DISTRIBUTION AND ABUNDANCE OF FISHES IN RELATION TO

BARRIERS: IMPLICATIONS FOR MONITORING STREAM

RECOVERY AFTER BARRIER REMOVAL.

By

Cory Gardner

B.S. Wildlife Ecology, University of Maine 2007.

A THESIS

Submitted in Partial Fulfillment of the

Requirements for the Degree of

Master of Science

Department of Wildlife Ecology

The University of Maine

December, 2010

Advisory Committee:

Stephen M. Coghlan Jr., Assistant Professor of Wildlife Ecology, Co-Advisor

Joseph Zydlewski, Assistant Professor of Wildlife Ecology, Co-Advisor

Kevin Simon, Assistant Professor of Biology

Rory Saunders, Fishery Biologist, NOAA NMFS (ad hoc)

LIBRARY RIGHTS STATEMENT

In presenting this thesis in partial fulfillment of the requirements for an advanced degree

at The University of Maine, I agree that the Library shall make it freely available for

inspection. I further agree that permission for "fair use" copying of this thesis for

scholarly purposes may be granted by the Librarian. It is understood that any copying or

publication of this thesis for financial gain shall not be allowed without my written

permission.

DISTRIBUTION AND ABUNDANCE OF FISHES IN RELATION TO

BARRIERS: IMPLICATIONS FOR MONITORING STREAM

RECOVERY AFTER BARRIER REMOVAL.

By Cory Gardner

Thesis Co-Advisors: Stephen M. Coghlan Jr. and Joseph Zydlewski

An Abstract of the Thesis Presented

In Partial Fulfillment of the Requirements for the

Degree of Master of Science

in Wildlife Ecology

December 2010

Dams are ubiquitous in coastal Maine, and have altered both instream habitats and

the distribution and abundance of fishes in these habitats. These impacts are caused

mainly by a disruption of the natural hydrology, temperature regime, and habitat

connectivity due to physical barriers; the subsequent fragmentation limits the movement

of resident fishes, and prevents anadromous fishes from reaching historic spawning

habitat. Dam removal has become a common method for restoring streams to a more

natural state, but often the response of the fish community is not monitored rigorously.

Sedgeunkedunk Stream, a small tributary to the Penobscot River (Maine, USA)

has been the focus of a restoration effort, which includes the removal of two dams (at 0.6

and 5.3 km upstream of the Penobscot confluence). Currently sea lamprey (Petromyzon

marinus) is the only anadromous species known to spawn successfully in the system.

In this study we quantified fish community metrics, by conducting electrofishing

surveys, along a headwaters-to-mouth gradient in Sedgeunkedunk Stream, establishing

pre-removal baseline conditions and variability. We also described the distribution and

abundance of spawning of sea lamprey in Sedgeunkedunk Stream, in anticipation of

potential range expansion after the completion of the dam removals. Sea lampreys, and

their nests, were censused using daily stream surveys and a mark-recapture model.

Over three years prior to dam removal (2007-2009), the fish community in

Sedgeunkedunk Stream showed consistent trends in species richness and abundance, with

metrics always highest downstream of the lowest dam (minimums of 14 species and 2.7

fish/m2), always lowest immediately upstream of that dam (minimums of 5 species and

0.9 fish/m2), and generally recovering upstream towards the headwaters. Although

seasonal and annual variation in metrics within each site were substantial, patterns across

all sites were remarkably consistent regardless of sampling episode. Immediately after

dam removal, we saw significant decreases in richness and abundance (10 species, 0.7

fish/m2) below the former dam site and a corresponding increase in fish abundance at the

upstream site (5 species, 5.0 fish/m2), and for the first time, anadromous Atlantic salmon

were found upstream of the former dam site. No such changes were obvious in a

reference stream.

In 2008 forty-seven sea lamprey entered the stream, and constructed 31 nests; all

fish received PIT tags, which allowed us to identify and follow movements of

individuals. Individuals that arrived early in the run were active in the system longer, and

used more nests, than did late migrants. Sea lamprey frequently ascended to the first dam

encountered before swimming back downstream and beginning nest construction. Nests

Deleted: two or

Deleted: , 2008,

Deleted: Often, s

were distributed from head-of-tide to the lowermost dam; no spawners or nests were

observed in the tidally-influenced zone or upstream of this dam.

Our results show that by quantifying baseline conditions in a small stream before

restoration, the effects of stream restoration efforts on fish communities can be monitored

successfully. Based on the observed movements in the system, and the range of their

habitat use, we anticipate that spawning sea lamprey will re-colonize formerly

inaccessible habitat

Deleted: ,

Deleted: following dam removal,

ii

ACKNOWLEDGEMENTS

Funding for this project was provided by NOAA’s National Marine Fisheries Service, the

Maine United States Geological Survey Cooperative Fish and Wildlife Research Unit, the

University of Maine, and the Maine Agriculture and Forestry Experiment Station. We

also thank R. Cope, G. Goulette, D. Kircheis, T. Liebech, K. Mueller, J. Murphy, B.

