The 46 ponds and lakes on Melville Island represent wide-ranging limnological
conditions. When examined as mean or median values, most of the measured environmental
variables are within the ranges previously reported for other high Arctic lakes and ponds.
Individual sites, however, recorded the lowest specific conductivities and Ca2+ concentrations yet
reported from the Canadian high Arctic. These sites occur at high elevations and are in proximity
to the small permanent ice caps on Melville Island. The influence of surficial geology was
apparent in both the locally high Ca2+ and SO42- concentrations occurring in sites located on
calcareous glacial till, and near a gypsum outcrop, respectively.
Nutrient measurements indicate some very high TP values on Melville Island. Most of
these can be explained by the influence of re-suspended sediments, and are not likely
representative of nutrient-rich conditions. Likewise, Chla concentrations suggest that most sites
are oligotrophic. Various particulate nutrient ratios from sites on Melville Island suggest that all
the ponds and lakes experience both general nutrient deficiency, as well as nitrogen deficiency.
TN:TP molar ratios suggest that the majority of sites are P deficient (47%), while 33% could be
considered N deficient.
PCA shows that 55.2 % of the variation in the environmental data can be explained by
the first two axes. Axis one explains 31.5 % of the variance and represents primarily pH,
elevation, DOC, TdN, and conductivity and related variables, while axis two explains 23.7 % of
the variance and represents a gradient of other nutrients, and trace metals. Ponds and lakes
occurring in different bioclimatic zones did not fall along any recognizable patterns of the PCA
axes, except for those occurring in the most lushly vegetated zone. This likely reflects the limited
influence of vegetation type on limnological characteristics in most high Arctic bioclimatic zones.
The sites located in bioclimatic zone 4 had significantly higher DOC and TdN than all other lakes
and ponds. This may reflect the influence of both higher vegetation cover, as well as the presence
of legumes in proximity to the ponds.
The physical and chemical characteristics of lakes and ponds from Melville Island may
have implications for both their resident biota and in the way these lakes will respond to climate
change. For example, DOC is known to act as a UV filter that may have important effects for
heterotrophic bacteria, cyanobacteria, algae and zooplankton in these systems (e.g. RAE &
VINCENT 1998; RAUTIO & KORHOLA 2002). Furthermore, as the Arctic continues to warm as a
result of climate change, the catchments of lakes and ponds will likely become increasingly
vegetated with concomitant changes in water chemistry (e.g. increased DOC and nutrient
cycling). These effects may be most pronounced in regions within the high Arctic which, at
present, are especially depauperate in terrestrial vegetation. Likewise, the reduction of seasonal
ice cover within lakes will have major implications for pH (e.g. PSENNER & SCHMIDT 1992),
conductivity (e.g. DOUGLAS & SMOL 1999), and light regimes within lakes, which will likely
have associated effects on aquatic biota.
Acknowledgements
This project was funded through Natural Sciences and Engineering Research Council (NSERC)
grants to BEK, JPS and MSVD, and a Northern Scientific Training Program grant to BEK. We
are very grateful to X. Wang and D. Muir for water chemistry analyses at the Canadian Centre for
Inland Waters, Environment Canada. The Polar Continental Shelf Project (PCSP) provided
logistical support. Many thanks to J.R. Glew, D. Antoniades, and N. Michelutti for help in the
field, as well as to D. Antoniades, N. Michelutti, and D.S.S. Lim for graciously providing their
data. Finally, thank you to D. Antoniades, A. Poulain, K. Rühland, D. Selbie and anonymous
reviewers for helpful comments on the manuscript. This PCSP/ÉPCP contribution # 032-06.
References
ACIA. (2004): Impacts of a warming Arctic. Arctic Climate Impact Assessment. –
Cambridge University Press, Cambridge, UK. 146 pp. ANTONIADES, D., DOUGLAS, M.S.V., & SMOL, J.P. (2003a): Comparative physical and
chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick Island, NWT). – Arch. Hydrobiol. 158: 485 – 516.
ANTONIADES, D., DOUGLAS, M.S.V., & SMOL, J.P. (2003b): The physical and chemical limnology of 24 ponds and one lake from Isachsen, Ellef Ringnes Island, Canadian High Arctic. – Internat. Rev. Hydrobiol. 88: 519 – 538.
ANTONIADES, D., DOUGLAS, M.S.V., & SMOL, J.P. (2005): Quantitative estimates of recent environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. – J. Paleolimnol. 33: 349 – 360.
BARNETT, D.M., EDLUND, S.A., & DREDGE, L.A. (1977): Terrain characterization and evaluation: an example from eastern Melville Island. – Geol. Surv. Can. Paper 76-23. 18 pp.
BONILLA, S., VILLENEUVE, V., & VINCENT, W.F. (2005): Benthic and planktonic algal communities in a High Arctic lake: pigment structure and contrasting responses to nutrient enrichment. – J. Phycol. 41: 1120 – 1130.
CANADIAN ICE SERVICE (2002). Sea ice climatic atlas: northern Canadian waters 1971-2000. Minister of Public Works and Government Services Canada, Ottawa. 266 pp.
CHRISTIE, R.L. (1994): Introduction. – In: CHRISTIE, R.L., & MCMILLAN, N.J. (eds.): The Geology of Melville Island, Arctic Canada. – Geol. Surv. Can. Bull. 450: pp. 7 – 11.
DILLON, P.J., & RIGLER, F.H. (1974): Phosphorus-chlorophyll relationships in lakes. – Limnol. Oceanogr. 19: 767 – 773.
DOUGLAS, M.S.V., & SMOL, J.P. (1994): Limnology of high Arctic ponds (Cape Herschel, Ellesmere Island, N.W.T.). – Arch. Hydrobiol. 131: 410 – 434.
DOUGLAS, M.S.V., SMOL, J.P., & BLAKE JR., W. (1994): Marked post-18th century environmental change in high-Arctic ecosystems. – Science. 266: 416 – 419.
DOWNING, J.A., & MCCAULEY, E. (1992): The nitrogen-phosphorus relationship in lakes. – Limnol. Oceanogr. 37: 936 – 945.
DUGDALE, R.C., & DUGDALE, V.A. (1961): Sources of phosphorus and nitrogen for lakes on Afognak Island. – Limnol. Oceanogr. 6: 13 – 23.
EDLUND, S.A. (1994): The distribution of plant communities on Melville Island, Arctic Canada. – In: CHRISTIE, R.L., & MCMILLAN, N.J. (eds.): The Geology of Melville Island, Arctic Canada. – Geol. Surv. Can. Bull. 450: pp. 247 – 255.
ENVIRONMENT CANADA (1994): Manual of Analytical Methods – National Laboratory for Environmental Testing, Canadian Centre for Inland Waters. Burlington, Ontario, 2 vols., 1097 pp.
EPPLEY, R.W., HARRISON, W.G., CHISHOLM, S.W., & STEWART, E. (1977): Particulate organic matter in surface waters off Southern California and its relationship to phytoplankton. – J. Mar. Res. 35: 671 – 696.
FLANAGAN, K.M., MCCAULEY, E., WRONA, F., & PROWSE, T. (2003): Climate change: the potential for latitudinal effects on algal biomass in aquatic ecosystems. – Can. J. Fish. Aquat. Sci. 60: 635 – 639.
GOLDMAN, C.R. (1960): Primary productivity and limiting factors in 3 lakes of the Alaska Peninsula. – Ecol. Monogr. 30: 193 – 223.
GREGORY-EAVES, I., SMOL, J.P., FINNEY, B.P., LEAN, D.R.S., & EDWARDS, M.E. (2000): Characteristics and variation in lakes along a north-south transect in Alaska. –
Archiv. Hydrobiol. 147: 193 – 223. GUILDFORD, S.J., & HECKY, R.E. (2000): Total nitrogen, total phosphorus, and nutrient
limitation in lakes and oceans: is there a common relationship? – Limnol. Oceanogr. 45: 1213 – 1223.
HAMILTON, P.B., GAJEWSKI, K., MCNEELY, R.N., & LEAN, D.R.S. (2000): Physical, chemical, and biological characteristics of lakes from the Slidre River basin on Fosheim Peninsula, Ellesmere Island, Nunavut. – In: Garneau, M., & Alt, B.T. (eds.): Environmental response to climate change in the Canadian High Arctic. – Geol. Surv. Can. Bull. 539: 235 – 248.
HAMILTON, P.B., GAJEWSKI, K., ATKINSON, D.E., & LEAN, D.R.S. (2001): Physical and chemical limnology of 204 lakes from the Canadian Arctic Archipelago. – Hydrobiologia. 457: 133 – 148.
HARRISON, J.C. (1994): Melville Island and adjacent smaller islands, Canadian Arctic Archipelago, District of Franklin, Northwest Territories. Map 1844A, Scale 1:250 000. Geological Survey of Canada: Ottawa.
HECKY, R.E., CAMPBELL, P., & HENDZEL, L.L. (1993): The stoichiometry of carbon, nitrogen, and phosphorus in particulate matter of lakes and oceans. – Limnol. Oceanogr. 38: 709 – 724.
HINZMAN, L.D., BETTEZ, N.D., BOLTON, W.R., CHAPIN, F.S., DYURGEROV, M.B., FASTIE, C.L., GRIFFITH, B., HOLLISTER, R.D., HOPE, A., HUNTINGTON, H.P., JENSEN, A.M., JIA, G.J., JORGENSON, T., KANE, D.L., KLEIN, D.R., KOFINAS, G., LYNCH, A.H., LLOYD, A.H., MCGUIRE, A.D., NELSON, F.E., OECHEL, W.C., OSTERKAMP, T.E., RACINE, C.H., ROMANOVSKY, V.E., STONE, R.S., STOW, D.A., STURM, M., TWEEDIE, C.E., VOURLITIS, G.L., WALKER, M.D., WALKER, D.A., WEBBER, P.J., WELKER, J.M., WINKER, K., & YOSHIKAWA, K. (2005): Evidence and implications of recent climate change in northern Alaska and other Arctic regions. – Clim. Change. 72: 251 – 298.
HODGSON, D.A. (1989): Surficial materials (Queen Elizabeth Islands). – In: FULTON, R.J. (ed.): Quaternary Geology of Canada and Greenland. – Geol. Surv. Can. pp. 448-452.
JUGGINS, S., & TER BRAAK, C.J.F. (1992): CALIBRATE – a program for species- environment calibration by [weighted averaging] partial least squares regression. – University College, London.
KORHOLA, A., TIKKANEN, M., & WECKSTROM, J. (2005): Quantification of Holocene lake-level changes in Finnish Lapland using a cladocera-lake depth transfer model. – J. Palelimnol. 34: 175 – 190.
LEVINE, M.A., & WHALEN, S.C. (2001): Nutrient limitation of phytoplankton production in Alaskan Arctic foothill lakes. – Hydrobiologia. 455: 189 – 201.
LIM, D.S.S., & DOUGLAS, M.S.V. (2003): Limnological characteristics of 22 lakes and ponds in the Haughton Crater region of Devon Island, Nunavut, Canadian High Arctic. – Arct. Antarct. Alp. Res. 35: 509 – 519.
LIM, D.S.S., DOUGLAS, M.S.V., & SMOL, J.P. (2005): Limnology of 46 lakes and ponds on Banks Island, NWT, Canadian Arctic Archipelago. – Hydrobiologia. 545: 11 – 32.
LIM, D.S.S., DOUGLAS, M.S.V., SMOL, J.P., & LEAN, D.R.S. (2001): Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. – Internat. Rev. Hydrobiol. 86: 1 – 22.
MAGNUSON, J.J., ROBERTSON, D.M., BENSON, B.J., WYNNE, R.H., LIVINGSTON, D.M., ARAI, T., ASSEL, R.A., BARRY, R.G., CARD, V., KUUSISTO, E., GRANIN, N.G., PROWSE, T.D., STEWART, K.M., & VUGLINSKI, V.S. (2000): Historical trends in lake and river ice cover in the Northern Hemisphere. – Science. 289: 1743 – 1746.
MAJOROWICZ, J.A., SKINNER, W.R., & SAFANDA, J. (2005): Ground surface warming
history in northern Canada inferred from inversions of temperature logs and comparison with other proxy climate reconstructions. – Pure and Applied Geophysics. 162: 109 – 128.
MCNEELY, R.N., NEIMANIS, V.P., & DWYER, L. (1979): Water Quality Sourcebook: a guide to water quality parameters. Minister of Supply and Services Canada. Ottawa, Ontario, 88 pp.
METEOROLOGICAL SERVICE OF CANADA. (2002): Canadian Climate Normals 1971 – 2000. Environment Canada, Ottawa.
MICHELUTTI, N., DOUGLAS, M. S. V., LEAN, D. R. S. & SMOL, J. P. (2002a): Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Island, Arctic Canada. – Hydrobiologia 482: 1–13.
MICHELUTTI, N., DOUGLAS, M. S. V., MUIR, D.C.G., WANG, X., & SMOL, J. P. (2002b). Limnological characteristics of 38 lakes and ponds on Axel Heiberg Island, High Arctic Canada. – Internat. Rev. Hydrobiol. 87: 385 –399.
MICHELUTTI, N., WOLFE, A.P., VINEBROOKE, R.D., RIVARD, B., & BRINER, J.P. (2005): Recent primary production increases in Arctic lakes. – Geophysical Research Letters. 32: L19715 – doi: 10.1029/2005GL023693.
NEFF, J.C., & HOOPER, D.U. (2002): Vegetation and climate controls on potential CO2, DOC and DON production in northern latitude soils. – Global Change Biology. 8: 872 – 884.
OVERPECK, J. J., HUGHEN, K., HARDY, D., BRADLEY, R., CASE, R., DOUGLAS, M., FINNEY, B., GAJEWSKI, K., JACOBY, G., JENNINGS, A., LAMOUREUX, S., LASCA, A., MACDONALD, G., MOORE, J., RETELLE, M., SMITH, S., WOLFE, A., & ZIELINSKI, G. (1997): Arctic environmental change of the last four centuries. – Science. 278: 1251 – 1256.
PSENNER, R., & R. SCHMIDT. (1992): Climate-driven pH control of remote alpine lakes and effects of acid deposition. – Nature. 356: 781-783.
RAE, R., & W. F. VINCENT. (1998): Effects of temperature and ultraviolet radiation on microbial foodweb structure: potential responses to global change. – Fresh. Biol. 40: 747-758.
RAUTIO, M. & KORHOLA, A. (2002): Effects of ultraviolet radiation and dissolved organic carbon on the survival of subarctic zooplankton. – Polar Biology. 25: 460-468.
ROUSE, W. R., DOUGLAS, M. S. V., HECKY, R. E., HERSHEY, A. E., KLING, G. W., LESACK. L., MARSH, P., MCDONALD, M., NICHOLSON, B. J., ROULET, N. T. & SMOL, J. P. (1997): Effects of climate change on the waters of Arctic and SubArctic North America. – Hydrological Processes 11: 973 –902.
RÜHLAND, K., & SMOL, J.P. (1998): Limnological characteristics of 70 lakes spanning Arctic treeline from Coronation Gulf to Great Slave Lake in the Central Northwest Territories, Canada. – Internat. Rev. Hydrobiol. 83: 183 – 203.
SAKAMOTO, M. (1966): Primary productivity by phytoplankton communities in some Japanese lakes and its dependence on lake depth. – Arch. Hydrobiol. 62: 1 - 28.
SMOL, J.P., WOLFE, A.P., BIRKS, H.J.B., DOUGLAS, M.S.V., JONES, V.J., KORHOLA, A., PIENITZ, R., RUHLAND, K., SORVARI, S., ANTONIADES, D., BROOKS, S.J., FALLU, M.A., HUGHES, M., KEATLEY, B.E., LAING, T.E., MICHELUTTI, N., NAZAROVA, L., NYMAN, M., PATERSON, A.M., PERREN, B., QUINLAN, R., RAUTIO, M., SAULNIER-TALBOT, E., SIITONENI, S., SOLOVIEVA, N., & WECKSTROM, J. (2005): Climate-driven regime shifts in the biological communities of Arctic lakes. - Proc. Natl. Acad. Sci. U.S.A. 102: 4397 - 4402.
SOLOVIEVA, N., JONES, V.J., NAZAROVA, L., BROOKS, S.J., BIRKS, H.J.B., GRYTNES, J.A., APPLEBY, P.G., KAUPPILA, T., KONDRATENOK, B., RENBERG, I., & PONOMAREV, V.
(2005): Palaeolimnological evidence for recent climatic change in lakes from the northern Urals, Arctic Russia. – J. Paleolimnol. 33: 463 - 482.
STEWART, D.B. & BERNIER, L.M.J. (1982): An Aquatic Resource Survey of Islands Bordering Viscount Melville Sound. Land Use Information Series - Environment Canada Lands Directorate, Background Report. No. 2: 110 pp.
TER BRAAK, C. J. F. & ŠMILAUER, P. (1998): CANOCO and CANOPost, Version 4 for Windows. – Microcomputer Power: Ithaca, N.Y., USA.
TOZER, E.T. & THORSTEINSSON, R. (1964): Western Queen Elizabeth Islands, Arctic Archipelago – Geol. Surv. Can. Memoir 332: 242 pp.
VILLENEUVE, V., VINCENT, W.F., & KOMÁREK, J. (2001): Community structure and microhabitat characteristics of cyanobacterial mats in an extreme high Arctic environment: Ward Hunt Lake. - Nova Hedwigia 123: 199 - 224.
VITOUSEK, P.M., MOONEY, H.A., LUBCHENCO, J., & MELILLO, J.M. (1997): Human domination of Earth's ecosystems. - Science 277: 494 - 499.
WETZEL, R.G. (1983): Limnology. 2nd ed. – Saunders Publishing, Philadelphia, PA, 767 pp.
WOLFE, A.P., BARON, J.S., & CORNETT, R.J. (2001): Anthropogenic nitrogen deposition induces rapid ecological changes in alpine lakes of the Colorado Front Range (USA). – J. Paleolimnol. 25: 1 - 7.
Figure Captions Figure 1. Regional map of the Canadian Arctic Archipelago with the locations of both Melville Island and previous modern limnological studies to which references are made in the text. The inset map A) indicates the location of the Canadian High Arctic within Canada, and the circle on the main map B) indicates Melville Island, the focus of our study. Figure 2. Map of Melville Island, with sites differentiated according to previously defined a) bioclimatic zones (after EDLUND 1994), and b) bedrock geology (after HARRISON 1994). On both maps, numbers indicate the following geographical features: 1) Liddon Gulf, 2) Murray Inlet, 3) Purchase Bay, 4) Leopold Glacier, 5) unnamed ice caps, 6) Bridport Inlet, 7) Sabine Bay, 8) Hecla and Griper Bay. 2a) Bioclimatic zone 1 is the most sparsely vegetated region, while zone 4 has the greatest number and abundance of terrestrial vegetation, including woody shrubs (EDLUND 1994). Figure 3. Box plots showing variability of selected environmental variables from islands of the Canadian Arctic Archipelago. Solid lines indicate median values, dashed lines indicate mean values, whiskers represent 10th and 90th percentiles, and dots are 5th and 95th percentiles. Data sources are as follows: Prince Patrick Island (ANTONIADES et al. 2003a), Banks Island (LIM et al. 2005), Victoria Island (MICHELUTTI et al. 2002a), Bathurst Island (LIM et al. 2001), Devon Island (LIM & DOUGLAS 2003), Ellef Ringnes Island (ANTONIADES et al. 2003b), Axel Heiberg Island (MICHELUTTI et al. 2002b), Alert (ANTONIADES et al. 2003a). Figure 4. Principle components analysis (PCA) biplot of measured environmental variables (arrows) and sampling sites (symbols). The light lines represent variables that were run passively in the PCA. Sampling sites are differentiated into: 4a) dominant type of bedrock geology (after HARRISON 1994), or 4b) bioclimatic zones (after EDLUND 1994). See caption of Figure 2a and 2b for symbol legends. Figure 5. Box plots of selected variables showing sites from bioclimatic zone 4, the mostly lushly vegetated zone on Melville Island (n=16) versus all other bioclimatic zones combined (n=30). Solid lines indicate median values, dashed lines indicate mean values, whiskers represent 10th and 90th percentiles, and dots are 5th and 95th percentiles.
Figure 1.
Figure 2.
Melville
Prince
Patr
ick
Banks
Victori
a
Bathurs
t
Devon
Ellef R
ingne
s
Axel H
eibergAler
t
TP u
nfilt
ered
(μg/
L)
0102030405060708090
120140
Tota
l nitr
ogen
(mg/
L)
0.00
0.25
0.50
0.75
1.00
1.25
2.00
Melville
Prince
Patrick
Banks
Victori
a
Bathurs
t
Devon
Ellef R
ingne
s
Axel H
eibergAler
t
Ca
(mg/
L)
0
25
50
75
100
125200
DIC
(mg/
L)0
10
20
30
40
50
Chl
a(μ
g/L)
0
1
2
3
4
5
6
DO
C (m
g/L)
02468
10121416
pH
4.55.05.56.06.57.07.58.08.59.0
Spe
c. c
ond.
(μS
/cm
)
0150300450600750900
105015002000
Figure 3.
-1.0 1.0
Figure 4.
-1.0
1.0
SO4
SiO2
CHLa
POC
DOC
Mn
TdN
TP
TPF
pH
COND
TEMPLONG
LAT
ELEVAB
C
DE
F
G
H
I
JKL
M NO
P
QRS
T U
V
W
XY Z
AA
AB
AC
AD
AE
AF
AG
AH
AIAJ
AKAL
AM
AN
AO
AP
AQ
AR
AS
AT
Axis 1λ = 0.315
Axi
s 2 λ
= 0.
237
-1.0 1.0
-1.0
1.0
SO4
SiO2
CHLa
POC
DOC
Mn
TdN
TP
TPF
pH
COND
TEMPLONG
LAT
ELEVAB
C
DE
F
G
H
I
JKL
M NO
P
QRS
T U
V
W
XY Z
AA
AB
AC
AD
AE
AF
AG
AH
AIAJ
AKAL
AM
AN
AO
AP
AQ
AR
AS
AT
Axis 1λ = 0.315
Axis
2λ
= 0.
237
a)
b)
-1.0 1.0
-1.0
1.0
SO4
SiO2
CHLa
POC
DOC
Mn
TdN
TP
TPF
pH
COND
TEMPLONG
LAT
ELEVAB
C
DE
F
G
H
I
JKL
M NO
P
QRS
T U
V
W
XY Z
AA
AB
AC
AD
AE
AF
AG
AH
AIAJ
AKAL
AM
AN
AO
AP
AQ
AR
AS
AT
Axis 1λ = 0.315
Axi
s 2 λ
= 0.
237
-1.0 1.0
-1.0
1.0
SO4
SiO2
CHLa
POC
DOC
Mn
TdN
TP
TPF
pH
COND
TEMPLONG
LAT
ELEVAB
C
DE
F
G
H
I
JKL
M NO
P
QRS
T U
V
W
XY Z
AA
AB
AC
AD
AE
AF
AG
AH
AIAJ
AKAL
AM
AN
AO
AP
AQ
AR
AS
AT
Axis 1λ = 0.315
Axis
2λ
= 0.
237
a)
b)
Chl
a(μ
g/L)
0
1
2
3
48
TN:T
P ra
tio
20406080100120140160180
zone 4 others
TdN
(mg/
L)
0.00
0.25
0.50
0.75
1.00
zone 4 others
TP u
nfilt
ered
(μg/
L)
0
10
20
30
40
50
60120140
pH
7.007.257.507.758.008.258.508.759.00
Spe
c. c
ond.
(μS
/cm
)
010020030040050060070010001200
DIC
(mg/
L)
0
10
20
30
40
50
DO
C (m
g/L)
024681012141618
Figure 5.
Table 1. The chemical and physical parameters for 46 freshwater sites on Melville Island. The values in italics represent variables that were below the method detection limit and were replaced with half the detection limit. Bold values with asterisks indicate missing values that were replaced using a linear regression equation with a highly correlated variable. No variables were highly correlated to SRP, thus missing/suspicious values were replaced with the mean SRP value across all sites.
Sample Pond or Latitude Longitude ELEV SA pH CONDID Lake m asl ha μS/cm
MVA P 74° 45.434'N 110° 37.057'W 5.0 0.478 8.7 340MVB P 74° 45.270'N 110° 35.772'W 5.0 0.059 8.8 800MVC P 74° 45.447'N 110° 38.237'W 5.0 0.785 8.7 231MVD P 74° 45.860'N 110° 38.528'W 10.0 0.024 8.6 187MVE P 74° 48.270'N 110° 36.784'W 15.0 0.039 8.5 149MVF P 74° 48.207'N 110° 36.773'W 15.0 0.020 8.5 161MVG P 74° 47.478'N 110° 44.990'W 70.0 2.356 8.5 109MVH P 74° 45.688'N 110° 45.846'W 65.0 7.854 8.4 101MVI P 74° 45.983'N 110° 44.217'W 65.0 1.178 8.4 120MVJ P 74° 48.323'N 110° 48.732'W 65.0 1.257 7.8 18MVK P 74° 48.728'N 110° 47.416'W 70.0 2.945 8.1 88MVL P 74° 45.936'N 110° 39.913'W 5.0 0.031 8.5 318MVM P 74° 45.865'N 110° 39.793'W 15.0 0.126 8.5 292MVN P? 74° 39.367'N 111° 04.364'W 40.0 4.909 8.3 150MVO P? 74° 27.887'N 112° 15.15'W 121.9 5.498 8.0 23MVP P 74° 28.04'N 112° 18.69'W 152.4 0.079 8.0 18MVQ P? 74° 25.138'N 112° 51.359'W 274.3 7.069 8.1 70MVR L? 74° 33.689'N 113° 04.570'W 304.8 19.635 8.0 39MVS P? 74° 47.301'N 113° 12.205'W 243.8 3.142 8.1 48MVT P 74° 53.809'N 110° 46.241'W 76.2 0.332 8.2 79MVU P 74° 44.619'N 110° 44.995'W 70.0 31.416 8.1 112MVV P? 74° 37.999'N 111° 19.198'W 80.0 78.540 8.0 90MVW P? 75° 09.427'N 111° 55.432'W 213.4 7.069 7.7 21MVX P 75° 24.080'N 111° 46.290'W 228.6 0.079 7.9 56MVY P 75° 38.291'N 111° 31.048'W 109.7 0.008 7.7 13MVZ P 76° 01.188'N 112° 22.334'W 173.7 0.785 7.3 10MVAA P? 76° 12.280'N 114° 00.330'W 85.3 4.909 8.0 32MVAB P 76° 19.755'N 114° 09.392'W 70.1 19.635 7.9 1040MVAC P 76° 23.812'N 114° 04.243'W 82.3 0.283 8.2 51MVAD P? 76° 12.910'N 113° 25.307'W 91.4 0.385 8.1 43MVAE L 75° 50.410'N 114° 41.532'W 655.3 19.635 8.1 5MVAF L? 75° 29.174'N 114° 24.009'W 442.0 9.425 6.9 5MVAG P? 75° 29.749'N 113° 33.54'W 579.1 12.566 7.5 16MVAH P 74° 57.907'N 109° 10.562'W 3.0 1.021 8.1 245MVAI P 74° 57.491'N 108° 44.407'W 15.2 3.534 8.0 1230MVAJ P? 75° 27.120'N 110° 19.200'W 121.9 11.781 7.8 7MVAK P 75° 56.722'N 109° 19.519'W 61.0 0.159 7.9 28MVAL L 76° 12.605'N 109° 33.637'W 15.2 80.994 7.5 83MVAM P 76° 38.657'N 108° 55.793'W 106.7 0.008 8.4 520MVAN P? 76° 32.363'N 108° 43.071'W 61.0 12.566 7.8 39MVAO L 76° 23.840'N 108° 33.277'W 30.5 58.905 7.7 77MVAP L 75° 40.114'N 107° 00.473'W 274.3 0.785 7.1 5MVAQ P? 75° 53.673'N 106° 03.333'W 274.3 2.688 7.7 5MVAR P 75° 16.517'N 106° 19.807'W 182.9 1.767 8.0 25MVAS P 75° 08.256'N 107° 37.298'W 76.2 0.238 8.3 77MVAT P 75° 19.020'N 111° 25.184'W 120.0 0.785 8.1 39MEAN 127.1 9.082 7.8 157MEDIAN 76.2 1.512 8.1 74HIGH 655.3 80.994 8.8 1230LOW 5.0 0.008 6.8 5
Table 1. Continued. Sample TEMP TP TPF TdN-N TKN-N NH3-N NO2-N
ID ° Celsius μg/L μg/L mg/L mg/L mg/L mg/LMVA 8.0 10.1 7.7 0.598 0.66 0.015 0.001MVB 8.0 14.8 6.9 0.703 0.734 0.013 0.001MVC 7.0 9.9 3.8 0.587 0.604 0.01 0.001MVD 9.0 9.8 3.9 0.672 0.615 0.007 0.001MVE 7.5 12.1 32.2 0.769 0.702 0.013 0.001MVF 7.0 14.4 24.9 0.892 0.938 0.008 0.001MVG 8.5 13.4 13.1 0.309 0.311 0.007 0.001MVH 9.5 16.8 16.9 0.184 0.177 0.007 0.001MVI 9.5 8 6.1 0.21 0.202 0.009 0.001MVJ 5.0 14.7 4.6 0.201 0.179 0.007 0.001MVK 8.0 22.1 6.1 0.29 0.277 0.0025 0.001MVL 11.0 20.6 5.2 0.977 1.18 0.013 0.002MVM 11.5 12.4 5.7 0.966 1.14 0.009 0.001MVN 11.5 14.4 5.1 0.179 0.175 0.006 0.001MVO 6.5 9.1 4.4 0.131 0.113 0.011 0.001MVP 15.5 20.1 14.9 0.309 0.316 0.008 0.003MVQ 11.0 6.67* 4 0.134 0.108 0.008 0.001MVR 6.0 12.3 4.5 0.087 0.063 0.005 0.001MVS 15.0 8.6 7.1 0.245 0.221 0.009 0.001MVT 14.0 12.2 13.9 0.333 0.321 0.008 0.001MVU 11.0 11.8 11.6 0.239 0.218 0.013 0.002MVV 6.0 6.2 4.9 0.067 0.049 0.006 0.001MVW 10.0 24.4 9.4 0.146 0.096 0.011 0.002MVX 14.0 15.4 2 0.81 0.838 0.024 0.002MVY 15.5 11.6 9 0.351 0.347 0.006 0.006MVZ 13.0 8.4 6.1 0.062 0.032 0.01 0.002MVAA 13.0 45.4 22.7 0.274 0.26 0.01 0.001MVAB 9.0 23.5 9.7 0.213 0.186 0.011 0.001MVAC 13.5 13 10 0.611 0.644 0.03 0.001MVAD 14.5 13 10.7 0.403 0.394 0.015 0.001MVAE 4.0 12.4 8.9 0.046 0.007 0.008 0.001MVAF 3.0 16.8 15.3 0.03 0.007 0.006 0.0005MVAG 11.0 9.5 13.7 0.056 0.032 0.006 0.001MVAH 9.0 27.3 10 0.393 0.399 0.013 0.001MVAI 9.0 12.9 9.4 0.327 0.331 0.016 0.001MVAJ 7.5 14.2 8.3 0.04 0.02 0.009 0.002MVAK 6.5 27.4 25.6 0.601 0.59 0.012 0.002MVAL 2.5 73.1 22.4 0.125 0.106 0.019 0.002MVAM 5.5 8.8 10 0.406 0.412 0.015 0.001MVAN 7.5 20.9 10 0.119 0.093 0.016 0.001MVAO 5.5 123 7.3 0.181 0.087 0.013 0.003MVAP 2.5 10 5.5 0.032 0.007 0.009 0.001MVAQ 3.5 34.5 6.9 0.061 0.04 0.013 0.001MVAR 5.0 135 8.8 0.286 0.275 0.012 0.001MVAS 7.5 13.4 3.9 0.236 0.227 0.024 0.001MVAT 6.0 12.7 7.8 0.641 0.648 0.011 0.001MEAN 8.8 21.5 10.0 0.338 0.334 0.011 0.001MEDIAN 8.3 13.2 8.6 0.260 0.244 0.010 0.001HIGH 15.5 135.0 32.2 0.977 1.180 0.030 0.006LOW 2.5 6.2 2.0 0.030 0.007 <0.005 <0.001
Table 1. Continued. Sample CHLa POC PON DOC DIC Cl SO4
ID ug/L mg/L mg/L mg/L mg/L mg/L mg/LMVA 0.5 0.414 0.041 7.6 29.9 84.5 1.9MVB 0.5 0.443 0.045 9 41.5 271 21.8MVC 0.5 0.38 0.034 8 34.4 26.5 7.3MVD 0.5 0.346 0.03 8.8 23.7 29.7 1.8MVE 1 0.436 0.036 10 26.3 11.6 1MVF 0.05 0.404 0.033 14.1 27.9 14.6 1.3MVG 0.6 0.674 0.066 5.6 23 2.8 0.9MVH 0.9 0.437 0.054 4.7 16.1 6.78 1.5MVI 1 0.218 0.02 3.1 20.6 6.53 1.3MVJ 1.3 0.692 0.082 3.8 1.4 2.6 1MVK 0.8 0.473 0.055 4 14.3 4.59 1.3MVL 1.4 0.44 0.037 16.3 47.8 11.7 2.5MVM 0.9 0.524 0.04 16.3 45.1 11.5 1.7MVN 0.9 0.351 0.038 3.8 24.7 12.1 1.8MVO 1.3 0.315 0.034 2.7 2.4 2.24 2.2MVP 5.6 2.3 0.21 6.6 2.5 1.37 0.9MVQ 0.5 0.278 0.028 2.5 10 1.18 5.9MVR 0.4 0.254 0.021 1.7 6.2 0.82 2.1MVS 0.5 0.338 0.035 3 7.2 2.12 0.7MVT 0.5 0.299 0.032 4.6 15.4 2.2 0.2MVU 1.3 0.393 0.043 3.2 17.4 6.8 1.3MVV 1.9 0.162 0.018 1.7 13.4 3.46 2.2MVW 0.9 0.412 0.043 2.8 2.3 1.14 3.1MVX 0.5 0.372 0.043 9.4 9.5 2.2 0.6MVY 0.3 0.448 0.05 5.9 2.7 2.25 0.8MVZ 0.9 0.2 0.02 2 0.7 0.62 1.8MVAA 1.4 0.414 0.051 4.9 3.8 3.29 3MVAB 3.7 0.677 0.083 6.7 10.9 412 45.7MVAC 1.5 0.484 0.051 7.2 5.3 7.51 0.7MVAD 1 0.653 0.054 5.2 2 8.62 1.5MVAE 0.3 0.178 0.011 0.8 0.7 0.4* 28.3MVAF 0.05 0.336 0.033 1.1 0.6 0.47 0.4MVAG 0.05 0.238 0.022 1.5 2.3 0.46 2.3MVAH 1.3 1.1 0.122 5.6 10.9 77 17.3MVAI 0.5 0.448 0.06 3.3 13.5 588 47.6MVAJ 0.3 0.205 0.021 1.2 0.2 1.49 0.5MVAK 10.1 2.71 0.296 12.3 2.4 4.45 0.8MVAL 0.4 0.379 0.039 5 2.1 20.2 18.7MVAM 0.9 0.434 0.043 7.2 19.4 3.8 357MVAN 0.9 0.268 0.025 3 1.6 5.28 12.6MVAO 7.9 0.503 0.041 7.2 1.8 19 18.6MVAP 0.05* 0.198 0.017 1.3 0.3 0.64 0.7MVAQ 3.6 0.507 0.062 1.1 0.4 0.77 0.9MVAR 1.6 0.526 0.06 3.9 2.5 1.28 4.1MVAS 1.1 0.32 0.03 3 13.1 3.52 1.7MVAT 1.6 0.559 0.053 8.2 4.8 1.93 4.8MEAN 1 0.503 0.051 5.5 12.3 36.59 13.8MEDIAN 1 0.413 0.041 4.7 8.4 3.66 1.8HIGH 10.1 2.710 0.296 16.3 47.8 588.00 357.0LOW <0.1 0.162 0.011 0.8 0.2 0.40 0.2
Table 1. Continued. Sample Ca Mg Na K SiO2 SRP Al
ID mg/L mg/L mg/L mg/L mg/L mg/L mg/LMVA 28.8 20.3 43.7 2.1 0.26 0.0006 0.02MVB 45.9 42.2 140 7.2 0.2 0.0011 0.03MVC 29.1 21.2 14.4 2.8 0.07 0.001 0.01MVD 27 12.2 13.2 1.1 0.27 0.0007 0.01MVE 25.2 11.6 4.8 0.8 0.14 0.0085 0.01MVF 26.6 14.9 6.1 1.3 0.05 0.0015 0.01MVG 21 10.3 1.3 1.1 0.1 0.001 0.25MVH 13.8 8.9 2.5 1.8 0.11 0.0091 2.01MVI 19.2 9.9 2.4 1.2 0.11 0.0009 0.04MVJ 0.8 0.7 1.1 0.4 0.1 0.0001 0.37MVK 12.3 7.7 1.6 0.9 0.24 0.0006 0.23MVL 40.5 28.5 5.7 2.4 0.32 0.0008 0.03MVM 38.7 27 5.3 2 0.23 0.0006 0.02MVN 20.8 15 5.6 2.1 0.06 0.0004 0.92MVO 1.8 1.6 1.6 0.5 0.12 0.0001 0.04MVP 1.5 1.1 0.7 0.1 0.2 0.0007 0.03MVQ 10.7 4.8 0.6 0.3 0.12 0.0016 0.05MVR 5.8 3 0.4 0.3 0.23 0.0001 0.06MVS 5.6 3.6 0.9 0.7 0.19 0.0002 0.03MVT 12.5 7.3 1.2 0.9 0.33 0.0003 0.1MVU 16.1 8 2.6 1.1 0.12 0.0001 0.37MVV 12.4 5.7 2 0.6 0.07 0.0007 0.01MVW 2.6 1.2 0.6 0.8 0.55 0.0008 1.75MVX 8.6 4 1 0.9 0.21 0.0006 0.01MVY 3.3 0.8 1 0.4 0.37 0.0006 0.1MVZ 0.8 0.3 0.2 0.2 0.4 0.0008 0.03MVAA 4.4 1.3 1.6 0.9 0.11 0.0063 0.08MVAB 22.1 30.3 219 8.1 0.22 0.0013 0.27MVAC 6.3 1.7 4.7 0.3 0.13 0.0028 0.005MVAD 2.2 1.2 3.6 0.8 0.25 0.0017 0.01MVAE 0.1 0.1 0.1 0.1 0.06 0.0011 0.08MVAF 0.1 0.2 0.1 0.1 0.16 0.0003 0.21MVAG 0.9 1.5 0.3 0.7 0.08 0.0013 0.42MVAH 13.3 13.5 44.4 3.2 0.09 0.0028 2.43MVAI 31.4 48.1 297 14.1 0.02 0.0049 0.07MVAJ 0.5 0.5 2.1 0.2 0.04 0.0135 0.19MVAK 1.8 0.7 1.8 0.1 1.48 0.0079 0.005MVAL 2.2 2.1 14.7 2 3.31 0.009 1.96MVAM 179 6.8 4.2 1.2 2.92 0.0113 0.005MVAN 2.9 1.8 4 0.9 0.18 0.0028 0.09MVAO 3.8 2.7 13.2 1.7 3.23 0.0075 2.29MVAP 0.2 0.2 0.3 0.2 0.21 0.0028 0.42MVAQ 0.3 0.3 0.3 0.1 0.2 0.006 0.47MVAR 1.8 1.7 1.6 0.5 0.59 0.004 0.13MVAS 15.1 3.9 1.7 1.1 0.06 0.0028 0.03MVAT 5.2 2.7 1.5 0.4 0.27 0.0057 0.02MEAN 15.8 8.5 19.1 1.5 0.41 0.0028 0.34MEDIAN 7.5 3.8 1.9 0.9 0.20 0.001 0.07HIGH 179.0 48.1 297.0 14.1 3.31 0.014 2.43LOW 0.1 0.1 <0.2 <0.2 0.02 <0.0002 <0.01
Table 1. Continued. Sample Ba Cu Fe Li Mn Sr Zn
ID mg/L mg/L mg/L mg/L mg/L mg/L mg/LMVA 0.0109 0.001 0.121 0.002 0.0135 0.0719 0.0005MVB 0.0283 0.002 0.14 0.008 0.0062 0.173 0.0005MVC 0.0121 0.002 0.041 0.002 0.004 0.0457 0.0005MVD 0.0055 0.002 0.075 0.002 0.0057 0.0368 0.0005MVE 0.007 0.002 0.111 0.001 0.0034 0.0221 0.0005MVF 0.0043 0.002 0.041 0.001 0.001 0.0203 0.0005MVG 0.0056 0.002 0.331 0.001 0.0125 0.0162 0.001MVH 0.0117 0.003 1.65 0.004 0.0192 0.0202 0.004MVI 0.0049 0.001 0.053 0.001 0.0018 0.0208 0.0005MVJ 0.0039 0.002 0.852 0.001 0.0129 0.0028 0.002MVK 0.0047 0.002 0.438 0.001 0.018 0.0126 0.002MVL 0.0108 0.002 0.124 0.001 0.0033 0.0246 0.0005MVM 0.0089 0.002 0.104 0.001 0.0027 0.0293 0.0005MVN 0.009 0.001 0.779 0.002 0.0084 0.0317 0.002MVO 0.0008 0.002 0.065 0.001 0.006 0.0063 0.0005MVP 0.00025 0.001 0.826 0.0005 0.0061 0.0031 0.001MVQ 0.0063 0.001 0.149 0.0005 0.0055 0.0208 0.0005MVR 0.0028 0.001 0.086 0.001 0.0029 0.0094 0.0005MVS 0.0015 0.001 0.083 0.0005 0.0098 0.0055 0.0005MVT 0.0031 0.002 0.152 0.001 0.0041 0.0132 0.001MVU 0.0057 0.002 0.372 0.001 0.0054 0.0209 0.003MVV 0.0028 0.001 0.021 0.001 0.0024 0.015 0.0005MVW 0.0053 0.001 0.99 0.001 0.0124 0.007 0.003MVX 0.0028 0.001 0.185 0.002 0.0109 0.0124 0.001MVY 0.0013 0.001 0.213 0.0005 0.0052 0.0098 0.001MVZ 0.0033 0.0005 0.05 0.001 0.0015 0.0043 0.001MVAA 0.0146 0.0005 0.247 0.003 0.0017 0.024 0.001MVAB 0.0373 0.001 0.327 0.011 0.0045 0.2 0.001MVAC 0.0086 0.0005 0.145 0.001 0.0027 0.0199 0.0005MVAD 0.0072 0.0005 0.085 0.005 0.0006 0.0158 0.0005MVAE 0.0027 0.0005 0.128 0.0005 0.0053 0.00025 0.001MVAF 0.0017 0.002 0.364 0.001 0.0089 0.00025 0.001MVAG 0.0023 0.001 0.434 0.0005 0.0046 0.0087 0.002MVAH 0.0129 0.003 2.02 0.004 0.0274 0.083 0.005MVAI 0.0069 0.002 0.083 0.007 0.003 0.185 0.001MVAJ 0.0035 0.001 0.324 0.0005 0.0065 0.0028 0.002MVAK 0.003 0.001 0.2 0.001 0.0062 0.0128 0.002MVAL 0.0086 0.002 2.67 0.008 0.0167 0.0317 0.006MVAM 0.0254 0.0005 0.026 0.004 0.0055 1 0.001MVAN 0.0024 0.0005 0.099 0.006 0.0018 0.0218 0.0005MVAO 0.0133 0.006 5.39 0.01 0.0347 0.0481 0.013MVAP 0.0072 0.001 0.168 0.0005 0.0035 0.0017 0.004MVAQ 0.003 0.0005 0.918 0.0005 0.0102 0.0013 0.002MVAR 0.0021 0.002 0.273 0.001 0.0163 0.0083 0.001MVAS 0.0053 0.0005 0.146 0.0005 0.0094 0.0399 0.0005MVAT 0.0023 0.001 0.353 0.001 0.0052 0.0112 0.0005MEAN 0.0073 0.001 0.488 0.002 0.0078 0.0516 0.002MEDIAN 0.0053 0.001 0.160 0.001 0.0055 0.0181 0.001HIGH 0.0373 0.006 5.390 0.011 0.0347 1.0000 0.013LOW <0.0005 <0.001 0.021 <0.001 0.0006 <0.0005 <0.001
Table 1. Continued.
Sample Turbid? In/Out flow? Nostoc? Inverts?ID
MVA n n y fairy shrimp, Daphnia, red & brown chironomidsMVB n n n DaphniaMVC n n n DaphniaMVD n n y nMVE n n n nMVF n n n nMVG y n n nMVH y n y nMVI n n n nMVJ y n n copepodsMVK n n n fairy shrimp MVL n n y Daphnia, chironomidsMVM n n n fairy shrimpMVN n n n fairy shrimpMVO n n n nMVP n n n copepods, chironomidsMVQ n n y fairy shrimp (small)MVR n n n fairy shrimpMVS n n n nMVT n n y fairy shrimpMVU n n n fairy shrimpMVV n n n nMVW y n n nMVX n n y fairy shrimpMVY n n n fairy shrimp, Daphnia MVZ n n n nMVAA n n y fairy shrimpMVAB y n n nMVAC n n n fairy shrimpMVAD n n n copepods, fairy shrimp, DaphniaMVAE n n n nMVAF y inflow (small) n nMVAG n n n red chironomidMVAH n n n nMVAI n n n fairy shrimp, 2 sizesMVAJ n n n nMVAK n n y nMVAL y inflow n nMVAM n n y nMVAN n outflow n nMVAO y n n nMVAP n n n nMVAQ n n n nMVAR n n n nMVAS n n n nMVAT n n n fairy shrimp, Daphnia, chironmids
Table 2. Selected nutrient ratio parameters for 46 freshwater sites on Melville Island.
Sample TN:TP POC:PON PON:POP POC:POP POC:CHLa POC:Chla Na:KID molar molar molar molar mg/L:mg/L μmol/L:μg/L mg/L:mg/L
MVA 139.9 11.78 37.76 445 828.0 68.9 20.81MVB 111.7 11.48 12.59 145 886.0 73.8 19.44MVC 138.7 13.04 12.32 161 760.0 63.3 5.14MVD 158.3 13.45 11.24 151 692.0 57.6 12.00MVE 147.1 14.13 NA NA 436.0 36.3 6.00MVF 142.0 14.28 NA NA 8080.0 672.8 4.69MVG 61.9 11.91 486.32 5793 1123.3 93.5 1.18MVH 31.3 9.44 NA NA 485.6 40.4 1.39MVI 63.6 12.72 23.27 296 218.0 18.2 2.00MVJ 42.6 9.84 17.95 177 532.3 44.3 2.75MVK 34.5 10.03 7.60 76 591.3 49.2 1.78MVL 108.8 13.87 5.31 74 314.3 26.2 2.38MVM 179.3 15.28 13.20 202 582.2 48.5 2.65MVN 33.3 10.78 9.03 97 390.0 32.5 2.67MVO 40.1 10.81 15.99 173 242.3 20.2 3.20MVP 57.1 12.78 89.27 1141 410.7 34.2 7.00MVQ 45.0 11.58 23.18 269 556.0 46.3 2.00MVR 19.4 14.11 5.95 84 635.0 52.9 1.33MVS 72.0 11.27 51.58 581 676.0 56.3 1.29MVT 66.1 10.90 NA NA 598.0 49.8 1.33MVU 52.8 10.66 475.27 5067 302.3 25.2 2.36MVV 30.3 10.50 30.61 321 85.3 7.1 3.33MVW 17.1 11.18 6.34 71 457.8 38.1 0.75MVX 122.4 10.09 7.09 72 744.0 61.9 1.11MVY 76.4 10.45 42.51 444 1493.3 124.3 2.50MVZ 21.6 11.67 19.22 224 222.2 18.5 1.00MVAA 15.8 9.47 4.97 47 295.7 24.6 1.78MVAB 27.8 9.51 13.30 127 183.0 15.2 27.04MVAC 112.6 11.07 NA NA 322.7 26.9 15.67MVAD 77.7 14.11 51.90 732 653.0 54.4 4.50MVAE 10.2 18.88 6.95 131 593.3 49.4 1.00MVAF 8.3 11.88 48.63 578 6720.0 559.5 1.00MVAG 18.1 12.62 NA NA 4760.0 396.3 0.43MVAH 41.7 10.52 NA NA 846.2 70.5 13.88MVAI 66.3 8.71 37.90 330 896.0 74.6 21.06
Table 2. continued.
Sample TN:TP POC:PON PON:POP POC:POP POC:CHLa POC:Chla Na:KID molar molar molar molar mg/L:mg/L μmol/L:μg/L mg/L:mg/L
MVAJ 9.5 11.39 7.87 90 683.3 56.9 10.50MVAK 72.4 10.68 363.51 3882 268.3 22.3 18.00MVAL 5.0 11.34 1.70 19 947.5 78.9 7.35MVAM 112.8 11.77 NA NA 482.2 40.2 3.50MVAN 15.2 12.51 NA NA 297.8 24.8 4.44MVAO 4.0 14.31 0.78 11 63.7 5.3 7.76MVAP 10.8 13.59 8.35 113 3960.0 330.0 1.50MVAQ 7.9 9.54 4.97 47 140.8 11.7 3.00MVAR 5.7 10.23 1.05 11 328.8 27.4 3.20MVAS 43.9 12.44 6.98 87 290.9 24.2 1.55MVAT 120.8 12.30 23.91 294 349.4 29.1 3.75MEAN 60.2 11.8 53.7 609.8 987.5 82.2 5.7MEDIAN 44.4 11.5 13.2 160.6 544.2 45.3 2.9MAXIMUM 179.3 18.88 475.27 5793 8080 672.8 27.04MINIMUM 4 8.71 <0 <0 63.7 5.3 0.43
Tabl
e 3.
Pea
rson
cor
rela
tion
mat
rix w
ith B
onfe
rron
i-adj
uste
d pr
obab
ilitie
s. S
igni
fican
tly c
orre
late
d va
riabl
es a
re in
dica
ted
in b
old
(p<0
.01)
or i
talic
s (p<
0.05
). N
H3
NO
2C
lSO
4Si
O2
CH
La
POC
PON
DO
CD
ICT
KN
SRP
Ba
FeM
nSr
Ca
NO
2-0
.02
1C
l0.
234
-0.1
181
SO4
0.17
4-0
.109
0.55
81
SiO
20.
148
0.33
8-0
.119
0.22
41
CH
La
0.13
50.
220.
039
0.06
0.41
31
POC
0.11
20.
209
0.18
60.
009
0.27
60.
621
PON
0.12
60.
203
0.17
3-0
.003
0.23
70.
601
0.96
91
DO
C0.
255
0.16
0.44
60.
093
0.29
60.
226
0.55
30.
483
1D
IC-0
.03
-0.1
930.
489
0.09
3-0
.139
-0.2
110.
009
-0.0
470.
618
1T
KN
0.27
60.
008
0.37
1-0
.043
0.04
9-0
.022
0.34
20.
260.
882
0.73
1SR
P0.
339
0.00
30.
146
0.35
20.
218
0.27
0.15
70.
156
0.13
8-0
.132
0.02
81
Ba
0.23
9-0
.278
0.60
30.
501
0.11
8-0
.013
-0.0
34-0
.016
0.34
30.
482
0.25
50.
344
1Fe
-0.0
160.
247
-0.0
3-0
.014
0.24
20.
319
0.37
90.
41-0
.062
-0.3
25-0
.282
0.20
40.
045
1M
n-0
.004
0.12
30.
019
0.07
70.
274
0.18
30.
255
0.30
9-0
.046
-0.1
2-0
.194
0.05
90.
029
0.70
51
Sr0.
32-0
.058
0.61
80.
569
0.17
80.
074
0.19
0.22
70.
575
0.51
70.
397
0.25
60.
672
-0.1
45-0
.019
1C
a0.
151
-0.1
150.
526
0.29
8-0
.004
-0.0
810.
119
0.13
30.
653
0.76
80.
583
0.00
30.
534
-0.2
97-0
.068
0.86
91
Mg
0.03
7-0
.241
0.70
40.
307
-0.1
9-0
.146
0.08
40.
093
0.57
30.
844
0.56
7-0
.068
0.55
7-0
.204
-0.0
140.
758
0.89
6N
a0.
306
-0.0
780.
886
0.50
5-0
.007
0.10
20.
263
0.29
50.
550.
498
0.41
30.
224
0.62
50.
001
0.04
60.
813
0.66
5K
0.17
6-0
.149
0.83
50.
542
-0.1
16-0
.095
0.09
10.
138
0.41
50.
573
0.33
0.09
30.
660.
016
0.06
10.
763
0.67
3A
l0.
017
0.14
50.
134
0.23
0.28
10.
178
0.14
30.
168
-0.0
46-0
.181
-0.2
790.
265
0.28
0.78
30.
603
0.10
4-0
.069
Cu
-0.1
350.
111
0.29
10.
121
0.21
20.
201
0.18
30.
158
0.29
20.
206
0.10
50.
046
0.22
0.46
70.
494
0.19
0.18
1L
i0.
261
0.00
80.
661
0.66
50.
30.
228
0.15
50.
182
0.22
90.
034
-0.0
40.
331
0.56
70.
274
0.11
20.
571
0.24
9Z
n0.
040.
285
0.07
80.
204
0.44
10.
402
0.11
80.
109
-0.0
3-0
.262
-0.2
930.
302
0.23
0.72
20.
546
0.05
4-0
.173
TdN
0.30
60.
039
0.43
0.00
10.
097
0.04
60.
443
0.37
50.
931
0.67
70.
964
0.05
40.
275
-0.2
37-0
.132
0.50
20.
662
TP
0.16
0.16
60.
049
0.22
20.
467
0.43
30.
364
0.37
80.
117
-0.2
64-0
.104
0.42
20.
130.
655
0.45
70.
026
-0.1
86T
PF0.
235
-0.0
40.
143
0.15
20.
092
0.09
70.
345
0.33
10.
148
-0.1
640.
024
0.42
40.
108
0.20
5-0
.063
0.14
3-0
.01
pH0.
077
-0.2
320.
506
0.12
3-0
.218
-0.0
970.
147
0.09
50.
591
0.78
10.
634
-0.0
40.
349
-0.3
43-0
.087
0.54
70.
782
CO
ND
0.19
2-0
.215
0.77
10.
484
-0.0
17-0
.034
0.16
90.
183
0.61
50.
739
0.53
10.
052
0.65
3-0
.199
-0.0
220.
889
0.91
5E
LE
V-0
.177
0.03
8-0
.729
-0.2
5-0
.002
-0.0
16-0
.261
-0.2
32-0
.674
-0.7
14-0
.618
-0.1
24-0
.535
0.02
6-0
.046
-0.6
36-0
.648
TE
MP
0.07
20.
342
-0.0
34-0
.305
-0.1
37-0
.067
0.15
40.
176
0.26
90.
119
0.25
8-0
.26
-0.1
28-0
.109
-0.2
790.
155
0.27
5L
ON
G-0
.194
-0.0
25-0
.048
-0.1
18-0
.213
-0.2
73-0
.119
-0.1
46-0
.081
-0.1
01-0
.046
-0.4
17-0
.126
-0.1
52-0
.308
-0.2
24-0
.116
LA
T0.
418
0.15
2-0
.016
0.33
50.
465
0.25
80.
029
0.02
8-0
.067
-0.4
81-0
.208
0.53
20.
226
0.12
7-0
.128
0.03
4-0
.324
Tabl
e 3.
Con
tinue
d.
M
gN
aK
Al
Cu
Li
Zn
TdN
TP
TPF
pHC
ON
DE
LE
VT
EM
PL
ON
GN
a0.
771
K0.
840.
884
1A
l-0
.013
0.16
70.
221
Cu
0.31
10.
341
0.35
20.
584
1L
i0.
343
0.70
80.
691
0.40
80.
389
1Z
n-0
.139
0.11
70.
101
0.82
40.
711
0.45
61
TdN
0.59
80.
492
0.37
5-0
.236
0.15
0.03
6-0
.25
1T
P-0
.157
0.12
30.
058
0.50
70.
426
0.40
90.
582
-0.0
551
TPF
-0.0
660.
168
0.02
10.
283
0.03
40.
202
0.13
50.
075
0.22
11
pH0.
737
0.48
90.
434
-0.2
350.
08-0
.017
-0.3
50.
691
-0.2
4-0
.076
1C
ON
D0.
950.
855
0.87
20.
026
0.27
90.
514
-0.0
680.
596
-0.0
780.
040.
683
1E
LE
V-0
.738
-0.8
08-0
.694
-0.1
81-0
.436
-0.3
75-0
.097
-0.6
45-0
.054
-0.1
63-0
.596
-0.7
381
TE
MP
0.15
80.
021
0.06
7-0
.18
-0.2
05-0
.128
-0.2
650.
311
-0.2
360.
018
0.20
90.
127
0.00
91
LO
NG
-0.1
03-0
.216
-0.1
43-0
.216
-0.2
61-0
.091
-0.3
11-0
.063
-0.3
320.
002
-0.0
5-0
.11
0.29
80.
436
1L
AT
-0.4
48-0
.026
-0.1
170.
119
-0.1
90.
427
0.28
6-0
.195
0.34
20.
4-0
.466
-0.2
310.
24-0
.126
-0.0
29
CHAPTER 3
EVALUATING THE ROLE OF ENVIRONMENTAL AND SPATIAL VARIABLES ON DIATOM SPECIES
DISTRIBUTIONS ON MELVILLE ISLAND (CANADIAN HIGH ARCTIC)
BRONWYN E. KEATLEY, MARIANNE S.V. DOUGLAS, AND JOHN P. SMOL
This chapter has been submitted to the Canadian Journal of Botany: Keatley, B.E., M.S.V. Douglas, and J.P. Smol. In review. Evaluating the role of environmental and spatial variables on diatom species distributions on Melville Island (Canadian High Arctic). Submitted: 7 June 2007.
Abstract
Diatom species assemblages were identified and enumerated from the surface sediments of 45
lakes and ponds across a wide spectrum of spatial and environmental gradients on Melville
Island, Nunavut/N.W.T, High Arctic Canada. Whereas the most common taxa were similar to
those recorded elsewhere in the Canadian High Arctic, significant differences in assemblages
existed between sites located in the different bioclimatic zones of Melville Island. For example,
taxa recorded in the most lushly vegetated bioclimatic zone were similar to those found in richly
vegetated regions elsewhere in the Canadian Arctic Archipelago, and generally different from
diatoms in the poorly vegetated regions on Melville Island. Of the measured environmental
variables, pH, specific conductivity, surface area, elevation, and chlorophyll a explained
significant portions of the variance in diatom assemblage composition when diatoms were
considered at the scale of the entire island. However, only total dissolved nitrogen was an
important explanatory variable within the most lushly vegetated bioclimatic zone. Ecological
relationships between diatom species distributions and pH were moderately strong, as indicated
by weighted averaging transfer functions (r2boot = 0.432 to 0.746, RMSEP = 0.341 to 0.242).
Interestingly, spatial factors were of little importance, confirming that diatoms are not likely to be
dispersal limited, at least at the landscape scale explored in this study.
Introduction
Polar regions are particularly sensitive to climate change (ACIA 2004), and these
changes are already resulting in dramatic consequences for biological communities across the
circumpolar north (e.g. ACIA 2004; Smol et al. 2005). As long-term monitoring data are often
unavailable for remote Arctic regions, paleolimnological data are being increasingly used to
assess the rate, magnitude and direction of environmental changes (Douglas et al. 2004).
Diatoms, unicellular algae with siliceous cell walls, are one of the most commonly used proxy
indicators used in Arctic paleolimnological studies because they preserve well in lake sediments,
they are prevalent in most aquatic systems, and certain species often have well defined optima to
given environmental variables (Stoermer and Smol 1999). However, in order to use proxy
indicators such as diatoms effectively, it is important to better understand the ecological and
biogeographical characteristics of taxa (Smol 2002).
Over the past approximately two decades, a concerted effort has been made to better
understand the factors influencing diatom species distributions in the Canadian High Arctic. The
resulting body of work includes diatom calibration sets from nine regions of the Canadian High
Arctic (Fig. 1): Cape Herschel (Douglas and Smol 1993, 1995), Bathurst Island (Lim et al. 2001a,
b), Isachsen (Antoniades et al. 2004), Alert and, Mould Bay (Antoniades et al. 2005), Axel
Heiberg Island (Michelutti et al. 2006), Banks Island (Lim et al. 2007), Devon Island (Lim 2004)
and Cornwallis Island (Michelutti et al. in press). Although considerable progress has been made,
critical gaps remain in our understanding of diatom species distributions across the vast and
ecologically diverse High Arctic landscape, and this is especially true for the western High Arctic
(Fig. 1). For example, in the western portion of the Canadian High Arctic, a region for which
climate change research is particularly scarce, our understanding of modern diatom ecology is
based largely on two surveys (Antoniades et al. 2005; Lim et al. 2007), one of which is restricted
to a region less than 5 km2 (Antoniades et al. 2005). Data from these studies revealed large
differences in environmental characteristics and diatom species distributions, underscoring the
need for additional data from the western Arctic.
Here, we explore diatom species-environment relationships from Melville Island, which,
due to its large size (i.e. 42 149 km2), relatively high relief (from sea level to ~655 m above sea
level), multiple bioclimatic zones (Edlund 1994), and marked gradients of water chemistry
(Keatley et al. [2]), provides us with a unique study site in which to determine the environmental
factors influencing diatom species distributions in the western Canadian High Arctic.
The large geographical scale of our sampling regime on Melville Island (spanning nearly
the whole island) also provides a database which allows us to explore the relative roles of spatial
versus environmental factors in structuring diatom communities. Although dispersal is often
thought to be of minimal importance in influencing microscopic species distributions (e.g.
Kristiansen 1996; Finlay 2002; Finlay et al. 2002), as their small size allows them to reach new
habitats via wind, waterfowl, or other vectors, some researchers have argued that dispersal in
diatoms may be more important than initially thought (e.g. Kociolek and Spaulding 2000; Telford
et al. 2006). The lack of any direct human activities makes Melville Island an ideal location to
examine the relative importance of dispersal versus environmental factors in diatom species
distributions.
Our main study objectives are to: 1) describe the ecological characteristics of diatoms
from Melville Island; 2) elucidate which environmental and/or spatial factors are able to explain
diatom species distributions across this island; and 3) assess whether diatom species distributions
can be classified according to previously defined bioclimatic zones. We show that diatom
assemblages are significantly influenced by environmental, but not spatial, variables at the whole
island scale, with pH, specific conductivity, surface area, elevation and chlorophyll a
concentrations explaining significant portions of the diatom variance. Interestingly, diatoms
within the most lushly vegetated bioclimatic zone on Melville Island are significantly different
from those found in all other zones, and a significant proportion of the diatom species variance
within this zone can be explained by total dissolved nitrogen. Finally, we construct quantitative
transfer functions for pH, the most ecologically-relevant variable explaining diatom species
abundances on Melville Island.
Methods
Site description
A detailed site description of Melville Island and the limnological characteristics of the
45 study sites can be found in Keatley et al. ([2]). In brief, Melville Island is located in the
western Canadian High Arctic (Fig. 1), with the border between Nunavut and the Northwest
Territories running longitudinally through the middle of the island. It is the fourth largest island
of the Queen Elizabeth Islands (consisting of the Arctic islands north of the Parry Channel), and
represents five bioclimatic zones (Edlund 1994) and four physiographic regions (Tozer and
Thorsteinsson 1964). The nearest long-term weather station is located at Mould Bay, Prince
Patrick Island (~ 250 km from the centre of Melville Island), where average temperatures for July
and February are 4.0°C and -34.0°C, respectively, and the mean annual precipitation is 111.0 mm
(Environment Canada 2004). Melville Island is the only island in the western Arctic that
currently retains small permanent ice caps.
The surficial geology of the island is largely composed of weathered bedrock, which is
highly variable within regions. In general, the bedrock of eastern Melville Island is comprised of
sandstones, mudstones, limestones, shales, and evaporites (Barnett et al. 1977). Western Melville
Island is dominated by sandstones and carbonates, whereas the Dundas Peninsula in the
southwest portion of the island contains glacial deposits, including calcareous tills derived from
southern locations (Tozer and Thorsteinsson 1964; Hodgson and Vincent 1984).
Of the five previously described bioclimatic zones (Zones 0 through 4) found on Melville
Island (Edlund 1994), four (Zones 1 through 4) were encompassed by our sampling regime (Fig.
1). These zones are characterized both by their climate and vegetation, with Zone 4 representing
the most diverse and lushly vegetated regions, and Zone 0 representing areas completely devoid
of vegetation (Edlund 1994).
Sampling
Surface sediment diatoms were sampled from 46 lakes and ponds (Fig. 1c), chosen to
encompass a large range of physical and chemical characteristics (Keatley et al. [2]). Using
identical sampling protocols used in all our previous Arctic diatom surveys, we sampled the
uppermost ca. 0.5 cm of sediments by hand from the near-shore environment from each site, and
stored the sediment in Whirlpak® sample bags. It was impossible to obtain sediment from pond
MVAF, so we sampled the surfaces of rocks as the dominant substrate for diatom samples. The
samples were kept cool and dark until they were processed in the lab at PEARL, Queen’s
University, Kingston, Canada.
At the same time that the sediment samples were obtained, the lake or pond was sampled
for water chemical variables (e.g. pH, specific conductivity, nutrients, major ions, and trace
metals) and physical characteristics (e.g. temperature, elevation, morphometric parameters). The
physical and chemical limnological methodologies and characteristics of these sites has been fully
described elsewhere (Keatley et al. [2]). Table 1 provides a summary of some key limnological
variables, separated by bioclimatic zones. Briefly, the limnological characteristics from Melville
Island sites represent one of the largest environmental gradients sampled to date in the Canadian
High Arctic. Of particular interest is the distinctive water chemistry of the lakes and ponds
located in bioclimatic Zone 4 (the most richly vegetated zone on Melville Island), which are
characterized by high pH, dissolved organic carbon, and total dissolved nitrogen (Keatley et al.
[2]). In addition, our sampling regime also included several sites close to high elevation ice caps,
which exhibit the lowest specific conductivity (~5 µS/cm) and calcium (0.1 mg/L) yet recorded in
the Canadian Arctic (Keatley et al. [2]).
Diatom analyses
The sediment samples were treated using standard diatom preparation techniques
(Battarbee et al. 2001). Briefly, organic material was digested from the sediment samples using a
50/50 mixture of concentrated HNO3 and H2SO4 at 80 °C for ~8 hours. The resultant slurries
were rinsed with deionized water until they reached a neutral pH. Diatoms were distributed onto
pre-cleaned coverslips, and then mounted onto microscope slides with Naphrax®, a mounting
medium with a high refractive index. Diatoms from each site were enumerated and identified
using Krammer and Lange-Bertalot (1991), Krammer (2002) and Antoniades et al. (in press). A
minimum of 300 valves were identified from each site.