Perry, K. Ravana, D. Skall and T. Trinko for assistance in sampling. We thank our field

technicians Silas Ratten, Jacob Kwapiszeski, Ryan Haley, Scott Ouellette, Anthony

Feldpausch, Michael Picard, Meghan Nelson, and many others who helped out along the

way (University of Maine) for their hard work, and Rory Saunders (NOAA) for sparking

our interest in the role of sea lamprey in Sedgeunkedunk Stream. Mention of trade names

does not imply endorsement by the United States government.

Deleted: NOAA NMFS Maine Field Station for assistance with sampling,

Deleted: especially Kyle Ravana and Tara Trinko

iii

TABLE OF CONTENTS

ACKNOWLEDGMENTS………………………………………………………………...ii

LIST OF TABLES………………………………………………………………………...v

LIST OF FIGURES………………………………………………………………………vi

Chapter

1. DISTRIBUTION AND ABUNDANCE OF STREAM FISHES IN RELATION TO

BARRIERS: IMPLICATIONS FOR MONITORING STREAM

RECOVERY AFTER BARRIER REMOVAL……………………………………….......1

Introduction………………………………………………………………………..1

Methods……………………………………………………………………………6

Study Area…………………………………………………………………6

Study Design………………………………………………………………7

Fish Surveys……………………………………………………………….8

Abundance Estimates……………………………………………………...9

Community Assessment…………………………………………………..10

Results……………………………………………………………………………11

Discussion………………………………………………………………………..15

2. DISTRIBUTION AND ABUNDANCE OF ANADROMOUS SEA LAMPREY

(Petromyzon marinus) SPAWNERS IN A DAMMED STREAM: CURRENT STATUS

AND POTENTIAL RANGE EXPANSION

FOLLOWING BARRIER REMOVAL…………………………………………………24 Deleted: 3

iv

Introduction………………………………………………………………………23

Methods…………………………………………………………………………..28

Study area…………………………………………………......................28

Sea lamprey population estimate and behavioral evaluation……………28

Nest surveys and abundance estimation…………………………………29

Results……………………………………………………………………………30

Discussion………………………………………………………………………..33

REFERENCES…………………………………………………………………………..37

BIOGRAPHY OF THE AUTHOR………………………………………………………46

Formatted: Font: Not Italic

Formatted: Font: Not Italic

Formatted: Indent: Left: 0.5", Firstline: 0.5"

Deleted: 7

Deleted: 7

Deleted: 8

Deleted: 29

Deleted: 2

Deleted: 6

Deleted: 5

v

LIST OF TABLES

Table 1.1. Occurrence of species, before the removal of the Mill Dam, in study sites in

Sedgeunkedunk Stream and Johnson Brook, Penobscot Co., Maine. 2007-

2009. …….........................................................................................................11

Table1.2. Shannon Wiener diversity index values for fish community composition in sites

in Sedgeunkedunk Stream and Johnson Brook, Penobscot Co. ME. 2009. …..12

Table 1.3. Sorenson’s Similarity Index values comparing fish community

composition between sites in Sedgeunkedunk Stream and Johnson Brook,

Penobscot CO. ME. 2009…………………………………………………......13

Table 2.1. Number of live sea lamprey captured for the first time on each day of the

sampling period, the mean water temperature of that day and the mean number

of days observed active in the river, and mean number of total nests sea

lamprey that entered on that day used before they died, Sedgeunkedunk

Stream, Penobscot Co., ME 2008. …...............................................................31

Table 2.2. Smallest, largest and mean sizes of sea lamprey nests, and the maximum

number of sea lamprey observed on them in Sedgeunkedunk Stream,

Penobscot Co. ME. 2008. …………………………………………………...33

Deleted: 0

Deleted: 2

vi

LIST OF FIGURES

Figure 1.1. Study sites on Sedgeunkedunk Stream (S1-S5) and Johnson Brook (J1-J3),

Penobscot Co. ME. 2007-2009. ………………………………………………6

Figure 1.2. Species richness at sampling sits along stream gradient in Sedgeunkedunk

Stream and Johnson brook, Penobscot Co. Maine 2008-2009. …..…………12

Figure 1.3. Fish density (#/m2) at sites along stream gradient of Sedgeunkedunk Stream

and Johnson Brook, Penobscot Co. Maine. 2007-2009……………………...14

Figure 1.4. Blacknose dace density (#/m2) at sites along stream gradient of

Sedgeunkedunk Stream and Johnson Brook, Penobscot Co. Maine,

2007-2009. …………………………………………………………………..14

Figure 1.5. Length frequency distribution of blacknose dace at sites below and above

removed dam in Sedgeunkedunk Stream, Penobscot Co. Maine. 2009. …...15

Figure 1.6. Net gain of blacknose dace within size classes at site S1 and S2 following

dam removal in Sedgeunkedunk Stream, Penobscot Co. Maine. 2009. ……15

Figure 2.1. Location of Sedgeunkedunk Stream, Fields Pond, and the dams that were

removed during the Sedgeunkedunk Stream restoration. Penobscot Co.