Statistical analyses
Diatom counts were converted to relative abundances and screened for frequency of
occurrence. Any species that were present in >1% relative abundance in at least one site were
retained for further analyses. A detrended correspondence analysis (DCA) was used to assess the
length of the gradient encompassed by the diatom taxa, and to determine whether linear or
unimodal statistical techniques should be used to assess the relationships between environmental
variables and diatom species.
The similarity of diatom assemblages within and between bioclimatic zones was assessed
using the program PRIMER v5.2.9 (Clarke and Gorley 2001). An analysis of similarities
(ANOSIM, based on a Bray-Curtis similarity matrix) was used to determine whether the diatom
species assemblages were significantly different between pairs of bioclimatic zones. The
PRIMER program SIMPER (similarity of percentages) was used to determine which diatom
species were most typical of each zone.
Environmental variables
In order to assess which measured environmental variables were most closely related to
diatom species distributions, we used a series of constrained multivariate ordinations. Any
environmental variable that was below the method detection limit in >50% of the sites was
removed from further analysis. A Pearson correlation matrix was then used to assess pairs of
significantly correlated variables; this information was also used to truncate the dataset by
removing highly collinear variables. This subset of variables was subsequently entered into a
canonical correspondence analysis (CCA) to determine which of the measured environmental
variables best explained the diatom species distributions. All ordinations were performed in
CANOCO 4.5 (ter Braak and Šmilauer 2002).
Spatial variables
The latitude and longitude of each site were converted to x,y coordinates (in metres),
based on the Lambert Canadian Conic projection to reduce the distortion found at high latitudes
in the more commonly used Universal Transverse Mercator coordinate system (Lo and Yeung
2007). These coordinates were used to create a dataset of spatial variables derived from principal
coordinates of neighbour matrices (PCNM) using the program SpacemakeR in R Statistical
Environment, following the methods of Dray et al. (2006). Distance-based eigenvector maps
were used to create matrices of spatial variables according to predefined criteria for “neighbours”.
Because it was impossible to determine a priori which neighbourhood model would be most
realistic, we tested five different models (based on Delaunay triangulation, Gabriel graph, Sphere
of Influence, Relative Neighbourhood, and a distance-based criterion which listed sites as
neighbours only if they were within 50 km of each other). Based on previous concepts regarding
the dispersal of diatoms, we felt that the most connected model (i.e. based on Delaunay
triangulation) was likely the most realistic, but we nevertheless examined all of the models.
SpacemakeR generated a series of spatial variables that were subsequently used in a CCA
with forward selection to determine what proportion of the variance in species distributions could
be explained by spatial structure. Finally, both the environmental data and spatial data were used
together, using the program VARCAN (Peres-Neto et al. 2006), to assess the relative importance
of the measured environmental variables and the spatial variables. VARCAN is capable of
performing an adjustment for bias correction, as well as assessing the significance of the
proportion of variation explained by both environmental and spatial (Peres-Neto et al. 2006).
Exploratory statistical analysis and transfer functions
Variables identified as explaining significant portions of the diatom species variance
were each run in individually constrained CCAs in order to determine the explanatory power of
each particular variable and examine which variables were best suited to the construction of a
diatom transfer function. This resulted in a subset of variables that explained significant portions
of the diatom variance individually; each of these was then tested with a detrended CCAs
(DCCA) to assess whether linear or unimodal techniques were most appropriate.
As the DCCAs indicated gradient lengths of > 2 standard deviations, we used unimodal
techniques (Lepš and Šmilauer 2003) to assess the possibility of building transfer functions to
reconstruct pH and specific conductivity (the two most ecologically significant variables shaping
diatom distributions). We examined both weighted averaging (WA) and weighted averaging
partial least squares (WA-PLS) techniques, in the program C2 v1.43 (Juggins 2003) to assess
which model was most appropriate for the given environmental variable. We included taxa that
reached either at least 1% relative abundance in a minimum of three sites or greater than 10%
relative abundance in at least one site. We also screened taxa to assess whether or not they had:
a) any significant response, or b) a significant unimodal response, using the program Huisman-
Olff-Fresco (HOF v2.3; www.helsinki.fi/~jhoksane/, Huisman et al. 1993).
Results and Discussion
Ecological characteristics of diatoms from Melville Island
Although our initial dataset consisted of 46 ponds and lakes, no diatoms were found in
site MVV (discussed below). This resulted in a subset of 45 lakes and ponds, from which a total
of 293 species were identified. Just under half of all taxa identified occurred in less than 1%
relative abundance and were removed from the dataset. All ordinations and statistical analyses
are based on the 164 taxa that occurred in at least 1% relative abundance in at least one site. Table
2 lists all taxa present in at least 1% relative abundance in at least three sites, or at least 10%
relative abundance in at least one site.
No diatoms were found in site MVV, a fairly large lake located in a poorly vegetated
catchment that was mostly ice-covered at the time of sampling. The persistent ice cover on this
lake was not unique to site MVV, as many other lakes on Melville Island were similarly covered
with ice during our field season of July 2002. What is particularly interesting to note, however, is
the extremely low SiO2 concentration (0.07 mg/L) found in MVV. In temperate systems, diatoms
are known to become less abundant when silica concentrations in the water column fall below 0.5
mg/L (Wetzel 2001). Thus, it is possible that the extremely low silica in MVV may have been
partially responsible for restricting diatom growth in this pond. However, other factors must also
have been acting in concert to exclude diatoms from this site as the rest of the water chemistry
was relatively unremarkable compared to some other sites on Melville Island, some of which
recorded even lower SiO2 concentrations. Perhaps a combination of persistent ice cover, low
nutrients, and low SiO2 precluded diatoms from this site.
Our survey revealed several diatom taxa common among our study sites (Fig. 2), which
have also been recorded elsewhere throughout the Canadian Arctic. For example, Nitzschia
perminuta, found in almost every lake sampled on Melville Island, is one of the most ubiquitous
diatoms of the Canadian Arctic (Antoniades et al. 2005; Lim et al. 2007). Other common taxa
found on Melville Island include Fragilaria capucina, Achnanthidium minutissimum,
Psammothidium marginulatum, and Chaemaepinnularia soehrensis (Fig. 2). These species are
all relatively small, benthic diatoms that are common in circumneutral pH habitats in High Arctic
freshwater ecosystems including Alert, Ellesmere Island, Axel Heiberg Island, Cornwallis Island,
and Devon Island (Lim 2004; Antoniades et al. 2005; Michelutti et al. 2006; Michelutti et al. in
press).
Although the diatom species were overwhelmingly dominated by benthic taxa, one site
(MVO) was characterized by high relative abundances of small, planktonic Cyclotella spp., which
have been linked to reduced ice cover and stronger stratification regimes in Subarctic lakes (e.g.
Sorvari et al. 2002; Rühland et al. 2003). Interestingly, MVO is a fairly large lake located in a
lush grassy area that would be predicted to record increases in Cyclotella spp. with warmer
temperatures.
A few lakes within this calibration set were also notable due to their location near high
elevation ice caps and their very low specific conductivity (~5 µS/cm) and nutrients (MVAE,
MVAF; Keatley et al. [2]). These high elevation sites were characterized by diatom species such
as Caloneis aerophila, Eunotia spp., Staurosira construens var. venter, F. capucina and
Achnanthes spp. (especially Psammothidium helveticum and A. rupestris). Elsewhere in the High
Arctic, many of these species have been recorded in sites with relatively low pH (<7, especially
C. aerophila and Eunotia spp., Antoniades et al. 2004) and low specific conductivity (e.g. C.
aerophila, P. helveticum, F. capucina, Antoniades et al. 2005). A. rupestris has previously been
reported as an uncommon taxon, yet when present can reach high abundances (e.g. >25% relative
abundance in a given site, Antoniades et al. 2004). However, previous surveys have reported
widely divergent specific conductivity (603 µS/cm, Antoniades et al. 2004; 66 µS/cm, Michelutti
et al. 2006) and pH optima for this species (pH: 6.4, Antoniades et al. 2004; 7.9, Michelutti et al.
2006). On Melville Island, A. rupestris reached very high relative abundances (>50%) in these
high elevation and extremely low specific conductivity sites (5 µS/cm), suggesting that this taxon
has a low specific conductivity optimum.
Differences between bioclimatic zones
Significant differences in diatom species assemblages between bioclimatic Zone 4 (the
most lush zone) and each of the other zones were identified using ANOSIM (Table 3). The
largest differences were found, not surprisingly, between Zone 4 (the most lush) and Zone 1 (the
least lush zone, Table 3). There were no significant differences between diatom assemblages in
zones 3, 2, and 1 (Table 3). The diatom species data track the general trend of differences in
water chemistry between these bioclimatic zones (Keatley et al. [2]). As climatic factors
undoubtedly influence water chemistry, it is difficult to completely untangle these signals.
The program SIMPER was used to determine which species within each zone contributed
to both: a) the similarity within a zone, and b) the dissimilarity between zones. Interestingly, taxa
with the highest relative abundances in each zone were remarkably similar, with Nitzshia
perminuta, for example, reaching an average of between 6-10% relative abundance in each zone,
although it was least abundant in zone 1. Likewise, Psammothidium marginulatum was quite
common in all zones. The differences in species assemblages between zones are therefore being
influenced by the relatively rare species found within each zone. The results from SIMPER also
indicate that sites located in zone 4 have more species in common (average similarity = 35%)
than those located in zone 1 (average similarity = 14%).
Although each zone was typified by similar taxa, the species contributing most to the
dissimilarity between zones provides insights in to the characteristics of each zone. Because the
greatest dissimilarities were found between Zone 4 and Zone 1, we restrict our discussion to the
taxa from these two zones. The species found in Zone 4 that contributed the most to the
dissimilarity from Zone 1 are listed in Table 4, and include several Cymbella spp., Achnanthes
spp., Nitzschia spp., Navicula spp., and Denticula kuetzingii. Zone 4, by definition a lushly
vegetated region, was characterized by sites with higher pH, DOC, and total dissolved nitrogen
relative to the sites in other zones (Keatley et al. [2]). Based on water chemistry, Zone 4 sites
most resemble those of Mould Bay, Prince Patrick Island (Antoniades et al. 2003) and Banks
Island (Lim et al. 2005). Not surprisingly, the diatom assemblages between these three regions
also share some similarities. The diatoms of the Mould Bay sites, for example, were notable for
their higher relative abundances of Cymbella and Achnanthes spp. (Antoniades et al. 2005), as
compared to other High Arctic diatom calibration sets. Likewise, on the very lush Banks Island,
sites characterized by higher total nitrogen typically contained Cymbopleura angustata var.
spitzbergensis and Denticula kuetzingii (Lim et al. 2007), both species that contributed to the
dissimilarity between Zone 4 and Zone 1 on Melville Island.
Species from Zone 1 sites that contributed to the dissimilarity from Zone 4 sites included
Encyonema silesiacum, F. capucina, and Navicula vulpina. While these species have not
previously been noted for characterizing particularly poorly-vegetated regions, they are common
components of High Arctic diatom studies (e.g. Lim et al. 2001a; Antoniades et al. 2004;
Antoniades et al. 2005; Michelutti et al. 2006). However, due to the relatively few taxa shared
between the sites located in Zone 1, the use of the species mentioned above as discriminators is
somewhat limited.
Previous studies have suggested that small, benthic Fragilaria taxa (e.g., Staurosira
venter and Staurosirella pinnata, synonyms F. construens var. venter and F. pinnata,
respectively) are common pioneering species, found in cold, dilute, nutrient-poor systems (e.g.
Douglas and Smol 1999; Lim et al. 2007). On Melville Island, these taxa, although present in
zone 1, were also found in very high relative abundances in Zone 3 sites with low silica (< 0.2
mg/L; MVO, MVQ, MVAS). The presence of these small, benthic Fragilaria sensu lato taxa in
these sites provides evidence of their opportunistic nature; perhaps these taxa are able to exploit
habitats that are not commonly favourable for most other diatom species, whether the harsh
conditions are due to cold temperatures, low light, limited availability of nutrients or, as in this
case, low silica. Recent physiological studies of nutrient limitation of S. pinnata from alpine
lakes supports this hypothesis, in that S. pinnata were found to have low nutrient requirements for
N, P, and Si (Michel et al. 2006).
Multivariate ordinations
In order to determine the relative influence of both measured environmental variables and
spatial variables on diatom species distributions, we used a series of exploratory CCAs.
Interestingly, spatial variables explained very little of the variance of diatom species on Melville
Island (Appendix 2). Based on the CCA, individual spatial variables were sometimes selected as
explaining significant proportions of the diatom variation, depending on the spatial
neighbourhood model chosen. However, all spatial variables were less important than the
environmental variables of pH, specific conductivity and surface area. Indeed, when the
significance of the environmental versus spatial relationship was assessed (Peres-Neto et al.
2006), spatial variables could only explain a significant amount of the variation in the diatom
species data in one out of the five models tested (i.e. when sites located < 50 km apart were
designated as neighbours, Appendix 2). Although Beisner et al. (2006) examined the role of
environmental versus spatial variables on the structure of phytoplankton communities, no
previous diatom surveys, to our knowledge, have assessed the relative significance of PCNM-
derived spatial versus environmental factors in determining diatom species distributions, making
comparisons impossible. Our results suggest, however, that assessing measured environmental
variables is likely capturing most of the variance in diatom species distributions and that dispersal
related factors are not important for diatoms, at least at the landscape scale identified in this
study. Our data support the hypothesis that diatoms in Arctic regions do not experience dispersal
limitations, even in the absence of human-mediated transport.
The canonical correspondence analysis (CCA) indicated that five environmental variables
could significantly explain ~25% of the variance in diatom species (0.944 out of 3.789). All four
axes were significant as tested by a series of partial CCAs each constrained with the sample
scores as co-variables. Axis 1 (λ = 0.319) of the CCA explained 8.4% of the variation in the
diatom data and represents gradients of pH, surface area, and chlorophyll a (Fig. 3). Axis 2 (λ =
0.255) explained 6.7% of the variation in the diatom data and represents gradients of specific
conductivity and elevation (Fig.3).
Diatom species characteristic of higher pH and specific conductivity plot in the upper left
quadrant of the CCA, and include Amphora spp. (A. spitzbergensis, A. copulata, A. inariensis),
Caloneis silicula, Denticula tenue, D. kuetzingii, many Achnanthes spp. (Achnanthidium kriegeri,
Psammothidium bioretti, Eucocconeis flexella, Eucocconeis laevis,), Cymbella spp. (Cymbella
cleve-eulerae, Cymbopleura angustata var. spitzbergensis, Encyonopsis descripta, Cymbella
designata, Encyonema fogedii, Cymbella subaequalis), Navicula spp. (N. cryptocephala, N. sp. cf
chiarae, N. vulpina), Pinnularia balfouriana (synonym Hygropetra balfouriana), and Nitzschia
spp. (N. inconspicua, N. frustulum). Species plotting in the bottom right quadrant of the CCA are
more common in lower pH, lower specific conductivity and higher elevation sites on Melville
Island and include the only Cyclotella spp. (C. sp. aff. comensis, C. stelligera), small Achnanthes
spp. sensu lato (Psammothidium broenlundensum, P. helveticum, A. ingratiformis, P.
marginulatum, Rossithidium petersenii, P. ventralis), Eunotia praerupta, Fragilaria capucina,
Staurosirella pinnata, Cavinula pseudoscutiformis, Nitzschia perminuta, Pinnularia spp. (P.
grunowii, P. krammeri), and Stauroneis anceps.
Of the measured environmental variables, pH, specific conductivity and surface area
were the most important for explaining diatom variance (p < 0.01). This is a finding common to
other diatom surveys in the Canadian High Arctic (e.g. Antoniades et al. 2004; Antoniades et al.
2005; Michelutti et al. 2006), and is not unexpected as pH and related variables have been
described as the single most important controlling variable on diatom species composition in
freshwater systems (Battarbee et al. 2001). When based on water chemistry data alone, the sites
located within bioclimatic Zone 4 (the most lushly vegetated zone) also had higher total dissolved
nitrogen and related variables (e.g. DOC) than the other sites (Keatley et al. [2]). However, this
distinctiveness was not captured in the diatom data, which did not show a significant response to
any nutrients or related variables when all sites were included together in the CCA. This is likely
due to the overriding influence of pH and specific conductivity on many diatom species.
As the sites from Zone 4 contained significantly different diatom species assemblages
compared to each of the other zones based on our ANOSIM results (see above), we performed an
additional ordination restricted to only Zone 4 sites. In this case, the short gradient length of the
DCA (1.9 standard deviations) indicated that linear methods, such as redundancy analysis (RDA),
would be most appropriate. The RDA with forward selection identified total dissolved nitrogen
as the only significant explanatory variable in explaining diatom abundances in Zone 4 sites (p <
0.01, data not shown). The importance of total dissolved nitrogen as an explanatory variable for
diatom species abundances is similar to the relationships found in the more southerly and lush
Banks Island (Lim et al. 2007), and the well-vegetated, High Arctic Bathurst Island (Lim et al.
2001a). Interestingly, Antoniades et al. (2005) did not find a significant relationship between
diatom assemblages and nutrients (i.e. N, P) in their relatively lush Mould Bay ponds; instead,
DOC explained a significant portion of the variance. The results from Zone 4 suggest that, in
Arctic environments that are warmer and more favourable for vegetation growth, different factors
may control diatom species assemblages as compared to those from other regions.
Quantitative relationships between diatom species and environmental variables
The environmental variables that explained significant proportions of diatom variation
(pH, specific conductivity, surface area, elevation and chlorophyll a) were tested individually in a
detrended CCA (DCCA) to assess whether they could explain significant portions of the variation
in the diatom species data. The DCCA results showed that pH, specific conductivity, surface
area, and elevation each explained significant amounts of the variation in the diatom data (p <
0.01), whereas chlorophyll a did not. However, since pH had the strongest relationship and is
ecologically-relevant for paleoenvironmental reconstructions, we focus here on pH models.
Exploratory models relating specific conductivity to diatom species distributions were also
generated, however, due to the highly structured nature of the residuals and the large error
associated with our conductivity models, we provide them as supplementary material only
(Appendices 3, 4).
Because the gradient length of species turnover with respect to pH and specific
conductivity were relatively long (3.18 standard deviation units), we compared models using the
unimodal modelling techniques of weighted averaging (WA, with and without tolerance
downweighting, as well as with either classical or inverse deshrinking), and weighted averaging
partial least squares (WA-PLS, with 1 to 5 components). All models were validated using the
bootstrapping technique with Monte Carlo permutations (n = 1000). We also examined the
effects of more stringent cut-off criteria for our diatom data by assessing model performance
with: a) all taxa present in at least three sites or in one site if greater than 10% (n = 90); b) taxa
with significant responses to pH (n = 70); and c) only taxa with significant unimodal responses to
pH (n = 46).
None of the WA-PLS models performed significantly better than the WA models, and are
not discussed further. The results of the various WA models are given in Table 5.
The most robust pH model was constructed using WAtol inv for our less stringent
inclusion criteria, with cross-validated r2boot = 0.432 and 0.495 respectively, but when only
species with a significant unimodal response were used (n = 46), WAinv was the best model with
r2boot = 0.738 (Table 5, Fig. 4). Although the model with the more stringent cut-off criteria
produced a stronger relationship, this model also limited the number of species for which pH
optima could be estimated. In any case, our pH models performed as well as, or better, than those
previously generated from Axel Heiberg Island (Michelutti et al. 2006), Isachsen, Ellef Ringnes
Island (Antoniades et al. 2004), Alert, Ellesmere Island and Mould Bay, Prince Patrick Island
(Antoniades et al. 2005). Taken together, these data show that pH is an important explanatory
variable influencing diatom species in the Canadian High Arctic and provide important insights
into the pH preferences of High Arctic diatom taxa. In order to better visualize the relationships
between the most common taxa from Melville Island and pH, species response curves are
provided in Figure 5.
Summary
We characterized the diatom species distributions across Melville Island, NU/N.W.T, a
large geographical region covering five bioclimatic zones and broad environmental gradients in
the western Canadian High Arctic. While the most common species were similar to previously
reported High Arctic species, significant differences in species assemblages occurred between the
most lushly vegetated bioclimatic zone and all other bioclimatic zones. The diatom taxa
contributing to the dissimilarity of the most lush zone from others included species previously
reported to be common in other richly vegetated areas of the Canadian Arctic.
The most important factors explaining diatom species distributions included pH, specific
conductivity, surface area, elevation, and chlorophyll a. Within the most lush bioclimatic zone,
total dissolved nitrogen also explained a significant portion of diatom species distribution,
suggesting that nutrient dynamics may be more important in warmer, more richly vegetated
regions of the Arctic. Spatial variables did not explain a significant portion of the species data,
suggesting that dispersal was not limiting diatom species distributions on Melville Island.
Acknowledgements Funding for this research came from NSERC grants to B.E.K., M.S.V.D., and J.P.S., and an NSTP grant to B.E.K. Logistical support was provided by P.C.S.P. We thank J.R. Glew, N. Michelutti, and D. Antoniades for help in the field, G. Barber and D. Atkinson for help with geographical projections and N. Michelutti and K. Rühland for many useful comments on the manuscript. This is PCSP contribution # (TBA).
References ACIA. 2004. Impacts of a warming Arctic. Cambridge University Press, Cambridge, UK. Available online: http://amap.no/acia/
Antoniades, D., Douglas, M.S.V., and Smol, J.P. 2003. Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick Island, NWT). Archiv. Hydrobiol. 158: 485-516. doi: 10.1127/0003-9136/2003/0158-0485.
Antoniades, D., Douglas, M.S.V., and Smol, J.P. 2004. Diatom species-environment relationships and inference models from Isachsen, Ellef Ringnes Island, Canadian High Arctic. Hydrobiologia, 529: 1-18. doi: 10.1007/s10750-004-4942-8.
Antoniades, D., Douglas, M.S.V., and Smol, J.P. 2005. Benthic diatom autecology and inference model development from the Canadian High Arctic Archipelago. J. Phycol. 41: 30-45. doi: 10.1111/j.1529-8817.2005.04049.x
Antoniades, D., Hamilton, P.B., Douglas, M.S.V., and Smol, J.P. in press. Freshwater diatoms from the Canadian High Arctic. Iconographica Diatomologica.
Barnett, D.M., Edlund, S.A., and Dredge, L.A. 1977. Terrain characterization and evaluation: an example from eastern Melville Island. Geological Survey of Canada Report 76-023. Ottawa.
Battarbee, R.W., Jones, V.J., Flower, R.J., Cameron, N.G., Bennion, H., Carvalho, L., and Juggins, S. 2001. Diatoms. In Tracking Environmental Change Using Lake Sediments: Terrestrial, Algal, and Siliceous Indicators. Edited by J. P. Smol, H. J. B. Birks and W. M. Last. Kluwer Academic Publishers, Dordrecht, Netherlands. pp.155-202.
Beisner, B.E., Peres-Neto, P.R., Lindstrom, E.S., Barnett, A., and Longhi, M.L. 2006. The role of environmental and spatial processes in structuring lake communities from bacteria to fish. Ecology, 87: 2985-2991. doi: 10.1890/0012-9658(2006)87[2985:TROEAS]2.0.CO;2.
Clarke, K.R. and Gorley, R.N. 2001. Primer v 5.0. Primer-E Ltd., Plymouth Marine Laboratory, Plymouth, UK.
Douglas, M.S.V., and Smol, J.P. 1993. Fresh-water diatoms from High Arctic ponds (Cape Herschel, Ellesmere-Island, NWT). Nova Hedwigia, 57: 511-552.
Douglas, M.S.V., and Smol, J.P. 1995. Periphytic diatom assemblages from High arctic ponds. J. Phycol. 31: 60-69. doi: 10.1111/j.0022-3646.1995.00060.x
Douglas, M.S.V., and Smol, J.P. 1999. Freshwater diatoms as indicators of environmental change in the High Arctic. In The Diatoms: Applications for the Environmental and Earth Sciences. Edited by E.F. Stoermer and J.P. Smol. Cambridge University Press, Cambridge. pp. 227-244.
Douglas, M.S.V., Hamilton, P.B., Pienitz, R., and Smol, J.P. 2004. Algal indicators of environmental change in Arctic and Antarctic lakes and ponds. In Long-term environmental change in Arctic and Antarctic lakes. Edited by R. Pienitz, M.S.V. Douglas and J.P. Smol. Springer, Dordrecht. pp. 117-158.
Dray, S., Legendre, P., and Peres-Neto, P.R. 2006. Spatial modelling: a comprehensive framework for principal coordinate analysis of neighbour matrices (PCNM). Ecol. Model. 196: 483-493. doi: 10.1016/j.ecolmodel.2006.02.015
Edlund, S.A. 1994. The distribution of plant communities on Melville Island, Arctic Canada. In The Geology of Melville Island, Arctic Canada. Edited by R.L. Christie and N.J. McMillan. Geological Survey of Canada, Ottawa. pp. 247-255.
Environment Canada. 2004. Canadian Climate Normals 1971-2000. Environment Canada. http://www.climate.weatheroffice.ec.gc.ca/climate_normals/index_e.html,
Finlay, B.J. 2002. Global dispersal of free-living microbial eukaryote species. Science, 296: 1061-1063. doi: 10.1126/science.1070710
Finlay, B. J., Monaghan, E.B., and Maberly, S.C. 2002. Hypothesis: The rate and scale of dispersal of freshwater diatom species is a function of their global abundance. Protist, 153: 261-273. doi: 10.1078/1434-4610-00103.
Hill, M.O. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology, 54: 427-432.
Hodgson, D.A., and Vincent, J.-S. 1984. Surficial geology, Central Melville Island, Northwest Territories; Geological Survey of Canada, Map 1583A, scale 1:250 000.
Huisman, J., H. Olff, & L. F. M. Fresco, 1993. A hierarchical set of models for species response analysis. Journal of Vegetation Science, 4: 37-46.
Juggins, S. 2003. C2: software for ecological and palaeocological data analysis and visualisation, version 1.3. Department of Geography, University of Newcastle.
Keatley, B. E., Douglas, M.S.V., and Smol, J.P. [2]. Physical and chemical limnological characteristics of 46 lakes and ponds across environmental gradients on Melville Island, High Arctic Canada. Fundamental and Applied Limnology.
Kociolek, J.P., and Spaulding, S.A. 2000. Freshwater diatom biogeography. Nova Hedwigia, 71: 223-241.
Krammer, K. 2002. Diatoms of Europe: Cymbella. A.R.G. Gantner Verlag, Ruggell.
Krammer, K., and Lange-Bertalot, H. 1986-1991. Bacillariophyceae. In Sußwasswerflora von Mittleeuropa. Edited by H. Ettl, J. Gerloff, H. Heynig and D. Mollenhauer. Gustav Fischer Verlag, Stuttgart/Jena.
Kristiansen, J. 1996. Dispersal of freshwater algae - A review. Hydrobiologia, 336: 151-157. doi: 10.1007/BF00010829.
Lepš, J., and Šmilauer, P. 2003. Multivariate analysis of ecological data using CANOCO. Cambridge University Press, Cambridge.
Lim, D.S.S. 2004. Limnology and diatom paleoecology of diatoms on Banks Island, N.W.T., and Devon Island, Nunavut, Canadian Arctic. PhD thesis, Department of Geology, University of Toronto, Toronto, ON.
Lim, D.S.S., Douglas, M.S.V., and Smol, J.P. 2001a. Diatoms and their relationship to environmental variables from lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. Hydrobiologia, 450: 215-230. doi: 10.1023/A1017553112643.
Lim, D.S.S., Douglas, M.S.V., and Smol, J.P. 2005. Limnology of 46 lakes and ponds on Banks Island, NWT, Canadian Arctic Archipelago. Hydrobiologia, 545: 11-32. doi: 10.1007/s10750-005-1824-7.
Lim, D.S.S., Kwan, C., and Douglas, M.S.V. 2001b. Periphytic diatom assemblages from Bathurst Island, Nunavut, Canadian High Arctic: An examination of community relationships and habitat preferences. J. Phycol. 37: 379-392. doi: 10.1046/j.1529-8817.2001.03703379.x.
Lim, D.S.S., Smol, J.P., and Douglas, M.S.V. 2007. Diatom assemblages and their relationships to lakewater nitrogen levels and other limnological variables from 36 lakes and ponds on Banks Island, N.W.T., Canadian Arctic. Hydrobiologia, doi: 10.1007/s10750-007-0623-8.
Lo, C.P., and Yeung, A.K.W. 2007. Concepts and Techniques in Geographic Information Systems. Pearson Prentice Hall, Upper Saddle River, NJ.
Michel, T.J., Saros, J.E., Interlandi, S.J., and Wolfe, A.P. 2006. Resource requirements of four freshwater diatom taxa determined by in situ growth bioassays using natural populations from alpine lakes. Hydrobiologia, 568: 235-243. doi: 10.1007/s10750-006-0109-0.
Michelutti, N., Douglas, M.S.V., and Smol, J.P. 2006. Ecological characteristics of modern diatom assemblages from Axel Heiberg Island (High Arctic Canada) and their application to paleolimnological inference models. Can. J. Bot. 84: 1695-1713. doi: 10.1139/BO6-122.
Michelutti, N., Douglas, M.S.V., and Smol, J.P. in press. Evaluating diatom community composition in the absence of marked limnological gradients in the High Arctic: A surface sediment calibration set from Cornwallis Island (Nunavut, Canada). Polar Biology.
Peres-Neto, P.R., Legendre, P., Dray, S., and Borcard, D. 2006. Variation partitioning of species data matrices: Estimation and comparison of fractions. Ecology, 87: 2614-2625. doi: 10.1890/0012-9658(2006)87[2614:VPOSDM]2.0.CO;2.
Rühland, K.M., Smol, J.P., and Pienitz, R. 2003. Ecology and spatial distributions of surface-sediment diatoms from 77 lakes in the subarctic Canadian treeline region. Can. J. Bot. 81: 57-73. doi: 10.1139/BO3-005.
Smol, J.P. 2002. Pollution of lakes and rivers: a paleoenvironmental perspective. Arnold, London.
Smol, J.P., Wolfe, A.P., Birks, H.J.B., Douglas, M.S.V., Jones, V.J., Korhola, A., Pienitz, R., Rühland, K., Sorvari, S., Antoniades, D., Brooks, S.J., Fallu, M.A., Hughes, M., Keatley, B.E., Laing, T.E., Michelutti, N., Nazarova, L., Nyman, M., Paterson, A.M., Perren, B., Quinlan, R., Rautio, M., Saulnier-Talbot, E., Siitoneni, S., Solovieva, N., and Weckstrom, J. 2005. Climate-driven regime shifts in the biological communities of arctic lakes. Proc. Natl. Acad. Sci. U.S.A. 102: 4397-4402. doi: 10.1073/pnas.0500245102.
Sorvari, S., Korhola, A., and Thompson, R. 2002. Lake diatom response to recent Arctic warming in Finnish Lapland. Global Change Biology, 8: 171-181. doi: 10.1046/j.1365-2486.2002.00463.x.
Stoermer, E.F., and Smol, J.P. (eds). 1999. The diatoms: applications for the environmental and earth sciences. Cambridge University Press, Cambridge.
Telford, R.J., Vandvik, V., and Birks, H.J.B. 2006. Dispersal limitations matter for microbial morphospecies. Science, 312: 1015. doi: 10.1126/science.1125669.
ter Braak, C.J.F., and Šmilauer, P. 2002. CANOCO reference manual and CANOdraw for Windows user guide: Software for Canonical Community Ordination (Version 4.5).
Tozer, E.T., and Thorsteinsson, R. 1964. Western Queen Elizabeth Islands, Arctic Archipelago. Geological Survey of Canada Memoir 332. Ottawa.
Wetzel, R. G., 2001. Limnology. Academic Press, San Diego, CA, USA.
Figure captions
Figure 1. Map showing the location of (a) Melville Island in relation to Canada, (b) existing diatom calibration sets in the Canadian Arctic, and (c) the 45 lakes and ponds in this study. The numbers given in (b) correspond to the following studies: 1 = Mould Bay, Prince Patrick Island (Antoniades et al. 2005); 2 = Isachsen, Ellef Ringnes Island (Antoniades et al. 2003b); 3 = Axel Heiberg Island (Michelutti et al. 2006); 4 = Alert, Ellesmere Island (Antoniades et al. 2005); 5 = Cape Herschel, Ellesmere Island (Douglas and Smol 1993, 1995), 6 = Devon Island (Lim 2004); 7 = Cornwallis Island (Michelutti et al. in press); 8 = Bathurst Island (Lim et al. 2001a, b); 9 = Banks Island (Lim et al. 2007). Figure 2. Histograms of the common species found in the surface sediments of 45 Melville Island lakes and ponds. Only species present in at least 1% relative abundance in at least 10 sites are shown, arranged in order of DCA axis 1 species scores. The 45 study sites are also arranged in order of DCA axis 1 sample scores. Measured pH is plotted to the far right of the figure. Figure 3. Canonical Correspondence Analysis (CCA) of a) sites and environmental variables, and b) diatom species and environmental variables from Melville Island. See Table 2 for species code numbers. Figure 4. Relationship between observed and estimated pH values based on the WAtol inv model (n = 90) for a) bootstrapped pH values (r2
boot = 0.432), and b) bootstrapped pH residuals. Figure 5. Species response curves of common species (found in at least 1% relative abundance in at least 10 sites) in relation to pH. The number beside the species name corresponds to the HOF model that best fits the species distribution data (Huisman et al. 1993) and corresponds to the following models: I – no response, II – monotonic with a plateau at a theoretical maximum value, III – monotonic reaching a plateau less than the maximum value, IV – symmetric unimodal, V – skewed unimodal.
Figure 1.
FL
AM
MA
EC
DA
CB
PQ
IA
SG
AT
KS
AG
AK
AH
NA
DR
XT
HO
AA
UA
IJ
AB
YW
AR
AN
AQ
ZA
OA
FA
PA
LA
EA
J0
20
Psammoth
idium
chlid
anos
0Achna
nthes
scoti
ca
020
40
Navicu
lage
rloffi 20
4060
Psammoth
idium
marginu
latum
2040
Diades
misco
ntenta 20Psa
mmothidi
umve
ntrali
s
Nitzsc
hiapu
silla
020Frag
ilaria
capu
cina
0Amphora
inari
ensis
0Navicu
lacry
ptoce
phala
020
40Nitz
schia
perm
inuta
Encyo
nema
minutum
020Nitz
schia
incon
spicu
a
20Rossit
hidium
peter
senn
i
2040
Nitzsc
hiafru
stulum
0Eucoc
cone
islae
vis
020Ach
nanth
idium
minutis
simum
0Cymbe
llacle
ve-eu
lerae
0Navicu
lach
iarae
0Eucoc
cone
isfle
xella
020Nav
icula
soeh
rensis
67
89
pH
Rel
ativ
e ab
unda
nce
(%)
Figu
re 2
.
Fi
gure
3.
A
observed pH6.5 7.0 7.5 8.0 8.5 9.0
WA
toli
nv p
H e
stim
ates
6.5
7.0
7.5
8.0
8.5
9.0
r2boot = 0.432
B
observed pH6.5 7.0 7.5 8.0 8.5 9.0
WA
toli
nv p
H re
sidu
als
-1.0
-0.5
0.0
0.5
1.0
1.5
RMSEP = 0.341
A B
Figure 4.
Psammothidium chlidanos (IV)
0
5
10
15
20
25
30Eucocconeis flexella (IV)
0
2
4
6
8Eucocconeis laevis (V)
0
2
4
6
8
Psammothidium marginulatum (II)
Rel
ativ
e ab
unda
nce
(%)
0
10
20
30
40
50
60
70Achnanthidium minutissimum (II)
0
5
10
15
20
25Rossithidium petersenii (II)
0
2
4
6
8
10
12
14
Cymbella cleve-eulerae (II)
6.5 7.0 7.5 8.0 8.5 9.0
0
1
2
3
4
5Encyonema minutum (IV)
pH
6.5 7.0 7.5 8.0 8.5 9.0
0
2
4
6
8
10Chamaepinnularia soehrensis (IV)
6.5 7.0 7.5 8.0 8.5 9.0
0
4
8
12
16
20
Psammothidium scotica (IV)
0
2
4
6
8
10Psammothidium ventralis (I)
0
4
8
12
16
20Amphora inariensis (IV)
0
2
4
6
8
Figure 5.
Diadesmis contenta (IV)
0
5
10
15
20
25
30
35Fragilaria capucina (I)
0
5
10
15
20
25
30Navicula chiarae (IV)
0
2
4
6
8
10
12
Navicula cryptocephala (I)
Rel
ativ
e ab
unda
nce
(%)
0
2
4
6
8
10Navicula gerloffi (IV)
0
5
10
15
20
25
30
35
Nitzschia frustulum (IV)
6.5 7.0 7.5 8.0 8.5 9.0
0
10
20
30
40
50
Nitzschia inconspicua (I)
0
2
4
6
8
10
12
14
Nitzschia perminuta (II)
pH
6.5 7.0 7.5 8.0 8.5 9.0
0
5
10
15
20
25
30
35Nitzschia pusilla (IV)
6.5 7.0 7.5 8.0 8.5 9.0
0
1
2
3
4
Figure 5. continued.
Tabl
e 1.
Sum
mar
y of
sele
cted
lim
nolo
gica
l cha
ract
eris
tics f
or si
tes f
rom
Mel
ville
Isla
nd.
The
valu
es p
rese
nted
are
mea
n va
lues
for
each
bio
clim
atic
zon
e, w
ith m
ean
valu
es fo
r all
45 si
tes g
iven
at t
he b
otto
m.
Lim
nolo
gica
l det
ails
for e
ach
sam
plin
g si
te a
re d
escr
ibed
in
Kea
tley
et a
l. (2
007
[2])
.
EL
EV
SApH
CO
ND
TE
MP
TP
TdN
CH
La
DO
CD
ICSi
O2
m
asl
ha
μS/c
m° C
elsi
usμg
/L
mg/
Lµg
/Lm
g/L
mg/
Lm
g/L
Zone
4 m
ean
(n=1
5)
34.7
3.6
8.4
211.
78.
813
.7
0.5
0.8
7.9
26.3
0.2
Zone
3 m
ean
(n=1
0)
113.
22.
08.
118
1.9
9.9
28.3
0.
31.
44.
67.
80.
2Zo
ne 2
mea
n (n
=10)
15
1.8
14.8
7.8
136.
49.
725
.3
0.3
1.9
4.9
4.5
0.7
Zone
1 m
ean
(n=1
0)
259.
711
.77.
776
.67.
126
.2
0.2
1.8
3.6
3.4
0.7
All
site
s mea
n (n
=45)
12
6.4
7.49
37.
815
78.
821
.5
0.33
81
5.5
12.3
0.41
All
site
s med
ian
76.2
1.90
08.
174
8.3
13.2
0.
260
14.
78.
40.
20A
ll si
tes m
axim
um
655.
380
.994
8.8
1230
15.5
135.
0 0.
977
10.1
16.3
47.8
3.31
All
site
s min
imum
5.
00.
008
6.8
52.
56.
2 0.
030
<0.1
0.8
0.2
0.02
Tabl
e 2.
Lis
t of d
iato
m sp
ecie
s fou
nd in
Mel
ville
Isla
nd su
rfac
e se
dim
ent s
ampl
es in
>1%
rela
tive
abun
danc
e fr
om a
t lea
st th
ree
site
s, or
>10
% re
lativ
e ab
unda
nce
in a
t lea
st o
ne si
te.
Col
umn
head
ings
are
as f
ollo
ws:
N (t
he n
umbe
r of o
ccur
renc
es),
N2
(the
effe
ctiv
e nu
mbe
r of o
ccur
renc
es b
ased
on
Hill
’s N
2; H
ill 1
973)
, max
% (t
he m
axim
um re
lativ
e ab
unda
nce)
, pH
(est
imat
ed p
H o
ptim
um a
s ba
sed
on W
Ato
l inv
).
Cod
e N
ame
N
Max
%
N2
pH
1 Ps
amm
othi
dium
bio
retti
(Ger
mai
n) B
ukht
iyar
ova
& R
ound
14
4.
87
6 8.
16
2 Ps
amm
othi
dium
bro
enlu
nden
sum
(Fog
ed) H
amilt
on, A
nton
iade
s & S
iver
5
11.9
1 2
8.19
3
Psam
mot
hidi
um c
hlid
anos
(Hoh
n &
Hel
lerm
ann)
Lan
ge-B
erta
lot
32
28.0
6 15
7.
75
4 Eu
cocc
onei
s fle
xella
(Küt
zing
) Cle
ve
23
7.16
11
8.
43
5 Ps
amm
othi
dium
hel
vetic
um (H
uste
dt) B
ukht
iyar
ova
& R
ound
11
19
.06
3 8.
03
6 Ac
hnan
thes
ingr
atifo
rmis
Lan
ge-B
erta
lot
8 8.
00
5 8.
03
7 Ac
hnan
thid
ium
kri
eger
i (K
rass
ke) H
amilt
on, A
nton
iade
s & S
iver
15
5.
47
7 8.
06
8 Ps
amm
othi
dium
kry
ophi
lum
(Pet
erse
n) R
eich
ardt
11
4.
33
7 7.
93
9 Eu
cocc
onei
s lae
vis (
Øst
rup)
Lan
ge-B
erta
lot
23
7.49
12
8.
33
10
Psam
mot
hidi
um m
argi
nula
tum
(Gru
now
) Buk
htiy
arov
a &
Rou
nd
41
59.8
2 10
7.
82
11
Achn
anth
idiu
m m
inut
issi
mum
(Küt
zing
) Cza
rnec
ki
29
22.6
5 17
8.
30
12
Ross
ithid
ium
pet
erse
nii (
Hus
tedt
) Buk
htiy
arov
a &
Rou
nd
29
12.2
1 13
8.
22
13
Achn
anth
es ru
pest
ris K
rass
ke
4 63
.21
2 7.
61
14
Psam
mot
hidi
um ro
ssii
(Hus
tedt
) Buk
htiy
arov
a &
Rou
nd
12
9.14
8
7.83
15
Ps
amm
othi
dium
sp. [
cf. P
. sub
atom
oide
s (H
uste
dt) B
ukht
iyar
ova
& R
ound
] 8
7.79
3
7.99
16
Ps
amm
othi
dium
ven
tral
is (K
rass
ke) B
ukht
iyar
ova
& R
ound
29
13
.88
13
8.01
17
Ad
lafia
sp. [
cf A
. bry
ophi
la (P
eter
sen)
Mos
er &
Lan
ge-B
erta
lot
14
5.21
6
8.35
18
Am
phor
a co
pula
ta (K
ützi
ng) S
choe
man
& A
rchi
bald
12
1.
39
8 8.
29
19
Amph
ora
inar
iens
is K
ram
mer
16
6.
78
8 8.
22
20
Amph
ora
spitz
berg
ensi
s Van
Land
ingh
am
7 20
.67
3 8.
16
21
Cal
onei
s aer
ophi
la B
ock
10
28.4
0 3
7.63
22
C
alon
eis s
ilicu
la (E
hren
berg
) Cle
ve
19
2.80
12
8.
29
23
Cal
onei
s fas
ciat
a (L
ager
sted
t) C
leve
(gro
up 2
) 13
5.
10
8 8.
25
24
Cav
inul
a ja
erne
felti
i (H
uste
dt) D
. Man
n &
Stic
kle
7 9.
83
3 8.
11
25
Cha
mae
pinn
ular
ia g
andr
upii
(Pet
erse
n) L
ange
-Ber
talo
t & K
ram
mer
4
20.9
4 2
7.77
26
En
cyon
opsi
s sp.
[cf.
Nav
icul
a so
ehre
nsis
Kra
sske
] 16
17
.35
9 8.
53
27
Cyc
lote
lla c
omen
sis G
runo
w
1 14
.08
1 n/
a 28
C
yclo
tella
stel
liger
a (C
leve
& G
runo
w) V
an H
eurc
k 2
21.2
4 1
7.94
29
C
ymbo
pleu
ra a
mph
icep
hala
(Nae
geli)
Kra
mm
er
13
2.55
9
8.42
30
C
ymbo
pleu
ra a
ngus
tata
var
. spi
tzbe
rgen
sis K
ram
mer
23
5.
40
11
8.24
31
C
ymbe
lla b
otel
lus (
Lage
rste
dt) S
chm
idt
14
11.3
4 5
8.32
32
En
cyon
opsi
s ces
atii
(Rab
enho
rst)
Kra
mm
er
12
10.2
0 4
8.21
33
C
ymbe
lla c
leve
-eul
erae
Kra
mm
er
25
4.46
19
8.
36
34
Ency
onop
sis d
escr
ipta
(Hus
tedt
) Kra
mm
er
11
5.08
5
8.48
35
C
ymbe
lla d
esig
nata
Kra
mm
er
14
4.31
7
8.44
36
En
cyon
ema
foge
dii K
ram
mer
13
4.
99
7 8.
37
37
Ency
onem
a la
nge-
bert
alot
ii K
ram
mer
11
2.
18
8 8.
10
38
Cym
bopl
eura
lapp
onic
a (G
runo
w) K
ram
mer
5
12.3
7 2
8.05
39
En
cyon
ema
late
ns (K
rass
ke) D
. Man
n 15
6.
71
7 8.
14
40
Ency
onem
a m
inut
um (H
ilse)
D. M
ann
27
8.24
13
8.
14
41
Ency
onem
a si
lesi
acum
(Ble
isch
) D. M
ann
27
3.26
16
8.
21
42
Cym
bella
suba
equa
lis G
runo
w in
Van
Heu
rck
13
2.52
8
8.44
43
D
entic
ula
kuet
zing
ii G
runo
w
13
4.07
8
8.49
44
D
entic
ula
tenu
e K
ützi
ng
3 13
.03
1 8.
42
45
Dia
desm
is sp
. [cf
. D. c
onte
nta
(Gru
now
) D. M
ann]
26
32
.40
10
8.01
46
D
iade
smis
gal
lica
W. S
mith
1
23.7
4 1
n/a
47
Dia
desm
is p
arac
onte
nta
Lang
e-B
erta
lot &
Wer
um
4 6.
46
2 7.
58
48
Dia
tom
a te
nue
Aga
rdh
17
86.2
7 1
7.98
49
D
iplo
neis
sp. [
cf. D
. mar
gine
stri
ata
Hus
tedt
] 19
2.
80
12
8.36
50
D
iplo
neis
ocu
lata
(Bré
biss
on) C
leve
6
2.39
4
8.24
51
Eu
notia
bilu
nari
s (Eh
renb
erg)
Mill
s 8
3.03
4
7.49
52
Eu
notia
bor
eote
nuis
Nör
pel-S
chem
pp &
Lan
ge-B
erta
lot
6 19
.72
3 7.
59
53
Euno
tia sp
. [cf
. E. p
raer
upta
Ehr
enbe
rg]
22
3.20
14
7.
91
54
Frag
ilari
a ca
puci
na D
esm
aziè
res
31
26.0
7 15
7.
97
55
Frag
ilari
a ca
puci
na [c
f. F.
cap
ucin
a va
r. gr
acili
s (Ø
stru
p) H
uste
dt]
14
3.70
8
8.16
56
St
auro
sire
lla p
inna
ta (E
hren
berg
) Will
iam
s & R
ound
9
48.1
9 3
8.00
57
G
omph
onem
a sp
MV
AO
1
10.3
4 1
n/a
58
Gom
phon
ema
sp. a
ff e
xigu
um K
ützi
ng
1 17
.70
1 n/
a 59
N
avic
ula
chia
rae/
cryp
toce
phal
a 2
11.2
2 2
8.64
60
Nav
icul
a sp
. [cf
. N. c
hiar
ae L
ange
-Ber
talo
t & G
enka
l] 21
9.
94
12
8.41
61
N
avic
ula
cryp
toce
phal
a K
ützi
ng
27
8.80
14
8.
14
62
Nav
icul
a sp
. [cf
. Mic
roco
stat
us e
greg
rius
(Hus
tedt
) Lan
ge-B
erta
lot
5 4.
18
4 7.
84
63
Nav
icul
a ge
rlof
fi Sc
him
ansk
i 21
31
.42
3 7.
92
64
Nav
icul
a gr
egar
ia D
onki
n 9
2.33
7
7.91
65
N
avic
ula
ingr
ata
Kra
sske
1
11.1
1 1
n/a
66
Nav
icul
a ph
ylle
pta
Küt
zing
2
36.6
7 2
8.05
67
C
avin
ula
pseu
dosc
utifo
rmis
(Hus
tedt
) D. M
ann
& S
tickl
e 12
5.
36
7 8.
02
68
Sella
phor
a pu
pula
(Küt
zing
) Mer
esch
kow
sky
16
2.18
9
8.26
69
N
avic
ula
sp. [
cf. N
. ven
eta
Küt
zing
] 11
2.
93
7 8.
25
70
Nav
icul
a vu
lpin
a K
ützi
ng
16
3.68
10
8.
41
71
Nei
dium
affi
ne (E
hren
berg
) Pfit
zer
8 8.
57
4 7.
71
72
Nitz
schi
a cl
ausi
i Han
tzsc
h 7
5.81
4
7.78
73
N
itzsc
hia
diss
ipat
a va
r. m
edia
(Han
tzsc
h) G
runo
w
10
1.58
8
8.15
74
N
itzsc
hia
frus
tulu
m (K
ützi
ng) G
runo
w
29
43.2
0 9
8.22
75
N
itzsc
hia
hom
burg
iens
is L
ange
-Ber
talo
t 14
8.
46
4 7.
69
76
Nitz
schi
a in
cons
picu
a G
runo
w
21
12.0
9 7
8.22
77
N
itzsc
hia
pale
acea
Gru
now
8
3.70
5
7.91
78
N
itzsc
hia
perm
inut
a (G
runo
w) M
. Per
agal
lo
37
30.4
1 26
8.
16
79
Nitz
schi
a pu
ra H
uste
dt
18
3.10
12
8.
42
80
Nitz
schi
a pu
silla
Gru
now
18
3.
69
14
8.10
81
N
itzsc
hia
sp. [
cf. N
. suc
hlan
dtii
Hus
tedt
] 7
3.11
6
8.20
82
H
ygro
petr
a ba
lfour
iana
(Gru
now
ex
Cle
ve) K
ram
mer
& L
ange
-Ber
talo
t 5
59.8
7 2
8.43
83
Pi
nnul
aria
inte
rmed
ia (L
ager
sted
t) C
leve
7
5.71
4
7.59
84
Pi
nnul
aria
kra
mm
eri M
etze
ltin
29
2.68
19
8.
05
85
Pinn
ular
ia sp
MV
T 4
10.7
0 1
7.98
86
Pl
anot
hidi
um p
erag
allii
(Bru
n &
Hér
ibau
d) R
ound
& B
ukht
iyar
ova
5 21
.00
1 8.
14
87
Stau
rone
is a
ncep
s Ehr
enbe
rg
31
14.2
9 6
7.87
88
St
auro
neis
sp. [
cf. S
. pro
min
ula
(Gru
now
) Hus
tedt
] 6
2.70
4
7.77
89
Su
rire
lla m
inut
a B
rébi
sson
4
25.7
1 1
7.86
90
Ta
bella
ria
flocc
ulos
a (R
oth)
Kut
zing
14
7.
64
6 7.
75
Table 3. Analysis of similarity (ANOSIM) results. An * following a p value indicates pairs of zones containing significantly different diatom assemblages.
zones R statistic p value 4 vs 3 0.299 0.001* 4 vs 2 0.337 0.002* 4 vs 1 0.447 0.001* 3 vs 2 0.023 0.281 3 vs 1 0.097 0.045 2 vs 1 -0.02 0.59
Table 4. List of species contributing the most to the dissimilarity (Diss/SD; average dissimilarity/standard deviation) between Zones 4 and 1, based on an analysis of similarity percentages (SIMPER). Zone 4 Zone 1 Species Mean % Mean % Diss/SD Cymbella cleve-eulerae 1.85 0.34 1.54 Nitzschia perminuta 10.66 5.92 1.33 Psammothidium chlidanos 1.41 7.15 1.26 Nitzschia pura/palea 1.1 0.21 1.24 Pinnularia krammeri 0.32 0.3 1.22 Achnanthidium minutissimum 6.95 3.06 1.18 Chamaepinnularia soehrensis 6.27 0.02 1.17 Diploneis marginestriata 0.99 0 1.17 Navicula chiarae 3.1 0.61 1.09 Eucocconeis laevis 2 0 1.07 Rossithidium petersenni 1.83 0.45 1.07 Eunotia praerupta 0.51 0.5 1.06 Eucocconeis flexella 2.15 0.35 1.02 Encyonema silesiacum 0.57 0.64 1.01 Cymbopleura angustata var. sptizbergensis 0.87 0.54 0.97 Cymbella amphicephala 0.62 0.25 0.96 Caloneis silicula 0.65 0.24 0.95 Sellaphora bacillum 0.26 0.11 0.93 Fragilaria capucina 2.57 5.02 0.92 Navicula vulpina 0.61 0.67 0.91 Note: average dissimilarity between Zones 1 and 4 = 83.6 %
Table 5. Summary statistics of various weighted averaging models for pH with a) all species >1% relative abundance in three sites or >10% relative abundance in one site (n=90), b) all species with a significant response to pH (n=70), and c) all species with a significant unimodal response to pH (n=46). All values shown are cross-validated using bootstrapping techniques. “RMSEP” is the Root Mean Squared Error of Prediction. a) WAInv WACla WATol Inv WATol Cla
r2boot 0.344 0.368 0.432 0.446
Average bias boot -0.029 -0.039 -0.046 -0.055 RMSEP 0.350 0.372 0.341 0.351
b) WAInv WACla WATol Inv WATol Cla
r2boot 0.402 0.424 0.495 0.508
Average bias boot -0.030 -0.043 -0.046 -0.054 RMSEP 0.340 0.356 0.326 0.334
c) WAInv WACla WATol Inv WATol Cla
r2boot 0.738 0.742 0.746 0.748
Average bias boot -0.031 -0.037 -0.048 -0.057 RMSEP 0.230 0.237 0.242 0.246
CHAPTER 4
LIMNOLOGICAL CHARACTERISTICS OF A HIGH ARCTIC OASIS AND COMPARISONS ACROSS
NORTHERN ELLESMERE ISLAND
BRONWYN E. KEATLEY, MARIANNE S.V. DOUGLAS AND JOHN P. SMOL
This chapter is forthcoming in September 2007. Keatley, B.E., M.S.V. Douglas, and J.P. Smol. 2007. Limnological characteristics of a high Arctic oasis and comparisons across northern Ellesmere Island. Arctic, in press.
ABSTRACT
Rapidly warming temperatures in the Arctic are predicted to markedly alter the limnology of
tundra lakes and ponds. These changes include increases in aquatic production, pH, specific
conductivity, and nutrient levels. However, baseline limnological data from High Arctic regions
are typically restricted to single sampling events or to repeated samplings of a few select sites,
thereby limiting our ability to assess the influence of climatic change. We employ two techniques
to examine the influence of a warmer climate on High Arctic aquatic ecosystems. First we
compare limnological characteristics during July 2003 from 23 ponds and lakes from an
atypically warm High Arctic oasis on Ellesmere Island, to 32 ponds and lakes located across
northern Ellesmere Island, where climatic conditions are much cooler and more typical of High
Arctic environments. Second, we resample 13 sites originally analysed in 1963 to assess the
influence that 40 years of rising temperatures (as documented by meteorological records) have
had on the limnological characteristics of these freshwater ecosystems. The specific conductivity
values, as well as the concentrations of nutrients and related variables (especially dissolved
organic carbon, DOC), from the Arctic oasis sites are amongst the highest yet reported from the
Canadian High Arctic, and are significantly higher than those from the polar desert around
northern Ellesmere Island. Comparison of the modern and historical data indicated that most
oasis sites currently have higher pH than they did in 1963, consistent with documented warming
temperatures.
INTRODUCTION
The Canadian High Arctic is broadly classified as a polar desert due to the limited
precipitation and harsh annual climate of the region (Muc and Bliss, 1977). Given the vastness of
the High Arctic landscape, however, it is not surprising that climate is heterogeneous. Arctic
oases, regions of great biological production and diversity, are associated with greater availability
of local water sources relative to the surrounding polar desert and are generally found at small
scales (often less than 5 km2; Edlund and Alt, 1989). In the Canadian High Arctic, oases have
been identified on Devon Island (including Truelove Lowland, Bliss, 1977a), and on Ellesmere
Island including Eureka (Edlund and Alt, 1989), Alexandra Fiord (Freedman et al., 1994),
Tanquary Fiord and Lake Hazen (Edlund and Alt, 1989). Similar areas occur at Polar Bear Pass
on Bathurst Island, at Sherard Bay on Melville Island, and at Mould Bay on Prince Patrick Island
(Aiken et al., 1999 onwards). However, even among Arctic oases, the oasis of our study area at
Lake Hazen is strikingly warm and lush, particularly given its extreme location north of latitude
80°N.
Arctic oases are of particular interest to ecologists examining the effects of recent
climatic changes as they represent a glimpse of what the more typical polar desert ecosystems
might become under a warmer climate. By assessing the biological, physical and chemical
processes occurring in Arctic oases, we may better recognize the effects of climate change in
other Arctic regions. Because of their ecological importance and their uniqueness in the High
Arctic, Arctic oases have been relatively well-studied compared to their polar desert counterparts.
For example, terrestrial faunal (Bliss, 1977b; France, 1993) and botanical surveys (Muc and
Bliss, 1977; Soper and Powell, 1985; Henry et al., 1990) have been reported from Lake Hazen,
Truelove Lowland, and Alexandra Fiord (botanical only). However, aquatic biological research
from Arctic oases has largely been limited to a few lakes in the Lake Hazen area (zooplankton,
McLaren, 1964; non-diatom algae, Croasdale, 1973; cyanobacteria, Quesada et al., 1999) and to
three lakes at Truelove Lowland (Minns, 1977).
While Arctic oases are largely defined as regions of greater biological production and
diversity, little is known about the baseline limnological conditions that characterize lakes and
ponds from these regions. For example, limited limnological investigations were undertaken on
Truelove Lowland (Minns, 1977), and across northern Ellesmere Island (Hamilton et al. 1994;
Hamilton et al., 2001), including some sites in the oasis at Lake Hazen. More recent aquatic
work on dissolved organic carbon (DOC) and ultra violet (UV) penetration has been conducted
on Skeleton Lake in the Hazen oasis (Laurion et al., 1997). Also near Lake Hazen, a physical and
chemical limnological survey of ponds and lakes was carried out by the Department of Defence
Research Board (DRB, Canada) in 1963, with some additional observations in 1964 (Oliver and
Corbet, 1966). This valuable dataset includes seasonal measurements of important limnological
variables such as pH, specific conductivity, and major ions, but does not provide comparison data
to aquatic systems at similar latitudes outside of the Arctic oasis zone. Nonetheless, this early
1960’s dataset provides important reference data to assess whether these sites have changed over
the past ~40 years, a time of documented climate change in northern Ellesmere Island
(Environment Canada, 2004).
Excluding the oasis region of Lake Hazen on northern Ellesmere Island, previous
limnological survey data have been provided for aquatic systems near Alert, Ellesmere Island
(Antoniades et al. 2003a). Basic limnological data have also been provided for some lakes to the
south of Lake Hazen (Smith, 2002). In addition, detailed limnological analyses have been
undertaken in complex lakes along the northern coast of Ellesmere Island (Gibson et al., 2002;
Van Hove et al., 2006).
Our primary objective in this study is to characterize present-day limnological
characteristics from lakes and ponds on northern Ellesmere Island, including a large number of
sites located within a warm oasis region. Warm conditions have been linked to reduced ice cover,
longer growing seasons, higher pH and conductivity, and enhanced biological production (e.g.,
Douglas and Smol, 1999; Antoniades et al. 2005; Smol et al., 2005). However, these hypotheses
have not yet been tested from sites located on similar bedrock and at comparable latitudes. Hence
our goals are three-fold: 1) to provide baseline limnological data from sites located across
northern Ellesmere Island, both within and outside an Arctic oasis, and to compare these to other
Arctic regions; 2) to examine the hypothesis that oasis sites will have different limnological
characteristics than sites located outside the oasis; and 3) to assess differences between water
chemistry data from 1963 and 2003 from selected oasis sites.
METHODS
Site description
Our sampling regime took place on northern Ellesmere Island, largely, but not
exclusively, within Quttinirpaaq National Park (Fig. 1). Three physiographic regions exist within
the Park: the Grant Land Mountains covering 65% of the Park in the north, the Lake Hazen Basin
surrounding Lake Hazen, and the Hazen Plateau located between Lake Hazen and the southern
edge of Quttinirpaaq National Park (Bednarski, 1994). Four climatic zones can also be delineated
within the Park: 1) a cool marine climate in the northern coastal areas, 2) very cool regions
characterized by high elevation ice caps, 3) a marine climate in the south-eastern portion, and 4) a
continental climate of Lake Hazen and Tanquary Fiord (Thompson, 1994). The north coast
receives the greatest amount of precipitation, and the areas near Lake Hazen receive the least
(Thompson, 1994).
The Hazen Basin region experiences anomalously warm summer conditions due to its
continental location and its placement on the leeward side of the Grant Land Mountains (Gray,
1994). While average July daily temperatures at Eureka and Alert are 5.7˚C and 3.3˚C (1971-
2000 averages), respectively (Environment Canada, 2004), temperatures at the Lake Hazen camp
during our field work in July 2003 reached an average daily maximum of 16˚C with a minimum
as high as 9.6˚C. Average annual precipitation is 75.5 mm at Eureka (1971-2000) and153.8 mm
at Alert (Environment Canada, 2004). The summer melt periods are shortest for the north coast
(~3 weeks), last around 8 weeks near Alert, and last about 10 weeks at Lake Hazen (Thompson,
1994).
When defined by bioclimatic zone, the Lake Hazen region falls in Zone 4 (Edlund and
Alt, 1989), the most diverse botanical region in the high Arctic, which is dominated by shrubs
and sedges, and includes a large number of species (>100) that are typical of more southerly
Arctic locations (Edlund and Alt, 1989). Within the Lake Hazen oasis, however, there are also
some mountain sites that we consider ‘controls’ due to their relatively high elevation and lack of
catchment vegetation. Outside the oasis, study sites are located within a broad range of
vegetation zones, from low diversity Zone 0 sites (unvegetated) to Zone 3 sites (60-100 taxa,
prostrate shrub zone, dominated by Salix arctica and/or Dryas integrifolia; Edlund and Alt,
1989).
Geology
Northern Ellesmere Island is largely underlain by sandstones, limestones, and slates
(Christie, 1957; Christie, 1964). In the most northerly regions along the north coast, Precambrian
gneisses, schists and granitic rock dominate, while volcanic and sedimentary rocks, including
sandstones and limestones, underlie the northern interior regions (Christie, 1964). The north shore
of Lake Hazen, including the Hazen oasis, is composed of Permian, Triassic, Jura-Cretaceous and
Cenozoic sandstone and shale (Christie, 1964).
Sampling techniques
In July 2003, 55 ponds (< 2 m deep) and lakes (> 2 m deep) were sampled around
northern Ellesmere Island (Fig. 1). Of these, 23 sites were located in the Arctic oasis
immediately north of Lake Hazen, hereafter referred to as “oasis sites” and given unofficial
names EP1 through EP24. It should be noted that EP19 is Lake Hazen, and is kept separate from
all analyses due to its very large size (i.e. surface area ~ 54200 ha). Three of these sites (EP22,
23, 24) were located at relatively high elevations of > 850 m above sea level. Therefore, despite
their location in the warm oasis region, they serve as cooler controls within the oasis set. The
remaining 31 sites were selected from around the northern half of Ellesmere Island, to the north,
east, south and west of Lake Hazen (hereafter referred to as the “northern sites”, and given
unofficial names EPA through EPAE).
For each site, latitude, longitude, and elevation measurements were taken using a
handheld global positioning unit, the helicopter altimeter, and topographical maps. Water
temperature was recorded with a hand-held thermometer, and samples for total phosphorus
(unfiltered, TPu), trace metals (aluminum, Al; beryllium, Be; cadmium, Cd; chromium, Cr;
cobalt, Co; copper, Cu; iron, Fe; lead, Pb;, manganese, Mn; molybdenum, Mo; nickel, Ni;
vanadium, V; zinc, Zn; and silver, Ag), and major ions (calcium, Ca; magnesium, Mg; sodium,
Na; potassium, K; chloride, Cl; sulphate SO4) were retrieved using a pre-cleaned 125 mL sample
bottles from ~15 cm depth from the near shore area of each site. We have used identical
sampling techniques and analyses, as well as a similar time frame, to our previous limnological
investigations, allowing us to make comparisons amongst regions (Douglas and Smol, 1994; Lim
et al., 2001; Michelutti et al., 2002a; Michelutti et al., 2002b; Lim and Douglas, 2003; Antoniades
et al., 2003a; Antoniades et al., 2003b; Lim et al., 2005). A detailed description of the
methodologies used for water sampling are given in Appendix 1.
Additional water samples for pH, specific conductivity, filtered nutrients and related
variables (dissolved silica, SiO2; total phosphorus filtered, TPf; soluble reactive phosphorus, SRP;
nitrate-nitrogen, NO3-N; nitrate-nitrite-nitrogen, NO3-NO2-N; ammonia-nitrogen, NH3-N; total
Kjeldahl nitrogen (filtered), TKN; total dissolved nitrogen, TdN; particulate nitrogen, PON;
dissolved organic carbon, DOC; dissolved inorganic carbon, DIC; particulate carbon, POC; and
chlorophyll a, Chla) were taken with 1 L plastic Nalgene® bottles, rinsed three times with
pond/lake water. At base camp, pH and specific conductivity were measured the same day the
samples were obtained using a handheld Hanna pHep 3 meter and a YSI model 33 conductivity
meter, respectively. The dissolved and particulate fractions of the variables described above were
filtered on site following Environment Canada (1994) and details are given in Appendix 1. All
other analyses were performed at the National Water Research Institute (NWRI) in Burlington,
Ontario (Environment Canada), using protocols described in Environment Canada (1994).
Statistical analyses
Data were visually screened to assess normality of distribution using CALIBRATE 1.0
(Juggins and ter Braak, 1992). Any variables that were not normally distributed were
transformed using mostly logX, logX+1 or square root transformations. Variables whose
distributions could not be normalised were run passively in statistical analyses (i.e. they were
plotted onto the biplot after it was produced, and thus did not affect the results). A Pearson
correlation matrix with Bonferroni-adjusted probabilities was performed on the full dataset to
remove those variables which were highly correlated with each other, thereby reducing the
dataset to a more manageable size for ordination analyses.
A Principal Components Analysis (PCA) was run on the reduced dataset (by removing
highly correlated variables) to assess the important limnological gradients in the dataset using the
ordination program CANOCO 4.5 (ter Braak and Šmilauer, 2002).
Canonical Variates Analysis (CVA, also known as linear discriminant analysis), was used
to identify environmental variables that significantly discriminate between clusters of samples (in
this case, our oasis and northern sites) (Lepš and Šmilauer, 2003). Initially, a CVA was run for
each individual variable to assess whether it explained a significant portion of the variation
distinguishing the two groups. Any significant variables were retained. Next, highly correlated
variables were removed, according to the same variables that were kept in the PCA, and another
CVA was performed using forward selection to sequentially choose the most important
explanatory variables, given the presence of the other variables.
Comparison to historical data
The DRB water sampling of sites around Lake Hazen (Oliver and Corbet, 1966),
provides the earliest historical limnological survey data available in the Canadian high Arctic, and
thus provides a unique opportunity to assess changes in water chemistry on a regional scale over
40 years. Using site descriptions and locations from the DRB map, a subset of sites common to
both our study and the DRB study were identified. While we acknowledge that direct
comparisons of pH, specific conductivity and major ion concentrations are difficult due to
differences in both measurement techniques and seasonal sampling dates, we nonetheless make
use of this valuable historical dataset by providing a brief comparison.
RESULTS & DISCUSSION
Physical characteristics
The oasis sites consisted of 19 ponds and 4 small lakes (EP1, EP2, EP3, EP24; median
surface area (SA)oasis = 0.13 hectares). In contrast, less than one-third of the northern sites were
ponds (9 out of 31, median SAnorthern = 6 hectares). As would be expected based on their location
in the oasis and their smaller sizes, the oasis sites were much warmer (mean temp. = 15.7°C) than
the northern sites (mean temp. = 9.1°C). There was no significant elevational differences
between the two groups (meanoasis= 318 m, meannorthern = 289 m).
pH, specific conductivity, and major ions
The oasis and northern site were not significantly different with respect to pH values
(meanoasis = 8.23, meannorthern = 8.20, Tables 1, 2), and were similar to mean pH values elsewhere
in the Canadian Arctic including Devon (Lim and Douglas, 2003) and Bathurst (Lim et al., 2001)
islands, as well as Alert, Ellesmere Island (Antoniades et al., 2003a). The similar pH between
both our two groups of sites and between our study and those from previous surveys (Lim et al.,
2001; Antoniades et al. 2003a; Lim and Douglas, 2003) likely reflects the broadly similar bedrock
common to most of the sites.
Specific conductivity was significantly higher in the oasis sites (mean = 490 μS/cm) than
in the northern sites (mean = 245 μS/cm) (p = 0.022, Table 1, 2). Previous high Arctic
limnological surveys have reported mean specific conductivity ranging from ~100 μS/cm
(Victoria Island, Michelutti et al., 2002a; Bathurst Island, Lim and Douglas, 2003) to up to 405
μS/cm (Ellef Ringnes Island, Antoniades et al., 2003b), although specific conductivities >300
μS/cm generally reflect the influence of sea spray on coastal lakes and ponds (Michelutti et al.,
2002b, Antoniades et al., 2003b). While some of our northern sites included coastal ponds, all
our oasis sites are located inland and thus sea spray cannot be a factor for these elevated specific
conductivity values. In some sites, very high SO4 values contribute to high conductivity both in
the oasis (EP9, a very shallow site) and the northern (EPY, a small coastal site with gypsum
precipitates) datasets (Table 1a, b). Both these sites also had high Ca concentrations, suggesting
that local bedrock may have been important in influencing these values, as Ca and SO4 are known
to be very high in gypsiferous shale (McNeely et al., 1979).
Higher specific conductivities would, however, also be expected in smaller water bodies
under warmer conditions as increased evaporation would cause greater concentration of solutes in
the water column. During the summer months, prolonged solar radiation combined with clear
skies and warm temperatures characteristic of the Lake Hazen basin could result in enhanced
evaporation, further concentrating the solutes within the lakes and ponds. Although we do not
have seasonal data from our field season, previous work at Lake Hazen documented an average
drop in water levels by ~0.4 cm/day throughout the ice-free season (Oliver and Corbet, 1966).
This appears to be the case in our oasis sites; indeed, the subset of cool, poorly vegetated, high
elevation sites within the oasis region (EP22, 23, 24) had much lower conductivities (mean = 84
μS/cm) than the remaining low elevation oasis sites.
Concentrations of major ions (Ca, K, Mg, Na, SO4) were typically greater in oasis sites,
with K being significantly higher (meanoasis = 7.0 mg/L, meannorthern = 1.6 mg/L). Average K
concentrations elsewhere in the high Arctic range from 0.24 mg/L (Victoria Island; Michelutti et
al., 2002a) to 4.6 mg/L (Axel Heiberg Island; Michelutti et al., 2002b). Non-marine derived K is
often associated with exudates from plants (Prentki et al., 1980). As previously discussed, the
inland location of the oasis points to a terrestrial source of K; thus the relatively high
concentrations of K are likely indicative of the more highly vegetated catchments common in the
oasis. Indeed, our high elevation sites were distinctive in that they had an average K
concentration of 0.48 mg/L, less than 6% of that found in the oasis sites. More specifically, Na:K
ratios less than 2:1 may reflect enhanced terrestrial production (McNeely et al., 1979). Therefore,
the low ratio of Na:K (Na:Koasis = 1.5, Na:Knorthern = 3.8) in the oasis sites are likely indicative of
the more developed catchment vegetation. Once again, our high elevation oasis sites had
relatively higher Na:K (1.9) than the other oasis sites, reflecting the sparseness of catchment
vegetation. Average Na:K previously reported from across the Canadian Arctic range from 1.8
(Victoria Island; Michelutti et al. 2002a) to 18.4 (Alert; Antoniades et al., 2003a).
Nutrients and related variables
As expected, nutrients (TPu, TPf, SRP, TdN, TKN) and related variables (DOC, POC,
PON, SiO2) were significantly higher (p<0.05) in the oasis sites than in the northern sites (Table
1). When we compared only the ponds, most nitrogen fractions, as well as DOC and SiO2 were
significantly higher in the oasis. These high concentrations of TP and TdN in the oasis sites
indeed suggest that warmer conditions enhance nutrient export from the catchment into the lake
or pond. NH3 and Chla were not significantly different between zones.
TPu values for oasis sites (meanoasis = 11.3 μg/L) were most similar to those reported
from more southerly locations including Banks Island (18 μg/L, Lim et al., 2005), Bathurst Island
(12.7 μg/L, Lim et al., 2001), and Mould Bay, Prince Patrick Island (16.1 μg/L, Antoniades et al.,
2003a), the latter two of which include sites identified by Aiken et al. (1999 onwards) as potential
polar oases. Banks Island includes freshwater environments that occur in low, mid and high
Arctic ecozones, and Banks Island itself is one of the lushest islands in the Arctic Archipelago
(Lim et al., 2005). These relatively high TPu concentrations for the oasis sites once again are
indicative of their shared characteristics with other relatively warm, productive Arctic regions. It
should be noted, however, that while much higher TPu concentrations have been reported from
Arctic ponds and lakes than those we report for the oasis sites (see Lim et al., 2005 for a
summary), these have been attributed to sediment re-suspension rather than indicating high
production (Antoniades et al., 2003b).
When classified to trophic status based on TPu values (Wetzel, 1983), 48% of the oasis
sites were considered mesotrophic (i.e. TPu 10-30 μg/L, Table 2). TPu concentrations of the
northern sites (meannorthern = 7 μg/L) were more typical of aquatic habitats in the polar desert at
Axel Heiberg Island (mean = 4 μg/L, Michelutti et al., 2002b), Victoria Island (mean = 1.3 μg/L,
Michelutti et al., 2002a) and the Haughton Crater, Devon Island (mean = 3.7 μg/L, Lim and
Douglas, 2003). Only 19% of northern sites were mesotrophic or above (Table 1). The TPu
concentrations of the high elevation oasis sites (mean = 6 μg/L) were, once again, much lower
than those of the oasis area as a whole and even lower than the mean of the northern sites. Based
on the Wetzel (1983) TPu classification, the high elevation northern sites fall into the ultra-
oligotrophic (i.e. TPu < 5 μg/L, EP23, EP24) or oligo-mesotrophic (i.e. TPu 5-10 μg/L, EP22)
category.
Likewise, total N (TN) values for the oasis sites (meanoasis= 1.14 mg/L) exceed the
previously reported averages for Arctic islands (see summary in Lim et al., 2005), but are closest
to those reported from the lush regions of Mould Bay (0.616 mg/L, Antoniades et al., 2003a) and
Banks Island (0.499 mg/L, Lim et al. 2005). The high elevation oasis sites have a mean TN
concentration of 0.206 mg/L; suggesting that these high elevation sites are more similar to the
northern sites (meannorthern = 0.330 mg/L) than to those located within the oasis.
Interestingly, the TN:TPu ratios of the two groups of sites do not differ greatly (TN:TPu
meanoasis = 98, meannorthern = 67) and both groups are clearly limited by P (Downing and
McCauley, 1992). However, when we examine TPu versus TN graphically, we see that there is
little relationship between the two variables in either the full dataset (graph not shown) or in the
northern sites alone (Fig. 2b), but a positive linear relationship between them in the oasis sites
(Fig. 2a). This finding suggests that, in the northern sites, different mechanisms control nitrogen
and phosphorus delivery to the aquatic ecosystems, but that in the oasis sites the cycles of these
nutrients are linked. It is probable that autochthonous production is higher in the oasis sites (e.g.
Quesada et al. 1999). Similar to other High Arctic limnological surveys (see below), there is no
relationship between either TN or TPu to Chla.
Concentrations of DOC in the oasis sites (meanoasis = 17.3 mg/L) exceed the highest
previously reported mean values which were 6.7 mg/L for Mould Bay and 6.1 mg/L for Banks
Island by more than two-fold. DOC concentrations for the northern sites are similar to averages
for most other Arctic limnological surveys (meannorthern = 3.4 mg/L). The subset of the high
elevation oasis sites had even lower DOC than the northern sites (mean = 2.3 mg/L). As most
DOC is derived from catchment vegetation and aquatic mosses, and as the vegetation is much
richer in the oasis than outside, it is not a surprising result. What is especially noteworthy,
however, is the unprecedentedly high DOC concentrations from the oasis sites. This likely
reflects a few ponds that could possibly be considered wetlands due to their very shallow depths
and the mosses, grasses, and sedges growing throughout them.
Previously reported mean SiO2 concentrations for High Arctic lakes and ponds have
ranged from 0.41 mg/L (Melville Island, Keatley et al., [2]; and Mould Bay, Antoniades et al.,
2003a) to 1.69 mg/L (Axel Heiberg Island, Michelutti et al. 2002b). In our study, the oasis sites
had average SiO2 concentrations of 5.35 mg/L, while the northern sites had an average of 1.47
mg/L. While both our zones have high SiO2 concentrations, likely reflective of the bedrock
geology, the oasis sites greatly exceed previously reported Canadian High Arctic SiO2
concentrations. This may be attributable to the increased action of weathering due to enhanced
run-off during late spring snowmelt under these warmer oasis conditions. In addition, because
our SiO2 measurements were taken from unfiltered water samples, the high SiO2 values may also
reflect increased abundance of siliceous algae within the water samples of the more productive
oasis sites.
It is hypothesized that warmer conditions will result in higher nutrients and related
variables (e.g., Douglas and Smol, 1999), and consequently higher biological production. While
terrestrial production was indeed high in the oasis sites, there was no significant difference
between the two zones with respect to Chla, our proxy for autochthonous phytoplanktonic
production (meanoasis = 0.6 μg/L, meannorthern = 0.5 μg/L). Likewise, there was no relationship
between Chla and either TN or TPu, regardless of whether we examine the two zones together or
separately, or whether we examine the oasis sites with or without the high elevation sites. Chla
concentrations have similarly borne little resemblance to other typical indicators of high
production (such as high P and N concentrations) in other Canadian High Arctic limnological
surveys (Michelutti et al., 2002a; Michelutti et al., 2002b; Antoniades et al., 2003a; Antoniades et
al., 2003b; Lim et al. 2005). This has been attributed to discrepancies between measuring Chla in
the water column, whereas most of the primary production occurs in the periphytic habitat
(Vezina and Vincent, 1997; Villeneuve et al., 2001, Bonilla et al. 2005). Indeed, Quesada et al.
(1999), working on some of the same lakes (e.g. Skeleton Lake), have documented some of the
highest standing stocks of phytobenthos yet measured in polar regions.
Statistical results
The PCA ordination biplot of all sites (Fig. 3) indicates two main directions of variation
in the measured environmental data: axis 1 includes nutrients and related variables (TPu, TPf,
DOC, TdN, SiO2) as well as conductivity and major ions, and explains 52.9% of the variation in
the sites. Meanwhile, axis 2 represents a trace metal gradient and explains 16% of the variation
(Fig. 3). For the sake of clarity in the ordination plot (Fig. 3), we have chosen to remove some
highly correlated variables based on the Pearson correlation matrix (Table 3). For example, SiO2
has replaced the highly correlated variables of POC and PON, TdN represents both TKN and
TdN, and the metals U, V, Zn, Co, Cr, Be, Mg, Mn have been removed. The following
ecologically important variables could not be normalized and thus were plotted passively in the
ordination (Chla, DIC, K, SO4, Cl), along with the geograpahical variables (elevation, latitude,
longitude, temperature).
As expected, the oasis sites plot closer to each other than to the northern sites (Fig. 3),
and most of these lie along the left end of axis 1. This once again indicates that conductivity and
nutrients and related variables seem to distinguish the oasis sites even in the presence of all other
measured limnological variables. Some exceptions to this general trend include high elevation
oasis sites (EP22, 23, and 24) that were more dilute and less nutrient rich than most other oasis
sites (see above). These high elevation sites also had persistent ice cover and very little
vegetation in their catchments. The northern sites that plotted closest to our oasis sites on the
PCA (AE, E, H) tended to be small ponds with relatively rich vegetation when compared to the
rest of the northern sites. The two sites that plotted at the low end of axis 1 were the least
nutrient-rich and most dilute in the entire dataset were Lake Hazen, a very large lake, and EPO, a
pond located on top of a high mountain glacier with no vegetation, soil, or even rock in its
watershed.
In an attempt to quantitatively determine the main environmental gradients defining the
oasis and northern zones, a Canonical Variates Analysis (CVA) was performed to identify
environmental variables that could significantly discriminate between clusters of samples. Using
this method, only DOC explained a significant portion of the variation between the oasis and
northern sites (p = 0.001). However, DOC was also highly correlated to many nutrients and
related variables (including TPu, TPf, TdN, TKN, POC, and PON, Table 3, Fig. 3), and thus,
while DOC was the only significant variable retained in the analysis, it represents a number of
correlated water chemistry variables.
Historical data
Some of the sites we sampled at Lake Hazen had been part of a Defence Research Board
limnological study in 1963 (Oliver and Corbet, 1966). These historical data represent the earliest
available quantitative limnological data for the Canadian High Arctic. Instrumental temperature
records from Alert and Eureka, as well as proxy climate indicators from Alexandra Fiord
(Rayback and Henry, 2006) and glacier mass balance records from around north central
Ellesmere Island (Braun et al., 2004), indicate a relatively cool period in the 1960s compared to
the late-1990s and the early 21st century. Temperature records from the DND study indicate
average July 1963 temperatures of 6.6°C (Oliver and Corbet, 1966), compared to an average
temperature of 12.8°C during our field season in July 2003. Since limnological characteristics
such as pH and specific conductivity also change over the course of a growing season in High
Arctic lakes and ponds (Douglas and Smol, 1994), comparisons between the two datasets must be
made with caution. Nevertheless, there are no other Arctic regions with available water chemistry
data from the 1960s, and so a comparison, even at a basic level, is warranted.
Interestingly, in almost all sites, we see a slight increase in pH (Fig. 4a) in 2003 relative
to 1963. By examining the identical sites 40 years apart, we have removed any influence of
differences in geology. In our modern survey, recall that we did not record significant differences
between our pH values in the oasis and northern sites, and that this was likely because of the
over-riding influence of geology. By removing the influence of geology (i.e. resampling the
same sites), we may be more directly tracking limnological differences related to a longer
growing season that would be reflected in the warmer temperatures.
Specific conductivity showed no clear pattern between 2003 and 1963, but instead
appears to be related to sampling date (Fig. 4b). In general, specific conductivity is, not
surprisingly, much higher later in the growing season (Fig. 4b), although this pattern is not
without exception (see EP17, for example). Indeed, seasonal studies both elsewhere in the High
Arctic (Douglas and Smol, 1994), and during 1963 at the Hazen Camp (Oliver and Corbet, 1966)
have noted that specific conductivity increased in the majority of sites over the course of the
summer season due to evaporation. During the 1963 study at Hazen Camp, conductivity
fluctuated over an average range of 500 μS/cm, with some ponds drying up completely over the
course of the ice-free season (Oliver and Corbet, 1966). In our modern comparison, we
conducted our field sampling within a short time window of less than two weeks, and thus we
largely removed the seasonal effect of changes in conductivity. Changes in precipitation regime
would also influence conductivity. Although there has been a significant increase in total annual
precipitation at Eureka, there has been no clear trend in annual precipitation at Alert, the closest
meteorological station, over the last 50 years (Environment Canada, 2007). Concentrations of K
and SiO2 are both higher in most sites in 2003 compared to 1963 (Fig. 4c,d), but Ca, Mg, Na, Cl,
and SO4 all show complex patterns that are similar to those found with conductivity (data not
shown).
SUMMARY & CONCLUSIONS
We provide a limnological survey of aquatic habitats located throughout the diverse landscape of
northern Ellesmere Island and compare these to other High Arctic limnological surveys. The
concentrations of nutrients and DOC reported from the oasis ponds and lakes are amongst, and in
some cases are, the highest yet reported from the Canadian High Arctic. The oasis sites at Hazen
Camp are more similar to oasis sites located at Mould Bay, Prince Patrick Island, Banks Island,
and Bathurst Island (many hundreds of kilometres to the southwest) than to those located within a
few hundred kilometres on Ellesmere Island. Meanwhile, the northern Ellesmere lakes and ponds
from our dataset are more similar to those located within the polar deserts of Alert, Axel Heiberg
Island, and Devon Island.
We compared point samples of limnological characteristics between aquatic habitats
located within an Arctic oasis at Hazen Camp to those located outside this oasis area to determine
if these smaller, warmer water bodies had higher specific conductivity and increased nutrient
concentrations. Our comparisons indicate that smaller sites located in warmer and more lushly
vegetated Arctic regions have distinctive water chemistry, particularly with respect to nutrients
and related variables. In our dataset, these higher concentrations of nutrients and related variables
(particularly DOC and correlated variables) were significant despite differences in latitude,
elevation, and surface area between the oasis and northern sites. Interestingly, the three high
elevation oasis ponds were more similar to the polar desert sites than to the other Arctic oasis
ponds with respect to specific conductivity, and nutrients and related variables.
A comparison of water chemistry from a subset of the oasis sites that were first examined
in 1963 to data we collected in 2003 showed that most sites had higher pH in 2003 than they did
in 1963, consistent with documented warming temperatures. Comparisons of specific
conductivity, however, appear to be more related to sampling date.
In summary, aquatic ecosystems in this Arctic oasis have distinct water chemistry from
those located in the near-by polar desert. We associate this difference to increased catchment
vegetation, greater run-off from the watershed, and enhanced evaporation, all of which can be
linked to the warmer temperatures of the oasis. Under a continued Arctic warming scenario, our
results may represent a preview of how other Arctic freshwater systems might change.
ACKNOWLEDGEMENTS
This project was supported by NSERC grants to BEK, MSVD, and JPS. We thank PCSP for
logistical and field support, NSTP for a field research grant to BEK, and Parks Canada for
allowing us to use the Parks Canada base camp at Lake Hazen. Field sampling assistance was
also provided by S. Arnott. We thank A. Poulain, K. Rühland, N. Michelutti, and W. Vincent, as
well as two anonymous journal reviewers for comments on the manuscript. This is PCSP
contribution number 012-07.
REFERENCES AIKEN, S.G., DALLWITZ, M.G., CONSAUL, L.L., MCJANNET, C.L., BOLES, R.L., ARGUS, G.W., GILLETT, J.M., SCOTT, P.J., ELVEN, R., LEBLANC, M.C., GILLESPIE, L.J., BRYSTING, A.K., SOLSTAD, H., and HARRIS, J.G. 1999 onwards. Flora of the Canadian Arctic Archipelago: Descriptions, illustrations, identification, and information retrieval. Version: 1 October 2005. http://www.mun.ca/biology/delta/arcticf/. Accessed 10 March 2006. ANTONIADES, D., DOUGLAS, M.S.V., and SMOL, J.P. 2003a. Comparative limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick Island, NWT). Archiv für Hydrobiologie 158: 485-516. ANTONIADES, D., DOUGLAS, M.S.V., and SMOL, J.P. 2003b. Physical and chemical limnology of 24 ponds and one lake from Isachsen, Ellef Ringnes Island, Canadian High Arctic. International Review of Hydrobiology 88: 519-538. ANTONIADES, D., DOUGLAS, M.S.V., and SMOL, J.P. 2005. Quantitative estimates of environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. Journal of Paleolimnology 33: 349-360. BEDNARSKI, J. 1994. Geomorphology. In: Resource description and analysis: Ellesmere Island, National Park Reserve. National Resource Conservation Section, Prairie and Northern Region, Parks Canada, Department of Canadian Heritage, Winnipeg. 124 pp. BLISS, L.C. 1977a. Introduction. In: Bliss, L.C., ed. Truelove Lowland, Devon Island, Canada: A High Arctic ecosystem. Edmonton: University of Alberta Press. 1-12. BLISS, L.C. 1977b. Truelove Lowland, Devon Island, Canada: A High Arctic ecosystem. Edmonton: University of Alberta Press. 714 pp. BONILLA, S., VILLENEUVE, V., and VINCENT, W.F. 2005. Benthic and planktonic algal communities in a high Arctic lake: pigment structure and contrasting responses to nutrient enrichment. Journal of Phycology 41: 1120-1130. BRAUN, C., HARDY, D.R., and BRADLEY, R.S. 2004. Mass balance and area changes of four high Arctic plateau ice caps, 1959-2002. Geografiska Annaler Series A-Physical Geography 86A: 43-52. CHRISTIE, R.L. 1957. Geological reconnaissance of the north coast of Ellesmere Island, District of Franklin, Northwest Territories. Paper 56-9. Ottawa: Department of Mines and Technical Surveys. 40 pp. CHRISTIE, R.L. 1964. Geological reconnaissance of northeastern Ellesmere Island, District of Franklin. Memoir 331. Ottawa: Geological Survey of Canada. 79 pp. CROASDALE, H. 1973. Freshwater algae of Ellesmere Island, N.W.T. Ottawa: National Museums of Canada. 131 pp. DOUGLAS, M.S.V., and SMOL, J.P. 1994. Limnology of high Arctic ponds (Cape Herschel, Ellesmere Island, N.W.T.). Archiv für Hydrobiologie 131: 410-434.
DOUGLAS, M.S.V., and SMOL, J.P. 1999. Freshwater diatoms as indicators of environmental change in the high Arctic. In: Stoermer, E.F. and Smol, J.P., eds. The diatoms: Applications for the environmental and earth sciences. Cambridge: Cambridge University Press. 227-244. DOWNING, J.A., and MCCAULEY, E. 1992. The nitrogen-phosphorus relationship in lakes. Limnology and Oceanography 37: 936-945. EDLUND, S.A., and ALT, B.T. 1989. Regional congruence of vegetation and summer climate patterns in the Queen Elizabeth Islands, Northwest Territories, Canada. Arctic 42: 3-23. ENVIRONMENT CANADA 1994. Manual of analytical methods – National Laboratory for Environmental Testing, Canadian Centre for Inland Waters. Burlington, Canada. 2 vols, 1097 pp. ENVIRONMENT CANADA, 2004. Canadian Climate Normals 1971-2000. Environment Canada. http://www.climate.weatheroffice.ec.gc.ca/climate_normals/index_e.html, Accessed 2 April 2004. ENVIRONMENT CANADA, 2007. Climate data online. http://www.climate.weatheroffice.ec.gc.ca/climateData/canada_e.html Accessed 23 February 2007. FRANCE, R.L. 1993. The Lake Hazen trough - A late winter oasis in a polar desert. Biological Conservation 63: 149-151. FREEDMAN, B., SVOBODA, J., and HENRY, G.H.R. 1994. Alexandra Fiord - An ecological oasis in the polar desert. In: Svoboda, J. and Freedman, B. eds. Ecology of a polar oasis: Alexandra Fiord, Ellesmere Island, Canada. Toronto: Captus University Publications: 1-12. GIBSON, J.A.E., VINCENT, W.F., VAN HOVE, P., BELZILE, C., WANG, X., and MUIR, D. 2002. Geochemistry of ice-covered, meromictic Lake A in the Canadian High Arctic. Aquatic Geochemistry 8: 97-119. GRAY, D.R. 1994. Ecology. In: Resource description and analysis: Ellesmere Island, National Park Reserve. National Resource Conservation Section, Prairie and Northern Region, Parks Canada, Department of Canadian Heritage, Winnipeg. 72 pp. HAMILTON, P.B., LEAN, D.R.S., and POULIN, M. 1994. The physicochemical characteristics of lakes and ponds from the northern regions of Ellesmere Island. In: Hamilton, P.B. ed. Proceedings of the Fourth Arctic-Antarctic Diatom Workshop, Canadian Museum of Nature. Canadian Technical Report of Fisheries and Aquatic Sciences 1957: 57-63. HAMILTON, P.B., GAJEWSKI, K., ATKINSON, D.E., and LEAN, D.R.S. 2001. Physical and chemical limnology of 204 lakes from the Canadian Arctic Archipelago. Hydrobiologia 457: 133-148. HENRY, G.H.R., SVOBODA, J., and FREEDMAN, B. 1990. Standing crop and net production of sedge meadows of an ungrazed polar desert oasis. Canadian Journal of Botany 68: 2660-2667. JUGGINS, S. and TER BRAAK, C.J.F., 1992. CALIBRATE - a program for species-environment calibration by [weighted averaging] partial least squares regression. Environmental Change Research Centre, University College, London.
KEATLEY, B.E., DOUGLAS, M.S.V., and SMOL, J.P. [2]. Physical and chemical limnological characteristics across environmental gradients on Melville Island, Nunavut/N.W.T., High Arctic Canada. Fundamental and Applied Limnology. 168: 355-376. LAURION, I., VINCENT, W.F., and LEAN, D.R.S., 1997. Underwater ultraviolet radiation: development of spectral models for northern high latitude lakes. Photochemistry and Photobiology 65: 107-114. LEPŠ, J. and ŠMILAUER, P. 2003. Multivariate analysis of ecological data using CANOCO. London: Cambridge University Press. 282 pp. LIM, D.S.S., and DOUGLAS, M.S.V. 2003. Limnological characteristics of 22 lakes and ponds in the Haughton Crater region of Devon Island, Nunavut, Canadian High Arctic. Arctic Antarctic and Alpine Research 35: 509-519. LIM, D.S.S., DOUGLAS, M.S.V., SMOL, J.P., and LEAN, D.R.S. 2001. Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. International Review of Hydrobiology 86: 1-22. LIM, D.S.S., DOUGLAS, M.S.V., and SMOL, J.P. 2005. Limnology of 46 lakes and ponds on Banks Island, NWT, Canadian Arctic Archipelago. Hydrobiologia 545: 11-32. MCLAREN, I. A. 1964. Zooplankton of Lake Hazen Ellesmere Island and a nearby pond with special reference to copepod Cyclops scutifer Sars. Canadian Journal of Zoology 42: 613-629. MCNEELY, R.N., NEIMANIS, V.P., and DWYER, L. 1979. Water quality sourcebook: A guide to water quality parameters. Ottawa: Minister of Supply and Services Canada. 88 pp. MICHELUTTI, N., DOUGLAS, M.S.V., LEAN, D.R.S., and SMOL, J.P. 2002a. Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Island, Arctic Canada. Hydrobiologia 482: 1-13. MICHELUTTI, N., DOUGLAS, M.S.V., MUIR, D.C.G., WANG, X., and SMOL, J.P. 2002b. Limnological characteristics of 38 lakes and ponds on Axel Heiberg Island, High Arctic Canada. International Review of Hydrobiology 87: 385-399. MINNS, C. K. 1977. Limnology of some lakes on Truelove Lowland. In: Bliss, L.C. ed. Truelove Lowland, Devon Island, Canada: A High Arctic ecosystem. Edmonton: University of Alberta Press. 567-586. MUC, M., and BLISS, L.C. 1977. Plant communities of Truelove Lowland. In: Bliss, L.C. ed. Truelove Lowland, Devon Island, Canada: A High Arctic ecosystem. Edmonton: University of Alberta Press. 143-154. OLIVER, D. R., and CORBET, P.S. 1966. Aquatic habitats in a high arctic locality: the Hazen Camp study area, Ellesmere Island, N.W.T. Ottawa: Department of National Defence. 115 pp. PRENTKI, R.T., MILLER, M.C., BARSDATE, R.J., ALEXANDER, V., KELLY, J., and COYNE, P. 1980. Chemistry. In: Hobbie, J.E. ed. Limnology of tundra ponds, Barrow Alaska. Stroudsburg: Dowden, Hutchinson and Ross, Inc., 76-178.
QUESADA, A., VINCENT, W.F., and LEAN, D.R.S. 1999. Community and pigment structure of Arctic cyanobacterial assemblages: the occurrence and distribution of UV-absorbing compounds. FEMS Microbiology Ecology 28: 315-323. RAYBACK, S.A., and HENRY, G.H.R. 2006. Reconstruction of summer temperature for a Canadian High Arctic site from retrospective analysis of the dwarf shrub, Cassiope tetragona. Arctic Antarctic and Alpine Research 38: 228-238. SMITH, I.R. 2002. Diatom-based holocene paleoenvironmental records from continental sites on northeastern Ellesmere Island, high Arctic, Canada. Journal of Paleolimnology 27: 9-28. SMOL, J.P., WOLFE, A.P., BIRKS, H.J.B., DOUGLAS, M.S.V., JONES, V. J., KORHOLA, A., PIENITZ, R., RÜHLAND, K., SORVARI, S., ANTONIADES, D., BROOKS, S.J., FALLU, M.-A., HUGHES, M., KEATLEY, B.E., LAING, T.E., MICHELUTTI, N., NAZAROVA, L., NYMAN, M., PATERSON, A.M., PERREN, B., QUINLAN, R., RAUTIO, M., SAULNIER-TALBOT, E., SIITONENI, S., SOLOVIEVA, N., and WECKSTRÖM, J. 2005. Climate-driven regime shifts in the biological communities of arctic lakes. Proceedings of the National Academy of Sciences of the United States of America 102: 4397-4402. SOPER, J. H., and POWELL, J.M. 1985. Botanical studies in the Lake Hazen Region, northern Ellesmere Island, Northwest Territories, Canada. Ottawa: National Museums of Canada. 67 pp. TER BRAAK, C. J. F., and ŠMILAUER, P. 2002. CANOCO reference manual and CANOdraw for Windows user guide: Software for Canonical Community Ordination (Version 4.5). THOMPSON, W. 1994. Climate. In: Resource description and analysis: Ellesmere Island, National Park Reserve. National Resource Conservation Section, Prairie and Northern Region, Parks Canada, Department of Canadian Heritage, Winnipeg. 78 pp. VAN HOVE, P., BELZILE, C., GIBSON, J.A.E., and VINCENT, W.F. 2006. Coupled landscape-lake evolution in high Arctic Canada. Canadian Journal of Earth Sciences 43: 533-546. VEZINA, S., and VINCENT, W.F. 1997. Arctic cyanobacteria and limnological properties of their environment: Bylot Island, Northwest Territories, Canada (73 degrees N, 80 degrees W). Polar Biology 17: 523-534. VILLENEUVE, V., VINCENT, W.F., and KOMÁREK, J. 2001. Community structure and microhabitat characteristics of cyanobacterial mats in an extreme high Arctic environment: Ward Hunt Lake. Nova Hedwigia 123: 199-224. WETZEL, R. G., 1983. Limnology 2nd Edition. Philadephia: Saunders Publishing, Philadephia. 767 pp.
Figure captions. Figure 1. Location map of northern Ellesmere Island. Inset a) indicates Ellesmere Island within Canada. Inset b) shows the northern sites around Ellesmere Island. The dashed black line denotes the boundary of Quttinirpaaq National Park and the patterned areas within this boundary represent different climate regions based on Thompson (1994). The black star indicates the location of the oasis sites detailed in inset c). Inset c) details the oasis sites just north of Lake Hazen. Figure 2. Plot of total phosphorus unfiltered (TPu) versus total nitrogen (TN) in a) the oasis sites, and b) the northern sites. While there is little relationship between TPu and TN in the northern region, there is a clear positive relationship with TPu and TN in the oasis sites, suggesting that different factors control nutrient cycling within the two regions. Figure 3. Biplot of a principal components analysis (PCA) of measured limnological variables for all sites. Oasis sites are represented by filled circles and northern sites are represented by open circles. Lake Hazen is kept separate due to its extremely large size and is represented by a star. Axis 1 most closely represents nutrients and related variables, pH, and conductivity, and explains 52.9% of the variance in the dataset. Axis 2 most closely represents a gradient of metals and explains 16% of the variance in the dataset. The dashed lines represent variables that were run passively in the ordination. Figure 4. Histograms indicating the change in the values of selected limnological variables in 2003 relative to 1963 for a) pH, b) specific conductivity, c) K, and d) SiO2. Site names with * indicate sites for which identification was approximate.
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Table 1. Summary of selected limnological variables for the northern sites and Lake Hazen. Lake Hazen is isolated at the bottom due to its extremely large size. All sites were sampled in July 2003, specific days are given in the “Date” column. Full details for other limnological parameters (e.g. metals, rare earth elements) are available in Appendix 6. Non-standard abbreviations are as follows: elevation (elev, given in meters above sea level) and surface area (SA). Abbreviations for other parameters are given in the text.
ID Name Latitude Longitude Date elev SA Pond or pH condN W m ha Lake μS/cm
EPA Craig 81˚50.34’ 68˚51.13’ 12 152 1455.0 L 8.47 296EPB Appleby 81˚50.90’ 68˚15.49’ 12 366 48.8 L 8.23 232EPC Brainard 81˚45.81’ 68˚10.32’ 12 640 18.0 L 8.03 169EPD 81˚46.78’ 64˚32.96’ 12 274 49.1 L? 8.37 172EPE 81˚42.84’ 70˚05.33’ 12 259 1.1 P 9.00 379EPF 82˚25.20’ 68˚12.77’ 12 91 125.0 L 8.13 110EPG 82˚36.21’ 68˚12.25’ 12 91 14.7 L 8.20 122EPH 82˚36.23’ 68˚13.03’ 12 91 0.4 P 8.63 322EPI 82˚55.11’ 66˚51.88’ 12 15 2.4 P 8.53 75EPJ Ward Hunt 83˚05.30’ 74˚09.87’ 13 61 35.0 L 8.30 45
EPK Lake A 83˚00.74’ 75˚23.28’ 13 30 490.0 L 7.90 69EPL 82˚58.60’ 75˚24.70’ 13 46 99.0 L 8.27 56
EPM 82˚58.54’ 75˚11.44’ 13 213 0.3 P 8.63 102EPN Lake C2 82˚49.59’ 77˚56.24’ 13 30 165.0 L 7.73 30EPO 82˚16.00’ 77˚53.32’ 13 1006 6.3 L 7.47 45EPP 81˚36.17’ 73˚53.11’ 13 686 6.0 L? 8.17 102EPQ 81˚27.28’ 67˚22.21’ 17 579 2.4 P 8.87 138EPR 81˚18.71’ 65˚34.92’ 17 183 14.7 L 8.30 118EPS 81˚19.41’ 66˚25.08’ 17 335 2.6 L 8.33 131EPT 81˚47.53’ 70˚26.77’ 17 457 0.2 P 8.37 135EPU Carolyn 81˚17.96’ 70˚43.48’ 17 305 161.0 L 8.30 119EPV Nan 81˚13.15’ 72˚19.46’ 17 305 6.7 L 8.77 409EPW 81˚04.89’ 74˚20.67’ 17 518 114.0 L 8.30 109EPX 80˚55.91’ 76˚32.69’ 17 396 61.4 L 8.30 154EPY 80˚36.43’ 79˚42.96’ 17 1 0.1 P 8.23 1500EPZ 81˚00.50’ 78˚13.03’ 17 244 4.1 L 8.47 160
EPAA Kettle 81˚23.80’ 76˚47.14’ 18 200 7.7 L 8.73 500EPAB 81˚58.49’ 80˚04.14’ 18 122 0.6 L? 7.90 58EPAC 82˚05.44’ 81˚50.39’ 18 76 98.2 L 8.20 200EPAD 82˚05.79’ 82˚34.70’ 18 75 0.1 P 8.73 325EPAE 81˚42.78’ 82˚17.04’ 18 808 0.5 P 8.30 1200mean 279 96.5 8.20 245
median 213 7.7 8.30 135max 1006 1455.0 9.00 1500min 1 0.1 7.47 30
EP19 Lake Hazen 81˚49.37’ 71˚20.18’ 15 154 54200 L 7.73 68
Table 1. Continued.
ID Chla DOC DIC POC PON NH3 TKN TdN TN TPu TPfμg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L μg/L μg/L
EPA 0.05 3.1 31.4 0.204 0.030 0.016 0.179 0.191 0.216 13.7 2.3EPB 0.05 4.7 36.3 0.246 0.030 0.021 0.383 0.361 0.4155 6.7 2.2EPC 0.1 3.6 27.4 0.256 0.030 0.021 0.315 0.327 0.35 7.4 3.3EPD 0.7 3.3 24.4 0.231 0.029 0.036 0.287 0.293 0.3185 19.0 2.9EPE 0.6 12.6 47.2 0.509 0.047 0.031 0.774 0.752 0.826 8.5 5.8EPF 0.1 1.5 7 0.208 0.024 0.013 0.044 0.077 0.097 6.3 1.6EPG 0.1 0.8 14 0.403 0.039 0.008 0.054 0.095 0.14 2.9 2.1EPH 0.6 5.9 23.5 0.438 0.050 0.043 0.467 0.469 0.5195 10.9 9.8EPI 0.7 2.6 10.8 0.389 0.052 0.008 0.133 0.183 0.233 6.7 2.3EPJ 0.6 1 4.4 0.163 0.017 0.067 0.148 0.217 0.187 4.1 3.8
EPK 0.6 1.3 3.6 0.180 0.027 0.016 0.078 0.089 0.113 3.2 2.2EPL 0.6 1.4 8.5 0.187 0.026 0.044 0.208 0.178 0.249 4.1 3.0
EPM 1 4.3 16.1 0.345 0.042 0.040 0.378 0.405 0.4225 10.3 5.1EPN 0.6 2.4 2.8 0.145 0.021 0.008 0.049 0.067 0.0725 7.4 2.2EPO 0.5 0.7 3.2 0.083 0.016 0.027 0.065 0.128 0.134 1.8 1.4EPP 1.4 4.7 11.9 0.415 0.043 0.023 0.398 0.412 0.452 6.5 3.8EPQ 0.7 12.3 24.4 1.670 0.137 0.048 0.997 1.010 1.143 9.5 6.8EPR 0.05 1.1 17.9 0.357 0.021 0.007 0.056 0.064 0.089 2.4 1.6EPS 0.5 2.2 21.2 0.285 0.012 <0.005 0.068 0.107 0.118 2.1 1.5EPT 0.1 0.9 19.5 0.116 0.021 0.005 0.053 0.413 0.457 1.0 1.4EPU 0.6 0.9 19.3 0.324 0.038 0.005 0.040 0.057 0.098 2.1 1.0EPV 0.1 5 59.5 0.473 0.055 0.020 0.458 0.408 0.522 5.0 3.6EPW 0.05 0.6 14 0.343 0.042 0.030 0.084 0.159 0.205 2.5 1.3EPX 0.05 0.9 25.9 0.288 0.031 0.013 0.091 0.103 0.134 2.6 2.2EPY 0.2 2.2 15.5 0.611 0.059 0.035 0.234 0.159 0.2955 33.5 3.7EPZ 0.05 2.3 26.9 0.230 0.030 0.009 0.174 0.175 0.209 3.4 1.9
EPAA 0.5 4.8 23.3 0.292 0.042 0.043 0.403 0.395 0.453 11.7 4.7EPAB 1.6 1.7 4.8 0.221 0.027 0.005 0.080 0.079 0.1095 3.8 1.0EPAC 0.1 1 25.4 0.283 0.039 0.005 0.068 0.179 0.231 4.8 1.1EPAD 0.6 4.7 36.5 0.441 0.045 0.015 0.336 0.351 0.439 5.2 3.6EPAE 1.5 11.5 22.3 0.430 0.043 0.012 0.936 0.882 0.985 8.1 8.3mean 0.5 3.4 20.3 0.347 0.038 0.022 0.259 0.283 0.33 7.0 3.1
median 0.5 2.3 19.5 0.288 0.031 0.016 0.174 0.183 0.23 5.2 2.3max 1.6 12.6 59.5 1.670 0.137 0.067 0.997 1.010 1.14 33.5 9.8min 0.05 0.6 2.8 0.083 0.012 <0.005 0.040 0.057 0.07 1.0 1.0
EP19 0.05 1.1 9 0.156 0.016 0.0025 0.040 0.060 0.047 2.0 2.1
Table 1. Continued.
ID SRP SiO2 Ca K Na Mg Cl SO4 Al Feμg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L μg/L μg/L
EPA 0.5 2.61 44.70 2.44 4.28 17.90 2.39 56.20 25.1 56.0EPB 0.2 0.37 30.10 3.53 5.34 17.60 3.20 6.31 1.6 12.4EPC 0.6 0.73 26.50 1.03 1.61 8.16 0.96 5.14 3.0 17.3EPD 0.7 1.15 31.10 0.46 3.18 3.89 7.11 1.58 3.7 25.9EPE 0.8 3.21 19.10 9.75 11.20 46.70 15.10 52.40 7.3 36.1EPF 1.7 0.49 18.30 0.32 0.69 3.04 1.05 39.70 177.0 374.0EPG 0.6 0.88 20.50 0.26 0.20 5.76 0.29 20.70 30.3 48.2EPH 0.9 2.21 52.90 0.78 0.78 19.60 0.54 120.00 6.9 31.8EPI 1.9 0.47 12.30 0.11 0.42 3.23 1.00 4.29 2.7 18.9EPJ 0.7 0.16 5.53 0.07 0.23 0.73 0.50 0.27 3.8 5.8
EPK 0.1 0.21 4.89 0.41 7.83 1.99 14.70 2.86 12.8 23.2EPL 0.5 0.45 10.30 0.53 8.33 2.91 15.70 4.17 10.6 11.6
EPM 0.2 1.33 21.80 0.36 0.43 3.05 0.29 3.83 12.2 236.0EPN 1.0 0.20 3.21 0.07 0.72 0.69 1.24 1.51 6.9 12.1EPO 0.1 0.13 4.11 0.07 0.08 1.93 0.15 5.62 8.6 21.8EPP 0.6 3.96 19.40 0.69 0.92 1.95 1.50 3.72 69.4 528.0EPQ 1.6 4.80 30.00 0.12 0.52 5.60 0.36 0.30 20.3 486.0EPR 0.5 0.35 24.30 0.20 1.74 3.95 3.71 2.06 20.0 25.6EPS 0.5 0.49 28.30 0.17 0.38 4.85 1.22 5.38 13.6 12.7EPT 0.4 0.61 26.70 0.15 0.83 5.79 3.07 5.95 11.8 6.6EPU 0.2 1.12 19.90 0.30 0.60 7.44 0.74 3.54 35.2 60.1EPV 0.6 1.66 25.90 5.34 7.18 57.30 14.20 50.90 6.0 16.8EPW 0.2 0.57 18.70 0.18 0.16 4.99 0.30 7.50 52.1 90.0EPX 0.5 1.67 28.40 0.36 0.30 8.94 0.57 3.49 8.4 12.9EPY 3.1 3.62 451.00 1.80 17.30 27.90 29.40 1160.00 2.2 4.2EPZ 0.7 1.36 27.30 0.65 0.76 9.65 1.72 0.77 2.3 27.9
EPAA 0.7 0.18 30.00 8.48 25.60 41.70 32.40 166.00 14.0 45.8EPAB 0.6 0.36 6.36 0.16 0.17 0.94 0.17 1.29 4.0 12.9EPAC 0.4 1.26 33.90 0.48 1.54 12.00 2.06 26.00 26.5 47.9EPAD 1.1 3.32 38.70 0.68 1.17 27.80 1.03 68.10 73.1 201.0EPAE 2.5 5.69 248.00 9.97 4.26 76.00 13.20 827.00 18.4 62.2mean 0.8 1.47 43.94 1.61 3.51 14.00 5.48 85.70 22.3 83.0
median 0.6 0.88 25.90 0.41 0.83 5.76 1.24 5.38 11.8 25.9max 3.1 5.69 451.00 9.97 25.60 76.00 32.40 1160.00 177.0 528.0min 0.1 0.13 3.21 0.07 0.08 0.69 0.15 0.27 1.6 4.2
EP19 0.1 0.60 13.30 0.25 0.30 1.38 0.17 6.32 17.5 37.5
Table 1. Continued.
ID TN:TPU Na:K Field observations
EPA 15.77 1.75 75% ice coveredEPB 62.01 1.51 75% ice coveredEPC 47.30 1.56 80-90% ice covered, Nostoc ballsEPD 16.76 6.91 80% ice covered, EPE 97.18 1.15 organic crust, Nostoc sheetsEPF 15.40 2.16 glaciers at both ends, mostly ice covered, supsended silt,EPG 48.28 0.77 80-90% ice covered, EPH 47.66 1.00 no iceEPI 34.78 3.82 60% ice covered, tufts of green algae, Nostoc ballsEPJ 45.61 3.29 ~100% ice covered,
EPK 35.31 19.10 99% ice covered, Nostoc near shore, EPL 60.73 15.72 flows into EPK, ~100% ice covered,
EPM 41.02 1.19 light brownish algal mats floatingEPN 9.80 10.29 some bright green algal filaments, some NostocEPO 74.44 1.14 on top of a glacierEPP 69.54 1.33 very grassy catchmentEPQ 120.32 4.33 lots of evidence of animal activityEPR 37.08 8.70 75% ice coveredEPS 56.19 2.24 in valley between 2 mountains, green filamentous algae, inflow, desolateEPT 457.00 5.53 probably had a stream flowing through, but not when we sampled, EPU 46.67 2.00 90% ice coveredEPV 104.40 1.34 some foam on water, ice freeEPW 82.00 0.89 very desolate, 95% ice coveredEPX 51.54 0.83 some floating pancakes of ice, but mostly ice-free, copepods with eggsEPY 8.82 9.61 ~200 m from sea, small pond with white precipitate (gypsum?)EPZ 61.47 1.17 foam, mosses, some green algal filaments
EPAA 38.72 3.02 lots of zoops, flocculant sedsEPAB 28.82 1.06 50% ice covered, EPAC 48.13 3.21 30-40% ice covered, EPAD 84.42 1.72 seems to have had streams flowing in, but not when we sampledEPAE 121.60 0.43 loads of fairy shrimpmean 66.73 3.83
median 48.13 1.75max 457.00 19.10min 8.82 0.43
EP19 23.50 1.20
Tabl
e 2.
Sum
mar
y of
sele
cted
lim
nolo
gica
l var
iabl
es fo
r the
oas
is si
tes,
with
abb
revi
atio
ns a
s des
crib
e in
Tab
le 1
. P
valu
es re
pres
ent
the
resu
lts o
f t-te
sts (
assu
min
g un
equa
l var
ianc
e) b
etw
een
the
oasi
s and
nor
ther
n si
tes.
“N
S” in
dica
tes n
o si
gnifi
cant
diff
eren
ce.
Full
deta
ils fo
r oth
er li
mno
logi
cal p
aram
eter
s (e.
g. m
etal
s, ra
re e
arth
ele
men
ts) a
re a
vaila
ble
in A
ppen
dix
7.
ID
Nam
eL
atitu
deL
ongi
tude
Dat
eel
evSA
Pond
/pH
cond
Chl
aD
ICD
OC
POC
PON
NH
3
NW
mha
Lak
eμS
/cm
μg/L
mg/
Lm
g/L
mg/
Lm
g/L
mg/
LEP
1Sk
elet
on81˚4
9.79
8’71˚2
8.48
3’8
296
1.84
L8.
2017
50.
620
.75.
80.
529
0.08
40.
014
EP2
81˚4
9.84
5’71˚2
8.35
2’8
296
1.00
L8.
1718
00.
622
.84.
61.
240
0.25
90.
032
EP3
81˚4
9.88
4’71˚2
8.05
2’8
296
0.24
L8.
0718
70.
925
.94.
11.
490
0.33
40.
010
EP4
81˚5
0.72
1’71˚2
3.92
8’9
300
0.06
P8.
9323
51.
122
31.1
1.37
00.
131
0.14
5EP
581˚5
0.75
2’71˚2
3.84
9’9
297
0.16
P8.
7044
30.
132
.429
.62.
250
0.48
60.
063
EP6
81˚5
0.77
3’71˚2
4.81
5’9
250
0.06
P7.
9090
0.5
7.3
5.6
0.90
00.
168
0.00
9EP
781˚5
0.33
7’71˚2
0.06
0’9
230
1.40
P8.
4070
01.
531
.412
.50.
574
0.08
60.
054
EP8
81˚5
0.47
4’71˚1
9.18
7’9
220
0.02
P8.
4365
00.
742
.740
.11.
000
0.17
40.
034
EP9
81˚5
0.09
6’71˚1
8.53
9’9
210
0.34
P8.
4016
500.
0524
.535
.90.
485
0.04
20.
063
EP10
81˚4
8.96
5’71˚2
4.99
5’10
170
0.05
P8.
5036
20.
620
.38.
10.
449
0.05
60.
003
EP11
81˚4
8.65
0’71˚2
6.92
2’10
170
0.76
P8.
9017
21.
113
.79.
10.
511
0.05
70.
057
EP12
81˚4
8.59
0’71˚3
4.73
2’10
190
0.13
P8.
1713
000.
555
360.
693
0.07
50.
011
EP13
81˚4
8.71
0’71˚3
3.61
4’10
200
0.02
P8.
3735
50.
0542
.427
.21.
550
0.31
30.
010
EP14
81˚4
9.23
6’71˚3
2.27
9’10
210
0.01
P8.
8739
00.
0531
.331
.80.
847
0.08
20.
049
EP15
81˚4
9.89
9’71˚3
1.62
9’15
300
0.14
P8.
5356
00.
112
.29.
20.
769
0.14
50.
044
EP16
81˚4
9.73
7’71˚3
2.25
1’15
300
0.50
P8.
7310
001
2518
.20.
854
0.08
80.
027
EP17
81˚4
9.78
1’71˚3
0.85
5’15
300
0.05
P8.
5352
00.
0524
.824
.10.
538
0.04
90.
044
EP18
81˚4
9.30
6’71˚2
1.04
5’15
160
0.05
P8.
3347
00.
527
.413
.40.
482
0.05
50.
047
EP20
81˚4
9.53
3’71˚1
9.99
9’15
170
0.09
P8.
0388
01
24.5
21.6
0.62
60.
061
0.00
7EP
2181˚4
9.53
3’71˚1
9.99
9’15
170
0.07
P8.
3369
01.
126
.323
.30.
387
0.03
50.
023
EP22
81˚4
9.26
’71˚4
5.04
’16
853
1.60
P8.
4711
50.
69.
93.
90.
317
0.05
20.
003
EP23
81˚4
9.29
8’71
.45.
162’
1686
00.
11P
7.97
910.
058.
92.
30.
185
0.02
90.
018
EP24
81˚4
9.04
1’71˚4
6.88
6’16
870
2.80
L7.
5345
0.5
4.8
0.7
0.42
50.
077
0.00
9m
ean
318
0.50
8.23
490
0.6
24.2
17.3
0.80
30.
128
0.03
4m
edia
n25
00.
138.
4039
00.
624
.513
.40.
626
0.08
20.
027
max
870
2.80
8.93
1650
1.5
5540
.12.
250
0.48
60.
145
min
160
0.01
7.53
450.
054.
80.
70.
185
0.02
90.
003
P va
lue
NS
0.02
2N
SN
S0
00.
001
NS
Tabl
e 2.
Con
tinue
d.
ID
TK
NT
dNT
NT
PuT
PfSR
PSi
O2
Ca
KN
aM
gC
lSO
4A
lFe
mg/
Lm
g/L
mg/
Lμg
/Lμg
/Lμg
/Lm
g/L
mg/
Lm
g/L
mg/
Lm
g/L
mg/
Lm
g/L
μg/L
μg/L
EP1
0.31
40.
354
0.40
39.
55.
21.
34.
6138
.80
1.09
1.42
6.53
0.61
41.7
022
.729
.4EP
20.
296
0.34
60.
561
7.5
3.6
1.0
5.21
41.7
00.
851.
265.
590.
5830
.30
1.3
72.3
EP3
0.29
80.
275
0.63
455.
63.
20.
85.
7044
.30
0.98
1.41
6.04
0.56
33.5
020
.545
.4EP
42.
270
1.94
02.
4035
15.7
9.3
2.3
4.99
35.0
05.
702.
1317
.20
2.27
65.4
08.
267
.7EP
52.
100
1.82
02.
592
11.7
7.1
2.1
5.91
51.2
08.
073.
2932
.80
3.45
132.
004.
414
9.0
EP6
0.28
90.
316
0.45
959.
25.
51.
03.
1615
.00
0.08
0.73
1.83
0.12
10.1
04.
014
3.0
EP7
0.77
80.
746
0.86
911
.85.
20.
90.
9829
.60
13.2
068
.40
40.7
021
.60
260.
009.
343
.6EP
82.
040
1.91
02.
2165
24.1
8.5
2.5
7.21
83.2
010
.30
5.71
41.1
05.
3121
1.00
2.0
206.
0EP
92.
120
1.99
02.
167
16.0
9.6
4.2
12.3
034
4.00
21.8
029
.60
97.7
016
.70
1150
.00
13.8
359.
0EP
100.
318
0.32
80.
3765
9.5
5.6
1.0
2.11
60.3
03.
391.
9514
.90
1.04
131.
0042
.186
.5EP
110.
584
0.56
70.
6435
13.5
6.4
0.7
2.49
29.7
02.
171.
094.
120.
8432
.90
19.6
231.
0EP
121.
610
1.47
01.
6875
13.0
10.5
4.5
12.2
023
4.00
32.0
028
.70
87.1
015
.20
751.
002.
243
0.0
EP13
1.15
01.
380
1.46
559.
75.
83.
76.
5159
.80
5.37
3.81
17.2
03.
5451
.60
9.6
178.
0EP
141.
840
1.61
01.
9245
20.6
7.1
2.2
5.33
67.2
04.
834.
4722
.40
2.92
130.
004.
218
3.0
EP15
0.59
40.
645
0.74
1510
.44.
60.
31.
1441
.00
2.86
3.22
18.9
01.
1912
7.00
6.9
68.2
EP16
1.38
01.
210
1.51
412
.06.
91.
52.
6610
7.00
19.9
013
.70
117.
005.
3468
0.00
4.7
92.4
EP17
1.50
01.
430
1.55
87.
66.
11.
913
.90
90.5
03.
945.
7434
.20
1.56
262.
0012
.773
.5EP
180.
961
0.92
71.
036
17.1
7.0
0.8
2.30
64.9
06.
321.
9234
.50
1.42
189.
005.
911
7.0
EP20
1.32
00.
030
1.38
359.
46.
10.
88.
1919
1.00
8.32
4.79
38.1
04.
7649
4.00
8.5
610.
0EP
210.
992
0.90
71.
0295
7.8
6.4
0.9
11.6
014
8.00
7.49
2.77
25.4
01.
8932
9.00
3.0
274.
0EP
220.
156
0.17
90.
2105
8.5
2.4
0.3
1.62
22.4
00.
641.
211.
940.
6418
.90
23.7
41.5
EP23
0.16
70.
207
0.22
74.
92.
40.
11.
4218
.30
0.55
1.06
1.76
0.37
16.6
031
.452
.1EP
240.
087
0.10
90.
182
4.8
2.0
0.1
1.56
8.87
0.25
0.52
0.83
0.36
3.25
37.1
53.4
mea
n1.
007
0.90
01.
1411
.35.
91.
55.
3579
.38
6.96
8.21
29.0
44.
0122
3.92
12.9
156.
8m
edia
n0.
961
0.74
61.
039.
76.
11.
04.
9951
.20
4.83
2.77
18.9
01.
5613
0.00
8.5
92.4
max
2.27
01.
990
2.59
24.1
10.5
4.5
13.9
034
4.00
32.0
068
.40
117.
0021
.60
1150
.00
42.1
610.
0m
in0.
087
0.03
00.
184.
82.
00.
10.
988.
870.
080.
520.
830.
123.
251.
329
.4P
valu
e0
00
0.00
70
0.01
80
NS
0.00
5N
SN
SN
SN
SN
SN
S
Tabl
e 2.
Con
tinue
d.
ID
TN
:TPU
Na:
KFi
eld
Obs
erva
tions
EP1
42.4
21.
30m
oss b
ank,
flow
s int
o EP
2EP
274
.80
1.48
gras
sy m
oss b
ank,
flow
s int
o EP
3EP
311
3.30
1.44
som
e N
osto
cEP
415
3.09
0.37
lots
of N
osto
c an
d fa
iry sh
rimp
EP5
221.
540.
41so
me
Nos
toc,
lots
of f
airy
shrim
p an
d D
aphn
iaEP
649
.95
9.13
wet
land
in a
col
, EP
773
.64
5.18
full
of c
otto
n gr
ass
EP8
91.9
70.
55gr
assy
are
a, m
oss t
hrou
ghou
t, w
ater
slig
htly
col
oure
dEP
913
5.44
1.36
flood
ed g
rass
y m
eado
w, b
ig fa
iry sh
rimp,
scat
in w
ater
EP10
39.6
30.
58co
nnec
ted
to a
noth
er b
asin
, old
shee
ts o
f Nos
toc,
lots
of D
aphn
ia a
nd fa
iry sh
rimp
EP11
47.6
70.
50re
d th
roat
ed lo
on n
estin
g he
reEP
1212
9.81
0.90
som
e N
osto
c,
EP13
151.
080.
71N
osto
c flo
atin
g, b
ig fa
iry sh
rimp,
pro
babl
y co
uld
take
cor
e he
reEP
1493
.42
0.93
near
Blis
ter C
reek
, ver
y lu
sh, E
quis
etum
nea
rby,
cou
ld b
e ph
ytol
iths
EP15
71.3
01.
13lu
sh, b
eige
fila
men
tous
mat
eria
l on
aqua
tic m
osse
s, ca
ribou
and
mus
k ox
act
ivity
, fai
ry sh
rimp
and
Dap
hnia
EP16
126.
170.
69lo
ts o
f Dap
hnia
and
fairy
shrim
p, w
olf t
rack
s and
ani
mal
act
ivity
nea
rby
EP17
205.
001.
46lo
ts o
f tad
pole
shrim
p he
re, f
airy
shrim
p al
so h
ad e
ggs
EP18
60.5
80.
30lo
ts o
f tad
pole
shrim
p, fa
iry sh
rimp,
Dap
hnia
EP20
147.
180.
58ne
xt to
EP2
1 (b
ut n
ot a
ttach
ed),
big
fairy
shrim
pEP
2113
1.99
0.37
pond
bot
tom
mos
tly m
oss,
som
e se
dEP
2224
.76
1.89
high
ele
v, a
rctic
har
e an
d pt
arm
igan
, EP
2346
.33
1.93
lots
of l
ight
gre
en fi
lam
ento
us a
lgae
, muc
h sh
allo
wer
than
EP2
2,
EP24
37.9
22.
0880
% ic
e co
vere
d, ro
cky
botto
m,
mea
n98
.65
1.53
med
ian
91.9
70.
93m
ax22
1.54
9.13
min
24.7
60.
30P
valu
e0.
041
0.00
7
Table 3. Pearson correlation matrix with Bonferroni-adjusted probabilities. Significantly correlated variables are shown in bold (p<0.01) or italics (p<0.05).
pH COND SiO2 Ca POC PON DOC TdN TPu TPf Mg NaCOND 0.467 1SiO2 0.347 0.67 1Ca 0.366 0.92 0.772 1POC 0.454 0.519 0.729 0.525 1PON 0.292 0.391 0.631 0.371 0.92 1DOC 0.516 0.734 0.749 0.659 0.698 0.593 1TdN 0.593 0.599 0.606 0.512 0.597 0.554 0.8 1TPu 0.416 0.613 0.557 0.569 0.566 0.523 0.742 0.595 1TPf 0.494 0.665 0.688 0.612 0.649 0.55 0.879 0.775 0.751 1Mg 0.557 0.943 0.586 0.811 0.48 0.354 0.694 0.616 0.495 0.61 1Na 0.349 0.75 0.39 0.586 0.336 0.292 0.616 0.5 0.595 0.55 0.723 1TKN 0.563 0.74 0.734 0.66 0.708 0.616 0.943 0.842 0.753 0.9 0.718 0.624Al -0.06 -0.243 -0.089 -0.184 -0.18 -0.201 -0.324 -0.29 -0.315 -0.284 -0.23 -0.259Fe 0.228 0.372 0.626 0.414 0.518 0.44 0.593 0.395 0.398 0.514 0.316 0.164Chla 0.044 0.036 0.116 -0.008 0.147 0.111 0.237 0.096 0.203 0.262 -0.036 0.047DIC 0.511 0.658 0.525 0.541 0.448 0.357 0.576 0.545 0.337 0.44 0.772 0.547Cl 0.175 0.497 -0.001 0.327 0.035 0.002 0.187 0.18 0.331 0.259 0.469 0.76SO4 0.077 0.764 0.475 0.784 0.251 0.125 0.454 0.331 0.477 0.486 0.626 0.591K 0.232 0.728 0.469 0.608 0.351 0.262 0.667 0.567 0.445 0.603 0.719 0.695SRP 0.241 0.607 0.594 0.646 0.527 0.426 0.65 0.59 0.558 0.601 0.535 0.481
Table 3. Continued.
TKN Al Fe Chla DIC Cl SO4 KTKN 1Al -0.354 1Fe 0.533 0.268 1Chla 0.242 -0.023 0.19 1DIC 0.581 -0.295 0.233 -0.137 1Cl 0.253 -0.183 -0.139 0.075 0.272 1SO4 0.468 -0.209 0.208 0.103 0.223 0.568 1K 0.638 -0.275 0.405 0.147 0.555 0.477 0.702 1SRP 0.599 -0.251 0.367 -0.017 0.417 0.342 0.685 0.663
CHAPTER 5
EARLY-20TH CENTURY ENVIRONMENTAL CHANGES INFERRED USING SUB-FOSSIL DIATOMS FROM A SMALL POND ON MELVILLE ISLAND, N.W.T., CANADIAN HIGH ARCTIC
BRONWYN E. KEATLEY, MARIANNE S. V. DOUGLAS, AND JOHN P. SMOL
This chapter has been published separately: Keatley, B.E., M.S.V. Douglas, and J.P. Smol. 2006. Early-20th century environmental change inferred using sub-fossil diatoms from a small pond on Melville Island, N.W.T., Canadian High Arctic. Hydrobiologia. 533:15-26.
ABSTRACT Diatom-based paleolimnological studies are being increasingly used to track long-term
environmental change in arctic regions. Little is known, however, about the direction and nature
of such environmental changes in the western Canadian high Arctic. In this study, shifts in diatom
assemblages preserved in a 210Pb-dated sediment core collected from a small pond on Melville
Island, N.W.T., were interpreted to record marked environmental changes that had taken place
since the early 20th century. For most of the history of the pond recorded in this core, the diatom
assemblage remained relatively stable and was dominated by Fragilaria capucina. A major shift
in species composition began in the early-20th century, with a sharp decline in F. capucina and a
concurrent increase in Achnanthes minutissima. In the last ~20 years, further changes in the
diatom assemblage occurred, with a notable increase in the Nitzschia perminuta complex. The
assemblage shifts recorded at this site appear to be consistent with environmental changes
triggered by recent climatic warming.
INTRODUCTION
The Arctic is well-recognized to be especially sensitive to environmental change (Serreze
et al. 2000; Houghton et al. 2001). Due to a number of feedback mechanisms, such as snow
cover-albedo, proposed temperature increases are likely to be maximized in high-latitude regions.
Thus the Arctic comprises a critical reference area for environmental change. In the Canadian
high Arctic, logistical constraints and the short duration and poor spatial coverage of the few
meteorological records makes monitoring of this vast area particularly difficult. The lack of long-
term instrumental data in many ways precludes an accurate assessment of long-term
environmental change. Paleolimnological techniques, however, may provide an effective
alternative method of gathering records of past environmental conditions when traditional
monitoring data are not available (Smol 2002).
Unlike many other arctic proxy records, lakes and ponds are abundant throughout the
Canadian high Arctic, and thus offer the potential of greater regional synthesis. Diatoms (class
Bacillariophyceae), siliceous unicellular algae, are particularly useful paleoenvironmental
indicators because they are ubiquitous, they respond rapidly to changing conditions, and different
species often have distinct optima to given environmental variables (Stoermer & Smol 1999).
Observational data (Serreze et al. 2000) and proxy records (e.g., Kaufman et al. 2004)
indicate that the timing and nature of environmental changes are not synchronous across the
Arctic. For example, divergences in stable oxygen isotope, atmospheric dust, and glaciochemical
records between ice core records from the Devon Ice Cap (Devon Island), Penny Ice Cap (Baffin
Island), and Greenland (Camp Century and GISP2) suggest increasingly regional climatic
influences during the Holocene (Paterson et al. 1977; Fisher 1979; O’Brien et al. 1995;
Zdanowicz et al. 2000; Grumet et al. 2001).
To date, most paleolimnological records from the Canadian high Arctic are from eastern
regions, and have illustrated differences in the timing and magnitude of environmental change
(e.g. Douglas et al. 1994; Doubleday et al. 1995; Perren et al. 2003; Michelutti et al. 2003a;
Antoniades et al. 2005). For example, diatom assemblages from shallow ponds on Cape
Herschel, east-central Ellesmere Island (Figure 1, site G), remained relatively static and were
interpreted to record cool temperatures for several millennia up until the mid-19th century, when
the diatom assemblages underwent substantial changes that were attributed to climatic warming
(Douglas et al. 1994). Since this initial study, other high arctic paleolimnological investigations
have shown similar changes in diatom community structure since ~1850 (e.g., Doubleday et al.
1995; Gajewski et al. 1997; Wolfe 2000; Perren et al. 2003; Michelutti et al. 2003a; Antoniades et
al. 2005). These shifts have not been simultaneous, but rather appear to be at least partly related
to the local limnological conditions, such as lake size (e.g., Doubleday et al. 1995; Michelutti et
al. 2003a) and other variables. For example, diatom assemblages from small water bodies near
Isachsen, Ellef Ringnes Island (Figure 1, site B) experienced species turnover starting in the mid-
19th century, while those from a larger lake at Alert, Ellesmere Island (Figure 1, site C), only
began to shift after the mid-to late 20th century (Antoniades et al. 2005). Likewise, subtle diatom
assemblage changes at a large and deep high arctic lake (Char Lake, Cornwallis Island, Figure 1,
site K) only began to occur in the late-1980s (Michelutti et al. 2003a). These diatom community
shifts were correlated to warmer temperatures documented by nearby instrumental meteorological
records. Warming conditions are expected to result in a lengthening of the growing season and
enhanced autochthonous production due to a reduction in duration of ice cover; in turn, this is
expected to affect limnological variables such as pH, nutrients, and specific conductivity
(Douglas & Smol 1999). These studies suggest that larger water bodies of the high Arctic had a
greater thermal inertia which acted as a buffer against the early-onset of changes in diatom
communities (e.g., Doubleday et al. 1995; Michelutti et al. 2003a).
Paleolimnological research using abiotic proxies, such as varves, have also indicated
relatively recent and marked environmental changes in Canadian high Arctic lakes and have
similarly been correlated to climatic warming (Smith et al. 2004). Likewise, analyses of sediment
cores from five lakes on Svalbard suggested that climatic warming was a contributing factor to
changing diatom assemblages over the last ~150 years (Birks et al. 2004; Jones & Birks 2004).
Similarly, research programs in subarctic regions have recorded recent environmental changes
consistent with warming (e.g., Sorvari et al. 2002; Rühland et al. 2003a).
Despite a growing body of arctic paleoenvironmental literature, no detailed
paleolimnological investigations have yet been published from the vast western Canadian high
Arctic. Instrumental meteorological data are from this region are sparse and of short duration; the
nearest weather station is located at Mould Bay, Prince Patrick Island (Figure 1, site L), which
began collecting data in 1948. This lack of instrumental meteorological data hampers our ability
to assess past climatic and associated environmental changes from the western Canadian high
Arctic. Given the sensitivity of small ponds in the eastern high Arctic (Douglas et al. 1994), a
small pond on Melville Island was chosen to be the focus of a high-resolution paleoenvironmental
investigation. The goals of this study are to assess diatom-based paleolimnological changes from
a small pond on central Melville Island, to examine whether diatom assemblages changed in
composition and, if so, when and why these changes occurred.
STUDY SITE
With a surface area of 42 149 km2, Melville Island (Figure 1) is the 4th largest of the
Queen Elizabeth Islands, and the 7th largest of all Canadian Arctic Islands, yet it is uninhabited.
Melville Island is the only island in the western high Arctic that retains small permanent ice caps.
In the absence of local meteorological records, data from Mould Bay (76˚13’N, 119˚19’W),
Prince Patrick Island (Figure 1), located approximately 230 km to the northwest, was used to
estimate average February and July temperature of -34.0˚C and 4.0˚C, respectively, and a mean
annual precipitation of 111.0 mm (Meteorological Service of Canada 2002).
Pond MV-AT (unofficial name, 75˚19’N, 111˚25’W) is a small, shallow (maximum
depth = 0.40 m), alkaline (pH = 8.1), and very dilute (specific conductivity = 39 μS/cm) pond
located on central Melville Island, Northwest Territories (Figure 1, site A; Table 1). The pond
has moderately low nutrients (e.g., total phosphorus = 12.7 μg/L) and relatively high dissolved
organic carbon (8.2 mg/L) (Table 1). Surficial geology is composed of a veneer of weathered
bedrock (sandstone, siltstone, and shale) of the Griper Bay Formation (Hodgson & Vincent
1984). Pond MV-AT is located on an interfluve approximately 15 m above two braided rivers,
with a catchment dominated by grasses, sedges, and mosses.
MATERIALS AND METHODS
A modified Glew (1989) gravity corer (diameter = 7.82 cm) was used to collect a 21 cm
long core from the centre of the pond on July 24, 2002. The sediment core was sectioned into 0.5
cm intervals from 0-13 cm using a Glew (1988) extruder immediately after retrieval. This section
of the core was rich in organic matter and contained mosses. Below 13 cm, the sediment was
composed predominantly of fine minerogenic sediment, and thus the remainder of the core was
sectioned into 1 cm intervals.
Temperature, pH, and specific conductivity were measured in the field, and water
samples were sent to the Canadian Centre for Inland Waters (Burlington, Ontario, Canada) for
analysis of nutrients, trace metals, and major ions (Environment Canada 1994). Methodological
details of sampling are provided in Appendix 1.
210Pb dating was performed at the Paleoecological Environmental Assessment and
Research Lab (PEARL), Queen’s University, using gamma spectrometry (Appleby 2001).
Activity levels were converted to dates using the Constant Rate of Supply (CRS) method (Binford
1990).
Preparation of diatom samples followed standard techniques (Battarbee et al. 2001). A
minimum of 300 diatom valves were identified and enumerated at each interval. Diatom
identification followed Krammer & Lange-Bertalot (1991), Krammer (2000), Krammer (2002),
Lange-Bertalot (2001), and Antoniades (2004).
Diatom results were converted to percent relative abundance measures and plotted
against depth using the program C2 version 1.4 beta (Juggins 2003). To identify zones which
may be considered to have some statistical validity, optimal splitting combined with broken-stick
analysis was performed (Bennett 1996), using the program psimpoll 4.10 (Bennett 2002). A
detrended correspondence analysis (DCA) was applied down-core to estimate species turnover
(Birks 1998). DCA was performed using CANOCO version 4.0 (ter Braak & Šmilauer 1998).
Loss-on-ignition (LOI) analysis followed Dean (1974). Sediments were freeze-dried and
pre-weighed before combustion at 550˚C for 2 hours to provide a proxy for organic matter
content of the sediment (Heiri et al. 2001). Further combustion at 1000˚C for two hours provided
an estimate of carbonate content (Dean 1974).
RESULTS
Core Chronology
While all 35 sediment intervals were dated, the unsupported 210Pb profile was only
detected within the upper 7 cm of the sediment core. Gamma emissions were low, as is common
in high arctic lake sediments (Pienitz et al. 2004), but nonetheless the profile exhibited a roughly
exponential decay, indicating that a reliable geochronological profile could be established for the
pond’s recent history (ca. 200 years). Three intervals had higher 210Pb counts relative to the
intervals immediately above them, and thus were not used in the date calculations; such reversals
are not uncommon when emissions are very low.
Based on the CRS model, the calculated sedimentation rate for the upper ~2.5 cm was
high (0.0322 cm/yr), relative to other high arctic lakes (e.g., Douglas et al. 1994). Below this
level, the rate dropped off exponentially and between 5 - 6.5 cm, the sedimentation rate reached a
plateau at 0.0106 cm/yr (Figure 2).
Diatoms and LOI
A total of 81 diatom taxa were identified, but only 7 were considered common (i.e.,
present at >5% relative abundance in at least one interval). Broken-stick analysis based on
optimal splitting identified one split of significant note occurring between 5-5.5 cm (Figure 3).
From 21 cm to 6 cm (Zone 1), the diatom assemblage remained relatively constant (Figure 3), and
was dominated by Fragilaria capucina Desmazières (~40-50%), and secondarily by Achnanthes
minutissima Kützing (~20%).
Zone 2 begins between 5-5.5 cm (Figure 3). Between 5.5-1.0 cm, F. capucina underwent
a marked decrease to about 10%, whereas A. minutissima increased to ~50% (Figure 3). Between
3-0 cm, Nitzschia perminuta (Grunow) M. Peragallo increased from ~10% to 15% relative
abundance. At the top of the core, A. minutissima decreased to about 43% relative abundance,
with a concordant increase in N. perminuta.
Species turnover was low for most of the sediment profile, as the DCA axis 1 species
scores remained relatively constant between 1 and 1.25 standard deviation (SD) units from 21 –
5.5 cm (Figure 4). The total variance of DCA axis 1 was 1.396 SD units. Above 5.5 cm, a sharp
decline in the DCA species scores to about 0.25 SD units indicates a strong species shift (Figure
4). This trend continues until 0.5 cm, at which time there is an inflection in the opposite
direction, likely reflecting the resurgence of N. perminuta at the surface of the core (Figure 4).
This shift in direction should be treated with caution, however, since the top of the core contains
the least compacted sediment.
The percentage of LOI remained constant at about 17% between 21-13 cm, increased to
~20% between 13-5.5 cm, and increased further to ~50% beginning at 5 cm (Figure 3). There
was insufficient sediment available to determine a value for LOI at the very surface of the core
(i.e., between 0-0.5 cm). The percent carbonate content remained consistently low (~2%)
throughout the core.
DISCUSSION
The diatom assemblages from MV-AT remained stable for most of the pond’s history
captured by this sediment record, with the most striking changes occurring early in the 20th
century (~5.25 cm, Figure 3). This marks the delineation of the only clear zone based on a
broken-stick analysis (Bennett 1996). As is also evident from the DCA profile (Figure 4), a
marked shift in species turnover is underway at this time. While these changes are most dramatic
at this time, it is likely that the diatom assemblage began to change even earlier (~ 5.75 cm,
Figures 3 & 4).
Potential causes contributing to the recent shift in diatom assemblages include
atmospheric pollution, anthropogenic acidification, artificial nitrogen deposition, and/or climatic
change. Melville Island is completely uninhabited, and located far from local pollution sources.
While organic pollutants, such as polychlorinated biphenyls (PCBs), may be transported long
distances (Rose et al. 2004), it is unlikely that the shifts in diatom assemblages recorded here
would be triggered by such contamination. For example, sites that had direct and prolonged PCB
contamination in Labrador showed no diatom or chrysophyte response to this input (Paterson et
al. 2003).
Diatoms are known to be sensitive indicators of acidification (e.g., Siver et al. 2003).
However, there is no suggestion of acidification of this site as the current pH is 8.1 and the
diatom shift is away from taxa known to be acidophilic (e.g., Eunotia spp.) and towards species
commonly found in more alkaline arctic waters (Antoniades et al. 2004). The increase in
Nitzschia perminuta near the top of the core provides further evidence that pH was not declining
at this site, as N. perminuta has the highest pH optimum of the three dominant species in this
core, based on a training set from Ellef Ringnes Island, Arctic Canada (Antoniades et al. 2004).
Nitrogen deposition has been implicated as a driver of diatom assemblage change in the
Rocky Mountains of Colorado (Wolfe et al. 2001; Saros et al. 2003; Wolfe et al. 2003). As
Achnanthes minutissima is commonly found in high arctic lakes and ponds with relatively high
TN values, it is possible that some degree of anthropogenic nitrogen deposition may be affecting
MV-AT as well. It is unlikely, however, that this deposition could have initiated the shift in
diatom communities on Melville Island because the commencement of the major species shift
predates the onset of artificial N production by at least ~30 years (Vitousek et al. 1997). In
addition, pond MV-AT is currently highly P-limited (TN:TP = 50, Table 1). The presence of
Nostoc balls between 0 - 5 cm in our core also suggests that N levels have not been increasing
significantly over the 20th century, as Nostoc (cyanobacteria) are N-fixers and are especially
competitive under low N conditions. Thus, increases in N would not likely result in the marked
diatom changes recorded in this pond. Instead, these diatom changes are more likely related to
climatic warming, as discussed below.
Under conditions of warming, limnological variables such as nutrients, specific
conductivity, and pH are all expected to increase in a pond such as MV-AT, as a reduction in the
duration of ice cover allows for longer growing seasons and higher primary production (Douglas
& Smol 1999). While the three most dominant taxa in the surface of the core are not known to
have highly restrictive ecological niches (e.g., Lim et al. 2001b; Michelutti et al. 2003b),
autecological information gleaned from recent diatom calibration sets in the Arctic and Subarctic
(Lim et al. 2001a; Lim et al. 2001b; Rühland et al. 2003; Michelutti et al. 2003b; Lim 2004;
Antoniades et al. 2003) suggests that the most likely cause of these assemblage shifts are
environmental changes precipitated by climatic warming. For example, arctic diatom calibration
sets suggest that A. minutissima has a higher specific conductivity optimum than F. capucina
(Antoniades et al. 2004), and is more frequently found in sites with higher total nitrogen levels
(Lim et al. 2001a; Rühland et al. 2003) and higher dissolved organic carbon (DOC)
concentrations (Rühland et al. 2003). Likewise, N. perminuta occurs in sites with relatively high
pH and soluble reactive phosphorus levels on Devon Island (Lim 2004). Small Nitzschia taxa are
also often associated with high nutrient loadings in arctic lakes (e.g., Douglas & Smol 2000;
Michelutti et al. 2002; Douglas et al. 2004). N. perminuta, however, is not commonly associated
with any specific environmental variable on Bathurst Island (Figure 1, site M; Lim et al. 2001a;
Lim et al. 2001b), the central Canadian Subarctic (Rühland et al. 2003), or Ellef Ringnes Island
(Antoniades et al. 2004). As algal production is expected to increase with a lengthening of the
growing season, a shift toward diatom species that have affinities for higher nutrient, DOC, and
specific conductivity levels suggests that the environment experienced a warming trend at MV-
AT beginning in the early 20th century.
The other common species found in Figure 3 occur in very low abundances. They are
comprised by three Cymbella and one Eunotia spp. It is likely that the decrease in Eunotia
further supports a rising pH over time in this site, as this genus is characteristic of acidic waters.
Cymbella spp. are commonly associated with shallow water arctic habitats (Douglas & Smol
1995; Michelutti et al. 2003b). Furthermore, it should be noted that all of the 7 common taxa are
periphytic (Figure 3). F. capucina is known to be commonly associated with rocks, moss, and
sediment on Victoria Island (Figure 1, site N; Michelutti et al. 2003b), while N. perminuta and A.
minutissima are both commonly found on moss, rocks, and sediment on Victoria Island
(Michelutti et al. 2003b), Cape Hershel (Douglas & Smol 1995), and Bathurst Island (Lim et al.
2001b). Thus, the changes in the diatom assemblage do not represent a shift between planktonic
and benthic taxa.
Concurrent with the marked floristic shifts in this core, there was also a sharp rise in
%LOI near the 5 cm depth (Figure 3). Increased run-off from the catchment may result in a
greater amount of allochthonous organic matter washing-in to the pond. However, evidence for
enhanced autochthonous production is apparent in the correspondence between high %LOI values
of 30-50% and the presence of many preserved Nostoc (cyanobacteria) balls in the sediment core
(up until ~5 cm depth). The occurrence of Nostoc may be a further indicator of climatic
warming, as increased temperatures would lengthen the growing season.
Comparison to other paleoenvironmental records
The diatom profile from MV-AT exhibits species shifts that are consistent with recent
environmental change inferred from diatom-based paleolimnological records from other areas in
the Canadian high Arctic (Figure 1, Douglas et al. 1994; Doubleday et al. 1995; Michelutti et al.
2003; Perren et al. 2003; Antoniades et al. 2005). While none of the previously recorded diatom
profiles show similar assemblages to those dominating the MV-AT core prior to ~1919, the more
recent MV-AT communities share similarities to other sites. For example, pond I-F (Isachsen,
Ellef Ringnes Island) recorded increases in Nitzschia perminuta since ~1850 A.D. (Antoniades et
al. 2005). In Self Pond (Alert, Ellesmere Island), the increased relative abundance of Achnanthes
minutissima tracked increases of both pH and warming temperatures since ~1920 A.D.
(Antoniades et al. 2005). The similarities between these modern assemblages and those found in
MV-AT provide further evidence that the diatom changes in MV-AT may indicate recent
warming. Furthermore, the assemblages in other studies are different, but the nature of the
diatoms shifts are similar to that of MV-AT, and have been interpreted as a response to warming
(e.g., Douglas et al. 1994; Doubleday et al. 1995; Gajewski et al. 1997; Perren et al. 2003). The
onsets of these diatom shifts, however, have occurred at different times. The large diatom
assemblage shift in MV-AT appears to have started ~1919, which would be consistent with the
timing of diatom changes at both Sawtooth Lake (Fosheim Peninsula, Ellesmere Island, Figure 1,
site E; Perren et al. 2003), and Self Pond (Antoniades et al. 2005), but later than diatom
assemblage changes at ponds near Isachsen (Antoniades et al. 2005), and at Cape Herschel,
Ellesmere Island (Douglas et al. 1994). Assuming the dating of each of the sediment profiles is
reasonably accurate, the discrepancy in timing may be due to a regional difference in warming in
the western Canadian high Arctic.
The only available paleoenvironmental record for Melville Island is the glacier mass
balance data (Koerner 2002) from the Melville South Ice Cap (Figure 1, site I). Glaciers cover
~160 km2 of Melville Island (Ommanney 2002). Mass balance measurements from the Melville
South Ice Cap have been made annually (with some exceptions) by the Geological Survey of
Canada since 1964 (Ommanney 2002). Although these records show that there is little trend in
the mass balance data, each year since 1968 (with the exception of 1984, 1986, and 1991) has had
an average negative mass balance, indicating that melt had exceeded accumulation (Koerner
2002). This is consistent with the trend inferred from our diatom and %LOI data.
The nearest available long-term meteorological data are from Mould Bay, Prince Patrick
Island (Figures 1 & 5). Although this record extends back to only 1948, and hence does not
capture the inception of the large diatom and %LOI shifts recorded here, it may still be used to
assess more recent temperature and precipitation variations. Mean June-July-August temperature
data indicate that there is little trend in temperature between 1948-1996 (Figure 5, Meteorological
Service of Canada 2004), likewise, the diatom record does not show any major shifts after ~1950.
The increase in Nitzschia perminuta near the top of the core (Figure 3) is not captured in the
meteorological data, which ends in 1996. There does appear to be an increase in annual
precipitation values at Mould Bay (Figure 5, Meteorological Service of Canada 2004), but this is
not reflected in the diatom profile of MV-AT. A closer examination of the precipitation values
for June, July, and August (the months representing and shouldering the open water season)
indicate only a very slight increase in precipitation since 1948 at Mould Bay (data not shown).
Other abiotic proxy environmental records from the Canadian high Arctic also show 20th
century warming. For example, varved sediment records from Tuborg Lake, Ellesmere Island
(Figure 1, site D), were interpreted to show warming beginning after ~1865, and especially after
~1908 (Smith et al. 2004). Likewise, ice core melt records from the Devon Ice Cap (Figure 1,
site F) suggest warm temperatures since 1869, and particularly after 1925 (Koerner 1977).
Agassiz Ice Cap, Ellesmere Island (Figure 1, site H), melt layers also suggested that the 20th
century has been the warmest of the last millennium (Koerner & Fisher 1990), and analysis of ice
fabric, dirt and firn from Meighen Island ice cap (Figure 1, site J) suggested warm conditions
between 1884-1964 (the extent of the record) (Koerner & Paterson 1974). These records,
although all located significant distances from Melville Island (Figure 1), all suggest that regional
climatic warming accelerated into the 20th century. The dramatic diatom change at MV-AT
appears to be consistent, in both time and nature, to these paleoenvironmental records.
CONCLUSIONS Diatom species assemblages have changed markedly since the early 20th century in this
small pond on Melville Island, suggesting higher pH, specific conductivity, and nutrient levels.
These data, coupled with increases in % LOI (as a proxy for organic matter), imply that algal
production increased beginning around 1919, and are consistent with environmental changes
projected under a 20th century climate warming scenario. The direction of environmental change
in pond MV-AT is broadly similar to that found in both lakes and ponds throughout the eastern
Canadian high Arctic, the Canadian Subarctic, and Svalbard. The onset of these environmental
changes, however, is somewhat later than those of small ponds from eastern Ellesmere Island
(Douglas et al. 1994), but similarly timed to changes in Self Pond (Antoniades et al. 2005) and
Sawtooth Lake (Perren et al. 2003). As expected, the changes in diatoms from MV-AT occur
much earlier and more dramatically than the shifts apparent in relatively large lakes from the
eastern Canadian high Arctic (e.g., Michelutti et al. 2003). Future paleolimnological studies in
this region will help to refine and corroborate these interpretations.
ACKNOWLEDGEMENTS
This research was funded by Natural Sciences and Engineering Research Council (NSERC)
grants to BEK, MSVD, and JPS, and a Northern Scientific Training Program grant and a Queen’s
Graduate Award to BEK. Logistical support was provided by the Polar Continental Shelf Project
(PCSP). Assistance in the field was provided by J.R. Glew, N. Michelutti, and especially by D.
Antoniades. Input from three reviewers greatly improved this manuscript. Many thanks to K.
Rühland, D. Selbie, A. Harris, A. Strecker, and R. Bull for helpful comments. This is PCSP
contribution # 05004.
REFERENCES
Antoniades, D., 2004. The limnology, diatom autecology, and paleolimnology of lakes and ponds from Alert, Ellesmere Island, Isachsen, Ellef Ringnes Island, and Mould Bay, Prince Patrick Island, Canadian High Arctic. University of Toronto. PhD Thesis.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2004. Diatom species-environment relationships and inference models from Isachsen, Ellef Ringnes Island, Canadian High Arctic. Hydrobiologia 529: 1-18.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2005. Quantitative estimates of recent environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. Journal of Paleolimnology 33.
Appleby, P. G., 2001. Chronostratigraphic techniques in recent sediments. In W. M. Last & J. P. Smol (eds), Tracking Environmental Change Using Lake Sediments Volume 1: Basin Analysis, Coring, and Chronological Techniques. Kluwer Academic Publishers, Dordrecht: 171-203.
Battarbee, R. W., V. J. Jones, R. J. Flower, N. G. Cameron, H. Bennion, L. Carvalho & S. Juggins, 2001. Diatoms. In J. P. Smol, H. J. B. Birks & W. M. Last (eds), Tracking Environmental Change Using Lake Sediments: Terrestrial, Algal, and Siliceous Indicators. Kluwer Academic Publishers, Dordrecht: 155-202.
Bennett, K. D., 1996. Determination of the number of zones in a biostratigraphical sequence. New Phytologist 132: 155-170.
Bennett, K. D., 2002. Documentation for psimpoll 4.10 and pscomb 1.03. Department of Earth Sciences, Uppsala Universitet, Sweden.
Binford, M. W., 1990. Calculation and uncertainty analysis of 210Pb dates for PIRLA project lake sediment cores. Journal of Paleolimnology 3: 253-267.
Birks, H. J. B., 1998. Numerical tools in palaeolimnology - Progress, potentialities, and problems. Journal of Paleolimnology 20: 307-332.
Birks, H. J. B., V. J. Jones, & N. L. Rose, 2004. Recent environmental change and atmospheric contamination on Svalbard as recorded in lake sediments - synthesis and general conclusions. Journal of Paleolimnology 31: 531-546.
Dean, Jr. W. E., 1974. Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: comparison with other methods. Journal of Sediment Petrology 44: 242-248.
Doubleday, N. C., M. S. V. Douglas, & J. P. Smol, 1995. Paleoenvironmental studies of black carbon deposition in the High Arctic - a case-study from northern Ellesmere Island. Science of the Total Environment 161: 661-668.
Douglas, M. S. V., & J. P. Smol, 1995. Periphytic diatom assemblages from high arctic ponds. Journal of Phycology 31: 60-69.
Douglas, M. S. V.& J. P. Smol, 1999. Freshwater diatoms as indicators of environmental change in the High Arctic. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge: 227-244.
Douglas, M. S. V., & J. P. Smol, 2000. Eutrophication and recovery in the High Arctic: Meretta lake (Cornwallis Island, Nunavut, Canada) revisited. Hydrobiologia 431: 193-204.
Douglas, M. S. V., J. P. Smol, & W. Blake Jr, 1994. Marked post-18th century environmental change in high-arctic ecosystems. Science 266: 416-419.
Douglas, M. S. V., J. P. Smol, J. M. Savelle, & J. M. Blais, 2004. Prehistoric Inuit whalers affected arctic freshwater ecosystems. Proceedings of the National Academy of Sciences of the United States of America 101: 1613-1617.
Environment Canada, 1994. Manual of Analytical Methods – National Laboratory for Environmental Testing, Canadian Centre for Inland Waters. Burlington, ON.
Fisher, D. A., 1979. Comparison of 105 years of oxygen isotope and insoluble impurity profiles from the Devon Island and Camp Century ice cores. Quaternary Research 11: 299-305.
Gajewski, K., P. B. Hamilton, & R. McNeely, 1997. A high resolution proxy-climate record from an arctic lake with annually-laminated sediments on Devon Island, Nunavut, Canada. Journal of Paleolimnology 17: 215-225.
Glew, J. R., 1988. A portable extruding device for close interval sectioning of unconsolidated core samples. Journal of Paleolimnology 2: 241-243.
Glew, J. R., 1989. A new trigger mechanism for sediment samplers. Journal of Paleolimnology 2: 241-243.
Grumet, N. S., C. P. Wake, P. A. Mayewski, G. A. Zielinski, S. I. Whitlow, R. M. Koerner, D. A. Fisher, & J. M. Woollett, 2001. Variability of sea-ice extent in Baffin Bay over the last millenium. Climatic Change 49: 129-145.
Heiri, O., A. F. Lotter, & G. Lemcke, 2001. Loss on ignition as a method for estimating organic and carbonate content in sediments: reproducibility and comparability of results. Journal of Paleolimnology 25: 101-110.
Hodgson, D.A. & J.-S. Vincent, 1984. Surficial geology, Central Melville Island, Northwest Territories; Geological Survey of Canada, Map 1583A, scale 1:250 000.
Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden & D. e. Xiaosu, 2001. Climate Change 2001: The Scientific Basis Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, Cambridge.
Jones, V. J., & H. J. B. Birks, 2004. Lake-sediment records of recent environmental change on Svalbard: results of diatom analysis. Journal of Paleolimnology 31: 445-466.
Juggins, S., 2003. C2: software for ecological and palaeocological data analysis and visualisation, version 1.3. Department of Geography, University of Newcastle.
Kaufman, D. S., T. A. Ager, N. J. Anderson, P. M. Anderson, J. T. Andrews, P. J. Bartlein, L. B. Brubaker, L. L. Coats, L. C. Cwynar, M. L. Duvall, A. S. Dyke, M. E. Edwards, W. R. Eisner, K. Gajewski, A. Geirsdóttir, F. S. Hu, A. E. Jennings, M. R. Kaplan, M. W. Kerwin, A. V. Lozhkin, G. M. MacDonald, G. H. Miller, C. J. Mock, W. W. Oswald, B. L. Otto-Bliesner, D. F. Porinchu, K. Rühland, J. P. Smol, E. J. Steig, & B. B. Wolfe, 2004. Holocene thermal maximum in the western Arctic (0-180°W). Quaternary Science Reviews 23: 529-560.
Koerner, R. M., 1977. Devon Island Ice Cap: core stratigraphy and paleoclimate. Science 196: 15-18.
Koerner, R. M., 2002. Glaciers of high arctic islands. In R. S. Williams & J. G. Ferrigno (eds), Glaciers of North America. U.S. Geological Survey Professional Paper 1386-J-1, J111-J146.
Koerner, R. M., & D. A. Fisher, 1990. A record of Holocene summer climate from a Canadian high-Arctic ice core. Nature 343: 630-631.
Koerner, R. M., & W. S. B. Paterson, 1974. Analysis of a core through the Meighen Ice Cap, Arctic Canada, and its paleoclimatic implications. Quaternary Research 4: 253-263.
Krammer, K., 2000. Diatoms of Europe: The Genus Pinnularia. A.R.G. Gantner Verlag, Ruggell.
Krammer, K., 2002. Diatoms of Europe: Cymbella. A.R.G. Gantner Verlag, Ruggell.
Krammer, K.& H. Lange-Bertalot, 1986-1991. Bacillariophyceae. In H. Ettl, J. Gerloff, H. Heynig & D. Mollenhauer (eds). Gustav Fischer Verlag, Stuttgart.
Lange-Bertalot, H., 2001. Diatoms of Europe. Navicula sensu stricto, 10 genera separated from Navicula sensu lato, Frustulia. A.R.G. Gantner Verlag, Ruggell.
Lim, D.S.S., 2004. Limnology and diatom paleoecology of diatoms on Banks Island, N.W.T., and Devon Island, Nunavut, Canadian Arctic. University of Toronto. PhD Thesis.
Lim, D. S. S., M. S. V. Douglas, & J. P. Smol, 2001a. Diatoms and thier relationship to environmental variables from lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. Hydrobiologia 450: 215-230.
Lim, D. S. S., C. Kwan, & M. S. V. Douglas, 2001b. Periphytic diatom assemblages from Bathurst Island, Nunavut, Canadian High Arctic: An examination of community relationships and habitat preferences. Journal of Phycology 37: 379-392.
Meteorological Service of Canada 2002. Canadian Climate Normals 1971-2000. Environment Canada, Ottawa.
Meteorological Service of Canada 2004. Climate Data Online. Environment Canada, Ottawa. http://www.climate.weatheroffice.ec.gc.ca/climateData/canada_e.html, accessed December 10, 2004.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2002. Tracking recent recovery from eutrophication in a high arctic lake (Meretta Lake, Cornwallis Island, Nunavut, Canada) using fossil diatom assemblages. Journal of Paleolimnology 28: 377-381.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2003a. Diatom response to recent climatic change in a high arctic lake (Char Lake, Cornwallis Island, Nunavut). Global and Planetary Change 38: 257-271.
Michelutti, N., A. J. Holtham, M. S. V. Douglas, & J. P. Smol, 2003b. Periphytic diatom assemblages from ultra-oligotrophic and UV transparent lakes and ponds on Victoria Island and comparisons with other diatom surveys in the Canadian Arctic. Journal of Phycology 39: 465-480.
O'Brien, S. R., P. A. Mayewski, L. D. Meeker, D. A. Meese, M. S. Twickler, & S. I. Whitlow, 1995. Complexity of Holocene climate as reconstructed from a Greenland ice core. Science 270: 1962-1964.
Ommanney, C. S. L., 2002. Glaciers of Canada. In R. S. Williams & J. G. Ferrigno (eds), Glaciers of North America. U.S. Geological Survey Professional Paper 1386-J-1, J27-J82.
Paterson, A. M., A. A. Betts-Piper, J. P. Smol, & B. A. Zeeb, 2003. Diatom and chrysophyte algal response to long-term PCB contamination from a point-source in northern Labrador, Canada. Water Air and Soil Pollution 145: 377-393.
Paterson, W. S. B., R. M. Koerner, D. A. Fisher, S. Johnsen, H. B. Clausen, W. Dansgaard, P. Bucher, & H. Oeschger, 1977. An oxygen-isotope climate record from the Devon Island ice cap, arctic Canada. Nature 266: 508-511.
Perren, B. B., R. S. Bradley, & P. Francus, 2003. Rapid lacustrine response to recent High Arctic warming: A diatom record from Sawtooth Lake, Ellesmere Island, Nunavut. Arctic Antarctic and Alpine Research 35: 271-278.
Pienitz, R., M. S. V. Dougas & J. P. Smol, 2004. Long-Term Environmental Change in Arctic and Antarctic Lakes. Springer, Dordrecht.
Rose, N. L., C. L. Rose, J. F. Boyle, & P. G. Appleby, 2004. Lake-sediment evidence for local and remote sources of atmospherically deposited pollutants on Svalbard. Journal of Paleolimnology 31: 499-513.
Rühland, K., A. Priesnitz, & J. P. Smol, 2003a. Paleolimnological evidence from diatoms for recent environmental changes in 50 lakes across Canadian arctic treeline. Arctic Antarctic and Alpine Research 35: 110-123.
Rühland, K. M., J. P. Smol, & R. Pienitz, 2003b. Ecology and spatial distributions of surface-sediment diatoms from 77 lakes in the subarctic Canadian treeline region. Canadian Journal of Botany 81: 57-73.
Saros, J. E., S. J. Interlandi, A. P. Wolfe, & D. R. Engstrom, 2003. Recent changes in the diatom community structure of lakes in the Beartooth Mountain Range, U.S.A. Arctic, Antarctic, and Alpine Research 35: 18-23.
Serreze, M. C., J. E. Walsh, F. S. Chapin III, T. Osterkamp, M. Dyurgerov, V. Romanovsky, W. C. Oechel, J. Morison, T. Zhang, & R. G. Barry, 2000. Observational evidence of recent change in the northern high-latitude environment. Climatic Change 46: 159-207.
Siver, P. A., R. Ricard, R. Goodwin, & A. E. Giblin, 2003. Estimating historical in-lake alkalinity generation from sulfate reduction and its relationship to lake chemistry as inferred from algal microfossils. Journal of Paleolimnology 29: 179-197.
Smith, Jr. S. V., R. S. Bradley, & M. B. Abbott, 2004. A 300 year record of environmental change from Lake Tuborg, Ellesmere Island, Nunavut, Canada. Journal of Paleolimnology 32: 137-148.
Smol, J. P., 2002. Pollution of lakes and rivers: a paleoenvironmental perspective. Arnold, London.
Sorvari, S., A. Korhola, & R. Thompson, 2002. Lake diatom response to recent Arctic warming in Finnish Lapland. Global Change Biology 8: 171-181.
Stoermer, E. F.& J. P. Smol, 1999. The diatoms: applications for the environmental and earth sciences. Cambridge University Press, Cambridge.
ter Braak, C. J. F., 1998. CANOCO and CANOPost, Version 4 for Windows. Microcomputer Power, Ithaca, NY.
Vitousek, P. M., J. D. Aber, R. W. Howarth, G. E. Likens, P. A. Matson, D. W. Schindler, W. H. Schlesinger, & D. G. Tilman, 1997. Human alteration of the global nitrogen cycle. Ecological Applications 7: 737-750.
Wolfe, A. P., 2000. A 6500-year record from southwestern Fosheim Peninsula, Ellesmere Island, Nunavut. In M. Garneau & B. T. Alt (eds), Environmental Response to Climate Change in the Canadian High Arctic. Geological Survey of Canada, Ottawa: 249-256.
Wolfe, A. P., J. S. Baron, & R. J. Cornett, 2001. Anthropogenic nitrogen deposition induces rapid ecological changes in alpine lakes of the Colorado Front Range (USA). Journal of Paleolimnology 25: 1-7.
Wolfe, A. P., A. C. Van Gorp, & J. S. Baron, 2003. Recent ecological and biogeochemical changes in alpine lakes of Rocky Mountain National Park (Colorado, USA): a response to anthropogenic nitrogen deposition. Geobiology 1: 153-168.
Zdanowicz, C. M., G. A. Zielinski, C. P. Wake, D. A. Fisher, & R. M. Koerner, 2000. A Holocene record of atmospheric dust deposition on the Penny Ice Cap, Baffin Island, Canada. Quaternary Research 53: 62-69.
Figure captions.
Figure 1. Map showing location of sites discussed in this paper. The oval on the inset map shows location of Canadian High Arctic. Sites are as follows: A) MV-AT, Melville Island; B) Isachsen, Ellef Ringnes Island; C) Alert, Ellesmere Island; D) Tuborg Lake, Ellesmere Island; E) Fosheim Peninsula, Ellesmere Island; F) Devon Ice Cap, Devon Island; G) Cape Hershel, Ellesmere Island; H) Agassiz Ice Cap, Ellesmere Island; I) Melville Island ice caps, Melville Island; J) Meighen Ice Cap, Meighen Island; K) Char Lake, Resolute Bay, Cornwallis Island; L) Mould Bay, Prince Patrick Island; M) Bathurst Island; N) Victoria Island. Figure 2. Sedimentation rate for dated sections of the MV-AT core, as calculated based on the Constant Rate of Supply (CRS) method described by Appleby (2001) and Binford (1990). Figure 3. Diatom profile showing the common diatom species found in pond MV-AT. Individual species with >5% relative abundance in at least one sample were retained for the profile; “other” is a sum of all other diatoms found in each interval. Dates are based on 210Pb dating using a Constant Rate of Supply model. Percent loss-on-ignition (%LOI 550) is expressed as a % of combustion at 550˚C, and is a proxy for organic matter content of the sediment. Percent carbonates (%LOI 1000) is expressed as a percentage of dry weight combusted at 1000˚C. Zones are based on optimal splitting and broken-stick analysis. While the most marked species change occurs at ~5.25 cm, the shift in diatom assemblage appears to have begun earlier (~5.75 cm). Figure 4. Detrended Correspondance Analysis (DCA) of diatom species scores versus depth. The DCA axis 1 species scores are scaled in Standard Deviation (S.D.) units, and provide an estimate of species turnover. Figure 5. Mean June-July-August temperature data and annual precipitation data from Mould Bay, Prince Patrick Island (see Figure 1 for location) between 1948-1996 (Meteorological Service of Canada 2004). The smoothed line is a LOWESS curve with a span of 0.35.
Figure 1.
Figure 2.
Fi
gure
3
Figure 4.
Figure 5.
Table 1. Present-day physical and chemical characteristics of pond MV-AT were collected on July 24, 2002. Abbreviations are as follows: TP (total phosphorus, unfiltered), TN (total nitrogen), DOC (dissolved organic carbon), DIC (dissolved inorganic carbon), Chl a (chlorophyll a), POC (particulate organic carbon), PON (particulate organic nitrogen), SRP (soluble reactive phosphorus), TKN (total Kjeldhal nitrogen).
Parameter Value
Latitude 75˚19’N Longitude 111˚25’W Maximum depth 0.40 m Elevation 120 m asl Specific conductivity 39 μS/cm pH 8.1 TP 12.7 μg/L TN 640 μg/L Chl a 1.60 μg/L DOC 8.2 mg/L DIC 4.8 mg/L NH3 0.011 mg/L NO2 0.001 mg/L Cl 1.93 mg/L SO4 4.8 mg/L SiO2 0.27 mg/L POC 0.559 mg/L PON 0.053 mg/L SRP 0.0057 mg/L TKN 0.648 mg/L Al 0.02 mg/L Ba 0.0023 mg/L Cu 0.001 mg/L Fe 0.353 mg/L Li 0.001 mg/L Mn 0.0052 mg/L Sr 0.0112 mg/L Ca 5.2 mg/L Mg 2.7 mg/L Na 1.5 mg/L K 0.4 mg/L
CHAPTER 6
PROLONGED ICE COVER DAMPENS DIATOM COMMUNITY RESPONSES TO RECENT CLIMATIC
CHANGE IN HIGH ARCTIC LAKES
BRONWYN E. KEATLEY, MARIANNE S.V. DOUGLAS, AND JOHN P. SMOL
This chapter has been accepted for publication and is in press: Keatley, B.E., M.S.V. Douglas, and J.P. Smol. Prolonged ice cover dampens diatom community responses to recent climatic change in high Arctic lakes. Arctic Antarctic and Alpine Research.
ABSTRACT Numerous paleolimnological studies of high Arctic lakes and ponds have shown marked shifts in
both algal and invertebrate taxa within the past ~150 years that are consistent with recent climatic
warming. However, the magnitude and timing of changes are often non-uniform, with large, deep
lakes frequently exhibiting muted assemblage shifts relative to smaller ponds. The hypothesis
that duration and extent of ice cover exerts an overriding influence on habitat availability for
biota has been commonly invoked to explain these differences, and many studies indicate that
changes in ice cover are important drivers of recent biological changes. However, a detailed
paleolimnological comparison of two lakes from the same region that have similar water
chemistry but different ice cover regimes has not yet been attempted. Here we examine the
influence of prolonged ice cover on the rate, magnitude, and direction of fossil diatom species
shifts over time in two remarkably similar and adjacent Ellesmere Island lakes that mainly differ
in their periods of ice cover (by an estimated 7-10 days). These two lakes exhibit strikingly
different paleolimnological diatom profiles, despite their physical proximity, similar depths, and
nearly identical water chemistry. In the lake characterized by prolonged ice cover, we find little
evidence of diatom-inferred environmental change over its recent history, whilst diatom
assemblages have undergone dramatic changes in the lake with the shorter duration of ice cover.
This study supports the general hypothesis that changes in ice cover are a principle determinant of
shifting diatom assemblages in high Arctic lakes.
INTRODUCTION
High latitude regions are especially sensitive to climatic changes (ACIA, 2004) and
therefore represent critical reference areas for studies of long-term environmental change. As
long-term monitoring data are lacking with respect to both temporal and spatial scales, proxy
indicators of environmental change must be relied upon to reconstruct environmental histories in
this climatically sensitive area (Pienitz et al., 2004). The abundance of lakes and ponds in the
Canadian high Arctic allows for regional assessments of environmental change using
paleolimnological techniques. The accumulation of allochthonous and autochthonous biological,
chemical, and physical indicators in lacustrine sediments provides a rich archive of information
about past environments (Smol, 2002). Diatoms are siliceous unicellular algae (class
Bacillariophyceae) that are particularly useful environmental indicators because they are
ubiquitous, respond rapidly to changing conditions, and different species often have distinct
optima with respect to many important environmental variables (Stoermer and Smol, 1999).
Diatoms have been especially effective biomonitors in Arctic regions (Douglas and Smol, 1999;
Douglas et al., 2004; Solovieva et al., 2005), where other sources of proxy data are sometimes
lacking.
In a recent analysis of 55 paleolimnological profiles from circumpolar lakes and ponds,
Smol et al. (2005) summarized the often dramatic changes in biological microfossils, including
diatoms, which have occurred over the last ~150 years. Assemblage changes were found to be
consistent with climatic warming, with the greatest changes observed in the northernmost regions.
Interpretations based on diatoms are also consistent with studies using non-biological proxies
(e.g. Smith et al., 2004). Changes in lake ice cover were first proposed by Smol (1983) as a
major cause for some of the striking species changes, and this idea was developed further by
Smol (1988) and Douglas and Smol (1999). Briefly, during cooler periods, the persistence of ice
cover can reduce primary production by limiting light penetration into the lake. Under warmer
conditions, ice cover is reduced, more light is available for photosynthesis, and as a result more
habitat becomes available for colonization by algae. However, other factors such as lake depth,
also play a role in how changes in climate, manifested through changes in lake ice, could affect
diatom flora. For example, in relatively deep, extensively ice-covered lakes near Alert and
Resolute Bay (Fig. 1), diatom communities showed subtle shifts that began over the last few
decades (Doubleday et al., 1995; Michelutti et al., 2003; Antoniades et al., 2005). In contrast,
small, shallow ponds at Cape Herschel (Ellesmere Island) and Isachsen (Ellef Ringnes Island)
(Fig. 1) tracked relatively dramatic changes in diatoms beginning much earlier, in the late 19th or
early 20th century, respectively (Douglas et al., 1994; Antoniades et al., 2005). Likewise, across
the circumpolar north, many deeper lakes have recorded increases in planktonic species during
recent history (Sorvari et al., 2002; Rühland et al., 2003), whereas shallow ponds have been
characterized by shifts in taxa indicative of greater periphytic (aquatic plant) habitat availability
(Smol et al., 2005). Both the development of plankton (in deeper lakes) and periphyton (in
shallow ponds), as well as the delayed impact of warming in some locations, could be explained
by changes in the persistence of ice cover, and associated limnological changes in the physical,
chemical, and biological characteristics of deeper versus shallower lakes (Smol, 1988; Douglas
and Smol, 1999).
While a large diatom calibration set has been developed to infer ice cover duration in
subarctic lakes in Fennoscandia (Thompson et al., 2005), the ice cover hypothesis has yet to be
explicitly tested using a paleolimnological approach. In this paper, we test the hypothesis that ice
cover is the overriding influence on diatom communities in high Arctic lakes by examining
diatom assemblage shifts in two small, adjacent lakes, Skeleton Lake and Lake EP2, in northern
Ellesmere Island, which are very similar in all limnological respects (e.g. location, morphology,
and water chemistry variables that influence diatoms) other than duration of ice cover.
Although instrumental records of climate change do not exist for the study sites and the
Lake Hazen region of northern Ellesmere Island, previous diatom-based paleolimnological
investigations suggest that environmental changes have occurred at Alert (Doubleday et al., 1995;
Antoniades et al., 2005), the Hazen Plateau (Smith, 2002), and the Fosheim Peninsula (Wolfe,
2000; Perren et al., 2003). Likewise, glacier mass balance records (Braun et al., 2004), melt
records from the nearby Agassiz Ice Cap (Koerner and Fisher, 1990), and instrumental data from
Alert and Eureka weather stations (Meteorological Service of Canada, 2006) indicate a regional
recent warming in recent decades. Thus, while recent environmental change has occurred in this
region, ice cover continues to persist longer on one lake than the other due to local cooling caused
by shading from a nearby hill. If climate modulated changes in the duration and extent of ice
cover is truly the dominant factor driving diatom assemblage shifts in Arctic lakes, we predict the
greatest ecological changes to occur in the lake with the longest open water period.
SITE DESCRIPTION
The two study lakes, Skeleton Lake and EP2 (unofficial name), are small (estimated
maximum depths ~4 m and ~3 m, respectively) high Arctic lakes (81°50’N, 71°28’W), located on
the north side of Lake Hazen, Ellesmere Island, Nunavut, Canada. Skeleton Lake drains into EP2
(less than 20 m to the east) via Skeleton Creek, which in turn flows through a third pond (EP3)
and then into Lake Hazen (Fig. 2). Skeleton Lake has repeatedly been shown to become ice free
later than EP2 (Fig. 3; Oliver and Corbet, 1966; National Air Photo Archives; and field
observations, this study, 2003). This is likely due to the greater shading and protection from
winds on Skeleton Lake by Blister Hill, located to the south (Fig. 2; Oliver and Corbet, 1966).
Skeleton Lake has been noted for its luxuriant cyanobacterial mats (Quesada et al. 1999),
although similar studies have not been conducted on EP2.
For its latitude, the Hazen Basin region experiences anomalously warm summer
conditions due to its continental location and its placement on the leeward side of the Grant Land
Mountains (Thompson, 1994). In July 2003, for example, mean maximum daily temperatures at
Eureka and Alert were 10.9˚C and 4.3˚C, respectively (Meteorological Service of Canada, 2006),
while the mean maximum daily temperature at Lake Hazen camp was 16˚C. The area receives
very little precipitation (~95 mm annually; Thompson, 1994). The frost-free period at Lake
Hazen (Fig. 1) lasts 8-10 weeks and, as such, supports a greater abundance and diversity of
vegetation than the surrounding areas (Soper and Powell, 1985). Skeleton Lake, EP2, and EP3
are surrounded by wet meadows which are predominantly covered with Carex aquatilis var. stans
and Eriophorum spp., as well as Saxifraga, Lychnis, Salix, Stellaria, Arctagrostis, Polygonum,
and Equisetum, and many types of mosses (Soper and Powell, 1985). Bedrock consists of
Jurassic and Cretaceous sandstone and shale; this is overlain by glacial till, sand, gravel, talus and
soils of Pleistocene and more recent origin (Christie, 1964).
METHODS
Water chemistry measurements and water samples were taken from Skeleton Lake, EP2,
and EP3 within approximately one hour of each other on 7 July 2003. Temperature, pH, and
specific conductivity were measured in the field with a handheld thermometer, a handheld Hanna
pHEP pH meter and aYSI model 33 conductivity meter, respectively. Water samples were also
taken for total filtered phosphorus (TPf), total unfiltered phosphorus (TPu), nitrate-nitrite-
nitrogen (NO3-NO2-N), ammonia-nitrogen (NH3-N), total Kjeldahl nitrogen (TKN, filtered), total
dissolved nitrogen (TdN), particulate nitrogen (PON), dissolved organic carbon (DOC), dissolved
inorganic carbon (DIC), particulate carbon (POC), dissolved silica (SiO2), chlorophyll a (Chl a,
uncorrected for phaeophytin), calcium (Ca), magnesium (Mg), sodium (Na), potassium (K),
sulfate (SO4), and chloride (Cl). Laboratory analyses were performed at the National Laboratory
for Environmental Testing (NLET), Environment Canada, Burlington, ON, Canada, using
methods outlined in Environment Canada (1994). Details about sampling methods are given in
Appendix 1.
Sediment cores were obtained using a modified Glew (1989) corer (3.75 cm inner
diameter) from the deepest part of both Skeleton Lake (33 cm long) and EP2 (17 cm long) on 14
July 2003. Cores were sectioned on-site at 0.5 cm intervals using a Glew (1988) extruder. The
uppermost (~2 cm) sediments from the adjacent pond, EP3, were sampled by hand to allow for a
comparison of the recently deposited diatom assemblages between the three sites.
Subsamples from each 0.5 cm thick interval of both cores were analysed for 210Pb using
alpha spectrometry at MyCore Scientific, Deep River, ON, Canada. The 210Pb dates for Skeleton
Lake were calculated using the constant rate of supply (CRS) method (Appendix 7; Binford,
1990). In the sediments from EP2, the 210Pb activity levels were too low to provide a reliable age
model, so trace elements were used to identify anthropogenic pollution horizons as an
approximate marker for the onset of the industrial revolution at ~1850 AD. Trace element
analyses were carried out by Activation Laboratories Inc., in Ancaster, Ontario, Canada. A suite
of metals (including total Pb) were measured using a Perkin Elmer SCIEZ ELAN 6100
Inductively Coupled Plasma Mass Spectrometry (ICP/MS). International certified reference
materials USGS GXR-1, GXR-2, GXR-4 and GXR-6 were analyzed at the beginning and end of
each batch of samples. Internal control standards were analyzed every 10 samples and a duplicate
was run for every 10 samples. Total Hg was measured on a Perkin Elmer FIMS 100 cold vapour
Hg analyzer. Organic matter was estimated by loss on ignition (LOI) at 550°C for 2 hours (Heiri
et al., 2001) at PEARL in Kingston, Ontario, Canada.
Microwave digestion techniques were used to clean diatom frustules from the sediment
with concentrated HNO3 (Parr et al., 2004). Diatom slurries were rinsed with deionized water
until they reached a neutral pH, dried on pre-cleaned glass cover slips, and mounted on slides
with Naphrax® (high refractive index mounting medium). A minimum of 500 diatom valves was
identified and enumerated from each sample using a Leica DMRB microscope. Taxonomy
primarily followed Krammer and Lange-Bertalot (1986-1991), Krammer (2002), and Antoniades
et al. (in press). In order to examine whether rare species were tracking environmental changes in
these lakes, we used all species in a principal components analysis (PCA). We also performed a
detrended correspondence analysis (DCA) to examine species turnover in standard deviation (SD)
units (Lepš and Šmilauer, 2003). This ecologically-relevant interpretation provides an objective
comparison of species compositional change between the two sites.
RESULTS
Modern limnology
The physical and chemical variables measured for Skeleton Lake, EP2, and EP3 are
shown in Table 1. The variables of greatest limnological interest that influence diatom
assemblages, including pH, specific conductivity, and dissolved nutrients, are nearly identical
between the three lakes; notable exceptions to this include temperature and the particulate
fractions of carbon (POC) and nitrogen (PON), and NH3. However, when examined in the
context of 55 lakes and ponds from northern Ellesmere Island, a principal components analysis
(PCA), based on water chemistry variables that are expected to influence diatom assemblages,
indicates that they are more similar to each other than to 52 other sites from northern Ellesmere
Island (Fig. 4).
Diatom assemblages identified from the recently deposited sediments of the three lakes
are listed in Table 2. All three lakes are characterized by small, benthic diatoms and share high
abundances of some common taxa (e.g. small, benthic Fragilaria spp. sensu lato). The diatom
assemblage in the surface sediments from EP3 was the most diverse of the three lakes, and
consisted of small Achnanthes spp. (A. petersenii, A. minutissima), Brachysira zellensis,
Cymbella spp., Denticula spp., and small Nitzschia species (Table 2). Similar to EP2, the most
prevalent diatom in EP3 was Fragilaria brevistriata suggesting that some, as of yet, unidentified
characteristic about these water bodies currently favours the growth of F. brevistriata.
Paleolimnology
Unsupported 210Pb concentrations in the Skeleton Lake sediment core displayed a largely
exponential decline until reaching background levels at 4.75 cm depth, corresponding to an age of
~1877 AD (Fig. 5, Appendix 11). Since the 210Pb concentrations were too low in the EP2
sediment core to obtain reliable dates, we examined total Pb and Hg concentrations as an
approximate time horizon indicating the onset of airborne anthropogenic pollution. Sedimentary
concentrations of Pb and Hg in ponds from nearby Greenland have been shown to increase
starting in the early 19th and 20th centuries (Bindler et al., 2001a; 2001b). Thus, a rise in the Hg
and Pb concentrations at 9 cm depth is interpreted as corresponding to ~150 to 200 years BP (see
Appendix 12 for total Hg and total Pb for Skeleton Lake).
A list of common diatom taxa and the DCA axis 1 gradient lengths (an estimate of
change in species composition) of the sediment cores of Skeleton Lake and EP2 can be found in
Table 3. There is little change in the 33 cm Skeleton Lake core (Fig. 6a); this is reflected in the
short gradient length of DCA axis 1 (1.06 SD units, Table 3). Diatom assemblages are composed
of small, benthic taxa, dominated by Fragiliaria construens var. venter (~80-90%). Throughout
much of this core, F. construens var. venter co-occurs with F. pinnata (up to ~20%). In the top 2
cm (corresponding to approximately the last ~30 years), there is a small increase in F.
brevistriata (up to its maximum relative abundance of ~10%), as well as the near complete
disappearance of F. pinnata.
In nearby Lake EP2, which is unshaded, warmer, and experiences less prolonged ice
cover than Skeleton Lake (Fig. 6b), three periods of diatom change occur within the 17 cm
sediment core (Fig. 6c). These patterns in the dominant species are also reflected when all
species are incorporated in the DCA, which indicates a species turnover of 2.00 SD units (Table
3). Below 10 cm, F. pinnata dominated the record (up to ~80% relative abundance), and was
accompanied by other small benthic diatoms such as F. construens var. venter and F. brevistriata
(Fig. 6c). At ~9.25 cm, estimated to be ~100-150 years BP based on our geochemical proxies,
diatom assemblages diversified to include taxa such as Denticula kuetzingii, small Nitzschia
species (N. frustulum, N. inconspicua, and N. perminuta), and Cymbella spp. In the uppermost
sediments, between 0-2 cm, a small benthic diatom, F. brevistriata, becomes the dominate taxon.
DISCUSSION
Modern limnology
The similarity of the modern limnological variables amongst the three lakes is a function
of their similar bedrock, regional climate, and connectivity via Skeleton Creek. The notable
exceptions to these similarities are temperature and the particulate fractions of carbon (POC) and
nitrogen (PON), which are all higher in EP2 and EP3 than Skeleton Lake, and NH3, which is
highest in EP2 (Table 1). The lower temperature of Skeleton Lake can be attributed to the
persistent ice cover that was present at the time of sampling; this ice cover is in turn related to the
local shading of Skeleton Lake. Both POC and PON are measures of particulate matter in the
water column, and not directly indicative of nutrients that are available to photosynthetic
organisms. Higher particulate concentrations in EP2 and EP3 suggest that they have higher
concentrations of algal and detrital matter, which are indicative of higher production and/or
inflow energy in these systems, as would be expected with decreased ice cover (Smol, 1988).
Moreover, as water flows from Skeleton Lake through EP2 and then EP3, it is reasonable to
expect higher particulate matter downstream. Although NH3 concentrations are twice as high in
EP2 as Skeleton Lake, both values are low compared to the range of NH3 values for 24 ponds and
lakes located within ~10 km radius of our study site (range NH3 = 0.003 to 0.145 mg/L, Keatley
et al., [4]). In any case, as all three lakes are P limited, not N limited (Table 1), slight differences
in these nutrients are not likely to be biologically significant.
Modern diatom assemblages
Although there are some specific taxonomic differences between the diatom assemblages
present in the recently deposited sediments of Skeleton Lake, EP2, and EP3, they also exhibit
some important similarities (Table 2). For example, small benthic Fragilaria (sensu lato) taxa are
the most abundant diatoms in each of the three sites. Small Fragiliara taxa, especially F.
pinnata, are considered to be pioneering diatoms that can commonly exploit harsh conditions and
have been well documented as an indicator of cool conditions with short growing seasons in both
the Arctic (Douglas et al., 1994; Michelutti et al., 2003a), and in alpine regions (Lotter and
Bigler, 2000), and are frequently the dominant taxa found in pre-industrial polar lake sediments
(Smol et al., 2005). Ecological preferences that can distinguish these small Fragilaria taxa are
difficult to define; for example, Karst-Riddoch (2004) suggested that the distribution of small
benthic Fragilaria taxa (F. pinnata, F. construens var. venter, F. brevistriata) in many lakes
across Iceland was not related to measured environmental variables.
The differences amongst the three lakes can best be appreciated with respect to the higher
diversity of taxa representing different growth forms and life strategies. In Lake EP3, which
experiences the least shading and is the shallowest lake, the diatom assemblage consists of many
species known to have secondary growth characteristics (e.g. Cymbella taxa often have
mucilaginous stalks). Such diverse assemblages have been linked to regions with longer growing
seasons, as these more complex life forms would take longer to develop (Douglas and Smol,
1999).
Paleolimnology
Skeleton Lake and EP2 have very similar morphologies, water chemistry, are located
within ~20 m of each other, and are connected by Skeleton Creek, yet the historical composition
of diatom species assemblages is strikingly different between the two lakes. Although both lakes
must have experienced warming to some degree, Skeleton Lake has had more persistent ice cover
than EP2. Thus, it was expected that diatom assemblage shifts consistent with climate warming
would be recorded in EP2, but that only a muted diatom response to the same environmental
drivers would be apparent in Skeleton Lake due to shading that leads to a cooler microclimate and
extended ice cover (Smol, 1988).
In Skeleton Lake, the consistent domination of the diatom assemblages by small, benthic
Fragilaria taxa is likely indicative of cool conditions characterized by extensive ice cover (e.g.
Smol, 1988; Douglas and Smol, 1999; Lotter and Bigler, 2000). In the most recent sediments, the
subtle increase in F. brevistriata is similar to that found in the very recent sediments, representing
the warmest years on record, of Char Lake, a large, mostly ice-covered high Arctic lake on
Cornwallis Island (Michelutti et al., 2003a). Thus, Skeleton Lake sediments indicate a relatively
muted response to recent regional warming, consistent with its more extended ice cover.
In Lake EP2, on the other hand, shifts in diatom assemblages from those indicative of
cool, harsh conditions towards a diversification of taxa indicative of more diverse substrate
availability (e.g. moss epiphytes such as Denticula kuetzingii; Lim et al., 2001; Michelutti et al.,
2003b), and larger, more complex frustules (e.g. Cymbella taxa) are consistent with climate
warming and reduced ice cover. Such an assemblage shift is consistent with climatic warming
and reduced ice cover duration that would have led to a longer growing season for aquatic
macrophytes and algae, as well as enhanced nutrient export from the catchment (Smol, 1988;
Douglas and Smol, 1999). These types of diatom species changes are consistent with many other
diatom changes seen in lakes and ponds throughout the Arctic (e.g. Douglas et al., 1994;
Michelutti et al., 2003a; Antoniades et al., 2005; Smol et al., 2005).
Alternative hypotheses
Other factors, such as acid precipitation, atmospheric deposition of anthropogenic
nitrogen or other nutrients or pollutants, and/or some other unmeasured phenomenon, may
potentially be proposed as alternate drivers explaining the diatom assemblage changes in Skeleton
Lake and EP2. A comparison of the diatom profiles between the two lakes addresses these other
factors.
The remote location of our study sites, far from point sources of pollution of any type,
identifies atmospheric deposition as the major pathway of pollutant transport to these sites. Due
to the proximity of our sites, both lakes are subjected to identical types and quantities of
atmospheric deposition. Nonetheless, if some type of atmospheric deposition was affecting the
lakes differentially, we should see differences in our water chemistry data, especially with respect
to pH, dissolved nitrogen, and sulphate. However, both Skeleton Lake and EP2 are presently
nearly identical with respect to these variables (Table 1); with the exceptions of POC, PON, and
NH3, which have been discussed above. Furthermore, historical water chemistry measurements
indicate that Skeleton Lake and EP2 have remained circumneutral to alkaline and, if anything,
have slightly increased in pH since 1964 (Oliver and Corbet, 1966). These historical data also
confirm that the water chemistry of Skeleton Lake and EP2 were very similar in 1964 (Oliver and
Corbet, 1966), just as it was during our field work in 2003.
Enrichment from atmospheric nitrogen has been linked to diatom community shifts in
lakes in the Rocky Mountains, USA (e.g. Wolfe et al., 2001; Saros et al., 2003). In Arctic lakes
on Baffin Island, nitrogen deposition may have acted in concert with climate changes to cause
diatom shifts after 1950, although diatom community change commenced in the mid-19th century
(Wolfe et al., 2006). Although our extremely remote study sites must have received
comparatively less atmospheric N input than those in the continental USA, persistent ice on
Skeleton Lake could lead to differences in the timing of N delivery to the two lakes, thereby
eliciting unique ecological responses. However, the diatom changes in the very remote Lake EP2
began over the last ~100-150 years, based on our approximate chronological estimate, while
atmospheric N deposition in comparatively impacted lakes in the continental USA only become
an ecologically important source of N after the mid- to late 20th century (e.g. Wolfe et al., 2001;
Saros et al., 2003). Thus, even if differences in timing of delivery of N deposition were affecting
diatom communities, this could only explain much more recent changes, and in any case, would
ultimately still be traced back to differences in ice cover regime.
With respect to other types of pollution, previous paleolimnological studies have shown
that no changes in diatom assemblages have been recorded in lakes that have experienced point-
source pollution from PCBs (Paterson et al., 2003). Thus, differences in acidification, amount of
nitrogen deposition, and pollution cannot explain the types of diatom responses recorded in the
Skeleton Lake and EP2 cores.
The results from the sediment cores from Skeleton Lake and EP2, as well as the surface
sediment diatoms from EP3, all suggest that ice cover plays an important role in dampening
diatom community shifts to environmental change. This is likely due to the role of ice cover in
restricting the growing season for both algae and their potential substrates (e.g. mosses and
grasses). However, due to microclimatic differences, the ice cover continues to persist longer at
Skeleton Lake than at EP2. Thus, the changes recorded in the diatom records may reflect the
crossing of some type of ice cover threshold in EP2 that has not yet been reached in Skeleton
Lake, as the climatic changes have not yet been great enough to reduce ice cover to the point at
which dramatic diatom shifts are occurring (Fig. 7). Furthermore, these results underscore the
importance of careful site selection and contextual interpretation in paleolimnological studies, as
local conditions may greatly influence ecological responses to regional environmental change
(Smol et al., 2005).
CONCLUSIONS
Although Skeleton Lake and EP2 are nearly identical with respect to water chemistry,
morphology, and are located within ~ 20 m of each other, the two lakes have recorded strikingly
different diatom histories. The major difference between the two sites is the longer length of ice
cover on Skeleton Lake relative to EP2. Both the muted diatom assemblage change in Skeleton
Lake and the marked changes in EP2 suggest that ice cover plays an important role in influencing
biological assemblages in polar lakes.
ACKNOWLEDGEMENTS
Funding for this project came from Natural Sciences and Engineering Research Council of
Canada (NSERC) grants to BEK, MSVD, and JPS, and from a Northern Scientific Training
Program (NSTP) grant to BEK. We would also like to thank the Polar Continental Shelf Project
and Parks Canada for logistical support. We are grateful to the National Laboratory for
Environmental Testing (NLET) at Environmental Canada for water chemistry analysis, S. Arnott
and J. Glew for assistance in the field, and J.R. Glew for drafting the detailed inset map. Many
thanks to D. Antoniades, N. Michelutti, and the anonymous journal reviewers for constructive
comments on the manuscript, and to one of the journal reviewers for providing Figure 4. This is
PCSP contribution number 010-07.
REFERENCES
ACIA, 2004: Impacts of a warming Arctic. Arctic Climate Impact Assesment. Cambridge: Cambridge University Press, 140 pp.
Antoniades, D., Douglas, M.S.V., and Smol, J.P., 2005: Quantitative estimates of recent environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. Journal of Paleolimnology, 33: 349 - 360.
Antoniades, D., Hamilton, P.B., Douglas, M.S.V., and Smol, J.P., in press: Freshwater diatoms from the Canadian High Arctic, submitted to Diatomologica Iconographica.
Bindler, R., Renberg, I., Anderson, N.J., Appleby, P.G., Emteryd, O., and Boyle, J., 2001a: Pb isotope ratios of lake sediments in West Greenland: inferences on pollution sources. Atmospheric Environment, 35: 4675 - 4685.
Bindler, R., Renberg, I., Appleby, P.G., Anderson, N.J., and Rose, N.L., 2001b: Mercury accumulation rates and spatial patterns in lake sediments from west Greenland: A coast to ice margin transect. Environmental Science and Technology, 35: 1736 - 1741.
Binford, M.W., 1990: Calculation and uncertainty analysis of 210Pb dates for PIRLA project lake sediment cores. Journal of Paleolimnology, 3: 253 - 267.
Braun, C., Hardy, D.R., and Bradley, R.S., 2004: Mass balance and area changes of four High Arctic plateau ice caps, 1959-2002. Geografiska Annaler Series A-Physical Geography, 86A: 43 - 52.
Christie, R.L., 1964: Geological reconnaissance of northeastern Ellesmere Island, District of Franklin. Ottawa: Geological Survey of Canada Memoir 331: 79 pp.
Doubleday, N.C., Douglas, M.S.V., and Smol, J.P., 1995: Paleoenvironmental studies of black carbon deposition in the High Arctic - a case-study from northern Ellesmere Island. Science of the Total Environment, 161: 661 - 668.
Douglas, M.S.V., Hamilton, P.B., Pienitz, R., and Smol, J.P., 2004: Algal indicators of environmental change in Arctic and Antarctic lakes and ponds. In Pienitz, R., Douglas, M.S.V., and Smol, J.P. (eds.) Long-term environmental change in Arctic and Antarctic lakes. Dordrecht: Springer, 117 - 158.
Douglas, M.S.V., and Smol, J.P., 1999: Freshwater diatoms as indicators of environmental change in the High Arctic. In Stoermer, E.F. and Smol, J.P. (eds.) The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge: Cambridge University Press, 227 - 244.
Douglas, M.S.V., Smol, J.P., and Blake Jr, W., 1994: Marked post-18th century environmental change in high-arctic ecosystems. Science, 266: 416 - 419.
Environment Canada, 1994: Manual of Analytical Methods National Laboratory for Environmental Testing, Canadian Centre for Inland Waters. Burlington, Canada, 2 vols., 1097 pp.
Glew, J.R., 1988: A portable extruding device for close interval sectioning of unconsolidated core samples. Journal of Paleolimnology, 2: 241 - 243.
Glew, J.R., 1989: A new trigger mechanism for sediment samplers. Journal of Paleolimnology, 2: 241 - 243.
Heiri, O., Lotter, A.F., and Lemcke, G., 2001: Loss on ignition as a method for estimating organic and carbonate content in sediments: reproducibility and comparability of results. Journal of Paleolimnology, 25: 101 - 110.
Hill, M.O., 1973: Diversity and evenness: a unifying notation and its consequences. Ecology, 54: 427 - 432.
Jackson, C.I., 1959: Operation Hazen: The Meteorology of Lake Hazen, N.W.T. Part 1: Analysis of the observations. Ottawa: Defence Research Board, 194 pp.
Joynt, E.H. III, and Wolfe, A.P., 2001: Paleoenvironmental inference models from sediment diatom assemblages in Baffin Island lakes (Nunavut, Canada) and reconstruction of summer water temperatures. Canadian Journal of Fisheries and Aquatic Sciences, 58: 1222 - 1243.
Karst-Riddoch, T.L., 2004: Sedimentary diatoms from sensitive alpine (northwest Canada) and subarctic (Iceland) lakes as indicators of environmental and climatic conditions. PhD thesis, Department of Biology, Queen’s University, Kingston, Canada. 194 pp.
Keatley, B.E., Douglas, M.S.V., and Smol, J.P., [4]. Limnological characteristics of a high Arctic
oasis and comparisons across northern Ellesmere Island. Arctic. in press. Koerner, R.M. and Fisher, D.A., 1990: A record of Holocene summer climate from a Canadian
high-Arctic ice core. Nature, 343: 630 - 631.
Krammer, K., 2002: Diatoms of Europe: Cymbella. Ruggell: A.R.G. Gantner Verlag, 584 pp.
Krammer, K. and Lange-Bertalot, H., 1986-1991: Bacillariophyceae, Stuttgart: Gustav Fisher Verlag., 2485 pp.
Lim, D.S.S., Kwan, C., and Douglas, M.S.V., 2001b: Periphytic diatom assemblages from Bathurst Island, Nunavut, Canadian High Arctic: An examination of community relationships and habitat preferences. Journal of Phycology, 37: 379-392.
Lepš, J., and Šmilauer, P., 2003: Multivariate Analysis of Ecological Data using CANOCO, Cambridge: Cambridge University Press, 269 pp.
Lotter, A.F. and Bigler, C., 2000: Do diatoms in the Swiss Alps reflect the length of ice-cover? Aquatic Sciences, 62: 125 - 141.
Meteorological Service of Canada, 2006: Monthly Climate Data. Environment Canada http://www.climate.weatheroffice.ec.gc.ca/climateData/monthlydata_e.html , Accessed July 2006
Michelutti, N., Douglas, M.S.V., and Smol, J.P., 2003a: Diatom response to recent climatic change in a high arctic lake (Char Lake, Cornwallis Island, Nunavut). Global and Planetary Change, 38: 257 - 271.
Michelutti, N., Douglas, M.S.V., and Smol, J.P., 2003b: Periphytic diatom assemblages from ultra-oligotrophic and UV transparent lakes and ponds on Victoria Island and comparisons with other diatom surveys in the Canadian Arctic. Journal of Phycology, 39: 465-480.
Oliver, D.R. and Corbet, P.S., 1966: Operation Hazen: Aquatic habitats in a high arctic locality: the Hazen Camp study area, Ellesmere Island, N.W.T. Ottawa: Department of National Defence. 126 pp.
Parr, J.F., Taffs, K.H., and Lane, C.M., 2004: A microwave digestion technique for the extraction of fossil diatoms from coastal lake and swamp sediments. Journal of Paleolimnology, 31: 383 - 390.
Paterson, A.M., Betts-Piper, A.A., Smol, J.P., and Zeeb, B.A., 2003: Diatom and chrysophyte algal response to long-term PCB contamination from a point-source in northern Labrador, Canada. Water Air and Soil Pollution, 145: 377 - 393.
Perren, B.B., Bradley, R.S., and Francus, P., 2003: Rapid lacustrine response to recent High Arctic warming: A diatom record from Sawtooth Lake, Ellesmere Island, Nunavut. Arctic Antarctic and Alpine Research, 35: 271 - 278.
Pienitz, R., Douglas, M.S.V., and Smol, J.P., 2004: Long-Term Environmental Change in Arctic and Antarctic Lakes. Dordrect: Springer, 550 pp.
Quesada, A., Vincent, W. F., and Lean, D.R.S., 1999: Community and pigment structure of Arctic cyanobacterial assemblages: the occurrence and distribution of UV-absorbing compounds. FEMS Microbiology Ecology, 28: 315-323.
Rühland, K., Priesnitz, A., and Smol, J.P., 2003: Paleolimnological evidence from diatoms for recent environmental changes in 50 lakes across Canadian Arctic treeline. Arctic Antarctic and Alpine Research, 35: 110 - 123.
Saros, J.E., Interlandi, S.J., Wolfe, A.P., and Engstrom, D.R., 2003: Recent changes in the diatom community structure of lakes in the Beartooth Mountain Range, USA. Arctic Antarctic and Alpine Research, 35: 18 - 23.
Smith, I.R., 2002: Diatom-based holocene paleoenvironmental records from continental sites on northeastern Ellesmere Island, high Arctic, Canada. Journal of Paleolimnology, 27: 9 - 28.
Smith, Jr.S.V., Bradley, R.S., and Abbott, M.B., 2004: A 300 year record of environmental change from Lake Tuborg, Ellesmere Island, Nunavut, Canada. Journal of Paleolimnology, 32: 137 - 148.
Smol, J.P., 1983: Paleophycology of a high arctic lake Cape Herschel, Ellesmere Island. Canadian Journal of Botany, 61: 2195 - 2204.
Smol, J.P., 1988: Paleoclimate proxy data from freshwater Arctic diatoms. Verh. int. Ver. Limnol., 23: 837 - 844.
Smol, J.P., 2002: Pollution of lakes and rivers: a paleoenvironmental perspective. London: Arnold, 280 pp.
Smol, J.P., Wolfe, A.P., Birks, H.J.B., Douglas, M.S.V., Jones, V.J., Korhola, A., Pienitz, R., Ruhland, K., Sorvari, S., Antoniades, D., Brooks, S.J., Fallu, M.A., Hughes, M., Keatley, B.E., Laing, T.E., Michelutti, N., Nazarova, L., Nyman, M., Paterson, A.M., Perren, B., Quinlan, R., Rautio, M., Saulnier-Talbot, E., Siitoneni, S., Solovieva, N., and Weckstrom, J., 2005: Climate-driven regime shifts in the biological communities of arctic lakes. Proceedings of the National Academy of Sciences of the United States of America, 102: 4397 - 4402.
Solovieva, N., Jones, V.J., Nazarova, L., Brooks, S.J., Birks, H.J.B., Grytnes, J.A., Appleby, P.G., Kauppila, T., Kondratenok, B., Renberg, I., and Ponomarev, V., 2005: Palaeolimnological evidence for recent climatic change in lakes from the northern Urals, Arctic Russia. Journal of Paleolimnology, 33: 463 - 482.
Soper, J.H. and Powell, J.M., 1985: Botanical studies in the Lake Hazen Region, northern Ellesmere Island, Northwest Territories, Canada. Ottawa: Publications in Natural Sciences: National Museums of Canada, 67 pp.
Sorvari, S., Korhola, A., and Thompson, R., 2002: Lake diatom response to recent Arctic warming in Finnish Lapland. Global Change Biology, 8: 171 - 181.
Stoermer, E.F. and Smol, J.P. (eds), 1999: The diatoms: applications for the environmental and earth sciences. Cambridge: Cambridge University Press, 469 pp.
Thompson, W., 1994: Climate. In: Resource description and analysis: Ellesmere Island National Park Reserve. National Resource Conservation Section, Prairie and Northern Region, Parks Canada, Department of Canadian Heritage, Winnipeg. 78 pp.
Thompson, R., Price, D., Cameron, N., Jones, V., Bigler, C., Rosen, P., Hall, R.I., Catalan, J.,
Garcia, J., Weckstrom, J., and Korhola, A., 2005: Quantitative calibration of remote mountain-lake sediments as climatic recorders of air temperature and ice-cover duration. Arctic Antarctic and Alpine Research, 37: 626 - 635.
Wolfe, A.P., 2000: A 6500-year record from southwestern Fosheim Peninsula, Ellesmere Island, Nunavut. In Garneau, M. and Alt, B.T. (eds.) Environmental Response to Climate Change in the Canadian High Arctic. Ottawa: Geological Survey of Canada, Bulletin 529, 249 - 256.
Wolfe, A.P., Baron, J.S., and Cornett, R.J., 2001: Anthropogenic nitrogen deposition induces rapid ecological changes in alpine lakes of the Colorado Front Range (USA). Journal of Paleolimnology, 25: 1 - 7.
Wolfe, A.P., Cooke, C.A., and Hobbs, W.O., 2006: Are current rates of atmospheric nitrogen deposition influencing lakes in the Eastern Canadian Arctic? Arctic Antarctic and Alpine Research, 38: 465 - 476.
Figure captions. Figure 1. Map showing the locations of the study site (star) and the other sites (numbered) mentioned in the text: 1. Alert, 2. Hazen Plateau, 3. Agassiz Ice Cap, 4. Fosheim Peninsula, 5. Cape Herschel, 6. Isachsen, Ellef Ringnes Island, 7. Char Lake, Cornwallis Island. Figure 2. Topographical map detailing the local topography near Skeleton Lake, EP2 and EP3. Figure 3. Ice-off dates for Skeleton Lake (solid bars) and Lake EP2 (hatched bars). The data from the 1960s is from Oliver and Corbett (1966), while the 2003 data is from our own field observations. These data are also corroborated by archival air photos that show ice persisting on Skeleton Lake after Lake EP2 has become ice-free. Figure 4. Principal components analysis (PCA) biplot based on measured water chemistry variables that are considered to influence diatom assemblages. The arrows represent the measured environmental variables, while the open circles represent 52 lakes and ponds across northern Ellesmere Island. Skeleton Lake is represented by the solid square, EP2 is represented by the open square, and EP3 is represented by the solid circle. The proximity of these three sites represents their highly similar water chemistry. Figure 5. a) Age-depth model for Skeleton Lake based on CRS model (Binford, 1990), and b) Total 210Pb activities as estimated by alpha spectroscopy. The dashed line indicates estimated supported 210Pb. Figure 6a. Diatom profile of Skeleton Lake showing taxa present in at least at least one interval with a relative abundance of >3%. See Table 2 for synonyms for some of the common taxa. Percent loss-on-ignition (%LOI; an estimate of organic matter) and PCA axis 2 sample scores (PCA2; a summary of change occurring in rare taxa) are presented at the right side of the profile. Figure 6b. Photographs of ice cover on Skeleton Lake and Lake EP2, indicating the physical proximity of the two lakes, Blister Hill, and a nearby pingo for reference (photographs taken 7 July 2003). Figure 6c. Diatom profile of EP2 showing only species that are present in at least 3% relative abundance in at least one interval. See Table 2 for synonyms for some of the common taxa. Percent loss-on-ignition (% LOI), total Pb and total Hg (both expressed per gram organic carbon), and PCA axis two sample scores (PCA2) are presented at the right side of the profile. The rise in total Pb and Hg are interpreted to mark the onset of anthropogenic pollution (mid-19th to early 20th century). Figure 7. Schematic diagrams illustrating possible diatom responses to changing ice cover conditions in the two study lakes. Although the length of the ice-free season has likely increased in both lakes, it has yet to reach a critical threshold in Skeleton Lake, resulting in a muted diatom response. In Lake EP2, reduction in ice cover has crossed this critical threshold and resulted in marked diatom changes.
Figure 1.
Figure 2.
27-Jun
02-Jul
07-Jul
12-Jul
17-Jul
22-Jul
27-Jul
1963 1964 2003
Year
Dat
e
Figure 3.
-0.4 1.0
-1.0
1.0
pH
conductivity
Chla
DICDOC
NH3
TKN
TdNTN
TPu
TPf
SRP
SiO2
Ca
K
Na
Mg
ClSO4
Figure 4.
Year AD
1860
1880
1900
1920
1940
1960
1980
2000
Dep
th (c
m)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
A
Total 210Pb (pCi/g)
0.51.01.52.0
Dep
th (c
m)
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
32.5B
Figure 5.
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32Depth (cm)
20F. pinn
ata 2550
100
F. con
strue
nsva
r.
vente
rcomple
x
20F. brev
istria
ta
F. pse
udoc
onstr
uens
2040
Rel
ativ
e ab
unda
nce
(%)
%
LOI
-20
2
PCA2
75
Depth (cm)
Rel
ativ
e ab
unda
nce
(%)
%pp
b/g
Cpp
m/g
C
F. con
strue
nsva
r.
vente
r F. pinn
ataN. s
oehre
nsis
A. minu
tissim
a
C. ang
ustat
a
Nitzsc
hiaco
mplex
C. silic
ulaA. p
eterse
nii
D. kue
tzing
ii F. brev
istria
ta
LOI
Hg
Pb
PCA2
0 2 4 6 8 10 12 14 16
2020
4060
8020
2040
2025
5075
100
2040
1575
135
2040
60-2
02
A C
Skel
eton
Lak
e
ping
o
Blis
ter
Hill
Skel
eton
Cre
ek
EP2
B
Fi
gure
6.
ice-free period
chan
ge in
dia
tom
as
sem
blag
eSkeleton Lake
~1850 2003
small
large
several weeks
several months
ice-free period
chan
ge in
dia
tom
as
sem
blag
e
EP2
~1850
2003
small
large
several weeks
several months
Figure 7.
Table 1. Selected limnological characteristics for the three study lakes. Skeleton Lake EP2 EP3Latitude (N) 81˚49.798’ 81˚49.845’ 81˚49.884’Longitude (W) 71˚28.483’ 71˚28.352’ 71˚28.052’Date sampled 08/07/2003 08/07/2003 08/07/2003Depth (m) ~4 ~3 ~2Temperature (°C) 6 13.5 12pH 8.2 8.2 8.1Specific conductivity (μS/cm) 175 180 187
SiO2 (mg/L) 4.61 5.21 5.7Na (mg/L) 1.42 1.26 1.41Ca (mg/L) 38.8 41.7 44.3K (mg/L) 1.09 0.85 0.98Mg (mg/L) 6.53 5.59 6.04SO4 (mg/L) 41.7 30.3 33.5Cl (mg/L) 0.61 0.58 0.56Particulate organic carbon (POC; mg/L) 0.529 1.24 1.49Particulate nitrogen (PON; mg/L) 0.084 0.259 0.334Dissolved organic carbon (DOC; mg/L) 5.8 4.6 4.1Dissolved inorganic carbon (DIC; mg/L) 20.7 22.8 25.9NO3-NO2 (mg/L) 0.005 0.006 <0.005NH3 (mg/L) 0.014 0.032 0.01Total Kjeldahl nitrogen (TKN; mg/L) 0.314 0.296 0.298Total dissolved nitrogen (TdN; mg/L) 0.354 0.346 0.275Total phosphorus (Tpu; mg/L) 0.0095 0.0075 0.0056Total phosphorus filtered (TPf; mg/L) 0.0052 0.0036 0.0032Chlorophyll a (μg/L) 0.6 0.6 0.9Total N : Total P (molar) 61 112 159
Table 2. Diatom taxa (>3% relative abundance) from the top 2.5 cm intervals from the sediment cores of Skeleton Lake and EP2 and the surface sediment (~2 cm) diatoms from EP3.
Relative abundance
(%) Skeleton Lake Fragilaria brevistriata 11.26 Fragilaria construens var. venter 85.78 EP2 Denticula kuetzingii 7.55 Fragilaria brevistriata 83.88 Nitzschia complex 4.95 EP3 Achnanthes minutissima 7.24 Achnanthes petersenii 5.36 Brachysira zellensis 3.22 Cymbella angustata 5.36 Cymbella microcephala 4.83 Cymbopleura sp. 8.85 Denticula tenuis 4.29 Eunotia praerupta 4.29 Fragilaria brevistriata 19.03 Fragilaria pinnata 5.90 Navicula chiarae 4.29 Nitzschia complex 7.24
Tabl
e 3.
Com
mon
dia
tom
taxa
(>3%
rela
tive
abun
danc
e) fo
und
in th
e se
dim
ent c
ores
from
Ske
leto
n La
ke a
nd E
P2.
Num
bers
in
pare
nthe
ses a
fter t
he la
ke n
ame
repr
esen
t the
leng
ths o
f the
DC
A a
xis 1
gra
dien
ts, i
n st
anda
rd d
evia
tion
units
, N is
the
num
ber o
f oc
curr
ence
s, N
2 is
the
effe
ctiv
e nu
mbe
r of o
ccur
renc
es (H
ill, 1
973)
and
Max
% is
the
max
imum
rela
tive
abun
danc
e. S
ynon
yms f
or
taxa
with
revi
sed
nam
es a
re li
sted
in th
e fa
r rig
ht c
olum
n.
N
N
2 M
ax
%
Syno
nym
Sk
elet
on L
ake
(1.0
6 SD
)
Frag
ilari
a br
evis
tria
ta G
runo
w in
Van
Heu
rck
33
29.6
12
Ps
eudo
stau
rosi
ra b
revi
stri
ata
(Gru
now
in V
an H
eurc
k)
Will
iam
s & R
ound
Fr
agila
ria
cons
true
ns v
ar. v
ente
r (Eh
renb
erg)
Gru
now
in V
an
Heu
rck
33
33.0
97
St
auro
sira
con
stru
ens v
ar. v
ente
r (Eh
renb
erg)
Ham
ilton
Fr
agila
ria
pinn
ata
Ehre
nber
g 33
27
.1
18
Stau
rosi
rella
pin
nata
(Ehr
enbe
rg) W
illia
ms &
Rou
nd
Frag
ilari
a ps
eudo
cons
true
ns
26
18.8
10
Ps
eudo
stau
rosi
ra p
seud
ocon
stru
ens
EP2
(2.0
0 SD
)
Ac
hnan
thes
min
utis
sim
a K
ützi
ng
14
10.5
13
Ac
hnan
thid
ium
min
utis
sim
um (K
ützi
ng) C
zarn
ecki
Achn
anth
es p
eter
seni
i Hus
tedt
14
11
.0
5 Ro
ssith
idiu
m p
eter
seni
i (H
uste
dt) B
ukht
iyar
ova
&
Rou
nd
Cal
onei
s sili
cula
(Ehr
enbe
rg) C
leve
12
10
.8
4 C
alon
eis s
ilicu
la (E
hren
berg
) Cle
ve
Cym
bella
ang
usta
ta
13
10.9
5
Cym
bopl
eura
ang
usta
ta v
ar. s
pitz
berg
ensi
s Kra
mm
er
Den
ticul
a ku
etzi
ngii
Gru
now
16
13
.4
24
Den
ticul
a ku
etzi
ngii
Gru
now
Frag
ilari
a br
evis
tria
ta G
runo
w in
Van
Heu
rck
17
12.3
98
Ps
eudo
stau
rosi
ra b
revi
stri
ata
(Gru
now
in V
an H
eurc
k)
Will
iam
s & R
ound
Fr
agila
ria
cons
true
ns v
ar. v
ente
r (Eh
renb
erg)
Gru
now
in V
an
Heu
rck
10
8.8
10
Stau
rosi
ra c
onst
ruen
s var
. ven
ter (
Ehre
nber
g) H
amilt
on
Frag
ilari
a pi
nnat
a Eh
renb
erg
17
13.1
79
St
auro
sire
lla p
inna
ta (E
hren
berg
) Will
iam
s & R
ound
N
avic
ula
soeh
rens
is K
rass
ke
11
8.2
6 C
ham
aepi
nnul
aria
soeh
rens
is (K
rass
ke) L
ange
-Ber
talo
t N
itzsc
hia
com
plex
16
12
.7
32
Nitz
schi
a co
mpl
ex
CHAPTER 7
IMPACTS OF SEABIRD-DERIVED NUTRIENTS ON WATER QUALITY AND DIATOM SPECIES
ASSEMBLAGES FROM CAPE VERA, DEVON ISLAND, CANADIAN HIGH ARCTIC
BRONWYN E. KEATLEY, MARIANNE S.V. DOUGLAS, JULES M. BLAIS, MARK MALLORY, AND
JOHN P. SMOL
Abstract
Allochthonous nutrient subsidies play an important role in shaping biotic communities,
particularly in naturally oligotrophic ecosystems such as the Canadian Arctic. Seabirds have been
identified as one vector capable of transporting significant quantities of nutrients from marine to
terrestrial environments. However, the influence of seabird-derived nutrients on water quality
and ecosystem functioning of Arctic freshwater systems has been largely unexplored. Here we
sampled 24 ponds on Cape Vera, Devon Island (Arctic Canada), an area that is home to over 10
000 nesting pairs of northern fulmars, in order to explore the influence of seabirds on water
quality and diatom ecology. Our data reveal distinctive water chemistry changes (i.e. high pH,
nutrients) and diatom assemblages (i.e. characterized by extremely low diversity), related to
varying degrees of bird influence. Using δ15N as an ecological tracer, we show that a small but
significant portion of the variance in the diatom species data can be explained by seabird-derived
nutrients. Although the generation of robust quantitative models predicting δ15N from fossil
diatom data was limited by the overwhelming dominance of a few taxa, this study provides
evidence that seabird-derived nutrients play a critical role in influencing the water quality of
Arctic ponds. Interestingly, diatom assemblage composition does not respond in a simple fashion
to seabird-derived nutrients, in contrast to diatom-nutrient relationships recorded in many
temperate regions.
Introduction
Allochthonous nutrient subsidies (e.g. from fertilizer, sewage, fish, or birds) can play
significant roles in food web functioning across diverse environments (Polis et al. 1997; Polis et
al. 2004). Generalizations concerning the ways ecosystems respond to nutrient inputs, however,
are difficult to make, in part due to the differences in temporal and spatial scales, trophic
interactions within food webs, and differences in the types of nutrient fluxes (Polis et al. 1997;
Vanni et al. 2004; Ellis 2005). Previous research has shown, for example, that production of
terrestrial plants and the abundance of beetles and arthropods are greater on areas influenced by
seabird activities than those without (Polis & Hurd 1996; Sánchez-Piñero & Polis 2000).
Conversely, terrestrial production may be reduced in areas of high seabird activity (see Ellis
2005) and plant species richness can be suppressed in regions receiving large quantities of seabird
guano (Anderson & Polis 2004; Wait et al. 2005). Similarly divergent results have been found
for microbial communities (Tscherko et al. 2003) and abiotic aspects (Ellis 2005) of seabird-
influenced soils.
Seabirds that feed in marine environments, but nest on land, represent important links
between marine, terrestrial, and, by extension, freshwater environments. Indeed, on a global
basis, seabirds are estimated to consume a similar volume of fish to that harvested by commercial
fishing operations (Brooke 2004). Yet, while seabirds have been used as monitors of
environmental change in both marine (reviewed in Diamond & Devlin 2003) and terrestrial
(reviewed in Ellis 2005) ecosystems, few studies have assessed the impact of seabird-derived
nutrients on freshwater quality or biota (Izaguirre et al. 1998; Harding et al. 2004). In temperate
and subarctic regions, non-marine birds, including ducks, geese, and gulls, have been shown to
deposit significant amounts of nutrients into freshwater lakes (Manny et al. 1994; Kitchell et al.
1999; Payne & Moore 2006). In Arctic regions, only a few studies have documented the
transport of nutrients and pollutants into freshwater ponds and lakes via seabird guano (Evenset et
al. 2004; Blais et al. 2005; Evenset et al. 2007).
As many Arctic lakes and ponds are naturally oligotrophic (Vincent & Hobbie 2000),
nutrient subsidies from allochtononous sources can significantly alter their water quality and
biological communities, although ecological responses can be complex. For example, Bonilla et
al. (2005) found that nutrient additions stimulated phytoplanktonic, but not benthic, communities
in a High Arctic lake. Diatoms (unicellular algae of the class Bacillariophyceae) have been used
to track nutrient additions in a wide range of freshwater habitats (reviewed by Hall & Smol
1999), and previous work in Arctic regions has shown that diatoms have responded, albeit in a
manner different from responses found in temperate regions, to nutrient enrichment from sewage
(Douglas & Smol 2000; Michelutti et al. 2002a), the decay of whale carcasses (Douglas et al.
2004a), and salmon (Gregory-Eaves et al. 2003). Stable isotopes of N (δ15N) are commonly
employed to track marine-derived eutrophication, as upper trophic level consumers in marine
foodwebs are enriched in δ15N compared to their terrestrial counterparts (Finney et al. 2000;
Robinson 2001; Douglas et al. 2004a). Seabirds, however, represent a far more common nutrient
source to Arctic lakes and ponds than salmon or whales as the Canadian Arctic is home to over 10
million seabirds (Mallory & Fontaine 2004). Recently, Blais et al. (2005) demonstrated that
seabirds are important vectors of marine-derived nutrients and contaminants in a series of 11
ponds located near Cape Vera, Devon Island, Arctic Canada, but the significance of seabird-
derived nutrients on freshwater biota remains unknown.
Here, we build upon the Blais et al. (2005) study by extending the original 11 study sites
to encompass 24 ponds across a seabird-derived nutrient gradient near Cape Vera, Devon Island
in order to: 1) determine the influence of seabirds on water chemistry of freshwater Arctic ponds
by comparing limnological changes in ponds across this gradient of seabird influence (as well as
to other High Arctic ponds unaffected by seabirds); 2) examine the distributions and patterns of
community structure of diatoms across this gradient of seabird influence; 3) determine which
environment variables can best explain diatom distributions in this region. This study also
provides autecological diatom data for a companion paleolimnological study that extends the
results of the modern limnological survey to an historical context (Keatley et al. [8]).
We found that sites located within a range of seabird influence at Cape Vera exhibited
highly distinctive water chemistry, exemplified by elevated nutrients and related variables, as
well as by high trace metal concentrations. The diatom assemblages recorded in surface
sediments of the ponds are unlike those found elsewhere in the High Arctic and are characterized
by high dominance of very few taxa. Finally, we determined that δ15N (a proxy for seabird
influence) can explain a small but significant portion of the variation in diatom species
distributions, although, surprisingly, the most abundant taxon showed no direct relationship to
seabird-derived nutrients.
Site description
Cape Vera (76°15’N, 89°15’W; Figure 1) is located at the northwestern portion of Devon
Island, Nunavut, in the Canadian High Arctic. Cape Vera is characterized by cold winters, cool
summers, and little precipitation. Long-term meteorological data do not exist for Cape Vera, but
measurements from the nearest weather station at Grise Fiord, Ellesmere Island, indicate mean
January and July temperatures of -31.8°C and 3.4°C, respectively, and a mean annual
precipitation of 167 mm (Meteorological Survey of Canada 2007).
The ~250 m high dolostone and limestone cliffs (Mayr et al. 1998) at Cape Vera house a
large colony (~10 000 breeding pairs) of northern fulmars (Fulmarus glacialis), which are
medium-sized petrels. These birds feed near the top of the marine food chain on squid, capelin,
crustaceans (Hatch & Nettleship 1998), and are also known to scavenge polar bear kills in the
High Arctic; as such, their tissues are elevated in 15N, and have δ15N values averaging between
13-17‰ (reviewed in Mallory 2006). The northern fulmars arrive at Cape Vera to breed in early
May, during which time they travel as far as ~250 - 400 km away to feed. They incubate their
nests at various times throughout the summer season and depart in September for their over-
wintering grounds, thought to be in the North Atlantic (Mallory 2006). Beneath the nesting cliffs
at Cape Vera is a small foreland that contains several freshwater ponds, which, depending on
their proximity to the bird colony and local hydrographic features, receive varying amount of
seabird subsidies (including guano, regurgitated stomach oil, feathers, eggshells and carcasses).
Northern fulmars return to their breeding sites with extremely high fidelity, and, unlike many
other Arctic breeding seabirds, lay eggs at the same time each year, regardless of weather
conditions (Mallory 2006). In the Canadian Arctic, the northern fulmar colony at Cape Vera does
not appear to have changed in colony size since the 1970s (Gaston et al. 2006), although these
data are admittedly based on only a few surveys performed at irregular intervals.
We selected 24 ponds located across a “bird-affected” gradient near the colony at Cape
Vera (Figure 1). Thirteen of the twenty-four ponds (unofficial names: CV5-10, 12-15, 20, 31)
were in the immediate vicinity of the most densely populated section of the colony, while four
sites (CV1-4) were situated just to the north of this heavily-used section. Seven of our sites
currently receive no run-off from the bird-colony (Figure 1).
Methods
Sampling
We conducted our sampling during the first two weeks of July over a three year period
(2004, 2005, 2006); a strategy consistent with our other water quality surveys. Some sites were
sampled in multiple years (e.g. CV5, 6, 7, 8, 9, 10, 12, 13, 14, 15), but due to helicopter
availability, some sites were only sampled once during the three year period (e.g. CV1-4, 11, 22-
24). Water samples were collected from all ponds for water chemistry analyses, whereas surface
sediment samples were taken for diatom identification and enumeration. The protocols we used
were identical to those from our previous sampling years in other Arctic regions (Lim et al. 2001;
Michelutti et al. 2002b; Michelutti et al. 2002c; Lim & Douglas 2003; Antoniades et al. 2003a;
Antoniades et al. 2003b; Lim et al. 2005; Keatley et al. [2]; Keatley et al. [4]; Michelutti et al. in
press a), and sampling procedures followed those described in Environment Canada
(Environment Canada 1994). Specific conductivity (YSI model 33 conductivity meter), pH
(handheld Hanna pHEP meter), and temperature (handheld thermometer) were measured each
day at our base camp, while all other water chemistry analyses (major ions (Ca, Mg, Na, K, Cl,
SO4), nutrients (total phosphorus unfiltered (TPu), total phosphorus filtered (TPf), total dissolved
nitrogen (TdN), total Kjeldahl nitrogen (TKN, filtered), ammonia-nitrogen (NH3-N), nitrate-
nitrogen (NO3-N), nitrate-nitrite-nitrogen (NO3-NO2-N), particulate nitrogen (PON), trace metals
(Ag, Al, As, B, Ba, Be, Bi, Cd, Co, Cr, Cu, Fe, Ga, La, Li, Mn, Mo, Ni, Pb, Rb, Sb, Se, Sr, Tl, U,
V, Zn), dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), particulate carbon
(POC), chlorophyll a) were performed at the National Laboratory for Environmental Testing
(Environment Canada 1994). Details on sampling methodology can be found in Appendix 1.
Surface sediments were sampled by hand from the top ~1 cm from bottom of each pond,
placed into Whirlpak® bags, and kept dark and cold (~4°C) until analysed for δ15N and diatoms.
Diatom frustules were prepared for identification following standard techniques with a MARSX
microwave digester (Parr et al. 2004). Digested material was rinsed several times with deionized
water until a neutral pH was reached, and the resultant slurries were plated on microscope slides
using Naphrax®, a high refractive mounting medium. Surface sediments were also analysed for
% C and N, and stable isotopes of δ15N at the G.G. Hatch Stable Isotope Laboratory, Ottawa,
Canada. Elemental analyses were performed for C and N using a Vario EL III (Elementar,
Germany), while stable isotopes of N were measured using a continuous-flow DeltaPlus
Advantage isotope ratio mass spectrometer (ThermoFinnigan, Germany) coupled with a ConFlo
III. Isotope data was normalized using previously calibrated internal standards, and analytical
precision is +/-0.2 ‰.
A minimum of 300, and often > 500, diatom valves were enumerated and identified from
each sediment sample using primarily Antoniades et al. (in press), Krammer (2002) and Krammer
and Lange-Bertalot (1991).
Statistical analyses
A Pearson’s correlation matrix with Bonferroni-adjusted probabilities was generated
using Systat 9.0 (SPSS 1999) to identify pairs of significantly correlated variables because highly
collinear variables can unduly influence multivariate ordinations. All multivariate statistical
analyses were performed using CANOCO version 4.5 (ter Braak & Šmilauer 2002). Measured
environmental variables were analysed using a series of Principal Components Analyses (PCA) to
assess main directions of variation in the dataset, and to validate our initial visual ranking across
the gradient of seabird influence. Since we had multi-year environmental data from some sites,
we used this to semi-quantitatively assess whether inter-year variation within a site was greater
than inter-pond variation within a sampling period (one sample per pond per year). For all
subsequent analyses, however, we used a single sample year. A PCA was used to determine the
main directions of variation in the measured environmental variables amongst the sites. To better
explore the impact of seabird activity on the most ecologically-relevant variables (e.g.,
phosphorus, nitrogen, carbon, pH, specific conductivity, chlorophyll a), we compared the
differences between the Cape Vera sites to other Devon Island sites far from seabird influence
(Lim & Douglas 2003; Lim 2004). In an attempt to identify how unique the Cape Vera ponds
were with respect to other Arctic limnological surveys, we performed a PCA of our measured
environmental variables (pH, specific conductivity, TPu, TPf, TNu, DOC) with those from
previous studies throughout the Canadian Arctic (Lim et al. 2001; Michelutti et al. 2002b;
Michelutti et al. 2002c; Lim & Douglas 2003; Antoniades et al. 2003a; Antoniades et al. 2003b;
Lim et al. 2005; Keatley et al. [2]; Keatley et al. [4]; Michelutti et al. in press a).
Diatom species counts were converted to relative abundances and these data were
analysed using a detrended correspondence analysis (DCA) to assess the gradient length of the
species data, and to determine whether linear or unimodal statistical techniques were more
appropriate. As the DCA indicated a short gradient (<2 standard deviation units), linear
ordination techniques (PCA and RDA) were further used to assess patterns in the diatom species
distributions (Lepš & Šmilauer 2003). In order to compare patterns of diatom diversity between
sites, Hill’s N2 (a measure of species evenness and essentially the number of very abundant
species) was calculated (Hill 1973).
A series of constrained ordinations (RDAs) were used to determine whether measured
environmental variables could explain significant portions of the variance in diatom species
distributions. First, the correlations between environmental variables were assessed using a
Pearson correlation with Bonferroni-adjusted probabilities. In this manner, the environmental
dataset was truncated to reduce the influence of highly collinear environmental variables. Next,
the species data were constrained to the measured environmental variables, and these were
assessed using forward selection techniques to determine which environmental variables could
explain significant portions of the species data using a series of RDAs. We constructed predictive
models to explore the feasibility of quantitatively inferring the most important measured
environmental variable using regression-calibration techniques in C2 version 1.3 (Juggins 2003).
Results and Discussion
Water chemistry
A summary of the measured environmental variables for each Cape Vera site is given in
Table 1. For the sake of brevity, we limit our discussion of water chemistry variables to pH,
specific conductivity, nutrients and related variables, and some metals, due to their ecological
interest. Full details of all water chemistry data can be found in Appendix 5. In the PCA biplot of
measured environmental variables, the sites we a priori identified as being most influenced by
seabirds plotted in the right-hand side of the diagram and were characterized by elevated pH,
conductivity, production (Chla), nutrients and related variables (N, P), and δ15N, our isotope
proxy of bird influence (Figure 2). In contrast, our control sites plotted on the left-hand site of the
ordination plot (Figure 2). The distinctive water chemistry of the most heavily-affected ponds
becomes even more apparent when compared to sites from the nearby Haughton Crater region
(Figure 3), which shares similar climate and geological features, yet is not influenced by seabird
activities (Lim & Douglas 2003).
While pH values across the High Arctic are generally reflective of alkaline conditions
(Hamilton et al. 2001), the seabird-influenced ponds at Cape Vera had the highest pH yet
recorded in the Canadian Arctic (10.75, Table 1). For comparison, the mean pH of 36 lakes and
ponds elsewhere on Devon Island was 8.3, despite sharing similar climatic and geological
features (Lim 2004). The only comparably high pH values recorded in the Canadian Arctic come
from small, hyper-eutrophic sewage ponds near Resolute Bay (Douglas & Smol 2000). The
elevated pH values at Cape Vera likely reflect the enhanced aquatic production stimulated by the
nutrients from seabird wastes. In temperate regions, for example, shallow rock pools influenced
by gull guano are known to exceed pH 10 on a daily basis, linked to high photosynthetic activity
(Loder et al. 1996).
Specific conductivity, although relatively high compared to elsewhere on Devon Island,
was not as high as might have been expected given the proximity of the ocean to many of these
ponds (Cape Vera range: 85 – 312 µS/cm; mean = 160 µS/cm; Table 1). The relatively low
conductivity, coupled with the large concentrations of DIC and Ca as the dominant ions, suggest
that, despite their proximity to the sea, the ponds at Cape Vera are not receiving large quantities
of marine inputs. Coastal ponds elsewhere in the Canadian Arctic often have recorded specific
conductivity values > 1000 µS/cm and are driven by high concentrations of Cl and Na (e.g.
Antoniades et al. 2003a).
Interestingly, trace metal concentrations in the ponds at Cape Vera were almost all above
method detection limits, which is unusual because metal concentrations rarely exceed detection
limits in the High Arctic. For example, of the nine High Arctic water chemistry surveys
completed to date, only sites located on northern Ellesmere Island had metal concentrations that
were as high as, or higher than, those from Cape Vera (Keatley et al. [4]). On northern Ellesmere
Island, the elevated metal concentrations may be related to geological sources. That the ponds
closest to the bird colony at Cape Vera had the highest metal concentrations is consistent with an
ornithogenic source of metals (Table 1). Indeed, previous studies have found that certain metals
are known to bioaccumulate in animals (e.g. Rb, Campbell et al. 2005; Cd, Cu, Hg, Zn, Liu et al.
2006).
The nutrient concentrations of the Cape Vera ponds were highest in the ponds closest to
the bird colony (Table 1, Figure 2). The high production in the bird-influenced ponds (as
estimated by chlorophyll a concentrations, Chlamean: 9.0 µg/L, Table 1) was accompanied by,
although not significantly correlated to, high phosphorus and nitrogen (e.g. TPu: range 4 – 207
µg/L, mean: 36.1 µg/L; TPf: 2.2-72.0 µg/L, mean: 14.3 µg/L; TdN: 0.072 – 1.53 mg/L, mean:
0.375 mg/L). In the context of over 400 lakes and ponds our labs have sampled from the
Canadian Arctic Archipelago, the Cape Vera sites include several of the most nutrient rich Arctic
freshwater ecosystems (Figure 4). A few other Arctic sites have equally high, or higher, TPu, but
these have been previously attributed to re-suspended particulate matter rather than to available
phosphorus, and thus not considered to be truly eutrophic systems (Antoniades et al. 2003b;
Michelutti et al. 2007a; Keatley et al. [2]).
In general, sites in the High Arctic that are characterized by relatively high production,
lush vegetation, and high nutrients also have high DOC (Lim et al. 2001; Antoniades et al. 2003a;
Lim et al. 2005; Keatley et al. [2]; Keatley et al. [4]). Interestingly, DOC concentrations at Cape
Vera were not nearly as high as we might have expected (range: 0.6 – 12.6 mg/L; mean: 3.0
mg/L), given the lush vegetation that surrounds many of these ponds. This can be most easily
seen by the composite PCA, in which the sites from the Arctic oasis near Lake Hazen are clearly
distinguished by their high DOC (Figure 4). This interesting finding might be due to a
decoupling of the dissolved organic matter cycle, whereby the high terrestrial and aquatic
production at Cape Vera exists, in spite of very harsh climatic conditions, as a function of the
external nutrient subsidies deposited by birds. In contrast, localized sites of relatively high
production elsewhere in the Arctic Archipelago are generally found in concert with warm
microclimatic conditions that are likely more conducive to decomposition of terrestrial organic
matter and thus higher DOC. The relatively low DOC in the Cape Vera sites corroborates our
evidence from δ15N that the nutrients in the ponds are generated from the seabirds rather than
terrestrial plant matter.
Diatom assemblages
A total of 95 taxa were identified from the 24 study sites at Cape Vera, although 44% of
these were considered very rare (<1% relative abundance). The assemblages were dominated by
small, benthic diatoms (Figure 5, Table 2), a typical finding in shallow Arctic ponds (Douglas et
al. 2004b). The dominant taxa are all common components of diatom assemblages in ponds in
the Canadian High Arctic (Antoniades et al. in press). However, their distributional patterns are
quite distinctive at Cape Vera. For example, many sites were dominated by small Nitzschia spp.
(often >90% relative abundance) and Cymbella sensu lato spp. (Figure 5). Elsewhere in the High
Arctic, the Cymbella spp. found at Cape Vera (e.g. C. cleve-eulerae, C. botellus, Encyonema.
minutum) are generally not found at >10% relative abundances and are typically associated with
oligotrophic, low electrolyte waters (Antoniades et al. in press). On the other hand, small
Nitzschia spp. (e.g. N. perminuta, N. frustulum (recorded by Antoniades et al. (in press) as N.
alpina)) are perhaps the most ubiquitous and abundant freshwater diatoms (max abundance 68%
and 26%, respectively) found in the Canadian High Arctic (Antoniades et al. in press), although
morphological similarities often make it difficult to distinguish these two taxa and hence
appreciate their true abundances and environmental preferences. For example, Michelutti et al.
(2003) found very high relative abundances (often > 90%) of the Nitzschia complex (N.
frustulum, N. perminuta, N. libertruthii, N. inconspicua) in the periphytic samples of ponds
receiving sewage near Resolute Bay, Nunavut. The extremely high relative abundance of these
small Nitzschia taxa at Cape Vera (often > 90%) suggests that N. frustulum and N. perminuta do
exceedingly well under high nutrient, high pH conditions. Interestingly, none of the most
common species found at Cape Vera were common components of the diatom flora of lakes and
ponds elsewhere on Devon Island (Lim 2004), further highlighting the uniqueness of the Cape
Vera sites.
In the PCA biplot, sites with more species in common plot closer together (e.g. CV9, 9a,
10), while those with very different species assemblages plot further away from each other
(Figure 6). Species associated with the higher δ15N sites included Encyonema minutim, E. fogedii,
and E. silesiacum, while N. frustulum plotted amongst the higher DOC sites (Figure 6). Species
characteristic of lower δ15N sites included small Achnanthes sensu lato spp. (Achnanthidium
minutissimum, Psammothidium marginulatum, Rossithidium petersenii, Achnanthes kriegeri,
Eucocconeis laevis), Denticula kuetzingii, Diatoma tenuis, Cymbella botellus, C. cleve-eulerae,
C. designata, Encyonopsis microcephala, and Cymbopleura angustata (Figure 6). Not
surprisingly, the control sites at the Cape Hawes (CV22-24) sites plotted closely together based
on the species data (Figure 6).
The diatom assemblage diversity is extremely low in most of the bird-influenced sites, as
indicated by the Hill’s N2 number (Table 2; N2 mean: 7.28; N2 range 1.0 – 15.0; Hill 1973).
High Arctic ponds are generally species-poor (mean Hill’s N2 values range from 6.72 (Cape
Herschel, Ellesmere Island; Douglas & Smol 1993) to 13.09 (Alert, Ellesmere Island; Antoniades
et al. 2005), but the overwhelming dominance by only a few taxa at Cape Vera is unusual.
Previous Arctic limnological studies have linked increased diversity of diatom assemblages with
the increased habitat heterogeneity that occurs as a result of increased moss and macrophyte
growth, which in turn has been attributed to recent warming (e.g. Douglas & Smol 1999). Thus,
one might have expected to find an increase in diatom diversity with high nutrient concentrations
and the presence of relatively diverse substrates (including rocks, sediments, mosses, and
filamentous algae) at Cape Vera. However, only a weak negative relationship (R2 = 0.081, p =
0.177) was found between diatom diversity and seabird-derived nutrients, while no relationship
was found between diatom diversity and primary production (as estimated by chlorophyll a) in
these sites. Previous ecological studies have suggested that plant and algal diversity (or richness)
follows a hump-shaped curve along a productivity gradient (Dodson et al. 2000; Mittelbach et al.
2001; Groner & Novoplansky 2003), but we find no evidence of such a relationship with respect
to diatom species assemblages at Cape Vera, perhaps reflecting dominance by other algal groups.
Species-environment relationships
Three measured environmental variables explained significant portions of the variance in
the diatom data based on the RDA: δ15N, 13.5%, specific conductivity 10.5%, and dissolved
organic carbon (DOC) 10.2% (Figure 7). Only the first axis of the RDA was significant,
however, and the low eigenvalue of this axis suggests that other factors not captured by the RDA
are also influencing diatom species distributions (Axis 1 λ1= 0.168, p = 0.008; Figure 7).
Therefore, in order to explore which environmental factor was most important for explaining the
variation in diatom species distributions in the absence of δ15N, we performed an RDA without
δ15N. In this more restricted analysis, filtered phosphorus (TPf) explained a significant amount of
the variation in the diatom species data (12.4%). Together, these two findings suggest that
nutrients, which are associated with seabirds, play a role in shaping diatom communities at Cape
Vera, but the relationship is not simple and not directly related to nutrients.
To explore which species are most closely associated with the δ15N gradient, we
performed an additional RDA constrained only to δ15N (Table 4). The RDA axis 1 species scores
(a measure of how well the species can be explained by the sole environmental variable) indicate
that N. frustulum, although by far the most dominant taxon in this dataset, has little relationship to
δ15N, the measured environmental variable that explained the most variance in the diatom species
data (Table 4). The high relative abundance of this taxon in the control sites at Cape Hawes and
CV11, which all exhibit low δ15N, underscores the lack of a relationship between δ15N and N.
frustulum. Instead, the relationship between diatom species distributions and δ15N must be
influenced by species associated with low δ15N sites, such as Cymbella botellus, C. designata,
Cymbopleura angustata, and Psammothidium marginulatum (RDA axis 1 species scores > +0.5),
rather than those characteristic of the high δ15N sites (RDA axis 1 score < - 0.5: Encyonema
minutum). Furthermore, although δ15N explains the largest percentage of diatom species
distributions in our dataset, the high dominance of relatively few species at Cape Vera precluded
the generation of a robust transfer function.
That the relationships between the most dominant diatom taxa at Cape Vera are not easily
explainable by patterns of nutrients across such an extreme nutrient gradient further supports
studies suggesting that Arctic diatoms do not respond to nutrient enrichment in a simple fashion,
unlike diatoms from temperate regions (Hall & Smol 1999). While paleolimnological studies of
Arctic lakes receiving large quantities of human sewage have shown distinct diatom responses,
these changes have been very different from those recorded in temperate regions and generally
are indicative of species associated with periphytic habitats rather than high nutrient
concentrations (Douglas & Smol 2000; Michelutti et al. 2002a; Michelutti et al. in press b).
These subtle, and sometimes delayed, diatom responses to eutrophication have been attributed to
the overriding influence of climate (Douglas & Smol 2000; Michelutti et al. in press b). In the
ponds closest to the northern fulmar colony, harsh climatic conditions have not prevented the
development of extremely high standing stocks of mosses, large masses of filamentous algae
(qualitatively identified as Cosmarium sp., Zygnema spp., Ulothrix sp., and Tetraspora sp.) and
cyanobacteria (Nostoc commune and Lyngbya sp.). It is possible that these algae have
outcompeted diatoms in the nutrient-rich ponds at Cape Vera, and therefore precluded any
marked diatom species changes associated with nutrient enrichment. Evidence for differential
nutrient requirements between the phytoplankton and benthos in Arctic lakes is provided by
Bonilla et al. (2005), who found that phytoplankton (including Chrysophyceae and cyanobacteria)
respond much more strongly to nutrient enrichment than do benthic communities, which are
generally not nutrient limited.
Conclusions
Ponds affected by seabird-derived nutrients at Cape Vera are characterized by distinctive
water chemistry (i.e. high pH and nutrients) as compared to ponds less affected by seabird-
nutrients, and compared to lakes and ponds elsewhere in the Canadian Arctic Archipelago.
Surface sediment diatoms were overwhelmingly dominated by very few taxa in almost all ponds,
regardless of seabird influence; nevertheless, a significant portion of the variance in the diatom
species data can be explained by δ15N. Although the generation of robust quantitative models
predicting δ15N from fossil diatom data was stymied by the dominance of a taxon indifferent to
seabird derived nutrients, this study provides evidence that seabird-derived nutrients play a
critical role in shaping the water quality and also, in a more complex manner, the biological
communities of these ponds.
Acknowledgements
We would like to thank NSERC, PCSP, NSTP, CWS for funding and logistical support, and Paul
Hamilton for aid in identifying the non-diatom algae. This fieldwork was completed with the
generous assistance of I. Gregory-Eaves, A. Fontaine, N. North, M. Wayland, N. Michelutti, L.
Kimpe, K. Foster, H. Liu, J. Akearuk, and M. Falconer . This manuscript was greatly improved
by comments from N. Michelutti. This is PCSP/ÉPCP contribution # (TBD).
References
Anderson, W. B., & G. A. Polis, 2004. Allochthonous Nutrient and Food Inputs: Consequences for Temporal Stability. In G. A. Polis, M. E. Power & G. R. Huxel (eds), Food Webs at the Landscape Level. University of Chicago Press, Chicago: 82-95.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2003a. Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick Island, NWT). Archiv für Hydrobiologie 158: 485-516.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2003b. The physical and chemical limnology of 24 ponds and one lake from Isachsen, Ellef Ringnes Island, Canadian High Arctic. International Review of Hydrobiology 88: 519-538.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2005. Benthic diatom autecology and inference model development from the Canadian High Arctic Archipelago. Journal of Phycology 41: 30-45.
Antoniades, D., P. B. Hamilton, M. S. V. Douglas & J. P. Smol, in press. Freshwater diatoms from the Canadian High Arctic. Iconographica Diatomologica.
Blais, J. M., L. E. Kimpe, D. McMahon, B. E. Keatley, M. L. Mallory, M. S. V. Douglas, & J. P. Smol, 2005. Arctic seabirds transport marine-derived contaminants. Science 309: 445.
Bonilla, S., V. Villeneuve, & W. F. Vincent, 2005. Benthic and planktonic algal communities in a High Arctic Lake: Pigment structure and contrasting responses to nutrient enrichment. Journal of Phycology 41: 1120-1130.
Brooke, M. D., 2004. The food consumption of the world's seabirds. Proceedings of the Royal Society of London Series B-Biological Sciences 271: S246-S248.
Campbell, L. M., A. T. Fisk, X. W. Wang, G. Kock, & D. C. G. Muir, 2005. Evidence for biomagnification of rubidium in freshwater and marine food webs. Canadian Journal of Fisheries and Aquatic Sciences 62: 1161-1167.
Diamond, A. W., & C. M. Devlin, 2003. Seabirds as indicators of changes in marine ecosystems: Ecological monitoring on Machias Seal Island. Environmental Monitoring and Assessment 88: 153-175.
Dodson, S. I., S. E. Arnott, & K. L. Cottingham, 2000. The relationship in lake communities between primary productivity and species richness. Ecology 81: 2662-2679.
Douglas, M. S. V., & J. P. Smol, 1993. Fresh-Water Diatoms from High Arctic Ponds (Cape Herschel, Ellesmere-Island, NWT). Nova Hedwigia 57: 511-552.
Douglas, M. S. V.& J. P. Smol, 1999. Freshwater diatoms as indicators of environmental change in the High Arctic. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge: 227-244.
Douglas, M. S. V., & J. P. Smol, 2000. Eutrophication and recovery in the High Arctic: Meretta lake (Cornwallis Island, Nunavut, Canada) revisited. Hydrobiologia 431: 193-204.
Douglas, M. S. V., J. P. Smol, J. M. Savelle, & J. M. Blais, 2004a. Prehistoric Inuit whalers affected arctic freshwater ecosystems. Proceedings of the National Academy of Sciences of the United States of America 101: 1613-1617.
Douglas, M. S. V., P. B. Hamilton, R. Pienitz, & J. P. Smol, 2004b. Algal indicators of environmental change in arctic and antarctic lakes and ponds. In R. Pienitz, M. S. V. Douglas & J. P. Smol (eds), Long-term environmental change in Arctic and Antarctic lakes. Springer, Dordrecht: 117-158.
Ellis, J. C., 2005. Marine birds on land: a review of plant biomass, species richness, and community composition in seabird colonies. Plant Ecology 181: 227-241.
Environment Canada, 1994. Manual of Analytical Methods – National Laboratory for Environmental Testing, Canadian Centre for Inland Waters. Burlington, Canada.
Evenset, A., J. Carroll, G. N. Christensen, R. Kallenborn, D. Gregor, & G. W. Gabrielsen, 2007. Seabird guano is an efficient conveyer of persistent organic pollutants (POPs) to Arctic lake ecosystems. Environmental Science & Technology 41: 1173-1179.
Evenset, A., G. N. Christensen, T. Skotvold, E. Fjeld, M. Schlabach, E. Wartena, & D. Gregor, 2004. A comparison of organic contaminants in two high Arctic lake ecosystems, Bjørnøya (Bear Island), Norway. The Science of the Total Environment 318: 125-141.
Finney, B. P., I. Gregory-Eaves, J. Sweetman, M. S. V. Douglas, & J. P. Smol, 2000. Impacts of climatic change and fishing on Pacific salmon abundance over the past 300 years. Science 290: 795-799.
Gaston, A. J., M. L. Mallory, H. G. Gilchrist, & K. O'Donovan, 2006. Status, trends and attendance patterns of the northern fulmar Fulmarus glacialis in Nunavut, Canada. Arctic 59: 165-178.
Gregory-Eaves, I., J. P. Smol, M. S. V. Douglas, & B. P. Finney, 2003. Diatoms and sockeye salmon (Oncorhynchus nerka) population dynamics: Reconstructions of salmon-derived nutrients over the past 2,200 years in two lakes from Kodiak Island, Alaska. Journal of Paleolimnology 30: 35-53.
Groner, E., & A. Novoplansky, 2003. Reconsidering diversity-productivity relationships: directness of productivity estimates matters. Ecology Letters 6: 695-699.
Hall, R. I., & J. P. Smol, 1999. Diatoms as indicators of lake eutrophication. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, New York: 128-168.
Hamilton, P. B., K. Gajewski, D. E. Atkinson, & D. R. S. Lean, 2001. Physical and chemical limnology of 204 lakes from the Canadian Arctic Archipelago. Hydrobiologia 457: 133-148.
Harding, J. S., D. J. Hawke, R. N. Holdaway, & M. J. Winterbourn, 2004. Incorporation of marine-derived nutrients from petrel breeding colonies into stream food webs. Freshwater Biology 49: 576-586.
Hatch, S. A.& D. N. Nettleship, 1998. Northern fulmar (Fulmarus glacialis). In A. Poole & F. Gill (eds), The Birds of North America, No. 361. The Birds of North America, Inc., Philadelphia, PA: 1-32.
Hill, M. O., 1973. Diversity and evenness: a unifying notation and its consequences. Ecology 54: 427-432.
Izaguirre, I., A. Vinocur, G. Mataloni, & M. Pose, 1998. Phytoplankton communities in relation to trophic status in lakes from Hope Bay (Antarctic Peninsula). Hydrobiologia 370: 73-87.
Juggins, S., 2003. C2: software for ecological and palaeocological data analysis and visualisation, version 1.3. Department of Geography, University of Newcastle.
Keatley, B. E., M. S. V. Douglas, & J. P. Smol, [2]. Physical and chemical limnological characteristics of 46 lakes and ponds across environmental gradients on Melville Island, High Arctic Canada. Fundamental and Applied Limnology 168: 355-376.
Keatley, B. E., M. S. V. Douglas, & J. P. Smol, [4]. Limnological characteristics of a high Arctic oasis and comparisons across northern Ellesmere Island. Arctic.
Kitchell, J. F., D. E. Schindler, B. R. Herwig, D. M. Post, M. H. Olson, & M. Oldham, 1999. Nutrient cycling at the landscape scale: The role of diel foraging migrations by geese at the Bosque del Apache National Wildlife Refuge, New Mexico. Limnology and Oceanography 44: 828-836.
Krammer, K., 2002. Diatoms of Europe: Cymbella. A.R.G. Gantner Verlag, Ruggell.
Krammer, K.& H. Lange-Bertalot, 1991. Bacillariophyceae. In H. Ettl, J. Gerloff, H. Heynig & D. Mollenhauer (eds). Gustav Fischer Verlag, Stuttgart.
Lepš, J.& P. Šmilauer, 2003. Multivariate analysis of ecological data using CANOCO. Cambridge University Press, London.
Lim, D.S.S., 2004. Limnology and diatom paleoecology of diatoms on Banks Island, N.W.T., and Devon Island, Nunavut, Canadian Arctic. University of Toronto. PhD Thesis.
Lim, D. S. S., M. S. V. Douglas, J. P. Smol, & D. R. S. Lean, 2001. Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. International Review of Hydrobiology 86: 1-22.
Lim, D. S. S., & M. S. V. Douglas, 2003. Limnological characteristics of 22 lakes and ponds in the Haughton Crater region of Devon Island, Nunavut, Canadian High Arctic. Arctic Antarctic and Alpine Research 35: 509-519.
Lim, D. S. S., M. S. V. Douglas, & J. P. Smol, 2005. Limnology of 46 lakes and ponds on Banks Island, NWT, Canadian Arctic Archipelago. Hydrobiologia 545: 11-32.
Liu, X. D., S. P. Zhao, L. G. Sun, X. B. Yin, Z. Q. Xie, L. Honghao, & Y. H. Wang, 2006. P and trace metal contents in biomaterials, soils, sediments and plants in colony of red-footed booby (Sula sula) in the Dongdao Island of South China Sea. Chemosphere 65: 707-715.
Loder, I. T. C., B. Ganning, & J. A. Love, 1996. Ammonia nitrogen dynamics in coastal rockpools affected by gull guano. Journal of Experimental Marine Biology and Ecology 196: 113-129.
Mallory, M. L., 2006. The Northern Fulmar (Fulmarus glacialis) in Arctic Canada: ecology, threats, and what it tells us about marine environmental conditions. Environmental Reviews 14: 187-216.
Mallory, M. L., & A. J. Fontaine, 2004. Key marine habitat sites for migratory birds in Nunavut and the Northwest Territories. Canadian Wildlife Service, Ottawa. Occasional Paper 109.
Manny, B. A., W. C. Johnson, & R. G. Wetzel, 1994. Nutrient Additions by Waterfowl to Lakes and Reservoirs - Predicting Their Effects on Productivity and Water-Quality. Hydrobiologia 280: 121-132.
Mayr, U., T. de Freitas, G.H. Eisbacher, & J.Wm. Kerr, 1998. Geology, Prince Alfred Bay-Cardigan Strait and parts of adjacent sheets, Northwest Territories. Geological Survey of Canada, Ottawa. Map 1924a; Scale 1:250 000.
Meteorological Service of Canada, 2007: Monthly Climate Data. Environment Canada http://www.climate.weatheroffice.ec.gc.ca/climateData/monthlydata_e.html, Accessed February 2007
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2002a. Tracking recent recovery from eutrophication in a high arctic lake (Meretta Lake, Cornwallis Island, Nunavut, Canada) using fossil diatom assemblages. Journal of Paleolimnology 28: 377-381.
Michelutti, N., M. S. V. Douglas, D. C. G. Muir, X. Wang, & J. P. Smol, 2002b. Limnological characteristics of 38 lakes and ponds on Axel Heiberg Island, High Arctic Canada. International Review of Hydrobiology 87: 385-399.
Michelutti, N., M. S. V. Douglas, D. R. S. Lean, & J. P. Smol, 2002c. Physical and chemical limnology of 34 ultra-oligotrophic lakes and ponds near Wynniatt Bay, Victoria Island, Arctic Canada. Hydrobiologia 482: 1-13.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2003. Tracking recovery in a eutrophied high Arctic lake (Meretta Lake, Cornwallis Island, Canadian Arctic) using periphytic diatoms. Verh. int. Ver. Limnol. 28: 1533-1537.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, in press a. Evaluating diatom community composition in the absence of marked limnological gradients in the high Arctic: A surface sediment calibration set from Cornwallis Island (Nunavut, Canada). Polar Biology.
Michelutti, N., M. H. Hermanson, J. P. Smol, P. J. Dillon, & M. S. V. Douglas, in press b. Delayed response of diatom assemblage changes to sewage inputs in an Arctic lake. Aquatic Sciences.
Mittelbach, G. G., C. F. Steiner, S. M. Scheiner, K. L. Gross, H. L. Reynolds, R. B. Waide, M. R. Willig, S. I. Dodson, & L. Gough, 2001. What is the observed relationship between species richness and productivity? Ecology 82: 2381-2396.
Parr, J. F., K. H. Taffs, & C. M. Lane, 2004. A microwave digestion technique for the extraction of fossil diatoms from coastal lake and swamp sediments. Journal of Paleolimnology 31: 383-390.
Payne, L. X., & J. W. Moore, 2006. Mobile scavengers create hotspots of freshwater productivity. Oikos 115: 69-80.
Polis, G. A., W. B. Anderson, & R. D. Holt, 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Reviews of Ecological Systems 28: 289-316.
Polis, G. A., & S. D. Hurd, 1996. Linking marine and terrestrial food webs: allochthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. The American Naturalist 147: 396-423.
Polis, G. A., M. E. Power & G. R. Huxel, 2004. Food Webs at the Landscape Level. University of Chicago Press, Chicago.
Robinson, D., 2001. δ15N as an integrator of the nitrogen cycle. Trends in Ecology and Evolution 16: 153-162.
Sánchez-Piñero, F., & G. A. Polis, 2000. Bottom-up dynamics of allochthonous input: direct and indirect effects of seabirds on islands. Ecology 81: 3117-3132.
SPSS, 1999. SYSTAT 9. SPSS Inc., Chicago, IL, USA.
ter Braak, C. J. F.& P. Šmilauer, 2002. CANOCO reference manual and CANOdraw for Windows user guide: Software for Canonical Community Ordination (Version 4.5).
Tscherko, D., M. Bolter, L. Beyer, J. Chen, J. Elster, E. Kandeler, D. Kuhn, & H. P. Blume, 2003. Biomass and enzyme activity of two soil transects at King George Island, maritime Antarctica. Arctic Antarctic and Alpine Research 35: 34-47.
Vanni, M. J., D. L. DeAngelis, D. E. Schindler & G. R. Huxel, 2004. Overview: Cross-Habitat Flux of Nutrients and Detritus. In G. A. Polis, M. E. Power & G. R. Huxel (eds), Food Webs at the Landscape Level. University of Chicago Press, Chicago: 3-11.
Vincent, W. F.& J. E. Hobbie, 2000. Ecology of Arctic lakes and rivers. In M. Nuttall & T. V. Callaghan (eds), The Arctic: Environment, People, Policy. Harwood Academic Publishers, Amsterdam: 197-232.
Wait, D. A., D. P. Aubrey, & W. B. Anderson, 2005. Seabird guano influences on desert islands: soil chemistry and herbaceous species richness and productivity. Journal of Arid Environments 60: 681-695.
Figure captions.
Figure 1. Map of study location at Cape Vera, Devon Island. Insets showing location of: a) Devon Island within Canada; b) Cape Vera on Devon Island; and c) sites located throughout Cape Vera. The isoclines are not drawn to scale. The gradient indicates the approximate concentration of northern fulmars along the cliffs at Cape Vera, with the most birds occurring within the southern third of the colony. Figure 2. Principal Components Analysis (PCA) biplot of measured environmental variables and Cape Vera sites. The “control sites” (CV11, 16, 17, 18, 22, 23, 24) are denoted with open circles while the rest of the ponds are denoted with solid circles. Figure 3. Box plots comparing selected measured environmental variables (TPu, TPf, TN, DOC, pH, chla) between Cape Vera ponds (n = 24, this study) and ponds located in the nearby Haughton Crater, Devon Island (n = 22, Lim & Douglas 2003). Figure 4. Composite PCA biplot of selected measured environmental variables common to our labs’ previous limnological surveys in the Canadian Arctic Archipelago. References for each study are as follows: Melville Island (Keatley et al. [2]), Mould Bay, Prince Patrick Island (Antoniades et al. 2003a), Banks Island (Lim et al. 2005), Victoria Island (Michelutti et al. 2002a), Isachsen, Ellef Ringnes Island (Antoniades et al. 2003b), Bathurst Island (Lim et al. 2001), Cornwallis Island (Michelutti et al. in press a), Devon Island (Lim & Douglas 2003), Axel Heiberg Island (Michelutti et al. 2002b), northern Ellesmere Island (Keatley et al. [4]), Alert, Ellesmere Island (Antoniades et al. 2003). Figure 5. Histogram of dominant diatom species (>1% relative abundance in at least 5 sites) found in ponds from Cape Vera. Both the sites and the species are ordered according to their DCA axis 1 scores. Figure 6. Principal components analysis (PCA) biplot of species and sites from the surface sediments of Cape Vera, Devon Island. Figure 7. Redundancy analysis (RDA) biplot constrained to the three measured environmental variables that explained significant portions of the diatom species variance (δ15N, specific conductivity, and dissolved organic carbon (DOC)). Biplot 7a) presents the diatom species, and 7b) the sites in relation to the environmental variables.
b c
a
Figure 1.
Figure 2.
0
10
20
30
40
50
60
70
0.00
0.05
0.10
0.15
0.20
0.25
Cape Vera (n = 24)
Devon Crater (n = 22)
DO
C (m
g/L)
0
2
4
6
8
10
12
14
Cape Vera (n = 24)
Devon Crater (n = 22)
TNu
(mg/
L)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
TPu
(mg/
L)
Chl
a(µ
g/L)
0
10
20
30
40
50
60
70
0.00
0.05
0.10
0.15
0.20
0.25
Cape Vera (n = 24)
Devon Crater (n = 22)
DO
C (m
g/L)
0
2
4
6
8
10
12
14
Cape Vera (n = 24)
Devon Crater (n = 22)
TNu
(mg/
L)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
TPu
(mg/
L)
Chl
a(µ
g/L)
Figure 3.
-0.4 1.0
-0.6
1.0
cond
pH
TPu
TPf
TNu
DOC
Axis 1 λ=0.379
Axi
s 2 λ=
0.18
0
Mould Bay
N. EllesmereOasis Ellesmere
Cape VeraMelville
BanksVictoriaBathurstIsachsenDevonAxelAlert
Cornwallis
-0.4 1.0
-0.6
1.0
cond
pH
TPu
TPf
TNu
DOC
Axis 1 λ=0.379
Axi
s 2 λ=
0.18
0
-0.4 1.0
-0.6
1.0
cond
pH
TPu
TPf
TNu
DOC
Axis 1 λ=0.379
Axi
s 2 λ=
0.18
0
Mould Bay
N. EllesmereOasis Ellesmere
Cape VeraMelville
BanksVictoriaBathurstIsachsenDevonAxelAlert
Cornwallis
Mould Bay
N. EllesmereOasis Ellesmere
Cape VeraMelville
BanksVictoriaBathurstIsachsenDevonAxelAlert
Cornwallis
Cape VeraMelville
BanksVictoriaBathurstIsachsenDevonAxelAlert
Cornwallis
Figure 4.
5-05
7-04
9-04
4-04
31-0
617
-05
3-04
13-0
68-
0410
-04
18-0
59a
-56-
0420
-05
14-0
516
-05
2-04
1-04
12-0
515
-05
24-0
522
-05
23-0
511
-04
020
Microc
ostat
uskra
sske
i
0Nitzsc
hiapu
ra
020
Encyo
nemaminu
tum
0Navicu
lach
iarae
020
Encyo
nemafog
edii
0Encyo
nemasil
esiac
um
0Fragila
riaca
pucin
a
020
4060
8010
0
Nitzsc
hiape
rminu
ta
020
4060
80
Nitzsc
hiafru
stulum
020
Cymbe
llacle
ve-eu
lerae
0Navicu
lavu
lpina
020
40
Diatom
atenuis
0Psammoth
idium
marginu
latum
0Achna
nthidi
umminu
tissim
um
020
Cymbe
llabo
tellus
020
Cymbo
pleura
angu
stata
Pond name
Rel
ativ
e ab
unda
nce
(%)
5-05
7-04
9-04
4-04
31-0
617
-05
3-04
13-0
68-
0410
-04
18-0
59a
-56-
0420
-05
14-0
516
-05
2-04
1-04
12-0
515
-05
24-0
522
-05
23-0
511
-04
020
Microc
ostat
uskra
sske
i
0Nitzsc
hiapu
ra
020
Encyo
nemaminu
tum
0Navicu
lach
iarae
020
Encyo
nemafog
edii
0Encyo
nemasil
esiac
um
0Fragila
riaca
pucin
a
020
4060
8010
0
Nitzsc
hiape
rminu
ta
020
4060
80
Nitzsc
hiafru
stulum
020
Cymbe
llacle
ve-eu
lerae
0Navicu
lavu
lpina
020
40
Diatom
atenuis
0Psammoth
idium
marginu
latum
0Achna
nthidi
umminu
tissim
um
020
Cymbe
llabo
tellus
020
Cymbo
pleura
angu
stata
Pond name
Rel
ativ
e ab
unda
nce
(%)
5-05
7-04
9-04
4-04
31-0
617
-05
3-04
13-0
68-
0410
-04
18-0
59a
-56-
0420
-05
14-0
516
-05
2-04
1-04
12-0
515
-05
24-0
522
-05
23-0
511
-04
020
Microc
ostat
uskra
sske
i
0Nitzsc
hiapu
ra
020
Encyo
nemaminu
tum
0Navicu
lach
iarae
020
Encyo
nemafog
edii
0Encyo
nemasil
esiac
um
0Fragila
riaca
pucin
a
020
4060
8010
0
Nitzsc
hiape
rminu
ta
020
4060
80
Nitzsc
hiafru
stulum
020
Cymbe
llacle
ve-eu
lerae
0Navicu
lavu
lpina
020
40
Diatom
atenuis
0Psammoth
idium
marginu
latum
0Achna
nthidi
umminu
tissim
um
020
Cymbe
llabo
tellus
020
Cymbo
pleura
angu
stata
Pond name
Rel
ativ
e ab
unda
nce
(%)
Figu
re 5
.
-1.5 1.0
-1.0
1.0
P.margA.minu
C.anguC.desi
E.foge
E.late
E.micr
E.minu
C.sile
D.kuet
D.cont
D.tenu
D.marg
F.capu
M.kras
N.chia
N.phyl
N.vulp
N.diss
N.frust
N.perm
N.pura
1
2
34
5
6
78
99a
10
11
12
14
15
16
17
18
20
2223
24
1331
Axis 1 λ = 0.278
Axi
s 2 λ
= 0.
172
R.pete
A.krie
C.clevE.laev
N.pale
-1.5 1.0
-1.0
1.0
P.margA.minu
C.anguC.desi
E.foge
E.late
E.micr
E.minu
C.sile
D.kuet
D.cont
D.tenu
D.marg
F.capu
M.kras
N.chia
N.phyl
N.vulp
N.diss
N.frust
N.perm
N.pura
1
2
34
5
6
78
99a
10
11
12
14
15
16
17
18
20
2223
24
1331
Axis 1 λ = 0.278
Axi
s 2 λ
= 0.
172
R.pete
A.krie
C.clevE.laev
N.pale
-1.5 1.0
-1.0
1.0
P.margA.minu
C.anguC.desi
E.foge
E.late
E.micr
E.minu
C.sile
D.kuet
D.cont
D.tenu
D.marg
F.capu
M.kras
N.chia
N.phyl
N.vulp
N.diss
N.frust
N.perm
N.pura
1
2
34
5
6
78
99a
10
11
12
14
15
16
17
18
20
2223
24
1331
Axis 1 λ = 0.278
Axi
s 2 λ
= 0.
172
R.pete
A.krie
C.clevE.laev
N.pale
Figure 6.
b)
-1.0 1.0
-1.0
1.0
δ15N
cond
DOC
1
2
3
4
5
6
7
8
9
9a
10
11
12
14
15
16
17
18
2022
2324
13
31
Axis 1 λ = 0.132
Axis
2 λ
= 0.
051
a)
-1.0 1.0
-1.0
1.0
A.krie
E.laev
A.marg
A.minu
E.late
D.kuet
N.perm
R.pete
C.angu
C.bote
C.clev
E.foge
E.micr
E.minu
E.sile
D.cont
D.tenu
D.margF.capu
M.kras
N.chia
N.cryp
N.phyl
N.vulp
N.dissN.frus
N.pale
N.pura
C.desi
Axis 1 λ = 0.132
Axis
2 λ
= 0.
051
b)
-1.0 1.0
-1.0
1.0
δ15N
cond
DOC
1
2
3
4
5
6
7
8
9
9a
10
11
12
14
15
16
17
18
2022
2324
13
31
b)
-1.0 1.0
-1.0
1.0
δ15N
cond
DOC
1
2
3
4
5
6
7
8
9
9a
10
11
12
14
15
16
17
18
2022
2324
13
31
Axis 1 λ = 0.132
Axis
2 λ
= 0.
051
a)
-1.0 1.0
-1.0
1.0
A.krie
E.laev
A.marg
A.minu
E.late
D.kuet
N.perm
R.pete
C.angu
C.bote
C.clev
E.foge
E.micr
E.minu
E.sile
D.cont
D.tenu
D.margF.capu
M.kras
N.chia
N.cryp
N.phyl
N.vulp
N.dissN.frus
N.pale
N.pura
C.desi
a)
-1.0 1.0
-1.0
1.0
A.krie
E.laev
A.marg
A.minu
E.late
D.kuet
N.perm
R.pete
C.angu
C.bote
C.clev
E.foge
E.micr
E.minu
E.sile
D.cont
D.tenu
D.margF.capu
M.kras
N.chia
N.cryp
N.phyl
N.vulp
N.dissN.frus
N.pale
N.pura
C.desi
Axis 1 λ = 0.132
Axis
2 λ
= 0.
051
Figure 7.
Tabl
e 1.
Sel
ecte
d lim
nolo
gica
l dat
a fo
r pon
ds lo
cate
d ne
ar C
ape
Ver
a, D
evon
Isla
nd.
Si
teSa
mpl
eSa
mpl
eδ1
5N%
CpH
Con
dN
O3N
O2
N
H3
CH
La
Cl
SO4
DO
CD
ICC
aM
g K
Na
Dat
eco
de(‰
)%
(μS/
cm)
mg/
Lm
g/L
µg/L
mg/
Lm
g/L
mg/
Lm
g/L
mg/
Lm
g/L
mg/
Lm
g/L
CV
-126
/07/
2004
1-04
6.62
17.1
68.
4215
20.
056
0.01
25.
56.
381.
941
19.5
209.
050.
333.
28C
V-2
26/0
7/20
042-
044.
363.
778.
2814
00.
034
0.00
99.
74.
60.
740.
917
.818
.27.
80.
272.
43C
V-3
26/0
7/20
043-
0418
.41
1.45
9.63
240
1.12
0.01
645
.632
.211
.43.
918
.820
.611
.92.
4221
CV
-426
/07/
2004
4-04
14.3
80.
549.
9521
50.
280.
026
10.9
27.4
6.45
7.1
1516
.511
2.25
16.7
CV
-502
/07/
2005
5-05
9.36
10.6
46.
9511
60.
021
0.00
60.
92.
840.
60.
914
.614
.75.
050.
181.
78C
V-6
03/0
7/20
056-
059.
1315
.44
7.5
146.
50.
025
0.01
0.5
3.73
0.64
1.6
19.5
196.
260.
292.
37C
V-7
09/0
7/20
057-
0510
.58
0.50
7.95
134.
90.
020.
009
10.8
1.69
0.37
1.4
17.1
20.7
4.66
0.26
1.24
CV
-826
/07/
2004
8-04
16.3
426
.50
10.7
521
00.
011
0.01
358
.624
.32.
5712
.617
.127
.110
.31.
3413
.2C
V-9
09
/07/
2005
9-05
15.9
520
.57
8.95
121
0.07
90.
011
13.
090.
621.
613
.215
.75.
220.
31.
99C
V-9
a09
/07/
2005
9a-0
513
.02
42.1
49.
7512
6.5
0.06
30.
025
1.9
5.89
1.07
3.9
11.5
146.
20.
443.
44C
V-1
026
/07/
2004
10-0
413
.26
19.3
89.
4915
30.
048
<0.0
0535
.711
.71.
071.
215
.121
.77.
540.
46.
43C
V-1
126
/07/
2004
11-0
40.
748.
438.
7014
20.
017
<0.0
0515
.46.
990.
690.
917
.819
7.03
0.2
2.99
CV
-12
02/0
7/20
0512
-05
3.30
16.5
37.
6531
2.5
0.01
20.
035
246
.98.
734.
121
.617
.611
1.87
29.4
CV
-14
03/0
7/20
0514
-05
8.15
32.4
37.
116
9.2
0.00
70.
006
1.5
9.69
1.45
3.6
19.6
17.3
7.48
0.64
5.32
CV
-15
03/0
7/20
0515
-05
16.3
421
.84
7.55
167.
40.
016
0.01
60.
94.
951.
12.
421
.320
.77.
290.
373.
01C
V-1
604
/07/
2005
16-0
55.
9921
.93
7.65
125.
80.
045
0.00
80.
94.
41.
53.
016
.014
.87.
460.
252.
36C
V-1
704
/07/
2005
17-0
56.
2327
.15
7.35
84.7
0.02
10.
016
0.6
3.96
0.62
1.4
10.3
8.78
4.45
0.16
2.08
CV
-18
04/0
7/20
0518
-05
6.46
32.3
77.
8514
80.
019
0.00
50.
510
.91.
123.
016
.014
.67.
660.
254.
69C
V-2
009
/07/
2005
20-0
57.
6035
.73
10.0
517
30.
020.
018
0.9
14.8
1.69
6.2
12.5
17.8
8.09
0.84
7.94
CV
-22
10/0
7/20
0522
-05
1.34
16.4
17.
5512
5.4
0.01
40.
026
0.9
15.2
1.07
1.8
10.9
10.7
5.19
0.35
7.25
CV
-23
10/0
7/20
0523
-05
1.61
34.5
67.
213
6.5
0.01
70.
013
0.5
20.5
1.19
1.5
10.1
9.99
5.3
0.41
9.73
CV
-24
10/0
7/20
0524
-05
1.62
35.7
58
127.
30.
022
0.00
90.
65.
171.
141.
515
.714
.37
0.35
2.93
CV
-13
10/0
7/20
0613
-06
19.3
825
.68
10.0
023
0.6
0.14
70.
050.
513
.84.
495.
718
.320
.710
.10.
987.
28C
V-3
118
/07/
2006
31-0
611
.95
5.78
8.60
180
0.01
10.
029
0.8
5.15
0.98
1.6
18.1
21.6
5.9
0.28
2.9
Mea
n9.
2619
.70
8.45
161.
60.
089
0.01
78.
6311
.93
2.22
3.0
16.1
17.3
7.46
0.64
6.74
Med
ian
8.64
19.9
88.
1414
7.3
0.02
10.
013
0.95
6.69
1.11
1.7
16.6
17.7
7.38
0.35
3.36
Max
19.3
842
.14
10.7
531
2.5
1.12
00.
050
58.6
046
.90
11.4
012
.621
.627
.111
.90
2.42
29.4
0M
in0.
740.
506.
9584
.70.
007
<0.0
050.
501.
690.
370.
910
.18.
84.
450.
161.
24
Tabl
e 1.
con
tinue
d.
Si
teN
O2
POC
PON
SiO
2SR
P T
KN
T
dN
TPf
TPu
H
ill's
N:P
N o
r P
mg/
Lm
g/L
mg/
Lm
g/L
mg/
Lm
g/L
mg/
Lm
g/L
mg/
LN
2m
olar
limite
d?C
V-1
0.00
40.
197
0.02
0.3
0.00
110.
129
0.20
30.
0052
0.00
564.
478
.4p
C
V-2
0.00
20.
159
0.02
10.
170.
001
0.08
10.
132
0.00
230.
004
11.5
72.9
pC
V-3
0.02
84.
860.
863
0.91
0.00
150.
589
1.59
0.01
390.
19.
3555
.1p
CV
-40.
012
0.89
20.
136
2.64
0.00
170.
856
0.99
50.
0227
0.04
718.
2757
.9p
CV
-50.
003
0.34
20.
013
0.25
0.00
10.
093
0.11
40.
0057
0.00
948.
4729
.0ei
ther
CV
-60.
004
0.28
30.
017
0.31
0.00
070.
131
0.17
60.
0041
0.00
568.
2866
.2p
CV
-70.
003
0.74
20.
051
0.43
0.00
080.
184
0.21
10.
0052
0.02
698.
1220
.3ei
ther
CV
-80.
002
9.27
0.99
1.09
0.00
290.
937
0.82
30.
035
0.20
72.
6520
.1ei
ther
CV
-9
0.00
50.
323
0.02
40.
420.
0226
0.19
20.
293
0.02
860.
0353
6.37
17.9
n
CV
-9a
0.00
85.
620.
666
0.36
0.05
550.
404
0.44
80.
072
0.12
42.
7119
.6n
CV
-10
0.00
54.
80.
694
0.51
0.00
090.
243
0.30
10.
0117
0.01
833.
0811
5.3
pC
V-1
10.
001
1.33
0.23
40.
260.
0008
0.06
10.
072
0.00
220.
0169
13.3
739
.6ei
ther
CV
-12
0.00
40.
888
0.1
2.49
0.00
130.
448
0.43
10.
0134
0.02
615.
9146
.0ei
ther
CV
-14
0.00
20.
777
0.08
70.
090.
013
0.27
30.
286
0.01
640.
0318
6.37
24.7
eith
erC
V-1
50.
003
0.40
70.
033
0.51
0.00
070.
220.
240.
0042
0.01
047.
8455
.4p
CV
-16
0.00
30.
16<0
.005
0.37
0.00
280.
097
0.11
90.
0049
0.01
126.
6427
.4ei
ther
CV
-17
0.00
40.
50.
060.
140.
0011
0.15
40.
153
0.00
580.
0143
3.16
35.2
eith
erC
V-1
80.
004
0.23
0.00
50.
190.
001
0.13
90.
141
0.00
850.
0104
133
.6ei
ther
CV
-20
0.00
50.
479
0.04
90.
810.
009
0.65
70.
628
0.02
890.
0371
3.73
41.9
eith
erC
V-2
20.
004
1.21
0.11
70.
310.
0005
0.26
50.
261
0.00
290.
0359
11.7
723
.6ei
ther
CV
-23
0.00
30.
333
0.02
50.
210.
0003
0.21
30.
210.
0028
0.00
8313
.18
65.8
pC
V-2
40.
003
0.27
0.01
30.
250.
0003
0.14
70.
162
0.01
070.
0096
7.83
40.6
eith
erC
V-1
30.
005
0.32
0.04
52.
610.
0113
0.71
30.
909
0.03
530.
0488
5.92
39.7
eith
erC
V-3
10.
002
0.26
40.
027
0.33
0.00
830.
203
0.20
80.
0031
0.00
8414
.97
61.5
pM
ean
0.00
51.
444
0.18
70.
670.
0058
0.31
00.
379
0.01
440.
0355
7.29
45.3
2ei
ther
Med
ian
0.00
40.
443
0.04
90.
350.
0011
0.20
80.
226
0.00
720.
0176
7.24
40.1
8ei
ther
Max
0.02
89.
270
0.99
02.
640.
0555
0.93
71.
590
0.07
200.
2070
14.9
711
5.34
pM
in0.
001
0.15
9<0
.005
0.09
0.00
030.
061
0.07
20.
0022
0.00
401.
0017
.91
n
Tabl
e 2.
Pea
rson
’s c
orre
latio
n m
atrix
with
Bon
ferr
oni-a
djus
ted
prob
abili
ties.
Sig
nific
ant v
alue
s are
den
oted
in b
old
(p<0
.01)
or i
n un
derli
ned
italic
s (p<
0.05
).
δ15
N%
CpH
CO
ND
NO
3NO
2N
H3
CH
La
DO
CN
O2
PON
SiO
2SR
PT
KN
Cd
FePb
TdN
TPf
Tpu
δ15N
1%
C-0
.169
1pH
0.53
7-0
.076
1C
ON
D0.
328
-0.2
320.
446
1N
O3N
O2
0.49
2-0
.384
0.50
30.
251
1N
H3
0.26
3-0
.003
0.24
0.40
80.
215
1C
HL
a0.
243
-0.5
240.
538
0.34
0.33
9-0
.279
1D
OC
0.48
10.
137
0.63
70.
658
0.20
60.
567
0.19
31
NO
20.
485
-0.1
760.
384
0.28
10.
824
0.39
50.
151
0.38
11
PON
0.28
4-0
.133
0.58
20.
358
0.22
40.
096
0.67
20.
491
0.3
1Si
O2
0.45
-0.2
440.
628
0.76
10.
461
0.55
50.
282
0.68
30.
468
0.33
81
SRP
0.52
90.
241
0.47
20.
145
0.16
50.
322
-0.0
830.
453
0.17
0.18
20.
173
1T
KN
0.53
10.
061
0.67
0.66
20.
332
0.62
20.
230.
937
0.55
30.
572
0.73
80.
396
1C
d0.
579
-0.0
360.
502
0.38
10.
308
0.40
80.
180.
635
0.35
90.
418
0.49
60.
695
0.61
41
Fe0.
351
-0.2
240.
211
0.29
0.23
0.53
70.
016
0.47
40.
450.
376
0.45
30.
312
0.59
70.
731
1Pb
0.33
5-0
.402
0.41
10.
352
0.24
80.
205
0.42
50.
141
0.09
20.
489
0.42
70.
151
0.27
40.
477
0.49
81
TdN
0.61
8-0
.088
0.71
0.68
70.
563
0.58
50.
340.
864
0.70
20.
586
0.74
50.
392
0.95
30.
639
0.57
80.
334
1T
Pf0.
575
0.33
10.
669
0.35
10.
356
0.32
50.
121
0.75
0.49
40.
418
0.51
0.69
20.
742
0.74
10.
453
0.18
70.
722
1T
pu0.
461
0.05
20.
675
0.39
0.32
20.
334
0.40
60.
793
0.43
70.
715
0.52
70.
516
0.79
10.
747
0.58
40.
332
0.77
0.76
81
Table 3. Summary of dominant diatom species (at least 1% relative abundance in at least 3 sites) found in surface sediments of Cape Vera ponds, where “Code” is the shortened species name, “N” is the number of occurrences, “N2” is the effective number of occurrences (Hill 1973), and “Max %” is the maximum relative abundance of a given taxon.
Code N N2 Max %Achnanthidium kriegeri (Krasske) Hamilton, Antoniades & Siver A.krie 3 4.57 2.21Eucocooneis laevis (Østrup) Lange-Bertalot E.laev 3 2.88 3.42Psammothidium marginulatum (Grunow) Bukhtiyarova & Round P.marg 6 7.69 4.32Achnanthidium minutissimum (Kützing) Czarnecki A.minu 9 9.38 8.97Rossithidium petersenii (Hustedt) Round & Bukhtiyarova R.pete 3 6.22 4.32Cymbopleura angustata var spitzbergensis Krammer C.angu 5 4.37 17.32Cymbella botellus (Lagerstedt) Schmidt C.bote 8 7.95 16.30Cymbella cleve-eulerae Krammer C.clev 16 13.94 19.28Cymbella designata Krammer C.desi 4 5.43 2.18Encyonema fogedii Krammer E.foge 19 15.89 24.87Encyonema latens (Krasske) D. Mann E.late 4 7.59 2.56Encyonopsis microcephala (Grunow) Krammer E.micr 4 3.97 15.72Encyonema minutum (Hilse) D. Mann E.minu 14 11.73 26.74Encyonema silesiacum (Bleisch) D. Mann E.sile 5 12.35 3.30Denticula kuetzingii Grunow D.kuet 4 3.43 23.99Diadesmis contenta (Grunow ex. Van Huerck) D. Mann D.cont 3 2.48 14.50Diatoma tenuis Agardh D.tenu 8 7.17 43.11Diploneis marginestriata Hustedt D.marg 2 1.42 26.06Fragilaria capucina Desmazières F.capu 6 7.71 8.92Microcostatus krasskei (Hustedt) Johansen & Sray M.kras 7 6.74 19.44Navicula chiarae Lange-Bertalot & Genkal N.chia 4 4.71 3.55Navicula cryptocephala Kützing N.cryp 3 4.48 2.03Navicula phyllepta Kützing N.phyl 4 7.5 4.5Navicula vulpina Kützing N.vulp 7 9.81 4.30Nitzschia dissipata var media (Hantzsch) Grunow N.diss 3 5.97 3.05Nitzschia frustulum (Kützing) Grunow N.frus 23 19.99 87.45Nitzschia perminuta (Grunow) M. Peragallo N.perm 24 20.58 99.82Nitzschia paleacea Grunow in Van Heurck N.pale 3 3.7 3.1Nitzschia pura Hustedt N.pura 4 5 7.
9
825
Table 4. List of dominant diatom species arranged in order of their axis 1 species scores based on a redundancy analysis (RDA) constrained solely to δ15N.
RDA δ15N onlyaxis 1 scores
Cymbella botellus 0.6333Cymbella designata 0.628Cymbopleura angustata var spitzbergensis 0.6078Psammothidium marginulatam 0.5572Encyonopsis microcephala 0.4765Eucocooneis laevis 0.467Achnanthidium minutissimum 0.4564Nitzschia paleacea 0.3705Achnanthidium kriegeri 0.3174Rossithidium petersenii 0.2998Denticula kuetzigii 0.2864Diatoma tenuis 0.2562Cymbella cleve-eulerae 0.2311Diploneis marginestriata 0.1465Navicula vulpina 0.1334Nitzschia dissipata var media 0.1054Navicula chiarae 0.0788Navicula cryptocephala 0.0698Diadesmis contenta 0.0182Fragilaria capucina -0.0447Encyonema latens -0.1294Nitzschia perminuta -0.1305Navicula phyllepta -0.1342Encyonema silesiacum -0.2058Encyonema fogedii -0.2169Nitzschia frustulum -0.2169Nitzschia pura -0.2671Microcostatus krasskei -0.2803Encyonema minutum -0.6591
CHAPTER 8
TRACKING SEABIRD POPULATION DYNAMICS USING PALEOLIMNOLOGY: A CASE STUDY FROM
DEVON ISLAND, ARCTIC CANADA
BRONWYN KEATLEY, MARIANNE S.V. DOUGLAS, NEAL MICHELUTTI, JULES BLAIS, MARK
MALLORY, AND JOHN P. SMOL
ABSTRACT
Rapid environmental changes occurring in polar regions may pose a significant threat to the ~10
million seabirds currently inhabiting the Canadian Arctic. However, a paucity of long-term data
on seabird population dynamics makes proper management difficult. As top predators in the
marine environment, seabirds provide marine-derived nutrients (enriched in δ15N) to freshwater
habitats via their guano, causing trophic cascades that can be tracked using paleolimnological
techniques. Here, we analyze multiple proxies (δ15N, inferred-chlorophyll a, and diatom
assemblage composition) archived in the sediments of freshwater ponds to reconstruct population
dynamics of a colony of northern fulmars (Fulmarus glacialis) over the past few centuries at
Cape Vera, Devon Island, Nunavut, High Arctic Canada. Sedimentary δ15N and inferred-
chlorophyll a appear to track seabird activity, and indicate that seabirds have been present since
the formation of these ponds over the past ~200 years, and that their numbers have increased in
recent years. Interestingly, diatoms do not show any changes in species composition that are
consistent with shifts in δ15N and inferred-chlorophyll a. The lack of a species level response in
diatoms may be related to the dominance of chlorophytes and cyanobacteria in these systems.
INTRODUCTION
On a worldwide scale, seabirds exert comparable pressure on fish stocks to that of commercial
fisheries (Brooke 2004) and, because they feed in the marine environment and nest on land,
seabirds represent important linkages between marine and terrestrial food webs (Polis et al.
2004). For example, densely populated seabird colonies have been shown to transport significant
quantities of marine-derived nutrients and contaminants from the ocean to terrestrial sites via
their guano (Evenset et al. 2004; Blais et al. 2005; Keatley et al. [7]).
The Canadian Arctic is a critical migrating, nesting, and breeding habitat for ~10 million
marine birds (Mallory & Fontaine 2004), however, Arctic regions are likely to change
dramatically over the next few decades in response recent warming and other environmental
stressors (ACIA 2004). Such environmental changes are likely to influence seabird populations,
but understanding the magnitude or even the direction of seabird responses is unclear (Gaston &
Hipfner 1998; Grosbois & Thompson 2005; Gaston et al. 2005a; Gaston et al. 2005b). This
confusion directly relates to the lack of data on long-term, natural variability of seabird
populations. For example, seabird monitoring programs conducted under the auspices of the
Canadian Wildlife Service extend back only to the 1970s, and consist of either cursory aerial
surveys conducted at 5 – 10 year intervals, with intensive colony monitoring occurring at only a
few sites (Gilchrist et al. 2005). Other sources of long-term data, such as traditional ecological
knowledge (TEK), are often not available for species of seabirds that either roost in largely
inaccessible locations and/or are not frequent components of traditional diets (Gilchrist et al.
2005). Thus, alternative proxy data with which seabird population dynamics could be
reconstructed are highly desirable.
Paleolimnology offers the potential to track past seabird population dynamics as,
compared to un-impacted sites, seabird-influenced lakes and ponds have markedly different
limnological properties (i.e., high concentrations of nutrients), unique algal and invertebrate
species compositions, and distinct sedimentary geochemical signatures (Evenset et al. 2004; Blais
et al. 2005; Keatley et al. [7]; Evenset et al. 2007; Michelutti unpublished data). The freshwater
ponds located at Cape Vera on Devon Island, Arctic Canada represent ideal study sites to track
past seabird activity using paleolimnology. This area is home to ~10,000 nesting pairs of
northern fulmars (Fulmarus glacialis), which form dense nesting sites on cliffs that extend above
several small ponds. Recently, Keatley et al. ([7]) documented the influence of seabirds on the
ponds at Cape Vera, and Blais et al. (2005) showed a significant correlation between seabird
influence and concentration of contaminants in surface sediments.
Here, we utilize a multi-proxy approach to reconstruct the influence of seabird activity
over the past several centuries on a series of ponds at Cape Vera. In addition, we analyze two
ponds located near an abandoned colony, at Cape Hawes ~10 km from Cape Vera, but with no
present seabird activity. The proxies we have utilized to track seabird influence include (1) δ15N,
which serves as an ecological tracer for marine-derived nutrients where animals that feed high in
the trophic system, such as seabirds, have elevated δ15N ratios (Mallory 2006); (2) sedimentary
chlorophyll a, which tracks overall aquatic production; and (3) fossil diatom assemblages, which
are known to respond sensitively to changes in nutrient levels (Hall & Smol 1999). Indeed,
similar paleolimnological approaches have been successfully applied in Arctic regions to track
past abundances of sockeye salmon (Gregory-Eaves et al. 2003), as well as to assess the degree of
freshwater eutrophication from slaughtered whale carcasses (Douglas et al. 2004) and human
sewage (Douglas & Smol 2000; Michelutti et al. in press).
The main questions we address in this study are: (1) Do diatom species assemblages and
sedimentary-chla respond synchronously to seabird-derived nutrients (as measured by δ15N)
through time? (2) Have there been historical changes in the current northern fulmar colony at
Cape Vera? If so, in what direction are these changes? (3) Is there any paleolimnological
evidence to support the anecdotal evidence of an abandoned seabird colony at nearby Cape
Hawes?
We show that in most of the eight study ponds at Cape Vera, δ15N and inferred-
chlorophyll a were indicative of seabird-derived nutrient enrichment throughout the entire history
encapsulated in the cores, and had highest concentrations in the most recent sediments. As
expected, the control sites at Cape Hawes recorded low δ15N and chlorophyll a compared to the
Cape Vera sites, although one core recorded evidence of a once active seabird colony. Diatom-
based paleolimnological indices, however, do not show consistent patterns to either δ15N or
sedimentary inferred-chla, confirming previous research that diatom assemblages do not respond
in a simple manner to prolonged nutrient enrichment in High Arctic ponds.
SITE DESCRIPTION
Located in the Canadian High Arctic, Cape Vera, Devon Island, Nunavut (Fig. 1) is home
to a large colony (~10 000 breeding pairs) of northern fulmars (Fulmarus glacialis) (Gaston et al.
2006). Mallory (2006) provides a review of northern fulmar ecology in the Canadian High
Arctic. Current estimates suggest that the Cape Vera colony has not changed appreciably in size
since the 1970s, although this information is largely based on aerial surveys with which accurate
counts of the cryptically-coloured northern fulmar are difficult to obtain (Gaston et al. 2006).
Northern fulmars feed near the top of the marine food web and thus are enriched in δ15N with
average δ15N of ~13-17‰ (Mallory 2006).
The northern fulmars at Cape Vera nest in dolostone / limestone cliffs rising ~250 m
above sea level (Mayr et al. 1998). Stretching below these cliffs are a suite of freshwater ponds
that receive varying amounts of seabird-derived subsidies (e.g. guano, regurgitated stomach oil,
eggshells, feathers, carcasses) that significantly impact the water quality and biota (i.e. diatom
assemblages) within the ponds (Fig. 2; Keatley et al. [7]). Details on the modern limnology and
diatom distributions from these ponds are summarized in Keatley et al. [7]). Briefly, the ponds
most affected by seabirds are characterized by high pH, primary production, and nutrient-related
variables (Table 1). The diatom assemblages recorded at Cape Vera rank amongst the lowest in
the Canadian Arctic in terms of species diversity, and are dominated by small Nitzschia and
Cymbella taxa, some of which show significant relationships to seabird-derived nutrients (δ15N).
METHODS
We identified several ponds as coring candidates based on the criteria that they spanned
an apparent gradient of seabird influence and contained adequate sediment for coring. Sediment
cores were taken from 10 ponds during our field seasons in July 2005 and July 2006 using a
plexi-glass core tube from a Glew gravity corer (inner diameter = 3.75 cm; Glew et al. 2001). For
all but three ponds (CV12, CV13, CV24), replicate cores were obtained in order to perform multi-
proxy analyses. All cores were sectioned in the field at 0.5 cm intervals using a Glew (1988)
extruder, except CV5 which was sectioned into 1.0 cm intervals.
Diatom frustules were prepared for analysis using HNO3 in a MARSX microwave
digestion procedure (Parr et al. 2004) to remove organic material from each sample. The
resultant slurries were repeatedly rinsed with deionized water until a neutral pH was reached,
after which subsamples were strewn onto pre-cleaned cover slips and mounted on slides using
Naphrax®, a mounting medium with a high refractive index. Identifications primarily followed
Antoniades et al. (in press), Krammer and Lange-Bertalot (1991), and Krammer (2002).
Sediment samples from the cores were also analysed for stable isotopes of δ15N, as well
as for elemental analysis (C, N) at G.G. Hatch Laboratory, University of Ottawa, Ottawa, Canada.
The ratio of C/N can be used to assess whether sedimentary organic matter is largely of aquatic
(C/N <10) or terrestrial (C/N >20) origin (Meyers & Ishiwatari 1993). The elemental analyses
were conducted on a Vario EL III (Elementar, Germany), while stable isotopes were measured
using a continuous-flow DeltaPlus Advantage isotope ratio mass spectrometer (ThermoFinnigan,
Germany) coupled with a ConFlo III. Isotope data was normalized using previously calibrated
internal standards, and analytical precision is +/-0.2 ‰.
Sedimentary chlorophyll a (chla) was inferred using a new application of visible
reflectance spectroscopy (Das et al. 2005; Wolfe et al. 2006). Briefly, sediment spectra were
obtained using a FOSS NIR Systems Rapid Content Analyzer, and these data were converted to
chla estimations using the algorithm developed by Michelutti et al. (2005). This technique
effectively measures chla and its derivatives and is sensitive to chla changes on the order of 0 –
0.1 mg/g dry weight. Due to a shortage of sample material in some of the cores, inferred-chla
was analysed for replicate cores when necessary (CV9a, CV22). These replicate cores were
cross-validated using independent 210Pb dating where available.
The sediment cores were radiometrically dated using 210Pb and 137Cs from gamma
spectrometry (Appleby 2001) at either the University of Ottawa or at PEARL, Queen’s
University, Kingston. Both Constant Rate of Supply (CRS) and Constant Initial Concentration
(CIC) age models were explored to determine the most appropriate fit for each core, based on the
137Cs peak (Michelutti et al., submitted).
Statistical analyses
Diatom assemblage data are reported as relative abundances and dominant taxa (>3%
relative abundance) from each core were plotted in a stratigraphy using the program C2 (Juggins
2003). Species data were analysed using a variety of multivariate ordination approaches using
the program CANOCO version 4.5 (ter Braak & Šmilauer 2002). First, a detrended
correspondence analysis (DCA) was used to determine the length of the species gradient (or
amount of species turnover) along the main axis of variation in each core. Since the DCA axis 1
gradient lengths were short (i.e. < 2 standard deviation units), we used the linear ordination
technique of Principal Components Analysis (PCA) to summarize the main directions of variation
in the diatom data (Lepš & Šmilauer 2003).
RESULTS
Core summaries
We obtained sediment cores from eight ponds at Cape Vera (CV5, CV6, CV7, CV9,
CV9a, CV12, CV13, CV20) and two ponds at Cape Hawes (CV22, CV24). The sediment cores
ranged in length from 6 cm to 23 cm long. CIC models dating models were invalidated by non-
monotonic features of the 210Pb profiles in all cores, necessitating the use of CRS models
(Michelutti et al. submitted). However, agreement between the CRS dates and independent
markers (137Cs, PCB data) was only obtained for CV6, CV9a, and CV22. In CV7, CV9 and
CV20, dates are approximated based on 137Cs peaks (Michelutti et al. submitted). The other cores
were challenging to date. Details on all 210Pb profiles and model comparisons can be found in
Michelutti et al. (submitted). Given these challenges, we compare the paleoenvironmental data
with respect to depth for the purposes of this paper. Nevertheless, we suggest that we likely have
the full sedimentary record in these sites as, in most ponds, the core tube hit hard material (likely
bedrock or ice) that prevented the retrieval of longer sediment records.
Due to the wide variety of paleolimnological proxies and divergent responses with the
diatom data, ecosystem-level generalizations are difficult. However, in the interest of space
limitation, we summarize the main findings below, organized by type of proxy indicator. This
section is followed by a more detailed description of each core.
Stable Isotopes: δ15N
The lowest average δ15N (~2‰) was found in the two control sites at Cape Hawes
(CV24, CV22) and in CV12, the Cape Vera pond located furthest from the colony (Figs. 3-11).
The highest average δ15N (~12‰) was found in CV13, CV9, CV9a (Figs. 3-11). With the
exception of CV12, CV22, and CV24, all cores showed general patterns of δ15N enrichment
towards the tops of the cores with the greatest increase in the Cape Vera cores; this pattern was
much smoother in some cores than in others (Figs. 3-11).
Carbon/Nitrogen
Ratios of C/N in the Cape Vera cores were generally >20 in the bottom of most cores,
shifting towards values <10 near the top of the cores (Figs. 3-11). At Cape Hawes, the C/N ratios
are generally below 10 for most of the sedimentary record. Trends in C/N were generally not
consistent with δ15N, chla, or diatom proxies.
Inferred chlorophyll a
Due to a lack of sediment, chla was measured from only seven cores (CV5, CV6, CV9,
CV9a, CV20, CV22, CV24). Chla trends were generally correlated with those found for δ15N
(Figs. 3-11). The two control cores CV22 and CV24, as well as the snow pack-influenced CV5
had the lowest average chla (Figs. 3, 10, 11). At the other end of the spectrum, CV20 (~0.4
mg/g) had the highest chla, closely followed by CV9a, CV9, and CV6 (Figs. 4, 5, 6, 9). All cores
showed increases in chla in the uppermost sediments, although the more heavily affected ponds
exhibited more dramatic increases and these generally began earlier in the sedimentary records
(Fig.1, Figs. 3-11).
Diatoms
Diatom frustules were present throughout the entire length of all cores except CV7,
suggesting that diatoms were relatively well-preserved throughout the cores. While some cores
showed monotonic changes that are consistent with trends in δ15N and/or chla (e.g. CV9, CV13,
CV22), others indicated large fluctuations in diatom assemblage composition that bear little
relationship to either δ15N or chla (e.g. CV5, CV6, CV9a; Figs. 3-11).
Trends in diatom assemblage diversity, as estimated by Hill’s N2, indicated little relation
to patterns of main direction of species variation (PCA1), seabird-derived nutrients (δ15N) or chla
(Figs. 3-11). Diatom species diversity showed no relationship with time, as some cores displayed
greatest diversity in most recent sediments (e.g. CV5, CV9; Figs. 3, 5), while others had the
greatest diversity in the oldest sediments (e.g. CV6, CV9a; Figs. 4, 6). In general, the absolute
value of the Hill’s N2 number from site to site followed a trend of the most affected ponds
exhibiting lower diversity (i.e. CV9, CV20, CV13 (Hill’s N2 = ~5)) than the least affected ponds
(i.e. CV5, CV22 (Hill’s N2 = ~14)).
Detailed core descriptions – Cape Vera cores (CV5 – CV20)
Pond CV5 (Figs. 3, 12)
The 10 cm long core from CV5 represents at least ~100 years of sediment accumulation,
as estimated by the 210Pb date of 1929 at 8.25 cm (Michelutti et al. submitted). Nitzschia
frustulum and N. perminuta are overwhelmingly dominant, accounting for ~20-80% of the
assemblage at any one interval (Fig. 12). Other prominent features of this core included species
associated with higher δ15N sites based on our calibration set (e.g. Encyonema silesiacum, E.
fogedii, E. minutum; Fig. 12). Above ~3.5 cm, Nitzschia spp. decreases in abundance (~10 -
20%), while Navicula spp. (N. chiarae, N. phyllepta, N. salinarum, N. sp. aff veneta),
Microcostatus krasskei and Cymbella spp. sensu lato (~20%) become more important
components of the assemblage (Fig. 12). The species changes are summarized by the PCA axis 1
sample scores, which show non-monotonic changes throughout the core (Fig. 3). The δ15N is high
throughout this core (~7‰) and increases at the top (to ~10‰; Fig. 3). Elemental data shows little
change in %C (~12% throughout), while %N increases slightly at the top of the core (to a
maximum of ~1%); C/N remains above 20 for most of the record, although a recent decline (~15)
is apparent in the upper most sediments (Fig. 3). Inferred-chla is low throughout this core (0.05
mg/g), although a very small increase is apparent at the top of the core. Species diversity (Hill’s
N2) is high and increases towards the top of the core (~11 – 18).
Pond CV6 (Figs. 4, 13)
Pond CV6 drains pond CV5, yet the two sediment cores share little in common. The 12
cm long core from CV6 represents a much longer time period than the CV5 core, as 210Pb
background at 5 cm corresponds to an age of 1920 AD (Michelutti et al. submitted). Although
the entire record is dominated by Nitzschia frustulum (~30-70%), the bottom half of the core is
the most diverse (Hill’s N2 = ~12) and features species characteristic of low δ15N (e.g. Cymbella
botellus, C. cleve-eulerae, Achnanthidium minutissimum, Navicula vulpina, Cyclotella
pseudostelligera, and Fragilaria construens var. venter; Fig. 13). In the upper ~5.5 cm, diversity
decreases (Hill’s N2 = ~6) as Nitzschia frustulum reaches its maximum abundance and the low
δ15N species disappear (Figs. 4, 13). The δ15N remains steady at ~5‰ for most of the core, but
increase to ~10‰ in the top 2.5 cm. The relative stability of both %N is consistent at (~1%) and
%C (~16%) result in a relatively stable C/N ~20 (Fig. 4). The chla is characterized by a doubling
at the top from ~0.1 – 0.2 mg/g (Fig. 4).
Pond CV7 (Fig. 14)
Due to an absence of diatoms throughout much of the CV7 core, we only briefly describe
it here. At 20 cm long, the core from CV7 was amongst the longest retrieved and may represents
~ 200 years of sediment accumulation (137Cs peak at 3 cm = 1963 AD, Michelutti et al.
submitted). The oldest sediments in the assemblage are characterized by relatively few valves
representing Caloneis spp., Cocconeis spp., Cymbella botellus, C. cleve-eulerae, Encyonema
fogedii, Navicula chiarae, N. vulpina, Nitzschia perminuta, and an unidentified fragment,
possibly of marine origin (Fig. 14). No diatoms are present between 19 and 5.25 cm, although
diatoms are abundant in the uppermost ~5 cm (Fig. 14). Sediments above 3 cm depth were
characterized by greater relative abundances of small Nitzschia spp. and Psammothidium
marginulatum, and decreased abundances of Cymbella botellus (Fig. 14).
Pond CV9 (Figs. 5, 15)
Pond CV9 is situated in closest proximity to the bird colony; the 22 cm long core likely
represents ~200 years of accumulation (137Cs peak at 3 cm = 1963 AD, Michelutti et al.
submitted). Below ~14 cm depth, the core is heavily dominated by Nitzschia frustulum (up to
~70% relative abundance); this species is accompanied by small percentages of Cymbella cleve-
eulerae, Navicula vulpina, and Nitzschia perminuta (Fig. 15). Beginning at ~13.25 cm, Nitzschia
frustulum decreases in abundance (although never dropping below 30%) while Encyonema
fogedii and E. minutum (species associated with higher δ15N) become more abundant (Fig. 15).
The PCA1 exhibits a largely monotonic increase toward the top of the core (Fig. 5), and this is
mirrored by increases in δ15N (from ~9 - 17‰), %N (from ~0.5 - ~2%), %C (from ~0 – 16%),
C/N steadily decreasing (above 19 cm), chla (from 0.05 – 0.3 mg/g), and chironomid head
capsule abundance (Michelutti, unpublished data).
Pond CV9a (Figs. 6, 16)
Situated beside CV9, the 12.5 cm core from pond CV9a represents a slower
sedimentation rate, as 7 cm depth represents ~1900 AD (Michelutti et al. submitted). The greatest
diversity (Hill’s N2 = 14) in the sediment core occurs in the bottom sediments and is represented
by Cyclotella pseudostelligera, Encyonema minutum, Fragilaria construens var. venter, F.
pinnata, and Nitzschia frustulum (Fig. 16). The rest of the core is overwhelmingly dominated by
N. frustulum; these diatom assemblage changes are clearly captured by the fluctuations in the
PCA1 scores below 8 cm and by the static PCA1 values from 8 cm (~1910) to the top of the core
(Fig. 6). The δ15N was very high throughout the core (between 9-11‰), while C/N showed little
change (Fig. 6). CV9a had the highest %N (2-5%), %C (16-34%), and chla (0.1-0.5 mg/g) of all
the cores in our study (Fig. 6).
Pond CV12 (Figs. 7, 17)
Of the ponds located at Cape Vera, CV12 was the furthest from the colony. Fossil
diatoms in this 6.5 cm core were difficult to identify due to a large percentage of broken valves,
particularly in the bottom half of the core. Between 3 - 6.5 cm, the diatom assemblage was
characterized by what appeared to be central areas of large Neidium spp. and Navicula spp., and
to a lesser degree smaller Cocconeis spp. (Fig. 17). These may reflect marine influence, as several
Cocconeis spp. have been associated with high conductivity values in Arctic regions (Ng & King
1999). Beginning at ~2.25 cm, these species largely disappear and the core is dominated by
Nitzschia frustulum, N. perminuta and Denticula kuetzingii (Fig. 17). PCA1 scores exhibit a
steady increase towards the top of the core, while δ15N decreases monotonically from ~7‰ to
~1.5‰ (Fig. 7). Percent N and %C exhibit the opposite trend, increasing from ~0 - 2% and 12-
24%, respectively, at top of the core (Fig. 7). Aside from a peak at 4 - 4.5 cm, the C/N declines
towards the top of the core (Fig. 7). Diatom assemblage diversity decreases at the top of the core
(from Hill’s N2 of ~11 to ~5; Fig. 7).
Pond CV13 (Figs. 8, 18)
The presence of stone circles around CV13 suggests that early Arctic peoples were active
in the watershed of this pond. This 13.25 cm core is characterized by relatively low diversity
(Hill’s N2: ~5 – 7), although the upper samples are slightly more diverse (Fig. 8). In the bottom
half of the 13.25 cm core, Nitzschia frustulum (~60%) is the dominant diatom (Fig. 18). Above 4
cm, Cymbella cleve-eulerae, Encyonema silesiacum, E. minutum, and Navicula vulpina become
more common, while Diatoma tenuis largely disappears and Nitzschia frustulum drops to ~30%
(Fig. 18). Seabird-derived nutrients (δ15N) are extremely high throughout this record, and
experience increases in the upper sediments (from ~8 – 18‰; Fig. 8). Likewise, %N and %C
increase from ~0.5% to ~3%, and from ~14-24%, respectively, in concert with the PCA1 scores
and δ15N. C/N shows some fluctuations, but generally decreases from ~24 to ~10 at the top of the
core (Fig. 8).
Pond CV20 (Figs. 9, 19)
The CV20 core had one of the fastest sediment accumulation rates encountered at Cape
Vera, with ~40 years of accumulation contained in the uppermost 6.25 cm (137Cs peak at 6.25 cm
= 1963 AD, Michelutti et al. submitted). There are few changes in the dominant diatom
assemblages through time as Nitzschia frustulum remains relatively stable at ~80% relative
abundance and both Cymbella cleve-eulerae and Encyonema minutum are present at relatively
constant, yet low (<10%) abundances (Fig. 19). The muted changes that occur in the top of the
core include increased relative abundances of Cyclotella pseudostelligera and Nitzschia
perminuta, and decreased abundances of Cymbella silesiaca (Fig. 19). The δ15N remains
relatively constant throughout the core at ~7‰, while %N gradually increases from 3 to 4% at the
top (Fig. 9). The relative stability of %C (~40%) throughout the core results in an equally stable
C/N (~12; Fig. 9). Chla, while very high throughout the record, reaches its maximum
concentration of ~0.5 mg/g in the upper sediments (Fig. 9).
Cape Hawes cores (CV22 and CV24)
Pond CV22 (Figs. 10, 20)
In the CV22 core, 210Pb indicates an age of ~1950 AD at 5 cm (Michelutti et al.
submitted). This record is characterized by relatively high diversity (Hill’s N2 consistently >10;
amongst the highest in our study) and by large changes in diatom composition, as reflected in the
PCA1 scores (Figs. 10, 20). From the bottom to ~4 cm depth, the dominant taxa vary between
either combinations of Cocconeis spp. and Navicula vulpina or Nitzschia frustulum and Nitzschia
perminuta (Fig. 20). Above ~3 cm, N. frustulum and N. perminuta remain dominant, both
Achnanthidium minutissimum and Denticula tenuis reach their greatest relative abundances (up
to ~8%), but meanwhile Navicula vulpina decreases and Cocconeis spp. disappear (Fig. 20).
Seabird-derived nutrients, reflected by δ15N, decrease from ~9‰ in the bottom half of the core to
~1‰ in the top half of the core, a trend opposite to most of the Cape Vera cores (Fig. 10).
Percent N, %C, C/N, and chla, on the other hand, are very low and stable throughout most of the
core (~0%, 0%, ~8, and ~0 mg/g, respectively; Fig. 10). Inferred-chla exhibits a modest increase
in the uppermost sediments (Fig. 10).
Pond CV24 (Figs. 11, 21)
The 6.5 cm core from pond CV24 had an erratic 210Pb profile precluding the construction
of a reliable geochronology. The dominant diatom species in this core do not change through
time, and are continuously characterized by Nitzschia frustulum (60-70% relative abundance; Fig.
21). Seabird-influence was minimal to non-existent throughout this core, as evidenced by the
very low δ15N (~1‰; Fig. 11). At ~3.5% and ~36%, N and C were relatively high, leading to
C/N generally <10 (Fig. 11). Chla was low (~0.1 mg/g) and indicated a very small increase at the
top of the core (Fig. 11).
DISCUSSION
Geochemical evidence
The paleolimnological records of seabird influence in the ponds at Cape Vera and Cape
Hawes are complex. Concentrations of δ15N and chla support the hypothesis that seabirds have
influenced most of the ponds at Cape Vera (except CV12, the pond most distant from the cliffs,
see below), leading to high nutrient conditions and increased primary production, for the entire
history captured by the sediment cores (maximum of ~200 years). Seabirds generally choose to
nest and breed in cliffs that are inaccessible to predators (Mallory & Fontaine 2004), and thus it is
consistent with seabird behavioural traits that they would have colonized the cliffs at Cape Vera
even before the land beneath the cliffs isostatically rebounded from the ocean (i.e. not more than
~5000 years BP; Lowdon & Blake Jr. 1973; Blake Jr. 1975).
Ratios of C/N in the CV cores generally reflect the terrestrial origin of sedimentary
organic matter in the bottom of most cores (C/N>20), shifting towards values indicative of mostly
aquatic origin at the top of the cores (C/N<10; Meyers & Ishiwatari 1993). At Cape Hawes,
however, the C/N ratios are generally below 10 for most of the sedimentary record, suggesting
that they have never received significant amounts of terrestrial organic matter.
Inferred-chla and δ15N from the Cape Hawes sites confirm that few, if any, seabird-
derived subsidies currently reach these ponds, and that primary production is low. Indeed, these
two ponds located ~7 km from the nearest active seabird colony have limnological characteristics
typical of undisturbed High Arctic ponds. However, historically higher δ15N, sedimentary chla,
and chironomid abundance (Michelutti, unpublished data) prior to the 20th century in CV22
suggests that the cliffs at Cape Hawes may have been the location of a seabird colony in the past;
this would be consistent with CWS observations of jewel lichen, a commonly used indicator of
bird guano (Pielou 1994), still visible on the nearby cliff face. As CV24 is slightly further away
from the cliffs at Cape Hawes, the lack of an elevated δ15N earlier in its sedimentary record
suggests that it was not within range of the nutrient deposition from the seabird colony believed
to have existed at this site.
There is some evidence that hydrological factors may be responsible for the varying
responses in some of the cores within the Cape Vera sites. For example, CV9 is one of the
closest ponds to the bird cliffs and displays amongst the highest δ15N and chla concentrations,
which attain their greatest values in the most recent sediments (Fig. 5). In CV9, the main pattern
of diatom species variation and diatom diversity (Hill’s N2) follow the trends recorded by the
δ15N and chla. In contrast, CV12, which is the pond located furthest from the bird colony, has the
lowest δ15N (Fig. 7). The decrease of bird-derived nutrients in the most recent sediment of CV12,
concomitant with the increases recorded in CV9, imply that changes in drainage patterns have
certainly occurred over time in this region. However, the majority of ponds sampled show
similar trends over time with respect to bird-derived nutrients.
Diatom species composition
Unlike the δ15N and inferred-chla data, diatom-based paleolimnological data are less
easily interpretable with respect to impacts from marine-derived nutrients. Although the taxa
found in the sediment cores were generally encountered from the surface sediment calibration of
24 ponds near Cape Vera (Keatley et al. [7]), suggesting that the historical record covers similar
limnological conditions to those captured by our modern pond survey, shifts in assemblages, as
summarized by PCA axis 1 sample scores, show a wide variety of patterns that do not appear to
be related to proximity to the northern fulmar colony or to the trends recorded in δ15N and chla
(Figs. 3-11). Species assemblage changes do not show consistent patterns through time, whether
they are examined as shifts in species composition, the main directions of variation in all taxa
(PCA axis 1 scores), or by a diversity index (i.e. Hill’s N2). Likewise, the most common species,
small Nitzschia taxa, are alternately most dominant in the recent sediments of some ponds (e.g.
CV6), the oldest sediments of others (e.g. CV9), or may show little change throughout the
sediment cores (e.g. CV24). Moreover, Nitzschia frustulum attains extremely high relative
abundance (>50%) even in sites with very different modern bird influences. While the patterns of
diatom species and diversity changes are correlated to both δ15N and chla concentrations in some
ponds (i.e. CV9, CV2), when the diatom changes in all ten ponds are compared, the complex
responses suggest that species assemblage changes alone cannot provide robust historical
estimations of seabird-derived nutrients at Cape Vera.
Given that the Cape Vera ponds are clearly productive, are known to contain seabird
derived nutrients, and that our previous work calibrating diatom surface sediments to measured
ecological variables indicated a significant relationship between diatom species assemblage
variance and δ15N (Keatley et al. [7]), the lack of a diatom assemblage signal that agrees with the
other proxies of seabird influence was initially surprising. However, it is likely that a number of
factors, working in concert, may explain our results.
Previous studies of diatom-inferred eutrophication in Arctic lakes and ponds (Douglas &
Smol 2000; Michelutti et al. 2002; Douglas et al. 2004; Michelutti et al. in press) indicate that
diatoms respond quite differently to nutrient enrichment than those in temperate (Hall & Smol
1999), or even in subarctic regions (e.g. Gregory-Eaves et al. 2003). For example, diatom
assemblage changes in response to human sewage dumped directly into Arctic lakes were
relatively minor, compared to responses recorded in similarly-impacted temperate lakes. This
relatively muted diatom response in previous studies of nutrient-enriched Arctic lakes has been
attributed to the over-riding influence of prolonged ice cover and short growing seasons (Douglas
& Smol 2000; Michelutti et al. 2007). However, at Cape Vera, the cool conditions that
characterize even the summer months have not limited chlorophytes and some cyanobacteria
from achieving very high standing stocks in response to the elevated nutrient inputs from
seabirds. Therefore, it is possible that at Cape Vera, the diatoms are being outcompeted by other
taxa.
In addition, the diatom assemblages in the Cape Vera ponds are almost all dominated by
small Nitzschia frustulum and Nitzschia perminuta taxa. In the Canadian High Arctic, these taxa
are amongst the most common and abundant species found, and are thought to be environmental
generalists (Antoniades et al. in press). For example, taxa such as N. perminuta and N. frustulum
have been found to be the dominant taxa in both highly eutrophic sites (Michelutti et al. 2003)
and in ultraoligotrophic sites (Antoniades et al. in press). In our surface sediment calibration, we
found that δ15N could significantly explain diatom species distributions, yet the relationship
between δ15N and N. frustulum, one of the most dominant taxa, was weak (Keatley et al. [7]).
Regardless of the specific cause, the inability of diatom species assemblages to
consistently track changes in δ15N or chla at Cape Vera suggest that diatom species assemblages
should not be used in isolation to track changes in trophic status over time at Cape Vera. Given
that diatom identification and enumeration are often the most time consuming steps in
paleolimnological studies, their usefulness as paleo-proxies of seabird influence is limited.
Provided that a high degree of chronological control can be achieved, diatom concentration data,
biogenic silica, or some measure of diatom production would likely be a better indicator of
trophic status than diatom species changes. Our findings contrast sharply with those of temperate
regions, in which relatively small changes in nutrient input often result in dramatic shifts in
diatom assemblage composition (Hall & Smol 1999).
Reconstructing seabird populations
Sediment cores from Cape Vera and Cape Hawes reveal highly variable patterns of
diatom species assemblage change, despite relatively coherent trends in δ15N and chla
concentrations. At Cape Vera, the ponds most affected by seabird-derived nutrients (inferred by
δ15N) generally exhibit records of elevated δ15N throughout their entire histories, suggesting that
the northern fulmar colony has been a continuous source of marine-derived nutrients into these
ponds. In comparison, ponds at Cape Hawes have much lower δ15N and chla concentrations in
the recent sediments, consistent with the observed lack of seabird activity. However, one of the
Cape Hawes ponds records historically higher δ15N that corroborates observational evidence that
seabirds abandoned a colony at this site prior to the 20th century.
Recent increases in both δ15N and chla concentrations, as well as decreases in C/N
suggest that aquatic primary production has increased in recent years at Cape Vera, and that this
is likely due to an increase in seabird populations. However, increased primary production may
also be partially attributable to climate warming, as small increases in chla and decreases in C/N
are also apparent in Cape Hawes, despite the lack of a corresponding increase in δ15N. Indeed,
using the same techniques, Michelutti et al. (2005) have attributed increased chla in the recent
sediments of lakes from Baffin Island to climate-related increases in primary production.
Changes in climate, and especially sea ice conditions, are known to have significant repercussions
on some seabird population dynamics (e.g. Gaston et al. 2005), although the influence of
changing ice conditions on northern fulmars in the Canadian High Arctic is unclear. Nevertheless,
it is probable that climate plays a role in determining the success of northern fulmars at Cape
Vera, and that both shifts in seabird population dynamics and climate, acting both alone and in
concert, have resulted in increased production at Cape Vera.
Conversely, diatom-based paleolimnological records were complex and displayed
varying patterns through time in different ponds, several of which were inconsistent with both
δ15N and inferred-chla. At Cape Vera, the lack of diatom species assemblage response to nutrient
enrichment is related to the overwhelming dominance of a few taxa that do not show strong
relationships to δ15N. The diatom-based results contrast those from studies of eutrophication in
temperate regions, and suggest that diatom species in High Arctic freshwaters respond in a highly
complex manner to nutrient enrichment. This relationship may be related to the overwhelming
importance of the phytobenthos relative to the phytoplankton as primary producers in Arctic
regions.
ACKNOWLEDGEMENTS
We would like to thank NSERC, PCSP, NSTP, and CWS for funding and logistical support, and
Irene Gregory-Eaves, Alain Fontaine, Norm North, Mark Wayland, Linda Kimpe, Karen Foster,
Huijun Liu, Jason Akearuk, and Miles Falconer for assistance in the field.
REFERENCES
ACIA, 2004. Impacts of a warming Arctic. Cambridge University Press, Cambridge, UK.
Antoniades, D., P. B. Hamilton, M. S. V. Douglas & J. P. Smol, in press. Freshwater diatoms from the Canadian High Arctic. Iconographica Diatomologica.
Appleby, P. G., 2001. Chronostratigraphic techniques in recent sediments. In W. M. Last & J. P. Smol (eds), Tracking Environmental Change Using Lake Sediments Volume 1: Basin Analysis, Coring, and Chronological Techniques. Kluwer Academic Publishers, Dordrecht: 171-203.
Blais, J. M., L. E. Kimpe, D. McMahon, B. E. Keatley, M. L. Mallory, M. S. V. Douglas, & J. P. Smol, 2005. Arctic seabirds transport marine-derived contaminants. Science 309: 445.
Blake Jr, W., 1975. Radiocarbon age determinations and postglacial emergence at Cape Storm, southern Ellesmere Island, Arctic Canada. Geografiska Annaler Series A-Physical Geography 57: 1-71.
Brooke, M. D., 2004. The food consumption of the world's seabirds. Proceedings of the Royal Society of London Series B-Biological Sciences 271: S246-S248.
Das, B., R. D. Vinebrooke, A. Sanchez-Azofeifa, B. Rivard, & A. P. Wolfe, 2005. Inferring sedimentary chlorophyll concentrations with reflectance spectroscopy: a novel approach to reconstructing historical changes in the trophic status of mountain lakes. Canadian Journal of Fisheries and Aquatic Sciences 62: 1067-1078.
Douglas, M. S. V., & J. P. Smol, 2000. Eutrophication and recovery in the High Arctic: Meretta lake (Cornwallis Island, Nunavut, Canada) revisited. Hydrobiologia 431: 193-204.
Douglas, M. S. V., J. P. Smol, J. M. Savelle, & J. M. Blais, 2004. Prehistoric Inuit whalers affected arctic freshwater ecosystems. Proceedings of the National Academy of Sciences of the United States of America 101: 1613-1617.
Evenset, A., J. Carroll, G. N. Christensen, R. Kallenborn, D. Gregor, & G. W. Gabrielsen, 2007. Seabird guano is an efficient conveyer of persistent organic pollutants (POPs) to Arctic lake ecosystems. Environmental Science & Technology 41: 1173-1179.
Evenset, A., G. N. Christensen, T. Skotvold, E. Fjeld, M. Schlabach, E. Wartena, & D. Gregor, 2004. A comparison of organic contaminants in two high Arctic lake ecosystems, Bjørnøya (Bear Island), Norway. The Science of the Total Environment 318: 125-141.
Gaston, A. J., & M. Hipfner, 1998. The effect of ice conditions in northern Hudson Bay on breeding by Thick-billed Murres (Uria lomvia). Canadian Journal of Zoology 76: 480-492.
Gaston, A. J., H. G. Gilchrist, & J. M. Hipfner, 2005a. Climate change, ice conditions and reproduction in an Arctic nesting marine bird: Brunnich's guillemot (Uria lomvia L.). Journal of Animal Ecology 74: 832-841.
Gaston, A. J., H. G. Gilchrist, & M. L. Mallory, 2005b. Variation in ice conditions has strong effects on the breeding of marine birds at Prince Leopold Island, Nunavut. Ecography 28: 331-344.
Gaston, A. J., M. L. Mallory, H. G. Gilchrist, & K. O'Donovan, 2006. Status, trends and attendance patterns of the northern fulmar Fulmarus glacialis in Nunavut, Canada. Arctic 59: 165-178.
Gilchrist, G., M. Mallory, & F. Merkel, 2005. Can local ecological knowledge contribute to wildlife management? Case studies of migratory birds. Ecology and Society 10.
Glew, J. R., 1988. A portable extruding device for close interval sectioning of unconsolidated core samples. Journal of Paleolimnology 2: 241-243.
Glew, J. R., J. P. Smol, & W. M. Last, 2001. Sediment core collection and extrusion. In W. M. Last & J. P. Smol (eds), Basin Analysis, Coring, and Chronological Techniques. Kluwer Academic Publisher, Norwell: 73-106.
Gregory-Eaves, I., J. P. Smol, M. S. V. Douglas, & B. P. Finney, 2003. Diatoms and sockeye salmon (Oncorhynchus nerka) population dynamics: Reconstructions of salmon-derived nutrients over the past 2,200 years in two lakes from Kodiak Island, Alaska. Journal of Paleolimnology 30: 35-53.
Grosbois, V., & P. M. Thompson, 2005. North Atlantic climate variation influences survival in adult fulmars. Oikos 109: 273-290.
Hall, R. I., & J. P. Smol, 1999. Diatoms as indicators of lake eutrophication. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, New York: 128-168.
Juggins, S., 2003. C2: software for ecological and palaeocological data analysis and visualisation, version 1.3. Department of Geography, University of Newcastle.
Keatley, B. E., M. S. V. Douglas, J. M. Blais, M. Mallory, & J. P. Smol. [7]. Impacts of seabird-derived nutrients on water quality and diatom species assemblages from Cape Vera, Devon Island, Canadian High Arctic. PhD thesis chapter; in preparation.
Krammer, K., 2002. Diatoms of Europe: Cymbella. A.R.G. Gantner Verlag, Ruggell.
Krammer, K., & H. Lange-Bertalot, 1991. Bacillariophyceae. In H. Ettl, J. Gerloff, H. Heynig & D. Mollenhauer (eds). Gustav Fischer Verlag, Stuttgart.
Lepš, J., & P. Šmilauer, 2003. Multivariate analysis of ecological data using CANOCO. Cambridge University Press, London.
Lowdon, J. A.& W. Blake Jr, 1973. Geological Survey of Canada Radiocarbon Dates XIII. Ottawa, Geological Survey of Canada, Ottawa.
Mallory, M. L., 2006. The Northern Fulmar (Fulmarus glacialis) in Arctic Canada: ecology, threats, and what it tells us about marine environmental conditions. Environmental Reviews 14: 187-216.
Mallory, M. L., & A. J. Fontaine, 2004. Key marine habitat sites for migratory birds in Nunavut and the Northwest Territories. Canadian Wildlife Service, Ottawa.
Mayr, U., T. de Freitas, G.H. Eisbacher, & J.Wm. Kerr, 1998. Geology, Prince Alfred Bay-Cardigan Strait and parts of adjacent sheets, Northwest Territories; Geological Survey of Canada, Map 1924A, scale 1:250 000.
Meyers, P. A., & R. Ishiwatari, 1993. Lacustrine Organic Geochemistry - An Overview of Indicators of Organic-Matter Sources and Diagenesis in Lake-Sediments. Organic Geochemistry 20: 867-900.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2002. Tracking recovery in a eutrophied high Arctic lake (Meretta Lake, Cornwallis Island, Canadian Arctic) using periphytic diatoms. Verh. int. Ver. Limnol. 28.
Michelutti, N., M. S. V. Douglas, & J. P. Smol, 2003. Tracking recovery in a eutrophied high Arctic lake (Meretta Lake, Cornwallis Island, Canadian Arctic) using periphytic diatoms. Verh. int. Ver. Limnol. 28: 1533-1537.
Michelutti, N., M. H. Hermanson, J. P. Smol, P. J. Dillon, & M. S. V. Douglas, in press. Delayed response of diatom assemblage changes to sewage inputs in an Arctic lake. Aquatic Sciences.
Michelutti, N., A. P. Wolfe, R. D. Vinebrooke, B. Rivard, & J. P. Briner, 2005. Recent primary production increases in arctic lakes. Geophysical Research Letters 32: L19715-doi: 10.1029/2005GL023693.
Ng, S. L., & R. H. King, 1999. Development of a diatom-based specific conductivity model for the glacio-isostatic lakes of Truelove Lowland: implications for paleoconductivity and paleoenvironmental reconstructions in Devon Island lakes, NWT, Canada. Journal of Paleolimnology 22: 367-382.
Parr, J. F., K. H. Taffs, & C. M. Lane, 2004. A microwave digestion technique for the extraction of fossil diatoms from coastal lake and swamp sediments. Journal of Paleolimnology 31: 383-390.
Pielou, E. C., 1994. A Naturalist’s Guide to the Arctic. University of Chicago Press, Chicago.
Polis, G. A., F. Sánchez-Piñero, P. T. Stapp, W. B. Anderson & M. D. Rose, 2004. Trophic Flows from Water to Land: Marine Input Affects Food Webs of Islands and Coastal Ecosystems Worldwide. In G. A. Polis, M. E. Power & G. R. Huxel (eds), Food Webs at the Landscape Level. University of Chicago Press, Chicago: 200-216.
ter Braak, C. J. F., & P. Šmilauer, 2002. CANOCO reference manual and CANOdraw for Windows user guide: Software for Canonical Community Ordination (Version 4.5).
Wolfe, A. P., R. D. Vinebrooke, N. Michelutti, B. Rivard, & B. Das, 2006. Experimental calibration of lake-sediment spectral reflectance to chlorophyll a concentrations: methodology and paleolimnological validation. Journal of Paleolimnology 36: 91-100.
Figure captions.
Figure 1. Location of study site. A) Regional map of the Canadian Arctic with inset detailing location of the Canadian Arctic within Canada; B) Enlargement of Devon Island region with star identifying the location of Cape Vera; and C) Location of study ponds at Cape Vera, Devon Island. Figure 2. Photograph of a suite of ponds below the densely occupied cliffs at Cape Vera, Devon Island. Figures 3 – 11. Plots for each core summarizing overall changes in diatom assemblage composition based on Principal Components Analysis Axis 1 (PCA1) and PCA axis 2 (PCA2) sample scores, changes in diatom assemblage diversity based on Hill’s N2 (N2; Hill 1973), seabird-derived nutrients based on (δ15N (‰), sedimentary-inferred chlorophyll a concentrations (Chla mg/g dry weight), %N, %C, and C/N ratios. Figure 3. CV5 summary plot. Figure 4. CV6 summary plot. Figure 5. CV9 summary plot. Figure 6. CV9a summary plot. Figure 7. CV12 summary plot. Figure 8. CV13 summary plot. Figure 9. CV20 summary plot. Figure 10. CV22 summary plot. Figure 11. CV24 summary plot. Figures 12 – 21. Diatom stratigraphic plots for each core from the Cape Vera ponds. These plots include all species present in at least 3% relative abundance in at least one sample. Figure 12. CV5 diatom stratigraphy. Figure 13. CV6 diatom stratigraphy. Figure 14. CV7 diatom stratigraphy. Figure 15. CV9 diatom stratigraphy. Figure 16. CV9a diatom stratigraphy. Figure 17. CV12 diatom stratigraphy. Figure 18. CV13 diatom stratigraphy. Figure 19. CV20 diatom stratigraphy. Figure 20. CV22 diatom stratigraphy. Figure 21. CV24 diatom stratigraphy.
A
BC
A
BC
Fi
gure
1.
Figure 2.
0.25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla(m
g/g dw
)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
0.25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla(m
g/g dw
)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
0.25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla(m
g/g dw
)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
Fi
gure
3.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
3.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0%N
020
%C
010
2030
C/N
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
3.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0%N
020
%C
010
2030
C/N
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
3.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0%N
020
%C
010
2030
C/N
010
2030
C/N Fi
gure
4.
0.25
2.25
4.25
6.25
8.25
10.2
5
12.2
5
14.2
5
16.2
5
18.2
5
20.2
5
21.7
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0%N
020%C 0
816
2432
40
C/N0.
25
2.25
4.25
6.25
8.25
10.2
5
12.2
5
14.2
5
16.2
5
18.2
5
20.2
5
21.7
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0%N
020%C 0
816
2432
40
C/N0.
25
2.25
4.25
6.25
8.25
10.2
5
12.2
5
14.2
5
16.2
5
18.2
5
20.2
5
21.7
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0%N
020%C 0
816
2432
40
C/N
0%N
020%C 0
816
2432
40
C/N
Fi
gure
5.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
Depth (cm)
-1.0
0.0
1.0
2.0
3.0
4.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
3.0
PCA2
05
1015
20
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0
%N 020
40
%C
03
69
1215
C/N
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
Depth (cm)
-1.0
0.0
1.0
2.0
3.0
4.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
3.0
PCA2
05
1015
20
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0
%N 020
40
%C
03
69
1215
C/N
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
Depth (cm)
-1.0
0.0
1.0
2.0
3.0
4.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
3.0
PCA2
05
1015
20
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0
%N 020
40
%C
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0
%N 020
40
%C
03
69
1215
C/N
Fi
gure
6.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0
%N
020%C
080
160
240
320
C/N0.
25
1.25
2.25
3.25
4.25
5.25
6.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0
%N
020%C
080
160
240
320
C/N
Fi
gure
7.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-3.0
-1.0
1.0
3.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0
%N
020%C
05
1015
2025
C/N0.
25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-3.0
-1.0
1.0
3.0
PCA2
04
812
1620
N20
48
1216
20
δ15N (‰
)
0
%N
020%C
05
1015
2025
C/N
Fi
gure
8.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
ChlaC2
(mg/g
dw)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
ChlaC2
(mg/g
dw)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
Figu
re 9
.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
ChlaC2
(mg/g
dw)
0
%N
0
%C
0.0
1.0
2.0
3.0
4.0
5.0
C/N
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
ChlaC2
(mg/g
dw)
0
%N
0
%C
0.0
1.0
2.0
3.0
4.0
5.0
C/N
Figu
re 1
0.
0.25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
0.25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
-2.0
-1.0
0.0
1.0
2.0
PCA1
-2.0
-1.0
0.0
1.0
2.0
PCA2
04
812
1620
N2
04
812
1620
δ15N (‰
)
0.0
0.2
0.4
0.6
Chla (mg/g
dw)
0
%N
020
40
%C
0.0
2.4
4.8
7.2
9.6
12.0
C/N
Figu
re 1
1.
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
9.5
Depth (cm)
0Cymbe
llabo
tellus
0Psammoth
idium
chlid
anos
020
Encyo
nemaminu
tum
0Navicu
lach
iarae
0Encyo
nemasil
esiac
um0
20
Rel
ativ
e ab
unda
nce
(%)
Encyo
nemafog
edii
020
40Nitz
schia
perm
inuta
020
40
Nitzsc
hiafru
stulum
0Navicu
laph
yllepta
0Microc
ostat
uskra
sske
i
0Navicu
lasa
linaru
m
0Navicu
lasp
affv
eneta
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
9.5
Depth (cm)
0Cymbe
llabo
tellus
0Psammoth
idium
chlid
anos
020
Encyo
nemaminu
tum
0Navicu
lach
iarae
0Encyo
nemasil
esiac
um0
20
Rel
ativ
e ab
unda
nce
(%)
Encyo
nemafog
edii
020
40Nitz
schia
perm
inuta
020
40
Nitzsc
hiafru
stulum
0Navicu
laph
yllepta
0Microc
ostat
uskra
sske
i0.
5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
9.5
Depth (cm)
0Cymbe
llabo
tellus
0Psammoth
idium
chlid
anos
020
Encyo
nemaminu
tum
0Navicu
lach
iarae
0Encyo
nemasil
esiac
um0
20
Rel
ativ
e ab
unda
nce
(%)
Encyo
nemafog
edii
020
40Nitz
schia
perm
inuta
020
40
Nitzsc
hiafru
stulum
0Navicu
laph
yllepta
0Microc
ostat
uskra
sske
i
0Navicu
lasa
linaru
m
0Navicu
lasp
affv
eneta
Fi
gure
12.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
Depth (cm)
0Stauros
iraco
nstru
ensva
rvente
r
0Cyclot
ellaps
eudo
stellig
era
020
Cymbe
llabo
tellus 0Ach
nanth
idium
minutis
simum
0Cymbe
llacle
ve-eu
lerae
0
Rel
ativ
e ab
unda
nce
(%)
Encyo
nops
ismicr
ocep
hala
020
Navicu
lavu
lpina
020
Nitzsc
hiape
rminu
ta
020
4060
80
Nitzsc
hiafru
stulum
0Navicu
lach
iarae
0Encyo
nemafog
edii
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
Depth (cm)
0Stauros
iraco
nstru
ensva
rvente
r
0Cyclot
ellaps
eudo
stellig
era
020
Cymbe
llabo
tellus 0Ach
nanth
idium
minutis
simum
0Cymbe
llacle
ve-eu
lerae
0
Rel
ativ
e ab
unda
nce
(%)
Encyo
nops
ismicr
ocep
hala
020
Navicu
lavu
lpina
020
Nitzsc
hiape
rminu
ta
020
4060
80
Nitzsc
hiafru
stulum
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
Depth (cm)
0Stauros
iraco
nstru
ensva
rvente
r
0Cyclot
ellaps
eudo
stellig
era
020
Cymbe
llabo
tellus 0Ach
nanth
idium
minutis
simum
0Cymbe
llacle
ve-eu
lerae
0
Rel
ativ
e ab
unda
nce
(%)
Encyo
nops
ismicr
ocep
hala
020
Navicu
lavu
lpina
020
Nitzsc
hiape
rminu
ta
020
4060
80
Nitzsc
hiafru
stulum
0Navicu
lach
iarae
0Encyo
nemafog
edii
Fi
gure
13.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
14.2
5
15.2
5
16.2
5
17.2
5
18.2
5
19.2
5
Depth (cm)
020
Cocco
neis
sp 0Uniden
tified
fragm
ent
0Calone
issili
cula
0Navicu
lavu
lpina
020
Cymbe
llabo
tellus
020
40
Nitzsc
hiape
rminu
ta 020
Rel
ativ
e ab
unda
nce
(%)
Navicu
lach
iarae
0Navicu
laph
yllepta
020
Nitzsc
hiafru
stulum 0Cym
bella
cleve
-euler
ae
0Encyo
nemafog
edii
0Navicu
lasp
affv
eneta
0Navicu
laps
eudo
tenell
oides
0Encyo
nemaminu
tum
0Fragila
riaca
pucin
a
0Achna
nthidi
ummarg
inulat
um
0Microc
ostat
uskra
sske
i
0Navicu
lasp
.
(cf.N
. cryp
tocep
hala)
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
14.2
5
15.2
5
16.2
5
17.2
5
18.2
5
19.2
5
Depth (cm)
020
Cocco
neis
sp 0Uniden
tified
fragm
ent
0Calone
issili
cula
0Navicu
lavu
lpina
020
Cymbe
llabo
tellus
020
40
Nitzsc
hiape
rminu
ta 020
Rel
ativ
e ab
unda
nce
(%)
Navicu
lach
iarae
0Navicu
laph
yllepta
020
Nitzsc
hiafru
stulum
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
14.2
5
15.2
5
16.2
5
17.2
5
18.2
5
19.2
5
Depth (cm)
020
Cocco
neis
sp 0Uniden
tified
fragm
ent
0Calone
issili
cula
0Navicu
lavu
lpina
020
Cymbe
llabo
tellus
020
40
Nitzsc
hiape
rminu
ta 020
Rel
ativ
e ab
unda
nce
(%)
Navicu
lach
iarae
0Navicu
laph
yllepta
020
Nitzsc
hiafru
stulum 0Cym
bella
cleve
-euler
ae
0Encyo
nemafog
edii
0Navicu
lasp
affv
eneta
0Navicu
laps
eudo
tenell
oides
0Encyo
nemaminu
tum
0Fragila
riaca
pucin
a
0Achna
nthidi
ummarg
inulat
um
0Microc
ostat
uskra
sske
i
0Navicu
lasp
.
(cf.N
. cryp
tocep
hala)
Fi
gure
14.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
511
.25
12.2
513
.25
14.2
515
.25
16.2
517
.25
18.2
519
.25
20.2
521
.25
21.7
5
Depth (cm)
0Cymbe
llacle
ve-eu
lerae
020
4060
80
Nitzsc
hiafru
stulum
020
Navicu
lavu
lpina 0
Rel
ativ
e ab
unda
nce
(%)
Encyo
nops
ismicr
ocep
hala
020
Nitzsc
hiape
rminu
ta
0Encyo
nemasil
esiac
um
020
Encyo
nemafog
edii
020
Encyo
nemaminu
tum
020
Navicu
laph
yllepta
0Navicu
lacry
ptoce
phala
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
511
.25
12.2
513
.25
14.2
515
.25
16.2
517
.25
18.2
519
.25
20.2
521
.25
21.7
5
Depth (cm)
0Cymbe
llacle
ve-eu
lerae
020
4060
80
Nitzsc
hiafru
stulum
020
Navicu
lavu
lpina 0
Rel
ativ
e ab
unda
nce
(%)
Encyo
nops
ismicr
ocep
hala
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
511
.25
12.2
513
.25
14.2
515
.25
16.2
517
.25
18.2
519
.25
20.2
521
.25
21.7
5
Depth (cm)
0Cymbe
llacle
ve-eu
lerae
020
4060
80
Nitzsc
hiafru
stulum
020
Navicu
lavu
lpina 0
Rel
ativ
e ab
unda
nce
(%)
Encyo
nops
ismicr
ocep
hala
020
Nitzsc
hiape
rminu
ta
0Encyo
nemasil
esiac
um
020
Encyo
nemafog
edii
020
Encyo
nemaminu
tum
020
Navicu
laph
yllepta
0Navicu
lacry
ptoce
phala
Fi
gure
15.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
Depth (cm)
0Fragila
riaca
pucin
avargra
cilis
020
Stauros
irella
pinna
ta
020
Cyclot
ellaps
eudo
stellig
era0Cym
bella
cleve
-euler
ae0
20
Rel
ativ
e ab
unda
nce
(%)
Stauros
iraco
nstru
ensva
rvente
r
020
Encyo
nemaminu
tum
020
4060
80
Nitzsc
hiafru
stulum
020
4060
Nitzsc
hiape
rminu
ta0.
25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
Depth (cm)
0Fragila
riaca
pucin
avargra
cilis
020
Stauros
irella
pinna
ta
020
Cyclot
ellaps
eudo
stellig
era0Cym
bella
cleve
-euler
ae0
20
Rel
ativ
e ab
unda
nce
(%)
Stauros
iraco
nstru
ensva
rvente
r
020
Encyo
nemaminu
tum
020
4060
80
Nitzsc
hiafru
stulum
020
4060
Nitzsc
hiape
rminu
ta0.
25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
Depth (cm)
0Fragila
riaca
pucin
avargra
cilis
020
Stauros
irella
pinna
ta
020
Cyclot
ellaps
eudo
stellig
era0Cym
bella
cleve
-euler
ae0
20
Rel
ativ
e ab
unda
nce
(%)
Stauros
iraco
nstru
ensva
rvente
r
020
Encyo
nemaminu
tum
020
4060
80
Nitzsc
hiafru
stulum
020
4060
Nitzsc
hiape
rminu
ta
Fi
gure
16.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
Depth (cm)
020Nav
icula
sp. (l
arge c
entre
sonly)
0Cocco
neis
sp
020
4060
Neidium
sp. (l
arge c
entre
sonly)
0Navicu
lasp
.
(cfN. rh
ynch
ocep
hala
f. eleg
ans)
0
Rel
ativ
e ab
unda
nce
(%)
Diatom
atenuis
020Nav
icula
phylle
pta
020Den
ticula
kuetz
ingii
020
Nitzsc
hiape
rminu
ta
0Nitzsc
hiaco
mmuntat
a
020
4060
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
0.25
1.25
2.25
3.25
4.25
5.25
6.25
Depth (cm)
020Nav
icula
sp. (l
arge c
entre
sonly)
0Cocco
neis
sp
020
4060
Neidium
sp. (l
arge c
entre
sonly)
0Navicu
lasp
.
(cfN. rh
ynch
ocep
hala
f. eleg
ans)
0
Rel
ativ
e ab
unda
nce
(%)
Diatom
atenuis
020Nav
icula
phylle
pta
020Den
ticula
kuetz
ingii
020
Nitzsc
hiape
rminu
ta
0Nitzsc
hiaco
mmuntat
a
020
4060
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
0.25
1.25
2.25
3.25
4.25
5.25
6.25
Depth (cm)
020Nav
icula
sp. (l
arge c
entre
sonly)
0Cocco
neis
sp
020
4060
Neidium
sp. (l
arge c
entre
sonly)
0Navicu
lasp
.
(cfN. rh
ynch
ocep
hala
f. eleg
ans)
0
Rel
ativ
e ab
unda
nce
(%)
Diatom
atenuis
020Nav
icula
phylle
pta
020Den
ticula
kuetz
ingii
020
Nitzsc
hiape
rminu
ta
0Nitzsc
hiaco
mmuntat
a
020
4060
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
Fi
gure
17.
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
Depth (cm)
020
Diatom
atenuis 0
2040
60
Nitzsc
hiafru
stulum
020
Rel
ativ
e ab
unda
nce
(%)
Encyo
nemafog
edii
020
Nitzsc
hiape
rminu
ta
0Encyo
nemasil
esiac
um
020
Encyo
nemaminu
tum
020
Cymbe
llacle
ve-eu
lerae
0Navicu
lavu
lpina
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
Depth (cm)
020
Diatom
atenuis 0
2040
60
Nitzsc
hiafru
stulum
020
Rel
ativ
e ab
unda
nce
(%)
Encyo
nemafog
edii
020
Nitzsc
hiape
rminu
ta
0Encyo
nemasil
esiac
um
020
Encyo
nemaminu
tum
020
Cymbe
llacle
ve-eu
lerae
0.25
1.25
2.25
3.25
4.25
5.25
6.25
7.25
8.25
9.25
10.2
5
11.2
5
12.2
5
13.2
5
Depth (cm)
020
Diatom
atenuis 0
2040
60
Nitzsc
hiafru
stulum
020
Rel
ativ
e ab
unda
nce
(%)
Encyo
nemafog
edii
020
Nitzsc
hiape
rminu
ta
0Encyo
nemasil
esiac
um
020
Encyo
nemaminu
tum
020
Cymbe
llacle
ve-eu
lerae
0Navicu
lavu
lpina
Fi
gure
18.
0.25
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.25
6.75
7.25
7.75
Depth (cm)
0Encyo
nemasile
siacu
m0
2040
6080
Rel
ativ
e ab
unda
nce
(%)
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
020
Encyo
nemaminu
tum
020
Nitzsc
hiape
rminu
ta
0Encyo
nemafog
edii
0Cyclot
ellaps
eudo
stellig
era
0.25
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.25
6.75
7.25
7.75
Depth (cm)
0Encyo
nemasile
siacu
m0
2040
6080
Rel
ativ
e ab
unda
nce
(%)
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
020
Encyo
nemaminu
tum
020
Nitzsc
hiape
rminu
ta
0Encyo
nemafog
edii
0Cyclot
ellaps
eudo
stellig
era
Fi
gure
19.
0.25
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.25
6.75
7.25
7.75
Depth (cm)
0Cocco
neis
sp. s
mall
020
Cocco
neis
sp. la
rge
020
40
Navicu
lavu
lpina
0Navicu
ladig
itorad
iata
0Nitzsc
hiadis
sipata
varmed
ia
0
Rel
ativ
e ab
unda
nce
(%)
Cymbe
llacle
ve-eu
lerae
0Cymbe
llabo
tellus
0Achna
nthidi
umminu
tissim
um
0Dentic
ulaten
uis
020
40
Nitzsc
hiape
rminu
ta
020
40
Nitzsc
hiafru
stulum
0Rossit
hidium
peter
senii
0.25
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.25
6.75
7.25
7.75
Depth (cm)
0Cocco
neis
sp. s
mall
020
Cocco
neis
sp. la
rge
020
40
Navicu
lavu
lpina
0Navicu
ladig
itorad
iata
0Nitzsc
hiadis
sipata
varmed
ia
0
Rel
ativ
e ab
unda
nce
(%)
Cymbe
llacle
ve-eu
lerae
0Cymbe
llabo
tellus
0Achna
nthidi
umminu
tissim
um
0Dentic
ulaten
uis
020
40
Nitzsc
hiape
rminu
ta
020
40
Nitzsc
hiafru
stulum
0.25
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.25
6.75
7.25
7.75
Depth (cm)
0Cocco
neis
sp. s
mall
020
Cocco
neis
sp. la
rge
020
40
Navicu
lavu
lpina
0Navicu
ladig
itorad
iata
0Nitzsc
hiadis
sipata
varmed
ia
0
Rel
ativ
e ab
unda
nce
(%)
Cymbe
llacle
ve-eu
lerae
0Cymbe
llabo
tellus
0Achna
nthidi
umminu
tissim
um
0Dentic
ulaten
uis
020
40
Nitzsc
hiape
rminu
ta
020
40
Nitzsc
hiafru
stulum
0Rossit
hidium
peter
senii
Fi
gure
20.
0.25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
0Navicu
lavu
lpina
0Achna
nthidi
umminu
tissim
um0Enc
yono
psis
microc
ephe
la0Cym
bella
botel
lus0
2040
6080
Rel
ativ
e ab
unda
nce
(%)
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
020
Nitzsc
hiape
rminu
ta
0Dentic
ulaku
etzing
ii
0Cymbo
pleura
angu
stata
varsp
itzbe
rgens
is0.
25
1.25
2.25
3.25
4.25
5.25
Depth (cm)
0Navicu
lavu
lpina
0Achna
nthidi
umminu
tissim
um0Enc
yono
psis
microc
ephe
la0Cym
bella
botel
lus0
2040
6080
Rel
ativ
e ab
unda
nce
(%)
Nitzsc
hiafru
stulum
0Cymbe
llacle
ve-eu
lerae
020
Nitzsc
hiape
rminu
ta
0Dentic
ulaku
etzing
ii
0Cymbo
pleura
angu
stata
varsp
itzbe
rgens
is
Fi
gure
21.
Tabl
e 1.
Sum
mar
y of
lim
nolo
gica
l dat
a fr
om e
ach
pond
cor
ed a
t Cap
e V
era,
Dev
on Is
land
.
Site
Sa
mpl
e co
de
δ15N
%
C
pH
Con
d C
HL
a Si
O2
DO
C
DIC
PO
C
PON
T
dN
TPu
T
Pf
N:P
N
or
P
Dat
e
(‰)
%
(μ
S/cm
) µg
/L
mg/
L
mg/
L
mg/
L
mg/
L
mg/
L
mg/
L
mg/
L
mg/
L
mol
ar
limite
d
CV
-5
02/0
7/20
05
5-05
9.
36
10.6
4 6.
95
116
0.9
0.25
0.
9 14
.6
0.34
2 0.
013
0.11
4 0.
0094
0.
0057
29
.0
eith
er
CV
-6
03/0
7/20
05
6-05
9.
13
15.4
4 7.
5 14
6.5
0.5
0.31
1.
6 19
.5
0.28
3 0.
017
0.17
6 0.
0056
0.
0041
66
.2
p C
V-7
09
/07/
2005
7-
05
10.5
8 0.
50
7.95
13
4.9
10.8
0.
43
1.4
17.1
0.
742
0.05
1 0.
211
0.02
69
0.00
52
20.3
ei
ther
C
V-9
09
/07/
2005
9-
05
15.9
5 20
.57
8.95
12
1 1
0.42
1.
6 13
.2
0.32
3 0.
024
0.29
3 0.
0353
0.
0286
17
.9
n
CV
-9a
09/0
7/20
05
9a-0
5 13
.02
42.1
4 9.
75
126.
5 1.
9 0.
36
3.9
11.5
5.
62
0.66
6 0.
448
0.12
4 0.
072
19.6
n
CV
-12
02/0
7/20
05
12-0
5 3.
30
16.5
3 7.
65
312.
5 2
2.49
4.
1 21
.6
0.88
8 0.
1 0.
431
0.02
61
0.01
34
46.0
ei
ther
C
V-1
3 10
/07/
2006
13
-06
19.3
8 25
.68
10.0
0 23
0.6
0.5
2.61
5.
7 18
.3
0.32
0.
045
0.90
9 0.
0488
0.
0353
39
.7
eith
er
CV
-20
09/0
7/20
05
20-0
5 7.
60
35.7
3 10
.05
173
0.9
0.81
6.
2 12
.5
0.47
9 0.
049
0.62
8 0.
0371
0.
0289
41
.9
eith
er
CV
-22
10/0
7/20
05
22-0
5 1.
34
16.4
1 7.
55
125.
4 0.
9 0.
31
1.8
10.9
1.
21
0.11
7 0.
261
0.03
59
0.00
29
23.6
ei
ther
C
V-2
4 10
/07/
2005
24
-05
1.62
35
.75
8 12
7.3
0.6
0.25
1.
5 15
.7
0.27
0.
013
0.16
2 0.
0096
0.
0107
40
.6
eith
er
Mea
n
9.
13
21.9
4 8.
44
161.
37
2.00
0.
82
2.87
15
.49
1.05
0.
11
0.36
0.
04
0.02
34
.49
eith
er
Med
ian
9.25
18
.55
7.98
13
1.10
0.
90
0.39
1.
70
15.1
5 0.
41
0.05
0.
28
0.03
0.
01
34.3
5 ei
ther
M
ax
19.3
8 42
.14
10.0
5 31
2.50
10
.80
2.61
6.
20
21.6
0 5.
62
0.67
0.
91
0.12
0.
07
66.2
0 p
Min
1.
34
0.50
6.
95
116.
00
0.50
0.
25
0.90
10
.90
0.27
0.
01
0.11
0.
01
0.00
17
.91
n
CHAPTER 9
GENERAL DISCUSSION AND CONCLUSIONS
The heightened sensitivity of Arctic ecosystems to environmental change has been well
documented (ACIA 2004; Schindler & Smol 2006). Climate-related changes have already
resulted in major ecosystem consequences in Canadian High Arctic freshwaters, including
cascading trophic effects (Smol et al. 2005), as well as the complete desiccation of ponds (Smol
& Douglas 2007). High Arctic ecosystems, therefore, are key reference areas for studies of
global environmental change. The appropriate use of reference areas, however, requires a
thorough understanding of natural environmental variability and baseline conditions. Due to the
abundance of lakes and ponds in the Canadian High Arctic, paleolimnological techniques, based
on sound limnological data, have the potential to reconstruct historical patterns of environmental
change.
Modern limnology
Physical and chemical data from over 400 lakes and ponds have highlighted the
limnological variability that currently exists in the Canadian High Arctic and underscored the
need for more data from under-represented ecosystem types. This thesis has contributed new
limnological information from some of these under-studied areas, including the western High
Arctic (Chapter 2), High Arctic oases (Chapter 4), and from freshwater systems influenced by
seabird-derived nutrients (Chapter 7).
In general, the studies in this thesis showed that lakes and ponds located in more lushly
vegetated regions were often characterized by higher dissolved organic carbon (DOC), nutrients,
pH and specific conductivity (Chapters 2 and 4). The relationship between these environmental
variables and lush vegetation had previously been noted, albeit on a smaller scale (Antoniades et
al. 2003; Lim et al. 2005). The differences in environmental variables are likely due, at least in
part, to the underlying reasons for the lush vegetation, whereby microclimatic and topographic
features have resulted in conditions more amenable for plant growth (i.e. availability of water,
richer soils; Edlund & Alt 1989), which in turn leads to increased DOC and nutrient transport
from the catchment. Nutrient enrichment will increase photosynthesis where nutrients limit
primary production (Bonilla et al. 2005), and greater rates of photosynthesis ultimately result in
higher pH. DOC may also increase production by stimulating the microbial loop (i.e. Vincent &
Hobbie 2000), and by providing UV protection for phytoplankton and benthic microbial mats
(Rae & Vincent 1998). Collectively, the freshwater ecosystems found in the more lushly
vegetated regions had notably different limnological characteristics relative to other High Arctic
sites.
On the other hand, the Cape Vera sites, although lushly vegetated and replete with
nutrients, did not have particularly high DOC (Chapter 7). We suggest this may be due to the role
of allochthonous (seabird-derived) nutrients permitting the development of greater terrestrial
production in the absence of a favourable climate. As DOC is derived from the decomposition of
terrestrial organic matter, a slower rate of decomposition related to particularly harsh climatic
conditions would result in less DOC export to freshwater systems than would be otherwise
expected. Furthermore, the type of vegetation may also play a role in DOC cycling; for example,
Cape Vera was characterized by many mosses, yet few grasses and sedges. Mosses in Arctic
lakes have been shown to withstand cold temperatures and to decompose very slowly (Sand-
Jensen et al. 1999).
Previous Arctic limnological surveys (e.g. Lim et al. 2001; Michelutti et al. 2002;
Antoniades et al. 2003) have attributed the lack of a relationship between nutrients and
cholorphyll a measured in the water columns of High Arctic lakes and ponds to the dominance of
benthic production (Vezina & Vincent 1997; Villeneuve et al. 2001; Bonilla et al. 2005). Similar
to these previous studies, the water chemistry data from Melville Island (Chapter 2), northern
Ellesmere Island (Chapter 4), and Cape Vera (Chapter 7) also showed no clear relationship
between nutrients and chlorophyll a concentrations and that is likely attributable to similar factors
(i.e. the dominance of benthic production that is not captured by cholophyll a concentrations in
the water column). However, at Cape Vera, it is clear that several ponds are highly productive,
with luxuriant strands of filamentous algae and cyanobacteria, as well as relatively dense
populations of invertebrates (e.g. chironomids and cladocerans), so this lack of relationship
between measured nutrients and chlorophyll a is slightly more surprising. Due to a lack of upper
trophic level predators (e.g. fish), the Cape Vera ponds might present an ideal site to examine the
influence of grazing pressure on primary production in a High Arctic context, as has been found
in studies of Alaskan lakes (e.g. Hobbie et al. 1999).
Diatom ecology
Chapters 3 and 7 provided the first data on diatom species distributions across a range of
ecozones on Melville Island and from a suite of ponds influenced by seabird-derived nutrients at
Cape Vera, Devon Island, respectively. While diatom assemblages from all samples in this thesis
contained species that have previously been identified from High Arctic ponds and lakes
(Antoniades et al. in press), patterns of relative abundance were quite distinct. On Melville
Island, for example, similar taxa were found in ponds located in all four bioclimatic zones, yet the
relative proportions of the dominant taxa were different, leading to significant differences
between the most lushly vegetated bioclimatic zone and all other zones (Chapter 3). An
interesting conclusion drawn from this study is that spatial variables appear relatively
unimportant in structuring diatom species distributions, at least at the regional scale of Melville
Island.
Chapter 7 presents an examination of diatom species data from ponds across a gradient of
seabird influence. Significant relationships between diatom species distributions and seabird-
derived nutrients (inferred by δ15N) were identified. However, the high abundance of very few,
overwhelmingly dominant, diatom taxa in nearly all sites suggested that seabird-derived nutrients
are not the most important factor directly determining the most abundant diatom assemblages in
these ponds. The data from Chapter 7 add to the existing literature of other diatom-based studies
of nutrient enrichment (from sewage inputs and whale carcasses) in Arctic lakes and ponds
(Douglas & Smol 2000; Douglas et al. 2004; Michelutti et al. in press); together these data
indicate highly complex relationships between diatoms and nutrient enrichment. Interestingly,
our data contrast those of other studies that have suggested anthropogenic nitrogen deposition
may influence diatom assemblage shifts in Arctic lakes and ponds (Wolfe et al. 2006).
Paleolimnology
Chapters 5, 6, and 8 each present a different application of diatom-based paleolimnology.
The timing of environmental change recorded by diatoms from paleolimnological studies has
been attributed, in part, to the hypothesis that ice-cover is an important factor influencing diatom
assemblage shifts in High Arctic lakes and ponds (Smol 1983; Smol 1988; Douglas & Smol
1999). In Chapter 6, the difference in ice duration between two small, adjacent, and
limnologically similar lakes from northern Ellesmere was used as a test of the ice cover
hypothesis. The marked diatom changes in the lake with less ice cover, combined with the lack
of diatom change in the lake with more ice cover, strongly support the idea that the duration and
extent of ice cover impact the magnitude of the diatom response to environmental change
(Chapter 6).
The diatom record from a small pond on Melville Island provides the only
paleoenvironmental data available from this large western High Arctic island, and indicates
marked environmental change, consistent with climate warming, beginning in the early 20th
century (Chapter 5). The results from Chapter 5 are consistent with those from two small mid-
Arctic ponds on the more southerly Banks Island (Lim 2004). As small ponds are expected to be
more sensitive to environmental change than larger lakes (Smol et al. 2005), together these two
western Arctic diatom-based paleolimnological studies suggest a later onset of environmental
change as compared to small ponds from the eastern Canadian Arctic (Douglas et al. 1994;
Antoniades et al. 2005).
The diatom-based paleolimnological studies of ten ponds across a gradient of seabird
influence at Cape Vera, Devon Island, represent the first paleolimnological assessments of
seabird-influence in Arctic regions (Chapter 8). The diatom assemblage changes from the
sediment cores represent highly complex records, only some of which are consistent with other
proxies of seabird-influence. The conclusions drawn from this study suggest that changes in
diatom assemblages are not robust indicators of seabird-derived nutrients to High Arctic ponds;
rather, proxies tracking trophic sources of nutrients (i.e. stable isotopes) and estimates of overall
production (i.e. chlorophyll a) are likely more effective indicators. However, adequate
chronological control is of the utmost importance when using proxies based on concentration data
such as these.
Future directions
Much progress has been made in High Arctic limnology over the past several decades,
and the field is now at a very exciting point. While the continued collection and surveying of
modern limnological data would certainly supplement and therefore improve our understanding
of Arctic limnology, several other avenues of future limnological and paleolimnological research
also deserve attention.
With the advent of the excellent and much needed guide to High Arctic diatom taxonomy
and autecology provided by Antoniades et al. (in press), the seeds for a much more thorough
understanding of diatom autecology have been sown. Including the data from this thesis, both
limnological and diatom ecological information have been collected, using identical techniques,
from over 400 lakes and ponds from ten of the largest islands in the Canadian Arctic Archipelago.
Although complex taxonomic issues (i.e. splitting and grouping, revised names) make the
compilation of these data a daunting task, the resulting data would do much to further refine
diatom-based limnological and paleolimnological research in the High Arctic. Moreover, this
“super-calibration set” would rival any in the world by its size, scope, and, perhaps most
importantly, methodological consistency.
Furthering our understanding of Arctic limnology is critical, in light of the importance of
Arctic freshwater ponds and lakes as “hotspots of biodiversity”, as well as their sensitivity to
environmental change. Unlike some terrestrial ecosystems (Henry & Molau 1997), long-term
experimental data on the effects of warming in Arctic freshwater lakes are completely lacking
(ACIA 2005). Experimental data of this type would be particularly useful in High Arctic lakes
and ponds, where initial, shorter term studies have identified distinctive relationships relative to
temperate regions.
Perhaps the most intriguing research question arising from this thesis is that of the
complex relationship between diatom assemblages and nutrient enrichment in High Arctic lakes
and ponds: If other primary producers respond to nutrients, why is there no clear species
assemblage shift in diatom communities? Questions aimed at teasing apart the diatom-nutrient
relationship could be addressed in a series of controlled experiments in High Arctic ponds. Some
avenues that would be worth following include: a) Do diatom communities (composition or
biomass) change in response to known nutrient enrichment? b) Does the diatom response depend
on interactions with temperature and/or grazers? c) Is the diatom response, or lack thereof,
ecologically meaningful with respect to Arctic food webs? Such data would be useful in more
fully understanding the aquatic ecosystems of the High Arctic, and ultimately lead to a better
appreciation of the threats facing these sentinels of environmental change.
References
ACIA, 2004. Impacts of a warming Arctic. Cambridge University Press, Cambridge, UK.
ACIA, 2005. Arctic climate impact assessment. Cambridge University Press, New York.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2003. Comparative physical and chemical limnology of two Canadian High Arctic regions: Alert (Ellesmere Island, NU) and Mould Bay (Prince Patrick Island, NWT). Archiv für Hydrobiologie 158: 485-516.
Antoniades, D., M. S. V. Douglas, & J. P. Smol, 2005. Quantitative estimates of recent environmental changes in the Canadian High Arctic inferred from diatoms in lake and pond sediments. Journal of Paleolimnology 33: 349-360.
Antoniades, D., P. B. Hamilton, M. S. V. Douglas & J. P. Smol, in press. Freshwater diatoms from the Canadian High Arctic. Iconographica Diatomologica.
Bonilla, S., V. Villeneuve, & W. F. Vincent, 2005. Benthic and planktonic algal communities in a High Arctic Lake: Pigment structure and contrasting responses to nutrient enrichment. Journal of Phycology 41: 1120-1130.
Douglas, M. S. V.& J. P. Smol, 1999. Freshwater diatoms as indicators of environmental change in the High Arctic. In E. F. Stoermer & J. P. Smol (eds), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge: 227-244.
Douglas, M. S. V., & J. P. Smol, 2000. Eutrophication and recovery in the High Arctic: Meretta lake (Cornwallis Island, Nunavut, Canada) revisited. Hydrobiologia 431: 193-204.
Douglas, M. S. V., J. P. Smol, & W. Blake Jr, 1994. Marked post-18th century environmental change in high-arctic ecosystems. Science 266: 416-419.
Douglas, M. S. V., J. P. Smol, J. M. Savelle, & J. M. Blais, 2004. Prehistoric Inuit whalers affected arctic freshwater ecosystems. Proceedings of the National Academy of Sciences of the United States of America 101: 1613-1617.
Edlund, S. A., & B. T. Alt, 1989. Regional congruence of vegetation and summer climate patterns in the Queen Elizabeth Islands, Northwest Territories, Canada. Arctic 42: 3-23.
Henry, G. H. R., & U. Molau, 1997. Tundra plants and climate change: the International Tundra Experiment (ITEX). Global Change Biology 3: 1-9.
Hobbie, J. E., B. J. Petersen, N. Bettez, L. Deegan, W. J. O’Brien, G. W. Kling, G. W. Kipphut, W. B. Bowden, & A. E. Hershey, 1999. Impact of global change on the biogeochemistry of an Arctic freshwater system. Polar Research 18: 207-214.
Lim, D.S.S., 2004. Limnology and diatom paleoecology of diatoms on Banks Island, N.W.T., and Devon Island, Nunavut, Canadian Arctic. University of Toronto. PhD Thesis.
Lim, D. S. S., M. S. V. Douglas, & J. P. Smol, 2005. Limnology of 46 lakes and ponds on Banks Island, NWT, Canadian Arctic Archipelago. Hydrobiologia 545: 11-32.
Lim, D. S. S., M. S. V. Douglas, J. P. Smol, & D. R. S. Lean, 2001. Physical and chemical limnological characteristics of 38 lakes and ponds on Bathurst Island, Nunavut, Canadian High Arctic. International Review of Hydrobiology 86: 1-22.
Michelutti, N., M. S. V. Douglas, D. C. G. Muir, X. Wang, & J. P. Smol, 2002. Limnological characteristics of 38 lakes and ponds on Axel Heiberg Island, High Arctic Canada. International Review of Hydrobiology 87: 385-399.
Michelutti, N., M. H. Hermanson, J. P. Smol, P. J. Dillon, & M. S. V. Douglas, in press. Delayed response of diatom assemblage changes to sewage inputs in an Arctic lake. Aquatic Sciences.
Rae, R., & W. F. Vincent, 1998. Effects of temperature and ultraviolet radiation on microbial foodweb structure: potential responses to global change. Freshwater Biology 40: 747-758.
Sand-Jensen, K., T. Riis, S. Markager, & W. F. Vincent, 1999. Slow growth and decomposition of mosses in Arctic lakes. Canadian Journal of Fisheries and Aquatic Sciences 56: 388-393.
Schindler, D. W., & J. P. Smol, 2006. Cumulative effects of climate warming and other human activities on freshwaters of Arctic and subarctic North America. Ambio 35: 160-168.
Smol, J. P., 1983. Paleophycology of a high arctic lake Cape Herschel, Ellesmere Island. Canadian Journal of Botany 61: 2195-2204.
Smol, J. P., 1988. Paleoclimate proxy data from freshwater arctic diatoms. Verh. int. Ver. Limnol. 23: 837-844.
Smol, J. P., & M. S. V. Douglas, 2007. Crossing the final ecological threshold in high Arctic ponds. Proceedings of the National Academy of Sciences of the United States of America 104: 12395-12397.
Smol, J. P., A. P. Wolfe, H. J. B. Birks, M. S. V. Douglas, V. J. Jones, A. Korhola, R. Pienitz, K. Ruhland, S. Sorvari, D. Antoniades, S. J. Brooks, M. A. Fallu, M. Hughes, B. E. Keatley, T. E. Laing, N. Michelutti, L. Nazarova, M. Nyman, A. M. Paterson, B. Perren, R. Quinlan, M. Rautio, E. Saulnier-Talbot, S. Siitoneni, N. Solovieva, & J. Weckstrom, 2005. Climate-driven regime shifts in the biological communities of arctic lakes. Proceedings of the National Academy of Sciences of the United States of America 102: 4397-4402.
Vezina, S., & W. F. Vincent, 1997. Arctic cyanobacteria and limnological properties of their environment: Bylot Island, Northwest Territories, Canada (73 degrees N, 80 degrees W). Polar Biology 17: 523-534.
Villeneuve, V., W. F. Vincent, & J. Komárek, 2001. Community structure and microhabitat characteristics of cyanobacterial mats in an extreme high Arctic environment: Ward Hunt Lake. Nova Hedwigia 123: 199-224.
Vincent, W. F., and J. E. Hobbie, 2000. Ecology of Arctic lakes and rivers. In: M. Nuttal, and T. V. Callaghan (eds), The Arctic: Environment, People, Policy. Harwood, Amsterdam: 197-231.
Wolfe, A. P., C. A. Cooke, & W. O. Hobbs, 2006. Are current rates of atmospheric nitrogen deposition influencing lakes in the Eastern Canadian Arctic? Arctic Antarctic and Alpine Research 38: 465-476.
APPENDICES
Appendix 1. Detailed water chemistry methods used for all water sampling and temperature measurement during field work. Note that unless otherwise noted, all water samples were taken from the nearshore area where the average water depth was ~70 cm. All bottles were rinsed three times with lakewater prior to filling. Air temperature: Air temperatures were recorded twice daily (approximately 08:00h and 20:00h) from a thermometer kept inside a Stevenson screen. Water temperature: Water temperatures were measured at the time of sampling for water chemistry using a hand-held alcohol thermometer held ~10 cm below the water surface. pH: Samples for pH were taken from the lake water using a Nalgene bottle, rinsed 3 times with sample water. pH was measured back at the base camp at the end of each sampling day. Water samples were kept in the dark in a backpack from time of sampling until time of measurement. We used three hand-held pHEP 4 pH meters (Hanna) to measure each sample. All pH meters were calibrated to 7 and 10 pH buffer solution daily. We used the average pH from the three meters as our pH value, although often all three meters gave the same reading. Specific conductivity: We measured specific conductivity at the base came each day, using a subsample of the same water obtained for pH measurement. Specific conductivity was measured using a YSI model 33 conductivity meter. Chlorophyll a: Samples for chlorophyll a were filtered onto glass fibre filters (Whatman GF/C; diameter = 4.25 cm) under low vacuum pressure. Filters were placed immediately in Petri dishes, wrapped in aluminum foil, and then kept in the dark and as cold as possible until return to PCSP. Once at PCSP, filters were kept frozen until analysis at CCIW, Burlington, ON. Filtered nutrients (including total phosphorus filtered, all nitrogen fractions), dissolved organic carbon and dissolved inorganic carbon: water was filtered through Sartorius-type cellulose acetate filters (Brand: AcetatePlus Supported, Plain, pore size: 0.45 µm, filter diameter 4.7 cm). Filters were placed immediately in Petri dishes, wrapped in aluminum foil, and then kept in the dark and as cold as possible until return to PCSP. Once at PCSP, filters were kept frozen until analysis at CCIW, Burlington, ON. Particulate carbon and nitrogen: sample water is filtered through pre-ignited glass fibre filters (Whatman GF/C 2.5 cm). Filters were placed immediately in Petri dishes, wrapped in aluminum foil, and then kept in the dark and as cold as possible until return to PCSP. Once at PCSP, filters were kept frozen until analysis at CCIW, Burlington, ON.
Appendix 2. Output tables from various spatial neighbourhood models tested using VARCAN (Peres-Neto et al. 2006, see Chapter 3 for more details). Only the most restricted neighbourhood model indicated that spatial variables explained a significant portion of the diatom species distribution. The number of spatial variables tested for each model is given in the heading. The environmental variables included in the model were the same for each: elevation, surface area, pH, specific conductivity, temperature, total phosphorus, total phosphorus filtered, total dissolved nitrogen, chlorophyll a, DOC, SiO2, SRP, Al.
Delauney: 12 spatial variables retained
% explained % exp - adjustment probabilityenv 0.315 0.098 0.024env x space 0.098 0.084space 0.211 -0.013 0.898residual 0.375 0.831
Gabriel graph: 17 spatial variables retained
% explained % exp - adjustment probabilityenv 0.311 0.127 0.018env x space 0.103 0.055space 0.325 0.042 0.391residual 0.261 0.775
Sphere of Influence: 16 spatial variables retained
% explained % exp - adjustment probabilityenv 0.318 0.138 0.006env x space 0.095 0.044space 0.315 0.056 0.182residual 0.271 0.761
Relative neighbours: 17 spatial variables retained
% explained % exp - adjustment probabilityenv 0.325 0.158 0.002env x space 0.088 0.024space 0.334 0.065residual 0.253 0.752 0.178
50 000m: 8 spatial variables retained
% explained % exp - adjustment probabilityenv 0.32 0.124 0.004env x space 0.094 0.058space 0.181 0.067residual 0.405 0.751 0.026
Appendix 3. Diatom species response scatterplots versus specific conductivity from Melville Island, Nunavut/NWT. Roman numerals next to species names indicate the best HOF model. See Chapter 3 for details on HOF models.
Psammothidium chlidanos (II)
0
5
10
15
20
25
30Eucocconeis flexella (IV)
0
2
4
6
8Eucocconeis laevis (IV)
0
2
4
6
8
Psammothidium marginulatum (II)
0
10
20
30
40
50
60
70Achnanthidium minutissimum (IV)
0
5
10
15
20
25Rossithidium petersenii (I)
0
2
4
6
8
10
12
14
Psammothidium scotica (II)
Rel
ativ
e ab
unda
nce
(%)
0
2
4
6
8
10Psammothidium ventralis (II)
0
2
4
6
8
10
12
14
16Amphora inariensis (I)
0
2
4
6
8
Chamaepinnularia soehrensis (V)
0 300 600 900 1200 1500
02468
101214161820
Cymbella cleve-eulerae (IV)
specific conductivity (μS/cm)
0 300 600 900 1200 1500
0
1
2
3
4
5Encyonema minutum (I)
0 300 600 900 1200 1500
0
2
4
6
8
10
Appendix 3. Continued.
Diadesmis contenta (IV)
0
5
10
15
20
25
30
35Fragilaria capucina (V)
0
5
10
15
20
25
30Navicula chiarae (IV)
0
2
4
6
8
10
12
Navicula cryptocephala (I)
Rel
ativ
e ab
unda
nce
(%)
0
2
4
6
8
10Navicula gerloffi (I)
0
5
10
15
20
25
30
35Nitzschia frustulum (V)
0
10
20
30
40
50
Nitzschia inconspicua (I)
0 300 600 900 1200 1500
0
2
4
6
8
10
12
14Nitzschia perminuta (V)
specific conductivity (μS/cm)
0 300 600 900 1200 1500
0
5
10
15
20
25
30
35Nitzschia pusilla (I)
0 300 600 900 1200 1500
0
1
2
3
4
Appendix 4. Summary statistics, figures, and species optima of various weighted averaging models for diatom-inferred specific conductivity from surface sediments of Melville Island. Table 1a) all species >1% relative abundance in three sites or >10% relative abundance in one site (n=90), b) all species with a significant response to conductivity (n=55), and c) all species with a significant unimodal response to specific conductivity (n=30). All values shown are cross-validated using bootstrapping techniques. “RMSEP” is the Root Mean Squared Error of Prediction.
a)
WAInv WACla WATol Inv WATol Cla
r2boot 0.138 0.152 0.257 0.272
Average bias boot 44.560 48.941 74.910 84.390 RMSEP 250.110 249.923 249.336 250.231 b)
WAInv WACla WATol Inv WATol Cla
r2boot 0.156 0.176 0.467 0.488
Average bias boot 51.300 54.500 68.600 74.600 RMSEP 249.100 248.500 230.800 228.100 c)
WAInv WACla WATol Inv WATol Cla
r2boot 0.189 0.210 0.325 0.349
Average bias boot 3.380 5.040 18.300 32.130 RMSEP 217.300 312.500 201.000 250.920
Appendix 4. Figure 1. Scatterplot of: A) estimated versus observed specific conductivity, and B) residuals based on the weighted averaging technique with inverse deshrinking and tolerance downweighting.
observed cond (µS/cm)
WA
toli
nv e
stim
ated
con
d(µ
S/c
m)
observed cond (µS/cm)
WA
toli
nv c
ond
resi
dual
s
0 200 400 600 800 1000 1200
0
200
400
600
800
1000
1200r 2
boot = 0.257
0 200 400 600 800 1000 1200-1200
-1000
-800
-600
-400
-200
0
200RMSEP = 249
A B
Appendix 4. Table 2. Estimated specific conductivity optima for diatom species from the surface sediments of Melville Island. Species codes are the same as for Table 2 (Keatley et al. [3]).
Species Cond. Species Cond. Species Cond. code optimum code optimum code optimum
1 151 34 219 67 57 2 194 35 159 68 128 3 50 36 145 69 139 4 245 37 101 70 282 5 32 38 49 71 29 6 64 39 59 72 327 7 154 40 95 73 310 8 93 41 119 74 195 9 172 42 305 75 219
10 69 43 225 76 122 11 158 44 378 77 82 12 121 45 63 78 111 13 30 46 n/a 79 176 14 32 47 55 80 205 15 48 48 643 81 153 16 57 49 228 82 434 17 124 50 145 83 43 18 167 51 7 84 105 19 168 52 6 85 56 20 634 53 87 86 91 21 16 54 54 87 60 22 232 55 77 88 77 23 230 56 65 89 71 24 72 57 n/a 90 20 25 111 58 n/a 26 209 59 211 27 n/a 60 214 28 22 61 130 29 278 62 110 30 186 63 96 31 142 64 66 32 121 65 n/a 33 184 66 846
App
endi
x 5.
Raw
dia
tom
cou
nts f
rom
Mel
ville
Isla
nd su
rfac
e se
dim
ents
. M
elvi
lle Is
land
cal
ibra
tion
diat
oms
SITE
NAM
E (M
V- )
AB
CD
EF
GH
IJ
KL
MN
OP
QR
ST
UV
WX
YZ
AAAB
ACAD
AEAF
roAG
AHAI
AJAK
ALAM
ANAO
APAQ
ARAS
ATAc
hnan
thes
alta
ica
2Ac
hnan
thes
bro
enlu
nden
sis
00
700
00
30
00
00
40
00
00
00
00
00
00
00
00
00
60
20
00
00
00
00
00
Achn
anth
es h
olst
ii0
00
00
00
00
00
00
00
00
00
00
00
04
60
00
30
00
00
02
00
00
00
20
0Ac
hnan
thes
ingr
atifo
rmis
00
00
00
00
00
00
00
00
00
00
480
1318
00
200
010
00
00
00
00
07
00
08
190
Achn
anth
es k
riege
ri0
629
56
03
010
10
130
60
30
03
00
00
00
10
00
120
300
00
00
20
00
00
00
0Ac
hnan
thes
kry
ophi
la0
00
00
03
320
010
012
100
00
03
011
00
00
00
00
00
30
00
00
00
00
2612
00
0Ac
hnan
thes
mar
ginu
lata
108
7336
1617
082
1635
134
704
2519
910
712
2726
340
525
436
74
22
168
240
60
410
09
40
306
133
213
9Ac
hnan
thes
min
utis
sim
a11
021
4511
346
6528
043
756
116
780
018
293
2431
00
915
814
87
011
415
00
00
00
510
460
00
037
4043
Achn
anth
es p
eter
senn
i27
613
3413
40
80
1023
3910
773
34
815
20
2018
00
130
413
00
00
00
40
00
01
111
1933
Achn
anth
es s
cotic
a0
00
00
07
010
034
00
36
00
220
00
043
054
560
00
00
00
00
360
00
00
00
449
0Ac
hnan
thes
chl
idan
os8
95
2910
912
414
1110
3140
141
768
3143
2617
633
7812
62
4312
465
8513
336
4Ac
hnan
thes
cur
tissi
ma
12
Achn
anth
es d
aone
nsis
40Ac
hnan
thes
del
icat
ula
sp s
epte
ntrio
nalis
2Ac
hnan
thes
dev
ilens
is1
22
1114
Achn
anth
es fl
exel
la8
1410
72
3117
2317
5655
21
16
104
22
321
93
Achn
anth
es h
intz
ii2
Achn
anth
es la
evis
106
212
641
146
120
1963
88
10
06
58
130
68
100
120
00
00
00
00
00
00
00
00
40
Achn
anth
es v
entr
alis
22
22
00
996
764
200
68
212
02
1214
200
3738
3910
180
021
00
02
00
00
04
00
6120
96
Achn
anth
es h
elve
ticum
00
00
00
210
431
40
02
20
00
00
00
100
20
00
00
101
00
00
00
40
00
00
50
0Ac
hnan
thes
bio
retti
04
00
00
00
20
035
53
00
05
52
60
00
130
290
00
20
00
00
00
00
00
02
10
Achn
anth
es ri
cula
2Ac
hnan
thes
rupe
stris
126
335
278
Achn
anth
es ro
ssi
419
7Ac
hnan
thes
sac
cula
4Ac
hnan
thes
sp1
MVA
S2
Achn
anth
es s
p2 M
VAS
1Ac
hnan
thes
sto
lida
var s
chm
assm
anii
6Ac
hnan
thes
sub
atom
oide
s10
66
22
325
48Ad
lafi
bryo
phila
527
82
219
132
44
12
112
Amph
ora
inar
iens
is2
1226
82
29
26
112
62
138
Amph
ora
copu
lata
51
21
810
44
84
12
Amph
ora
duse
nii
42
122
32
22
6Am
phor
a fo
gedi
i26
710
Amph
ora
foge
dian
a29
25Am
phor
a ho
lsta
ica
44
24Am
phor
a lib
yca
52
47
Amph
ora
pedi
culu
s9
82
Amph
ora
spitz
berg
ensi
s3
66
222
6612
4Am
phor
a ve
neta
22
2C
alon
eis
silic
ula
tota
l0
60
60
122
64
12
103
120
00
142
00
00
06
00
02
00
00
82
00
014
00
00
04
Cal
onei
s cf
big
2C
alon
eis
aero
phila
26
109
62
144
1217
30C
alon
eis
hend
eyii
426
Cal
onei
s pu
lcha
2C
alon
eis
sp M
VK4
Cal
onei
s sm
all
190
00
60
110
190
00
176
00
20
02
00
20
06
00
00
00
20
00
00
362
00
00
00
Cav
inul
a co
cone
iform
is3
22
62
Cav
inul
a ja
ernf
eldi
i2
4759
22
74
Cav
inul
a ps
eudo
scut
iform
is23
Cha
mae
pinn
ular
ia g
andr
upii
00
00
00
014
00
00
00
00
00
00
00
00
00
00
00
00
028
00
094
010
00
00
00 0
Cha
mea
pinn
ular
ia s
ohre
nsis
228
102
6136
116
720
680
1610
063
80
00
00
04
00
00
02
01
00
00
00
00
00
00
00
04
App
endi
x 5.
con
tinue
d.
Mel
ville
Isla
nd c
alib
ratio
n di
atom
sSI
TE N
AME
(MV-
)A
BC
DE
FG
HI
JK
LM
NO
PQ
RS
TU
VW
XY
ZAA
ABAC
ADAE
AF ro
AGAH
AIAJ
AKAL
AMAN
AOAP
AQAR
ASAT
Cyc
lote
lla a
ntiq
ua3
22
1C
yclo
tella
ato
mus
(C g
lom
erat
a or
C p
seud
oste
llige
ra?)
892
Cyc
lote
lla b
odan
ica
var a
ff le
man
ica
2C
yclo
tella
oce
llata
1C
yclo
tella
trip
artit
a17
Cyc
lote
lla a
ff co
men
sis
(or C
aff
com
ta v
ar u
nipu
ncta
ta)
59C
ymbe
lla a
ffini
s5
16C
ymbe
lla a
mph
icep
hala
72
26
66
1614
64
218
2C
ymbe
lla a
ngus
tata
52
1015
107
64
256
211
44
22
66
22
3029
4C
ymbe
lla c
esat
ii4
97
22
1216
612
67
2C
ymbe
lla b
otel
lus
160
08
42
92
260
15
190
00
720
97
00
00
00
00
80
00
00
00
00
00
00
00
00
Cym
bella
cle
ve-e
uler
ae23
138
1915
219
88
016
1820
54
216
148
46
00
40
00
09
00
01
00
00
02
00
00
04
0C
ymbe
lla d
escr
ipta
24
00
215
260
40
06
390
30
22
00
00
00
00
00
00
00
00
00
00
00
00
00
00
Cym
bella
des
igna
ta2
104
1041
104
82
42
48
2C
ymbe
lla fo
gedi
i12
00
1224
01
02
08
105
02
016
00
02
00
00
00
00
00
00
00
00
00
00
00
029
5C
ymbe
lla g
auem
anii
24
Cym
bella
late
ns0
00
00
09
100
140
02
04
00
05
182
08
03
02
04
00
00
00
00
00
00
00
1039
8C
ymbe
lla la
nge-
bert
alot
ti4
84
62
82
22
87
Cym
bella
lapp
onic
a (a
ngus
tata
?)4
746
44
Cym
bella
min
uta
140
02
10
00
20
52
166
517
320
924
10
619
190
261
435
00
02
02
00
44
00
04
220
Cym
bella
obs
cura
22
Cym
bella
pro
xim
a3
Cym
bella
sile
siac
a9
48
57
47
04
04
12
12
160
26
20
00
134
313
017
170
00
20
05
04
00
00
03
0C
ymbe
lla s
p de
rmot
pla
te 3
5 #2
62
46
Cym
bella
sub
aequ
alis
1313
22
1110
112
44
22
2C
ymbe
lla s
ubsp
icul
a2
2C
ymbo
pleu
ra c
uspi
data
2C
ymbo
pleu
ra s
taur
onei
form
is2
22
22
16D
entic
ula
kuet
zing
ii21
30
244
1110
025
02
1129
00
06
00
90
00
00
00
00
00
00
00
00
02
00
00
00
0D
entic
ula
subt
ilis
15D
entic
ula
tenu
is0
00
69
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
092
00
00
00
0D
iade
smis
con
tent
a5
526
2713
923
362
1269
9874
558
7221
92
452
104
24
461
Dia
desm
is g
allic
a13
2D
iade
smis
par
acon
tent
a1
229
6D
iade
smis
per
pusi
lla2
22
66
2D
iato
ma
mon
ilifo
rmis
104
22
32
14D
iato
ma
tenu
is8
40
02
00
07
011
00
64
00
250
61
00
72
02
308
50
00
00
10
014
00
00
00
00
Dip
lone
is m
arge
nies
tria
ta6
62
25
912
1910
66
122
210
41
65
Dip
lone
is o
ccul
ata
124
24
1215
Dip
lone
is p
arm
a4
21
22
24
22
2En
cyon
ema
cf a
ff su
belg
inen
se2
Ency
onem
a sp
pla
te 3
5 #1
8-24
62
Ency
onem
a ve
ntric
osum
22
25Eu
notia
arc
us2
2Eu
notia
bilu
naris
24
22
62
1813
Euno
tia b
iluna
ris v
ar m
ucop
hila
2Eu
notia
bor
eote
nuis
00
00
00
00
08
00
00
00
00
00
00
00
00
00
00
4321
00
010
00
00
00
736
00
0Eu
notia
exi
gua
231
444
62
Euno
tia e
xigu
a fo
rmae
bid
ens
Hus
tedt
2Eu
notia
faba
3
Euno
tia fa
llax
var g
roen
land
ica
2Eu
notia
pra
erup
ta v
ar b
igib
ba4
Euno
tia p
raer
upta
40
04
20
00
85
214
140
011
00
30
00
20
226
40
12
212
00
00
00
02
07
10
06
Euno
tia p
seud
opec
tinal
is6
Euno
tia s
ubar
cuat
oide
s4
Falla
ceae
inso
ciab
ilis
21
Falla
ceae
pyg
mac
eae
43
Falla
ceae
sub
ham
ulat
a2
2Fa
llace
ae s
ublu
cidu
la1
App
endi
x 5.
con
tinue
d.
Mel
ville
Isla
nd c
alib
ratio
n di
atom
sSI
TE N
AME
(MV-
)A
BC
DE
FG
HI
JK
LM
NO
PQ
RS
TU
VW
XY
ZAA
ABAC
ADAE
AF ro
AGAH
AIAJ
AKAL
AMAN
AOAP
AQAR
ASAT
Frag
ilaria
alp
etris
2Fr
agila
ria c
apuc
ina
140
00
180
650
457
158
820
1035
4016
453
100
100
84
600
490
7226
032
140
00
00
02
120
316
515
Frag
ilaria
cap
ucin
a va
r aus
tria
ca24
30Fr
agila
ria c
apuc
ina
var g
raci
lis2
86
25
614
218
22
154
12Fr
agila
ria c
onst
ruen
s va
r ven
ter
212
Frag
ilaria
pin
nata
var
ven
ter(
?)14
6Fr
agila
ria p
inna
ta0
00
00
00
00
00
00
042
030
60
20
00
230
00
20
015
00
00
60
00
00
00
40
500
Frus
tulia
cra
ssin
ervi
a2
2G
eiss
leria
sch
oenf
eldi
i4
54
211
2G
eiss
leria
sp1
(MVN
)2
Gei
ssle
ria s
p D
erm
ot p
late
50
#19-
2311
4G
eiss
leria
sp
MVE
4G
eiss
leria
spe
c Fi
g16
plat
e 20
pag
e 14
86
Gom
phon
ema
angu
stat
um2
Gom
phon
ema
lage
rhei
mi
24
Gom
phon
ema
prod
uctu
m2
Gom
phon
ema
spp
aff e
xigu
um11
1G
omph
onem
a sp
aff
obsc
urum
16G
omph
onem
a sp
MVN
5G
omph
onem
a sp
MVA
O80
Han
tzsc
hia
amph
ioxy
s3
22
Hip
podo
ntic
a ar
ctic
a4
Lutic
ola
mut
ica
4Lu
ticol
a ni
valo
ides
2M
icro
cost
atus
kra
sske
i4
2M
ulle
ria g
ibbu
la2
Nav
icul
a ba
cillu
m6
42
22
Nav
icul
a cf
egr
egria
8N
avic
ula
cf rh
ynco
ceph
ala
1017
Nav
icul
a cf
sim
ilis
2N
avic
ula
chia
rae
220
40
3877
427
230
2517
4824
00
170
410
00
00
00
00
50
00
08
00
00
360
00
04
102
Nav
icul
a ch
iara
e/cr
ypto
ceph
ala
6647
Nav
icul
a ci
ncta
11
263
Nav
icul
a cl
emen
tis2
1N
avic
ula
com
plan
ata
1N
avic
ula
cryp
toce
phal
a2
217
420
231
88
743
132
536
225
1420
612
189
42
3620
Nav
icul
a di
gito
riadi
ata
4N
avic
ula
dolo
miti
ca (o
r Ach
nant
hes
dist
inct
a)6
Nav
icul
a eg
regr
ia8
214
1929
Nav
icul
a fr
agilo
ides
(?) s
mal
l, sk
inn y
23N
avic
ula
gerlo
ffi3
514
55
58
24
220
219
39
42
135
246
10N
avic
ula
genu
stria
ta6
Nav
icul
a gr
egar
ia3
116
1411
1416
62
Nav
icul
a hi
lliar
dii
102
42
4N
avic
ula
hilli
ardi
i und
ulat
e2
Nav
icul
a in
grat
a36
Nav
icul
a m
enis
culu
s6
234
Nav
icul
a m
utic
oide
s2
Nav
icul
a ph
ylle
pta
7322
0
App
endi
x 5.
con
tinue
d.
Mel
ville
Isla
nd c
alib
ratio
n di
atom
sSI
TE N
AME
(MV-
)A
BC
DE
FG
HI
JK
LM
NO
PQ
RS
TU
VW
XY
ZAA
ABAC
ADAE
AF ro
AGAH
AIAJ
AKAL
AMAN
AOAP
AQAR
ASAT
Nav
icul
a ps
eudo
scut
iform
is2
463
3315
339
72
321
3N
avic
ula
pseu
dote
nello
ides
(An
toni
ades
pla
te 4
9 #1
8)2
447
10N
avic
ula
pupu
la4
210
52
23
24
42
215
210
5N
avic
ula
pygm
aea
2N
avic
ula
rein
hard
ii15
3N
avic
ula
rhyn
coce
phal
a (d
erm
ot)
22
2N
avic
ula
salin
arum
62
Nav
icul
a cf
sal
inar
um b
ig6
Nav
icul
a sc
hmas
sman
ii19
Nav
icul
a sc
hmas
sman
ii bu
t ver
y sm
all a
nd n
ot a
s fa
t10
Nav
icul
a sm
all g
irdle
s M
VQ8
Nav
icul
a sp
ecie
s 1(
MVF
)4
Nav
icul
a sp
ecie
s M
VS2
Nav
icul
a sp
p48
DER
MO
T4
Nav
icul
a sp
34 p
late
51
derm
ot1
2N
avic
ula
sp15
pla
te 4
9 de
rmot
2N
avic
ula
sp24
-26
plat
e 49
der
mot
166
Nav
icul
a sp
18-2
3 pl
ate
49 d
erm
ot14
2N
avic
ula
spA
MVI
4N
avic
ula
spM
VAK
18N
avic
ula
spM
VD5
1N
avic
ula
spp
aff r
iedi
ana
25
25
Nav
icul
a sp
p af
f rei
nhar
dtii
92
2N
avic
ula
spp
aff v
enet
a4
63
94
29
616
217
Nav
icul
a sp
p af
f ven
erab
ilis
62
Nav
icul
a st
roem
ii5
2N
avic
ula
subh
amul
ata
6N
avic
ula
triv
alis
92
12
44
26
Nav
icul
a tu
scul
a23
27
12
22
22
22
102
Nav
icul
a vi
ridul
a va
r. lin
earis
2N
avic
ula
vulp
ina
1510
82
34
812
12
96
116
261
Nei
diop
sis
wul
fii2
83
2N
eidi
um a
ffine
14
26
182
84
16N
eidi
um a
ffine
var
long
icep
s2
13N
eidi
um a
mpl
iatu
m2
12
21
24
12
2N
eidi
um a
utsr
ium
3N
eidi
um b
ergi
i6
21
22
24
24
46
21
2N
eidi
um b
isul
catu
m2
Nei
dium
cal
vum
3N
eidi
um d
ecor
atum
19
Nei
dium
dec
orat
um v
ar b
ergi
i2
Nei
dium
dis
tinct
e-pu
ncta
tum
22
32
1N
eidi
um d
ubiu
m4
6N
eidi
um k
ozlo
wii
02
00
00
00
00
00
48
00
00
02
00
00
00
00
00
00
00
00
00
20
00
00
0N
eidi
um s
p1 M
VF1
0N
eidi
um s
p9 p
late
56
derm
ot2
Nei
dium
tem
perii
MVD
2N
itzsc
hia
amph
ibia
3
32
710
2N
itzsc
hia
accu
min
atum
8N
itzsc
hia
arch
ibal
di4
602
Nitz
schi
a cl
ausi
i3
23
89
1545
Nitz
schi
a co
mm
unta
ta2
22
14
104
Nitz
schi
a di
ssip
ata
var m
edia
63
84
44
115
22
Nitz
schi
a fr
ustu
lum
2211
1155
90
4811
170
3613
482
011
50
5414
80
50
318
20
30
167
00
218
00
02
305
00
04
019
18N
itzsc
hia
perm
inut
a35
5039
123
158
1718
610
244
3712
578
4824
380
3492
075
280
7076
120
3066
111
165
00
289
00
132
428
5826
00
6591
71
App
endi
x 5.
con
tinue
d.
Mel
ville
Isla
nd c
alib
ratio
n di
atom
sSI
TE N
AME
(MV-
)A
BC
DE
FG
HI
JK
LM
NO
PQ
RS
TU
VW
XY
ZAA
ABAC
ADAE
AF ro
AGAH
AIAJ
AKAL
AMAN
AOAP
AQAR
ASAT
Nitz
schi
a gr
acili
s4
10N
itzsc
hia
hom
burg
iens
is2
13
21
23
32
210
3846
1N
itzsc
hia
inco
nspi
cua
93
826
142
202
22
871
152
72
8112
22
6N
itzsc
hia
litto
ralis
6N
itzsc
hia
norm
anii
22
Nitz
schi
a ov
alis
1N
itzsc
hia
pale
a2
22
15N
itzsc
hia
pala
ecea
26
44
94
1912
Nitz
schi
a pa
laef
orm
is14
Nitz
schi
a pu
ra/p
alea
612
99
249
144
148
43
14
131
15
Nitz
schi
a pu
silla
87
1512
144
138
182
1215
1022
217
44
Nitz
schi
a re
cta
21
Nitz
schi
a su
blin
earis
2
172
44
44
419
Nitz
schi
a su
chla
ndii
148
210
276
7N
itzsc
hia
sp c
f aci
cula
ris6
Pinn
ular
ia a
ngus
tibor
ealis
2Pi
nnul
aria
bal
four
iana
247
37
1117
8Pi
nnul
aria
bic
eps
1Pi
nnul
aria
bor
ealis
var
lanc
eola
tat
22
Pinn
ular
ia b
ottn
ica
(lund
ii va
r bal
tica)
(gre
y bo
ok)
4Pi
nnul
aria
bra
ndel
iform
is2
22
Pinn
ular
ia b
rebi
sonn
i 2
22
31
4Pi
nnul
aria
bre
biso
nni v
ar b
icun
eata
2Pi
nnul
aria
bre
biso
nni s
ensu
lato
2Pi
nnul
aria
bre
vico
stria
ta2
Pinn
ular
ia c
f ang
lica
4Pi
nnul
aria
des
cres
cens
var
igno
rata
22
Pinn
ular
ia d
iver
gens
var
div
erge
ns18
Pinn
ular
ia fi
g 4
Plat
e61
2Pi
nnul
aria
spM
VT93
22
2Pi
nnul
aria
gru
now
ii2
22
102
2Pi
nnul
aria
inte
rmed
ia3
47
132
55
Pinn
ular
ia in
terr
upta
2Pi
nnul
aria
kra
mm
eri
10
20
47
02
60
30
52
90
00
610
00
65
43
162
20
40
03
05
52
23
02
20
3Pi
nnul
aria
lata
23
10Pi
nnul
aria
cf i
sost
auro
n va
r con
ifera
14Pi
nnul
aria
cf i
sost
auro
n va
r orie
ntal
is (p
142
Gre
y pi
nnul
aria
boo
k)13
35Pi
nnul
aria
cf m
inut
a6
Pinn
ular
ia m
icro
stau
ron
var b
rebi
sonn
i2
Pinn
ular
ia c
f mic
rost
auro
n4
Pinn
ular
ia o
bscu
ra2
4Pi
nnul
aria
str
epto
raph
e1
Pinn
ular
ia s
ubro
stra
ta2
Pinn
ular
ia c
f sub
rost
rata
7Pi
nnul
aria
cf p
laty
ceph
ala
(sm
all)
2Pi
nnul
aria
spp
aff
schr
oete
rae
22
Pinn
ular
ia s
p M
VK1
24Pi
nnul
aria
spM
VAL
28Pi
nnul
aria
sp
MVA
I3
Pinn
ular
ia v
asta
2Pl
acon
eis
clem
entis
24
6Pl
acon
eis
elgi
nens
is4
22
122
Plac
onei
s pl
acen
tula
22
4Pl
anot
hidi
um o
estr
upii
43
Plan
othi
dium
per
agal
li12
28
126
6
App
endi
x 5.
con
tinue
d.
Mel
ville
Isla
nd c
alib
ratio
n di
atom
sSI
TE N
AME
(MV-
)A
BC
DE
FG
HI
JK
LM
NO
PQ
RS
TU
VW
XY
ZAA
ABAC
ADA
EAF
rock
AG
AHA
IAJ
AK
ALAM
ANA
OAP
AQ
ARAS
ATSe
llaph
ora
sp4
2Se
llaph
ora
baci
llim
42
22
28
25
26
Stau
rone
is s
p M
VAA
22
Stau
rone
is a
ncep
s va
r sib
eric
a2
Stau
rone
is a
ncep
s5
22
41
44
48
62
62
214
45
102
46
410
412
42
21
152
Stau
rone
is p
hoen
icen
tero
n2
25
62
Stau
rone
is p
rom
inul
a4
122
16
15St
auro
neis
sm
ithii
22
2St
auro
neis
und
ata
237
556
Stau
rone
is s
p M
VAI
4St
epha
nodi
scus
min
utilu
s1
Step
hano
disc
us v
estib
ulus
3Su
rirel
la a
ngus
tata
25
22
102
Surir
ella
min
uta
22
199
2Ta
bella
ria fl
occu
losa
22
1720
24
145
310
32
453
Tabe
llaria
floc
culo
sa v
ar lo
ng2
Tryb
lione
lla d
ebili
s2
22
TOTA
L51
679
058
878
451
877
595
157
670
745
985
884
176
842
841
959
863
562
959
786
960
00
630
697
687
613
596
357
522
670
530
548
7062
760
050
755
644
970
632
177
459
569
477
958
132
4cy
sts
112
2012
310
415
67
1219
354
5812
5716
181
277
43
293
174
2418
22
18
269
79
28
4175
317
4fu
nky
frag
men
tsye
sye
s 6
yes
App
endi
x 6.
Tra
ce m
etal
wat
er c
hem
istry
dat
a fr
om n
orth
ern
Elle
smer
e Is
land
. A
ll un
its a
re µ
g/L.
IDO
ffic
ial N
ame
Al
As
BB
aB
eC
dC
oC
rC
uFe
Ga
La
Li
Mn
Mo
Ni
PbR
bSb
SrT
iU
VZ
nEP
AC
raig
Lak
e25
.10.
2721
.130
.00
0.00
60.
099
0.06
70.
166
1.31
56.0
0.01
30.
099
6.1
2.43
0.87
31.
430.
185
0.69
0.05
532
3.0
0.00
20.
7760
0.26
72.
57EP
BA
pple
by L
ake
1.6
0.53
24.1
32.2
0<
0.00
05<
0.00
050.
030
0.02
60.
3712
.40.
003
0.00
511
.29.
520.
086
0.42
0.02
60.
430.
023
348.
00.
001
0.11
400.
061
0.19
EPC
Bra
inar
d La
ke3.
00.
305.
426
.20
<0.
0005
<0.
0005
0.03
60.
028
0.46
17.3
0.00
20.
010
2.7
9.47
0.07
70.
980.
028
0.21
0.01
817
2.0
0.00
10.
1320
0.06
30.
72EP
D3.
70.
380.
85.
26<
0.00
050.
002
0.02
30.
016
0.54
25.9
0.00
20.
007
1.1
1.49
0.16
20.
360.
045
0.08
0.04
463
.3<
0.00
10.
0793
0.02
10.
86EP
E7.
31.
589.
36.
790.
002
0.02
00.
103
0.05
91.
2536
.10.
005
0.00
919
.62.
030.
620
0.84
0.05
90.
950.
170
80.5
0.00
21.
0400
0.17
02.
83EP
F17
7.0
0.24
1.0
4.50
0.01
00.
027
0.65
60.
203
0.66
374.
00.
044
0.15
22.
330
.60
0.17
81.
920.
558
0.50
0.05
513
7.0
0.00
30.
0969
0.09
61.
55EP
G30
.30.
041.
127
.70
0.00
60.
003
0.05
80.
100
0.46
48.2
0.00
80.
017
0.8
3.55
0.13
80.
270.
071
0.11
0.01
558
.40.
001
0.36
200.
031
0.71
EPH
6.9
0.48
2.1
30.5
00.
002
0.00
10.
089
0.04
12.
0431
.80.
002
0.00
73.
81.
700.
904
1.10
0.06
80.
180.
091
779.
00.
002
1.12
000.
033
0.45
EPI
2.7
0.15
1.5
3.55
0.00
10.
001
0.04
00.
024
0.68
18.9
0.00
10.
004
0.1
1.92
0.19
30.
310.
061
0.03
0.01
421
.9<
0.00
10.
1040
0.01
30.
71EP
JW
ard
Hun
t Lak
e3.
80.
221.
22.
220.
001
0.01
00.
013
0.04
00.
355.
80.
002
0.00
60.
11.
420.
031
0.14
0.59
90.
030.
010
5.2
<0.
001
0.04
530.
019
0.81
EPK
Lake
A12
.80.
363.
913
.20
0.00
30.
006
0.02
00.
053
0.42
23.2
0.00
50.
013
0.2
4.51
0.04
50.
150.
023
0.07
0.00
414
.0<
0.00
10.
0282
0.07
50.
42EP
L10
.60.
226.
519
.20
0.00
10.
003
0.01
50.
051
0.48
11.6
0.00
40.
022
0.3
1.66
0.08
20.
180.
047
0.07
0.00
832
.6<
0.00
10.
0632
0.20
21.
05EP
M12
.20.
391.
271
.80
0.00
30.
008
0.04
70.
078
1.18
236.
00.
006
0.03
10.
219
.00
0.08
30.
250.
253
0.07
0.02
131
.8<
0.00
10.
1070
0.09
61.
54EP
NLa
ke C
26.
90.
010.
95.
48<
0.00
05<
0.00
050.
011
0.02
50.
2512
.10.
002
0.00
80.
10.
860.
045
0.16
0.06
50.
020.
012
8.5
<0.
001
0.02
130.
024
0.40
EPO
8.6
0.01
0.7
1.26
0.00
2<
0.00
050.
029
0.02
70.
1321
.80.
001
0.02
50.
31.
110.
015
0.21
0.07
30.
050.
011
7.3
0.00
10.
1800
0.03
20.
52EP
P69
.40.
344.
47.
100.
007
0.01
80.
136
0.10
61.
0752
8.0
0.02
20.
334
1.6
28.2
00.
060
0.73
0.51
20.
240.
048
111.
00.
001
0.06
590.
167
3.60
EPQ
20.3
1.30
1.4
4.19
0.00
30.
003
0.17
70.
062
1.31
486.
00.
007
0.02
81.
216
.50
0.16
10.
540.
273
0.05
0.12
371
.5<
0.00
10.
1290
0.07
12.
04EP
R20
.00.
062.
831
.70
0.00
2<
0.00
050.
020
0.05
80.
2225
.60.
008
0.02
70.
71.
350.
046
0.21
0.10
50.
110.
016
139.
00.
001
0.08
310.
042
0.86
EPS
13.6
0.15
2.2
4.81
0.00
30.
007
0.01
00.
066
0.27
12.7
0.00
80.
019
0.4
0.35
0.19
50.
210.
071
0.12
0.03
440
.50.
001
0.13
200.
038
1.16
EPT
11.8
0.11
2.4
6.77
0.00
10.
001
0.00
50.
069
0.24
6.6
0.00
60.
007
0.3
0.10
0.15
90.
170.
054
0.05
0.02
528
.70.
001
0.13
900.
031
2.26
EPU
Car
olyn
Lak
e35
.20.
321.
73.
690.
016
0.05
40.
030
0.08
70.
2460
.10.
020
0.02
91.
11.
040.
118
0.17
0.11
90.
100.
038
56.0
0.00
10.
3410
0.04
60.
94EP
VN
an L
ake
6.0
1.31
5.2
8.95
0.00
20.
003
0.03
80.
027
1.00
16.8
0.00
40.
006
9.2
0.73
0.30
20.
460.
075
0.40
0.19
210
7.0
0.00
11.
0900
0.03
23.
15EP
W52
.10.
611.
18.
230.
008
0.01
20.
055
0.12
00.
4590
.00.
016
0.04
80.
81.
190.
066
0.33
0.22
70.
110.
049
47.1
0.00
10.
2720
0.06
11.
40EP
X8.
40.
333.
125
.70
0.00
30.
008
0.01
40.
056
0.33
12.9
0.00
30.
010
1.6
0.58
0.04
00.
280.
048
0.06
0.06
552
.30.
001
0.19
800.
036
0.11
EPY
2.2
0.58
41.5
17.2
00.
002
0.02
40.
031
0.06
30.
574.
20.
005
0.00
216
.90.
373.
120
0.61
0.05
40.
510.
067
6270
.00.
006
1.58
000.
147
4.48
EPZ
2.3
0.26
3.2
19.6
00.
001
0.00
80.
017
0.01
60.
6427
.90.
003
0.00
31.
40.
620.
037
0.30
0.11
20.
120.
016
68.7
0.00
10.
1790
0.01
70.
71EP
AA
Ket
tle L
ake
14.0
1.27
1.5
9.64
0.00
20.
010
0.06
10.
038
0.68
45.8
0.00
80.
012
9.4
5.23
0.51
20.
530.
086
0.99
0.12
722
4.0
0.00
20.
2350
0.10
60.
91EP
AB
4.0
0.15
1.3
8.21
0.00
20.
093
0.01
20.
023
0.85
12.9
0.00
30.
004
0.3
0.76
0.23
10.
390.
059
0.05
0.01
712
.30.
001
0.01
680.
017
0.75
EPA
C26
.50.
471.
78.
070.
003
0.02
30.
062
0.09
80.
5547
.90.
010
0.01
31.
61.
340.
250
0.57
0.07
00.
130.
099
85.5
0.00
10.
7860
0.06
50.
86EP
AD
73.1
1.39
2.1
41.1
00.
006
0.03
11.
240
0.19
62.
4320
1.0
0.02
10.
022
3.7
37.5
00.
874
2.16
0.26
60.
280.
115
108.
00.
002
2.25
000.
113
4.18
EPA
E18
.41.
0231
.232
.00
0.00
20.
039
0.06
20.
128
1.77
62.2
0.00
70.
008
38.6
12.6
00.
598
2.08
0.02
80.
990.
051
472.
00.
002
2.97
000.
072
3.25
mea
n22
.30.
486.
116
.67
0.00
30.
017
0.10
30.
069
0.75
83.0
0.00
80.
032
4.4
6.44
0.33
20.
600.
139
0.25
0.05
332
1.8
0.00
10.
4754
0.07
31.
48m
edia
n11
.80.
332.
18.
950.
002
0.00
80.
036
0.05
80.
5525
.90.
005
0.01
21.
21.
660.
159
0.36
0.07
10.
110.
038
68.7
0.00
10.
1390
0.06
10.
91m
ax17
7.0
1.58
41.5
71.8
00.
016
0.09
91.
240
0.20
32.
4352
8.0
0.04
40.
334
38.6
37.5
03.
120
2.16
0.59
90.
990.
192
6270
.00.
006
2.97
000.
267
4.48
min
1.6
0.01
0.7
1.26
<0.
0005
<0.
0005
0.00
50.
016
0.13
4.2
0.00
10.
002
0.1
0.10
0.01
50.
140.
023
0.02
0.00
45.
2<
0.00
10.
0168
0.01
30.
11
EP19
Lake
Haz
en17
.50.
191.
78.
220.
002
0.00
20.
022
0.06
10.
1637
.50.
010
0.03
81.
01.
300.
136
0.44
0.06
00.
170.
026
70.5
0.00
10.
1120
0.04
50.
49
App
endi
x 7.
Tra
ce m
etal
wat
er c
hem
istry
dat
a fr
om th
e oa
sis r
egio
n no
rther
n of
Lak
e H
azen
, Elle
smer
e Is
land
. A
ll un
its a
re µ
g/L.
IDO
ffic
ial N
ame
Al
As
BB
aB
eC
dC
oC
rC
uFe
Ga
La
Li
Mn
Mo
Ni
PbR
bSb
SrT
iU
VZn
EP1
Skel
eton
Lak
e22
.70.
323.
611
.80
0.00
10.
013
0.05
50.
063
0.70
29.4
0.00
80.
010
3.8
5.45
0.16
10.
530.
140
0.22
0.00
717
8.0
<0.
001
0.29
700.
090
2.30
EP2
1.3
1.31
4.1
14.0
00.
003
0.03
50.
049
0.05
60.
3672
.30.
006
0.01
23.
110
.90
0.08
70.
430.
043
0.15
0.00
217
3.0
<0.
001
0.09
870.
078
1.52
EP3
20.5
0.33
3.9
15.1
00.
003
0.00
60.
034
0.04
50.
2645
.40.
005
0.00
73.
77.
640.
124
0.41
0.03
10.
190.
002
192.
00.
001
0.13
200.
077
1.42
EP4
8.2
1.12
1.3
16.5
00.
003
0.01
80.
158
0.14
60.
8767
.70.
005
0.02
66.
610
.60
0.25
30.
660.
159
0.99
0.06
419
0.0
0.00
10.
0830
0.25
83.
01EP
54.
41.
549.
227
.70
0.00
30.
003
0.17
00.
153
0.65
149.
00.
004
0.02
012
.14.
840.
349
0.74
0.33
11.
150.
047
295.
00.
002
0.17
500.
225
3.85
EP6
4.0
0.10
4.3
6.04
0.00
20.
004
0.11
10.
057
1.28
143.
00.
001
0.09
01.
311
.20
0.06
70.
940.
058
0.06
0.00
462
.7<
0.00
10.
0071
0.04
13.
63EP
79.
31.
5211
9.0
9.32
0.00
10.
004
0.09
50.
069
0.80
43.6
0.00
50.
009
47.6
6.20
0.30
20.
640.
082
1.02
0.05
440
6.0
0.00
10.
2920
0.26
00.
83EP
82.
01.
7840
.639
.00
0.00
7<
0.00
050.
202
0.23
90.
6520
6.0
0.00
60.
009
24.7
3.50
0.17
91.
330.
071
1.72
0.07
271
7.0
0.00
10.
6660
0.39
22.
21EP
913
.82.
0076
.250
.40
0.00
10.
037
0.38
80.
097
1.71
359.
00.
006
0.01
773
.83.
526.
250
2.97
0.22
62.
350.
157
2410
.00.
009
12.5
000
0.15
26.
38EP
1042
.10.
4611
.341
.50
0.00
30.
007
0.09
60.
146
1.38
86.5
0.01
90.
123
10.2
9.24
0.69
61.
340.
153
0.66
0.04
967
6.0
0.00
51.
5000
0.10
50.
91EP
1119
.60.
571.
022
.30
0.00
10.
006
0.15
00.
045
1.17
231.
00.
007
0.08
73.
618
.00
0.31
80.
790.
248
0.37
0.03
037
6.0
0.00
20.
4500
0.04
46.
66EP
122.
21.
2516
.432
.20
0.00
10.
018
0.21
80.
062
0.88
430.
00.
002
0.01
251
.68.
681.
380
1.80
0.06
71.
890.
080
1630
.00.
002
1.73
000.
066
2.00
EP13
9.6
1.84
6.4
39.0
00.
005
0.00
50.
198
0.11
20.
8917
8.0
0.00
60.
034
13.7
14.5
00.
334
1.13
0.11
50.
740.
093
400.
00.
001
0.20
100.
234
1.41
EP14
4.2
1.66
3.7
27.1
00.
002
0.01
20.
353
0.11
90.
9718
3.0
0.00
20.
016
11.2
20.3
01.
060
1.20
0.09
10.
680.
092
436.
00.
001
0.74
000.
125
2.01
EP15
6.9
0.85
1.1
5.89
0.00
10.
009
0.09
60.
082
0.71
68.2
0.00
50.
025
4.7
4.17
0.34
70.
440.
186
0.42
0.03
921
9.0
0.00
20.
0887
0.25
54.
34EP
164.
72.
039.
017
.40
0.00
20.
007
0.20
00.
095
0.94
92.4
0.00
70.
018
33.7
6.49
0.38
51.
420.
068
2.02
0.07
810
20.0
0.00
30.
3880
0.16
22.
32EP
1712
.71.
107.
524
.80
0.00
20.
022
0.13
50.
243
0.57
73.5
0.00
30.
014
16.5
1.93
0.12
71.
330.
159
0.49
0.06
761
2.0
0.00
10.
5940
0.16
72.
25EP
185.
91.
3315
.836
.20
0.00
20.
001
0.23
60.
093
0.62
117.
00.
010
0.01
718
.925
.50
0.16
11.
510.
108
1.64
0.05
939
1.0
0.00
90.
2310
0.14
92.
56EP
208.
51.
5722
.318
.40
0.00
30.
009
0.20
80.
126
0.97
610.
00.
007
0.02
618
.86.
400.
383
2.47
0.14
11.
620.
148
769.
00.
008
0.59
800.
212
2.78
EP21
3.0
0.94
19.2
19.5
00.
001
0.00
80.
207
0.10
60.
8427
4.0
0.00
40.
009
13.4
8.46
0.54
41.
860.
104
1.17
0.07
157
6.0
0.00
40.
4520
0.08
63.
66EP
2223
.70.
522.
74.
040.
002
0.04
30.
034
0.04
90.
5841
.50.
007
0.07
50.
59.
140.
023
0.32
0.17
20.
150.
024
84.8
<0.
001
0.02
810.
046
2.07
EP23
31.4
0.13
1.6
5.16
0.00
20.
005
0.04
40.
062
0.60
52.1
0.00
80.
146
0.3
4.50
0.02
40.
310.
161
0.14
0.01
078
.30.
001
0.01
930.
036
1.67
EP24
37.1
0.07
1.4
4.30
0.00
30.
001
0.04
00.
079
0.26
53.4
0.01
00.
169
0.2
2.92
0.03
50.
270.
173
0.12
0.01
740
.40.
001
0.03
090.
055
0.81
mea
n12
.91.
0616
.621
.20
0.00
20.
012
0.15
10.
102
0.81
156.
80.
006
0.04
216
.38.
870.
591
1.08
0.13
40.
870.
055
518.
80.
002
0.92
620.
144
2.63
med
ian
8.5
1.12
6.4
18.4
00.
002
0.00
70.
150
0.09
30.
8092
.40.
006
0.01
811
.27.
640.
302
0.94
0.14
00.
680.
054
391.
00.
001
0.29
200.
125
2.25
max
42.1
2.03
119.
050
.40
0.00
70.
043
0.38
80.
243
1.71
610.
00.
019
0.16
973
.825
.50
6.25
02.
970.
331
2.35
0.15
724
10.0
0.00
912
.500
00.
392
6.66
min
1.3
0.07
1.0
4.04
0.00
1<
0.00
050.
034
0.04
50.
2629
.40.
001
0.00
70.
21.
930.
023
0.27
0.03
10.
060.
002
40.4
<0.
001
0.00
710.
036
0.81
App
endi
x 8.
Raw
dia
tom
cou
nts f
rom
MV
-AT
core
. M
VAT
core
cou
nts
Dep
th (c
m)
0.25
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.25
6.75
7.25
7.75
8.25
8.75
9.25
9.75
10.2
510
.75
11.2
511
.75
12.2
512
.75
13.5
14.5
15.5
16.5
17.5
18.5
19.5
20.5
Achn
anth
es b
iore
tti0
02
22
Achn
anth
es c
hild
anos
72
Achn
anth
es d
aone
nsis
1Ac
hnan
thes
flex
ella
1239
115
79
812
1413
71
33
82
82
25
07
12
53
41
22
83
50
Achn
anth
es c
f hel
vetic
a (th
in)
01
Achn
anth
es g
risch
una
6Ac
hnan
thes
krie
geri
42
Achn
anth
es la
evis
85
1111
113
1720
110
198
58
103
34
72
21
32
14
63
59
2Ac
hnan
thes
mar
ginu
lata
75
24
92
1210
27
105
136
613
24
14
32
42
Achn
anth
es m
inut
issi
ma
113
419
293
219
332
219
240
306
303
245
151
104
138
6114
488
116
7294
104
8065
8363
6093
9288
107
4698
5080
81Ac
hnan
thes
pet
erse
nii
634
176
104
611
411
1112
123
1511
1916
98
92
73
56
59
66
73
Achn
anth
es s
accu
la2
11
Achn
anth
es s
ubat
omoi
des
13
92
44
48
710
210
63
166
55
811
84
75
912
62
23
43
Achn
anth
es v
entra
lis0
41
22
17
42
23
23
45
12
3Am
phor
a in
arie
nsis
37
62
27
88
11
33
41
26
45
1Am
phor
a sp
17
42
Cal
onei
s si
licul
a2
90
05
09
95
27
1214
02
21
20
00
45
60
41
60
21
26
4C
alon
eis
sp1
11
Cyc
lote
lla a
ntiq
ua0
10
10
11
11
1C
yclo
tella
sp1
(8.5
-9)
12
Cym
bella
arc
tica
1C
ymbe
lla c
esat
ii5
113
64
1110
1317
220
1435
910
1719
1421
156
1511
610
277
1623
59
68
7C
ymbe
lla c
f gra
cilis
thin
12
Cym
bella
cf b
otel
lus
22
2C
ymbe
lla c
f. st
auro
neifo
rmis
00
42
2C
ymbe
lla d
escr
ipta
012
40
02
24
186
215
60
08
66
410
70
74
313
63
14
810
56
Cym
bella
gau
eman
ii0
00
21
Cym
bella
late
ns0
1314
418
63
24
63
24
23
3C
ymbe
lla m
esia
na3
Cym
bella
min
uta
1114
163
395
2116
2910
247
373
63
146
17
24
49
132
410
7C
ymbe
lla c
f obs
cura
2233
64
208
65
25
411
2512
2712
213
198
4216
2511
1613
1220
07
177
1523
Cym
bella
cf p
roxi
ma
21
1C
ymbe
lla o
bscu
ra0
00
2C
ymbe
lla p
erpu
silla
00
00
00
2C
ymbe
lla s
ilesi
aca
03
15
43
14
23
62
1C
ymbe
lla s
p11
Cym
bella
cle
ve-e
uler
ae c
mpl
x1
1610
521
66
96
10
00
40
03
02
20
20
21
20
32
01
23
1D
iato
ma
tenu
is0
00
10
22
04
15
42
1Eu
notia
arc
us2
21
41
26
12
41
11Eu
notia
exi
gua
01
23
3Eu
notia
mon
odon
24
Euno
tia p
raer
upta
213
96
176
87
36
03
02
87
74
316
46
1111
924
1312
1716
2228
1720
Frag
ilaria
cap
ucin
a va
r cap
ucin
a39
7723
4746
3344
4471
6411
516
520
417
523
827
825
525
422
328
317
116
622
017
911
520
615
117
023
857
180
119
211
81Fr
agila
ria c
apuc
ina
var g
raci
lis19
4120
25
28
74
144
42
61
42
12
Frag
ilaria
exi
gua
00
00
12
1Fr
agila
ria p
inna
ta2
App
endi
x 8.
con
tinue
d.
MVA
T co
re c
ount
sD
epth
(cm
)0.
250.
751.
251.
752.
252.
753.
253.
754.
254.
755.
255.
756.
256.
757.
257.
758.
258.
759.
259.
7510
.25
10.7
511
.25
11.7
512
.25
12.7
513
.514
.515
.516
.517
.518
.519
.520
.5G
irdle
s 1
10G
omph
onem
a la
gerh
eim
i0
2G
omph
onem
a m
inut
um0
41
45
22
23
14
42
67
21
24
21
4G
omph
onem
a sp
110
Nav
icul
a ch
iara
e0
23
14
32
212
24
11
2N
avic
ula
cf rh
ynco
ceph
ala
00
2N
avic
ula
cryp
toce
phal
a1
57
21
22
13
2N
avic
ula
dilu
vian
a1
Nav
icul
a ni
valis
1N
avic
ula
prot
ract
a0
00
00
02
Nav
icul
a pu
pula
022
1012
1616
54
53
512
1615
611
2119
1212
107
1112
813
107
163
153
208
Nav
icul
a st
roem
ii2
21
Nav
icul
a tu
scul
a0
00
03
14
31
15
4N
avic
ula
vulp
ina
05
91
42
24
22
31
21
11
Nav
icul
a sp
ecie
s 2
4N
av s
p1(0
.5-1
)2
Nav
sp1
2N
av s
p21
Nav
sp3
1N
av s
p41
Nei
dium
alp
inum
2N
eidi
um a
mpl
iatu
m2
21
22
44
21
58
154
46
21
45
57
79
42
813
75
Nei
dium
sp1
12
Nei
dium
bis
ulca
tum
1N
itzsc
hia
cf a
urar
iae
2N
itzsc
hia
inco
nspi
cua
332
1510
45
76
1010
102
64
14
45
22
27
44
24
Nitz
schi
a cf
per
min
uta
(hea
vy s
triae
)3
Nitz
schi
a fru
stul
um15
106
815
30
66
40
625
100
30
00
211
20
18
220
36
22
23
6N
itzsc
hia
perm
inut
a12
014
781
7216
178
6162
3835
5523
7641
4440
4421
2636
2614
2219
2313
1626
224
1814
2631
Nitz
schi
a pe
rmin
uta
long
04
04
00
00
00
00
00
30
00
00
00
00
00
00
20
00
00
Nitz
schi
a gi
rdle
s (1
.5-2
)8
2Pi
nnul
aria
ang
usta
13
71
63
74
46
146
84
86
88
1010
16
147
182
93
101
Pinn
ular
ia b
alfo
uria
na2
Pinn
ular
ia d
escr
esce
ns v
ar ig
nora
ta1
21
Pinn
ular
ia d
iver
gent
issi
ma
var s
ubro
stra
ta2
11
84
94
31
4Pi
nnul
aria
div
erge
ns v
ar d
iver
gens
13
2Pi
nnul
aria
gib
ba v
ar m
esog
ongy
la3
1Pi
nnul
aria
inte
rmed
ia3
2Pi
nnul
aria
stre
ptor
aphe
12
11
73
65
33
Pinn
sp1
7Pi
nn s
p24
1Pi
nn s
p31
Pinn
ular
ia in
terr
upta
00
21
12
11Pi
nn s
mal
l (6.
5-7)
4pi
nn s
p a
2pi
nn s
p b
(9.5
-10)
2sp
ecie
s a
1St
auro
neis
anc
eps
03
22
62
16
22
44
13
42
31
35
7St
auro
neis
pho
enic
ente
ron
03
57
35
51
19
21
33
29
23
15
76
41
Stau
rone
is c
f obt
usa
12
21
Surir
ella
sp1
00
1Ta
bella
ria fl
occu
losa
25
18
69
22
51
12
21
14
21
6TO
TAL
411
1003
587
468
769
448
485
567
589
497
468
432
658
399
561
555
596
476
487
575
415
374
455
367
296
512
380
431
538
176
444
342
470
312
Cys
ts56
107
94
114
44
24
43
App
endi
x 9.
Raw
dia
tom
cou
nts f
rom
the
Skel
eton
Lak
e (E
P1) s
edim
ent c
ore.
Sk
elet
on L
ake
(EP1
) cor
e 29
5-JP
S-03
Inte
rval
(cm
)0.
251.
252.
253.
254.
255.
256.
257.
258.
259.
2510
.25
11.2
512
.25
13.2
514
.25
15.2
516
.25
17.2
518
.25
19.2
520
.25
21.2
522
.25
23.2
524
.25
25.2
526
.25
27.2
528
.25
29.2
530
.25
31.2
532
.25
Achn
anth
es b
ahus
iens
is5
1Ac
hnan
thes
chi
ldan
os
3Ac
hnan
thes
cur
tissi
ma
13
Achn
anth
es in
grat
iform
is1
Achn
anth
es k
riege
ri1
2Ac
hnan
thes
min
utis
sim
a1
2Ac
hnan
thes
pet
erse
nii
23
24
26
21
85
33
Achn
anth
es v
entra
lis6
1C
alon
eis
silic
ula
22
Cyc
lote
lla a
ntiq
ua2
1C
ymbe
lla c
esat
tii2
2C
ymbe
lla tu
mid
ula
1D
entic
ula
kuet
zing
ii1
2D
iato
ma
mon
ilifor
mis
12
12
Dia
tom
a te
nuis
22
Frag
ilaria
bre
vist
riata
9413
216
315
4720
7616
1813
3339
1016
1511
927
917
1719
1925
2823
6110
145
3848
2739
Frag
ilaria
con
stru
ens
var v
ente
r com
plex
815
965
1170
503
637
397
825
778
653
847
704
763
1053
426
442
543
942
853
727
559
418
925
584
609
988
541
487
1115
547
643
614
489
636
Frag
ilaria
pin
nata
82
65
96
1026
97
1628
5053
9512
350
3728
2547
6763
2346
4254
4613
4121
1017
Frag
ilaria
pse
udoc
onst
ruen
s7
52
32
23
112
144
115
56
012
516
875
6526
101
212
Nav
icul
a ch
iara
e2
21
Nav
icul
a cr
ypto
ceph
ala
2N
avic
ula
gerlo
ffi1
Nav
icul
a pu
pula
11
2N
avic
ula
vulp
ina
12
1N
itzsc
hia
perm
inut
a2
16
2N
itzsc
hia
pusi
lla4
1114
13
23
42
11
2Pi
nnul
aria
bre
biso
nni
6Pi
nnul
aria
stre
ptor
aphe
23
Step
hano
disc
us m
ediu
s3
SUM
935
1123
1363
524
694
427
916
827
687
873
765
832
1133
497
556
688
1015
924
770
603
504
1019
706
667
1138
684
634
1279
608
728
697
527
704
Cys
ts9
48
11
11
21
12
23
42
Appendix 10. Raw counts from EP2 sediment core. EP2 core 297-JPS-03 Interval (cm)
0.25 1.25 2.25 3.25 4.25 5.25 6.25 7.25 8.25 9.25 10.25 11.25 12.25 13.25 14.25 15.25 16.25Achnanthes childanos 4 2 3Achnanthes flexella 1 4 2 1 1 1 1Achnanthes laevis 2Achnanthes marginulata 2 1 2Achnanthes minutissima 9 11 23 33 45 61 105 48 5 2 11 7 1 6Achnanthes petersenii 1 1 5 14 12 9 16 28 24 10 3 5 2 1Achnanthes scotica 1Achnanthes ventralis 2Amphora inariensis 2Brachysira zellensis 4 3 2 4 2 4 4 2Caloneis silicula 6 5 16 13 22 18 26 32 9 2 7 15Cyclotella antiqua 6 2 3 2 10 1 1Cyclotella stelligera 1Cymbella angusta 4 7 6 11 8 29 21 11 2 2 2 4 5Cymbella austriaca 2 2Cymbella bottellus 1 4 4 7 2 4 1Cymbella cesattii 3 6 5 10 10 2Cymbella cleve-eulerae 1 2 3 10 1 4 4 9 2Cymbella fogedii 2 1 6 10 6 12 9 1 1 2Cymbella microcephala (Nav soehensis) 2 14Cymbella minutum 5 2 6 4 11 1 1 2Cymbella obscurum 1 8 2 2 1 1 4 1 2Cymbella proxima 1 1Cymbella silesiaca 2Cymbella subaequalis 2Cymbella tumidula 2 1 1 8 6 9Cymbopleura sp nov? Dermot P34 2 9 2 10 10 4 3 2 2 2 2 2Denticula kuetzingii 5 41 117 123 128 145 82 61 100 77 17 26 14 11 11 8 10Diadesmis contenta 3 1 2 2Diatoma moniliformis 1 1 1Epithemia sorex 5 4 5 4 4 1Eunotia praerupta 1 14 7 8 3 14 6 4 2Fragilaria brevistriata 969 598 474 246 86 75 21 4 16 52 60 66 78 104 83 56 62Fragilaria capucina 1 1 2 4 6 17 18 5 3Fragilaria capucina var gracilis 1 2Fragilaria construens var venter 3 2 2 17 39 40 27 64 60 48 70Fragilaria pinnata 2 10 8 52 72 82 91 96 175 337 568 475 365 541 418 471 465Fragilaria pseudoconstruens football 7 2 19Gomphonema parvulum 2 1 9 2Gomphonema tricatum 1 2Navicula cf angusta 1Navicula cf incerta 2 4 2 1Navicula chiarae 3Navicula cryptocephala 4 1 1Navicula gerloffi 1 1 1 1Navicula pseudoscutiformis 2 1 2 6 2 2Navicula soehernsis 4 38 37 46 13 7 7 6 2 1 4Navicula tuscula 2 3Navicula vulpina 2 3 2 1Nitzschia alpina 2Nitzschia frustulum 16 19 28 4 1Nitzschia inconspicua 5 46 88 96 59 83 54 56 21 1 19 2 2 6Nitzschia palaceae 8 15 10 4 8 5Nitzschia perminuta 9 31 29 65 89 95 100 110 83 18 20 20 20 16 4 12Nitzschia pura 1 1Nitzschia pusilla 2Pinnularia balfouriana 4 2Pinnularia streptoraphe 2Stauroneis anceps 1SUM 988 676 728 629 598 602 585 587 833 758 735 658 578 783 599 596 671Cysts 1 0 2 3 6 6 6 5 10 6 1 1 1 5 3 0
1
1
2
2
3
Appendix 11. 210Pb data (based on alpha spectroscopy from MyCore Scientific, Inc. Deep River, ON) and CRS dates with errors for the Skeleton Lake sediment core.
midpoint year standard deviation Total 210 Pb
(cm) (AD) (years) (pCi/g) 0.25 2001 3.38 1.874780.75 1998 3.55 1.6030241.25 1994 3.805 1.5392161.75 1987 4.32 1.6545822.25 1980 5.01 1.2705132.75 1973 5.91 1.2623663.25 1963 7.63 1.1907383.75 1946 11.985 1.1490624.75 1877 133.14 0.979254
Appendix 12. Total Hg and total Pb for the Skeleton Lake sediment core.
0
5
10
15
20
25
30
35
40
45
50
0 5 10 15 20 25 30 35
depth in core (cm)
conc
entra
tion
Hg &
Pb
(tota
l in
seds
)
Hg (ppb)Pb (ppm)
0
20
40
60
80
100
120
140
160
180
0 5 10 15 20 25 30 35
depth in core (cm)
conc
entra
tion
Hg &
Pb
(exp
ress
ed r
elat
ive
to g
or
gani
c C
)
Hg (ppb)Pb (ppm)
0
2
4
6
8
10
12
0 5 10 15 20 25 30 35
depth in core (cm)
conc
entr
atio
n Hg
& P
b (e
xpre
ssed
rela
tive
to %
or
gani
c C
)
Hg (ppb)Pb (ppm)
App
endi
x 13
. R
aw c
ount
s fro
m C
ape
Ver
a su
rfac
e se
dim
ents
(mul
tiple
yea
r dat
a).
Site
CV1
CV2
CV3
CV4
CV5
CV5
CV6
CV6
CV7
CV8
CV9
CV9
CV9
aC
V10
CV1
1C
V12
CV1
3C
V14
CV1
5C
V16
CV1
7C
V18
CV2
0C
V22
CV2
3C
V24
CV1
3C
V31
Year
2004
2004
2004
2004
2004
2005
2004
2005
2004
2004
2004
2005
2005
2004
2004
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2006
2006
Achn
anth
es b
roen
lund
sis
3Ac
hnan
thes
chl
idan
os23
2Ac
hnan
thes
flex
ella
1050
Achn
anth
es k
riege
ri2
817
615
15Ac
hnan
thes
kry
ophi
la1
Achn
anth
es la
evis
21
820
Achn
anth
es m
argi
nula
ta3
175
22
252
1117
1116
Achn
anth
es m
inut
issi
ma
108
412
415
3151
6139
6Ac
hnan
thes
pet
erse
nii
22
53
62
819
16Ac
hnan
thes
sco
tica
2Ac
hnan
thes
ven
tral
is2
Aneu
mas
tus
tusc
ulus
2Am
phor
a co
pula
ta2
2Am
phor
a du
seni
i24
51
2Am
phor
a sp
itzbe
rgen
sis
5Am
phor
a ve
neta
92
Cal
onei
s si
licul
a2
3C
alon
eis
silic
ula
sp16
pla
te 2
09
2C
alon
eis
silic
ula
sp18
pla
te 2
04
Cal
onei
s sp
cf s
ilicu
la s
p9-1
4 pl
ate
2011
84
Cha
mae
pinn
ular
ia g
andr
upii
6C
occo
neis
sp
362
2C
occo
neis
sp2
1C
occo
neis
spC
V18
1C
ratic
ula
ambi
gua
525
Cyc
lote
lla a
ntiq
ua1
Cyl
indr
othe
ca g
raci
le1
Cym
bella
ang
usta
ta11
930
3782
12C
ymbe
lla a
rctic
a3
Cym
bella
bot
ellu
s7
25
811
29
231
1320
188
Cym
bella
cf e
xcis
a2
22
Cym
bella
cle
ve-e
uler
ae13
937
1524
197
1396
3333
1929
5510
2440
479
Cym
bella
des
crip
ta16
5C
ymbe
lla d
esig
nata
22
158
810
2C
ymbe
lla e
lgin
ense
2C
ymbe
lla fo
gedi
i12
2729
6498
3819
279
186
5311
820
2974
2213
1020
2029
Cym
bella
late
ns5
312
19
12
17
4C
ymbe
lla m
icro
ceph
ala
46
223
108
947
Cym
bella
min
uta
2727
235
1211
3014
73
747
1118
1032
212
716
Cym
bella
pau
cist
riatu
m2
10C
ymbe
lla p
roxi
ma
11
3C
ymbe
lla s
ilesi
aca
22
213
95
118
29
42
63
41
141
Cym
bella
sile
siac
a C
V14
164
Cym
bella
sub
aequ
alis
711
112
Cym
bella
sta
uron
eifo
rmis
92
Cym
bopl
eura
cus
pida
ta2
App
endi
x 13
. con
tinue
d.
Site
CV1
CV2
CV3
CV4
CV5
CV5
CV6
CV6
CV7
CV8
CV9
CV9
CV9
aC
V10
CV1
1C
V12
CV1
3C
V14
CV1
5C
V16
CV1
7C
V18
CV2
0C
V22
CV2
3C
V24
CV1
3C
V31
Year
2004
2004
2004
2004
2004
2005
2004
2005
2004
2004
2004
2005
2005
2004
2004
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2006
2006
Den
ticul
a ku
etzi
ngii
148
429
1119
Den
ticul
a su
btili
s2
Den
ticul
a te
nuis
911
Der
mot
sp1
7 pl
ate
4931
Dia
desm
is c
onte
nta
898
131
4D
iade
smis
par
acon
tent
a2
Dia
tom
a te
nuis
211
1717
58
24
266
653
415
212
Dip
lone
is m
arge
nies
tria
ta16
05
Dip
lone
is p
arm
a2
Ency
onem
a sp
4 pl
ate
283
112
Frag
ilaria
cap
ucin
a14
812
61
938
13
72
4033
Frag
ilaria
con
stru
ens
var v
ente
r1
Gei
ssle
ria s
choe
nfel
dii
212
Gei
ssle
ria s
p1
Gei
ssle
ria s
p19-
23 p
late
50
2G
eiss
leria
sp1
3,14
pla
te 5
04
Gom
phon
ema
lapp
onic
a de
rmot
10G
omph
onem
a pr
oduc
tum
18H
anna
ea a
rcus
102
41
Mic
roco
stat
us k
rass
kei
175
3330
25
9126
71
114
21N
avic
ula
chia
rae
214
115
755
169
Nav
icul
a ci
ncta
318
Nav
icul
a cr
ypto
ceph
ala
811
212
2N
avic
ula
cryp
toce
phal
a 2
4N
avic
ula
gerlo
ffi5
Nav
icul
a gr
egar
ia19
Nav
icul
a m
enis
culu
s2
Nav
icul
a ph
ylle
pta
102
42
72
102
517
Nav
icul
a ps
eudo
tene
lloid
es2
Nav
icul
a sa
linar
um5
34
214
18N
avic
ula
spp
aff r
einh
ardt
ii9
Nav
icul
a sp
p af
f ven
eta
26
26N
avic
ula
vulp
ina
2315
614
144
531
2512
48
11N
eidi
um a
mpl
iatu
m1
2N
eidi
um d
istin
cte-
punc
tatu
m6
23
1N
eidi
um k
ozlo
wii
2N
itzsc
hia
diss
ipat
a va
r med
ia22
12
27
17
5N
itzsc
hia
frus
tulu
m18
710
910
831
.316
.742
.550
248
326
416
221
228
288.
133
1.4
33.1
442
.79
368.
644
2.9
383.
738
3.6
044
6.8
286.
115
533
2.9
183
42N
itzsc
hia
perm
inut
a35
670
.419
720
38.
3312
011
061
.511
523
.752
.949
.214
5.9
174.
662
.86
44.2
181
.37
73.1
313
0.3
76.4
254
417
.19
82.8
810
412
7.1
7166
Nitz
schi
a ho
mbu
rgie
nsis
112
4N
itzsc
hia
pale
a15
34N
itzsc
hia
pala
ecea
125
1421
112
Nitz
schi
a pu
ra3
1834
702
154
314
Nitz
schi
a su
chla
ndii
204
Pinn
ular
ia b
rebi
sonn
i sp9
pla
te61
1014
Pinn
ular
ia h
umili
s8
Sella
phor
a ba
cillu
m1
Stau
rone
is a
ncep
s1
12
Stau
rone
is p
hoen
icen
tero
n1
1TO
TAL
721
614
510
469
460
394
795
684
468
476
633
574
450
553
687
617
066
072
167
454
654
558
364
168
067
947
537
0
Appendix 14. Raw diatom counts from CV5 core . CV5 core1
0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5Achnanthes broenlundensis 2 1Achnanthes chlidanos 13 14 4 3 2Achnanthes flexella 1 1Achnanthes girdles 4 14 6 2 10 28Achnanthes kreigeri 9Achnanthes kryophila 2Achnanthes marginulata 10 5Achnanthes minutissima 1 4 2 2 5 4 12Achnanthes petersenni 3 6 2 10 7 8 9 4 3Achnanthes scotica 2Adlafia bryophila 2 2Amphora copulata 2 2 2Amphora dusenii 2 4 6 1 1 2 2Caloneis sp cf silicula dermot 9-14 plate 20 4Caloneis sp cf silicula 2 4 2Cocconeis sp 3Cyclotella pseudostelligera 2 2Cymbella amphicephela 3 2 2Cymbella gauemanni 1 1Cymbella lange-bertalotti 2Cymbella latens 1 3 9 10 11 8 3 2Cymbella microcephala 7 1 4 2 1 2Cymbella silesiaca 42 41 22 27 22 20 34 52 21 32Cymbella large girdle 3Cymbella fogedii 118 83 80 64 91 75 77 63 37 42Cymbella minuta 100 23 18 18 33 36 43 64 19 24Cymbella bottelus 20 0 1 0 2 2 0 0 3 7Cymbella cleve-eulerae 8 6 8 13 7 16 6 13 8 12Cymbopleura cuspidata 4 3Diatoma tenuis 1 3 8 7 10 4 4 7 7Fragilaria capucina 4 4Hannaea arcus 2Hippodonta arctica 2Luticola mutica var ventricosa 2 2 2Microstarus krasskei 14 15 9 11 17 14 8 1Navicula chiarae 34 21 14 10 3 8 1 3 8 5Navicula cryptocephala 2 2 2 2 4Navicula phyllepta 7 19 19 4 15 16 7 6 10 3Navicula salinarum 15 13 6 6 10 2 3 2Navicula sp aff veneta 14 11 2 4 2Navicula sm girdles 2Navicula vulpina 1 5 2 8 9 6 4 3 11 9Navicula sp 29-38 plate 47 dermot 8Neidium distincte-punctatum 2 2 2Nitzschia homburgensis 2Nitzschia frustulum 44 85 114 169 238 236 175 148 76 241Nitzschia perminuta 95 68 88 67 204 223 201 168 226 64Nitzschia paleacea 3Nitzschia pura 3 10 4 1 2 2 3Pinnularia grunowii 2Pinnularia krammeri 1 3Pinnularia sp 2-4 plate 63 dermot 4Pinnularia vasta 1 2Stauroneis anceps 3 6 7Stauroneis phoenicenteron 8 2 2 2 1TOTAL 561 454 441 470 690 697 596 562 463 517cysts 7 5 3 9 6 4 4 4 2 2
Appendix 15. Raw diatom counts from CV6 core 1.
CV6 core10.25 1.25 2.25 3.25 4.25 5.25 6.25 7.25 8.25 9.25 10.25 11.25
Achnanthes flexella 4Achnanthes girdles 13 8 2 4 10 4Achnanthes kreigeri 2Achnanthes marginulata 2Achnanthes minutissima 4 2 6 3 2 1 7 15 7 10 1Achnanthes petersenni 2 2 2 3Achnanthes subatomoides 2Caloneis sp cf silicula 2 2 2 2 4Cyclotella pseudostelligera 6 1Cymbella angustata 1 8 4 2 3 3 2 7 6 2Cymbella amphicephela 2Cymbella botellus 4 3 12 2 27 23 82 15 69 77Cymbella cleve-eulerae 13 8 4 4 4 2 7 2 3 12 11 2Cymbella designata 7 6 4 2 1Cymbella latens 6 5 10 7 3 2 9 2Cymbella microcephala 8 8 14 17 5 5 6 14 12 6 38 13Cymbella proxima 2 2Cymbella silesiaca 2 9 3 6 7 2 2 2 5 7Cymbella fogedii 34 34 10 21 14 15 10 12 6 15 11 0Cymbella minuta 4 4 4 7 0 8 8 3 3 12 4Cymbella subaequalis 2 4 6 6 2 7 4 2Cymbella stauroneiformis 13 13 13 10 6 2 18 2Denticula kuetzigii 2 4Diadesmis contenta 2Diatoma tenuis 3 1 2 2 2 2 1 1Fragilaria capucina 6 12 4 5 9 4 13 1Fragilaria construens var venter 19Fragilaria pseudoconstruens 3Microstarus krasskei 3 4 2 1Navicula chiarae 23 11 16 11 13 2 3 7 2 23 3Navicula phyllepta 2Navicula sm girdles 2 7 2 4 4 2 2Navicula vulpina 11 12 24 19 33 26 58 45 61 20 54 26Neidium ampliatum 2 2 5 5 1 7 1 7Nitzschia dissipata var media 3Nitzschia frustulum 569 504 389 302 393 373 178 250 163 493 188 125Nitzschia perminuta 99 165 83 80 52 58 35 60 21 73 49 29Nitzschia sp cf acicularis 2Stauroneis phoenicenteron 1 3 1 1 2 1weird cv6 1TOTAL 795 809 606 509 578 538 364 460 406 747 471 369cysts 1 1 4 1 6 3 1 5 2 12 8 1
4
63
3
2
32
2
6
2
0
Appendix 16. Raw diatom counts from CV7 core 1.
CV7 core10.25 1.25 2.25 3.25 4.25 5.25 6.25-18.25 19.25
Achnanthes marginulata 7 11 4 12 2 NOAchnanthes minutissima 1 DIATOMSAchnanthes petersenni 2 2Amphora inariensis 2Caloneis silicula 2 9 4 2 4Caloneis silicula big 4Cocconeis sp 24Cymbella angustata 2 2Cymbella botellus 16 12 30 29 14 57 12Cymbella cleve-eulerae 67 31 30 19 27 30 4Cymbella designata 2 3 6 2Cymbella lange-bertalotti 6Cymbella latens 22 4 5 7 5 8Cymbella silesiaca 13 9 4 1Cymbella fogedii 41 21 63 49 25 24 4Cymbella minuta 13 21 30 21 14 4 0Cymbopleura cuspidata 2Diadesmis contenta 2 3Diploneis marginestriata 4 4 2 1Diploneis parma 6 8 4 2 1Fragilaria capucina 27 14 4 6 6 5Geissleria sp 19-23 plate 50 dermot 2Luticola mutica var ventricosa 2Microstarus krasskei 46 12 2 28 2Navicula chiarae 104 34 49 23 19 15 10Navicula cryptocephala 2Navicula phyllepta 4 6 14 23 24 2 2Navicula pseudotenelloides 4 6 16Navicula sp aff veneta 48 9 13 24 7 8 2Navicula sm girdles 2 2 8 2 6Navicula vulpina 6 12 9 10 2 4Navicula sp 29-38 plate 47 dermot 40 9 3 15Neidium ampliatum 5 4Neidium bergii 4 3Neidium distincte-punctatum 5 2Neidium kozlowii 2 2Nitzschia homburgensis 2 1Nitzschia frustulum 73 53 176 46 82 23 10Nitzschira perminuta 233 72 133 140 103 135 30Nitzschia palea 10 2 2 4Nitzschia pura 6Nitzschia sp cf acicularis 2unidentified (looks like peanut in the shell) 6Pinnularia biceps 2Pinnularia vasta 4 3 2 8Pinnularia large girdle CV7 4Sellaphora bacillum 4 2 7 2Stauroneis anceps 2 1TOTAL 766 333 639 519 395 341 0 121cysts 15 3 5 7 6 8
8
Appendix 17. Raw diatom counts from CV9 core 1.
CV9 core 10.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 17.5 18.5 19.5 20.5 21.5 22
Achnanthes marginulata 1Achnanthes petersenni 2Aneumastus tusculus 2Amphora dusenii 4 4Amphora inariensis 1 2Caloneis sp cf silicula 15 6 6 8 4 4 2 6 4 1 4 3 4 1 4 1 2Cocconeis sp 1 1 1 8 2 1 2Cymbella botellus 5 2 6 1 2 3Cymbella cleve-eulerae 2 1 3 6 4 6 19 14 11 15 5 14 9 12 12Cymbella latens 1 2 4 3 5 1 8 4 8 8 7 8Cymbella microcephala 5 7 13 5 5 7 6 2 5 1 7 6 3 6 2 4 23 6 15 3 1Cymbella fogedii total 60 102 85 83 84 92 59 49 50 56 33 46 28 33 10 22 9 15 21 22 11 11 11Cymbella minuta total 54 82 73 35 37 57 42 13 12 11 9 11 17 10 1 4 5 6 4 0 11 0 6Cymbella silesiaca 30 15 30 24 23 46 33 22 30 16 7 5 11 23 9 13 5 12 8 12 7 16 11Cymbella subaequalis 1Cymbella tynni 1Cymbella CV8 1Cymbopleura cuspidata 2Denticula kuetzigii 2 2Diadesmis contenta 2 10Diatoma tenuis 3 1 1Fragilaria brevistriata 2Fragilaria capucina 1 2 1Luticola muticaMicrostarus krasskei 8 19 14 4 2 2 2 1Navicula chiaraeNavicula cryptocephala 23 7 6 2Navicula menisculus 2Navicula muticola 1 1 1 1Navicula phyllepta 69 86 60 38 9 21 11 8 2 3 2 2 1 1 2Navicula salinarum 4 2Navicula trivalis 12 1Navicula tuscula 3 2Navicula vulpina 14 24 25 30 32 51 65 66 78 67 74 40 50 50 27 32 20 29 46 34 28 41 24Navicula large central area 5 1Neidium ampliatum 1 4 7 5 2 6 5 3 1 6 1 1 6Neidium distincte-punctatum 2 1 8 1 2 1 1 2 1 3Neidium kozlowii 1 1Nitzschia dissipata var media 3 2 5 3 4 5 2 4 5 8Nitzschia frustulum total 174 212 174 165 214 269 199 214 249 211 186 192 133 200 242 223 294 210 309 265 227 234 225Nitzschia perminuta total 61 80 32 20 18 60 29 40 22 53 27 34 51 45 28 14 38 49 21 38 11 64 22Nitzschia pusilla 4 2Pinnularia krammeri 1Sellaphora bacillum 1Stauroneis anceps 1Stauroneis phoenicenteron 4 4 2 4Big punctae CV8 1TOTAL 525 652 516 428 446 627 454 433 450 431 353 352 318 386 351 338 414 352 467 414 340 412 332cysts 6 5 3 2 2 3 4 1 3 1 2 4 7 5 4 5 21 6 8 18 2funky fragments 1 1 3 2 3 12 9 4 9 3 1 2
2
3
1
Appendix 18. Raw diatom counts from CV9a core 1.
CV9 core1 20050.25 1.25 2.25 3.25 4.25 5.25 6.25 7.25 8.25 9.25 10.25 11.25 12.25
Achnanthes marginulata 4 2 2 2 4Achnanthes minutissima 1Achnanthes ventralis 1Amphora inariensis 4Centric 1 1Cyclotella antiqua 1 1Cyclotella bodanica 2Cyclotella ocellata 1Cyclotella pseudostelligera 2 1 1 3 35Cymbella angustata 2Cymbella botellus 2Cymbella cleve-eulerae 2Cymbella fogedii 12 3 1 4Cymbella minuta 9 7 6 14 18 5 5 1 2 9 12 6Cymbella proxima 2 1 4 1 2Cymbella silesiaca 1 6 4 9 10 2 1 4Fragilaria capucina 2 2 1Fragilaria capucina var gracilis 12Fragilaria construens var venter 3 4 4 2 19Fragilaria pinnata 18Gomphonema productum 3Navicula trivalis 1Neidium ampliatum 2Nitzschia dissipata var media 5 2 8 5Nitzschia frustulum 473 573 504 581 440 309 293 256 46 46 87 77 45Nitzschia perminuta 76 138 148 146 150 150 182 89 54 11 59 11 10unidentified 1 1 1unidentified 2 1unidentified 3 3unidentified weird stick 2 1TOTAL 567 735 671 743 613 473 499 357 108 65 169 114 175cysts 32 17 14 28 6 7 4 2 1 297 6 8funky fragments 1 7 7 3 12 5
8
Appendix 19. Raw diatom counts from CV12 core 1.
CV12 broken broken broken broken broken0-0.5 1-1.5 2-2.5 3-3.5 4-4.5 5-5.5 6-6.5
Achnanthes girdle 4Achnanthes petersenii 2 2 1Amphora inariensis 4 2Amphora veneta 4 3Cocconeis sp 1 5 6 7 11Cocconeis (neodiminuta?) 3 2Cyclotella bodanica 1Big centric 3 1Cymbella cleve-eulerae 13 19 7 2 3Denticula kuetzigii 68 70 93 9 26 2 8Denticula tenuis 2Diatoma tenuis 4 10 16 12 3 1 2Diploneis didyma 4Diploneis big 3 1Fragilaria exigua 1 1Luticola mutica 3Microcostatus krasskei 6 6 1 1Navicula large centres 18 29 20Navicula rhyncocephala 7 27 5 2 1 1Navicula chiarae 1 2 2Navicula phyllepta 2 5 33 17 6 3Navicula vulpina 2Neidium large centres 33 73 95 72 7Nitzschia communtata 5 4 14 2 2Nitzschia frustulum 362 194 180 20 46 6 11Nitzschia palaecea 2Nitzschia perminuta 71 35 25 8 5 4Nitzschia pusilla 2TOTAL 536 362 444 153 209 130 150cysts 11 17 32 2 2 9 6funky fragments 3 14 13 10 27
1
App
endi
x 20
. Raw
dia
tom
cou
nts f
rom
CV
13 c
ore
1.
C
V13
0-0.
51-
1.5
2-2.
53-
3.5
4-4.
55-
5.5
6-6.
57-
7.5
8-8.
59-
9.5
10-1
0.5
11-1
1.5
12-1
2.5
13-1
3.5
Amph
ora
hols
taic
a4
44
22
Cal
onei
s si
licul
a2
Coc
cone
is s
p2
13
3C
ratic
ula
halo
phila
2C
yclo
tella
bod
anic
a1
Cym
bella
bot
ellu
s4
Cym
bella
cle
ve-e
uler
ae47
6574
3325
99
13
53
72
8C
ymbe
lla fo
gedi
i20
100
122
8962
5156
4147
5446
7858
43C
ymbe
lla m
inut
a12
749
544
1815
185
119
911
1811
Cym
bella
neo
cist
ula
62
Cym
bella
sile
siac
a14
3030
2012
37
717
154
76
Cym
bella
sm
all g
irdle
146
82
122
Dia
desm
is c
onte
nta
1D
iato
ma
tenu
is2
114
126
1930
2332
5820
5018
Frag
ilaria
cap
ucin
a4
Mic
roco
stat
us k
rass
kei
1N
avic
ula
rhyn
coce
phal
a2
23
21
Nav
icul
a vu
lpin
a11
1714
118
28
14
Nitz
schi
a fru
stul
um18
316
318
618
628
229
230
322
027
129
125
233
231
827
7N
itzsc
hia
perm
inut
a71
105
105
107
6042
3748
101
8164
9138
72TO
TAL
475
538
566
497
483
424
462
362
477
491
443
567
500
436
cyst
s11
227
62
115
156
716
67
Appendix 21. Raw diatom counts from CV20 core.
CV20 core 10.25 1.25 2.25 3.25 4.25 5.25 6.25 7.25
Cocconeis spCV20 1Cyclotella pseudostelligera 23 2 2Cymbella affinis 2Cymbella botellus 5 2 3 2Cymbella cleve-eulerae 25 30 36 21 23 29 20 25Cymbella fogedii 4 5 14 1 4Cymbella minuta 34 33 47 41 29 11 9 13Cymbella proxima 6 2 4 7 3Cymbella silesiaca 11 6 3 7 4 7 13 1Cymbella small girdle 6Fragilaria brevistriata 2Fragilaria capucina 2Fragilaria capucina var gracilis 1Fragilaria construens var venter 8 1 2Fragilaria pinnata 2 2Nitzschia frustulum 270 428 424 306 276 279 266 247Nitzschia perminuta 50 60 44 13 9 10 5 8TOTAL 430 579 572 389 350 342 325 328cysts 59 78 71 64 86 120 228 338
4
5
6
Appendix 22. Raw diatom counts from CV22 core.
CV22 core1 20050.25 1.25 2.25 3.25 4.25 5.25 6.25 7.25
Achnanthes flexella 12 2Achnanthes kriegeri 0 3 0 0 0 0 0 0Achnanthes laevis 0 1 3 0 2 0 1 0Achnanthes marginulata 0 6 0 0 0 0 0 0Achnanthes minutissima 58 23 26 15 10 13 16 5Achnanthes petersenni 11 3 4 3 0 3 20Amphora dusenii 2Amphora copulata 1 4Caloneis sp cf silicula 1Cocconeis sp 5 10 4 53 16 18 73Cocconeis small 8 7 4 3Cyclotella pseudostelligera 1Cymbella angustata 3 3 5 3Cymbella botellus 14 7 13 19 9 10 4 2Cymbella cleve-eulerae 13 22 5 4 9 13 12 2Cymbella descripta 7 5 2 6 3Cymbella designata 2 6 2Cymbella elginense 5Cymbella fogedii 17 11 6 4 3 13 7 2Cymbella microcephala 3 4 2 6 2 2Cymbella obscurum 1 1Cymbella proxima 1Cymbella silesiaca 2 2Cymbella subaequalis 7 4 2 1 2Cymbella tynni 2Cymbopleura cuspidata 2Denticula kuetzigii 8 1 1 7 6 4Denticula tenuis 31 21 8 5 7 2Diatoma tenuis 8 11 9 6 6 10 6 2Diploneis margenestriata 2Diploneis parma 1Fragilaria brevistriata 1Fragilaria capucina 4 2 3 4 5Fragilaria capucina var gracilis 1Large central 1 1Muelleria gibbula 1Navicula chiarae 10 6 2 12 2 2Navicula cryptocephala 2 4 2 6 4 3Navicula digitoradiata 4 17 5 2 2 1Navicula muticola 1 1Navicula phyllepta 1 5 4 3 4 2Navicula vulpina 3 14 76 14 117 23 32 130Neidium ampliatum 1 2 2Neidium distincte-punctatum 2 2Nitzschia clausii 2Nitzschia dissipata var media 8 6 6 4 4 16 11Nitzschia palaceae 7 5 2 4Nitzschia frustulum 314 251 159 177 66 189 143 29Nitzschia perminuta 383 208 210 173 57 176 180 46Nitzschia sublinearis 1 2Nitzschia suchlandii 6Orthoseira roseana 1Pinnularia krammeri 1Sellaphora bacillum 3 3Sp Dermot CV22 1 1unidentified 1 1TOTAL 924 649 582 474 382 530 479 371cysts 7 2 4 0 4 1 1 4funky fragments 14 4 49 2 15 35
0
5
2
3
5
2
21
Appendix 23. Raw diatom counts from CV24 core. CV24 core1
0.25 1.25 2.25 3.25 4.25 5.25Achnanthes marginulata 5Achnanthes minutissima 10 10 8 15 18 14Achnanthes petersenni 3Cymbella angustata 11 19 7 5 6 5Cymbella botellus 25 16 26 4 5 23Cymbella cleve-eulerae 26 12 15 8 5 9Cymbella descripta 1 2Cymbella designata 4Cymbella fogedii 9 13 12 7 6 13Cymbella microcephela 19 21 32 15 12 24Cymbella minuta 5 1 3Cymbella proxima 1Cymbella silesiaca 2 2 4Cymbella subaequalis 3 6 2 1 11Denticula kuetzingii 38 24 24 26 13 8Diatoma tenuis 1Encyonema sp18-24 plate 35 10 5 6 1 5Hannea arcus 2 3Navicula chiarae 1 4Navicula vulpina 5 8 6 6 14Nitzschia dissipata var media 1Nitzschia frustulum 311 366 263 209 267 268Nitzschia perminuta 79 76 47 65 32 37total valves 555 573 457 371 381 435cysts 1 4 9