Maine, USA. ………………………………………………………………...28

Figure 2.2 The cumulative proportion of sea lamprey entering Sedgeunkedunk Stream,

Penobscot Co. Maine. 2008. ………………………………………………...30

Figure 2.3. Frequency of the number of nests individual male, and female sea lamprey

were observed using during the duration of their activity in Sedgeunkedunk

Stream, Penobscot, Co., Maine. 2008. ………………………………………31

Deleted: 14

Deleted: 7

Deleted: 29

Deleted: 0

vii

Figure 2.4. Distribution sea lamprey nests, and percentage of substrate made up fines

(smaller than 2mm) along the stream gradient of Sedgeunkedunk Stream,

Penobscot Co., Maine. 2008. ………………………………………………..32

Deleted: 1

1

Chapter 1

DISTRIBUTION AND ABUNDANCE OF STREAM FISHES IN RELATION TO

BARRIERS: IMPLICATIONS FOR MONITORING STREAM

RECOVERY AFTER BARRIER REMOVAL.

Introduction

Resilience is defined as the ability of an ecosystem to return to a natural state

following a disturbance (Holling 1973). Often, resilience is dependent upon sufficient

species diversity persisting so that functional groups necessary to restore a natural state

remain (Elmqvist et al. 2003). Ecosystems which are subject to disturbances that occur

over large spatial and temporal scales are more sensitive to a lack of diversity (Elmqvist

et al. 2003). The temporal, and spatial, variability in species-specific responses to

disturbance are likewise important to maintaining ecosystem resilience (Elmqvist at el.

2003). Without a sufficient amount of species diversity the post-disturbance community

may attain different stable states than the one that preceded disturbance (Gunderson

2000).

Since the arrival of European settlers on the east coast of North America, human

population growth and development have led to myriad disturbances in most terrestrial

and aquatic ecosystems (Ehrlich and Holden 1971; Vitousek et al. 1997). Of interest here

Deleted:

2

is the construction of dams. For centuries, dams have been built to control flows, create

water supplies, and generate electricity in the coastal watersheds of New England and

elsewhere (Benke 1990). Dams caused changes in hydrology, sediment loading,

temperature regimes, and connectivity within and among ecosystems (Koster et al. 2007;

Petts 1980). Dams building was also associated with increases in industrial pollution.

While the physicochemical effects of dams on streams have been well documented, the

impact of dams on fish communities has not (Baldigo and Warren 2008; Quinn and Kwak

2003). There has been great recent emphasis on restoring stream ecosystems perturbed

by dams; however, it is unknown if these ecosystems have the ecological resilience to

return to a state that resembles preindustrial conditions after centuries of chronic

disturbance.

Dams may impact the distribution and abundance of fish communities by the

disrupting the natural hydrologic and thermal regimes and fragmenting habitat (Winston

et al. 1991). Impoundments convert lotic habitats to lentic (Martinez et al. 1994), and

may decrease sediment loading and increase water temperature downstream (Kinsolving

and Bain 1993; Ligon et al. 1995). Habitat fragmentation can prevent resident fish

species, which make relatively short movements, from accessing habitats that are

necessary in completing their life history (Schmetterling and Adams 2004) or

maximizing energy balance during critical periods (Hall 1972), thus disrupting energy

and material flows in the stream (Hall 1972).

Anadromous fishes are likely to suffer a greater impact from fragmentation

(Benke 1990) than would freshwater residents. Historically, many rivers and ponds in

Maine harbored spawning runs of numerous migratory species, such as Atlantic salmon

Deleted: were built during a period of increasing industrial pollution, as well as directly

3

(Salmo salar), alewife (Alosa pseudoharengus) sea lamprey (Petromyzon marinus) and

rainbow smelt (Osmerus mordax) (Saunders et al. 2006). In the Pacific Northwest,

anadromous fish such as Pacific salmon (Oncorhynchus spp.) can transport marine

derived nutrients and energy into otherwise oligotrophic freshwater ecosystems and

increase primary productivity and microbial decomposition rates of detritus (Brock et al.

2007; Durbin et al. 1979; Wipfli et al. 1998). Such increased productivity , in turn,

facilitates growth in resident fish species (Bilby et al 1996; Scheuerell et al. 2007; Wipfli

et al. 2003). It is likely that some of these relations exist, or existed, in systems on the

Atlantic coast as well. As dams have severed this marine-freshwater connectivity,

anadromous species have experienced worldwide decline (Limburg and Waldman 2009)

and freshwater systems have become more oligotrophic (Stockner et al. 2000). This

connection between anadromous communities and the resilience resident communities in

a perturbed system is poorly characterized.

Abundance, richness, and ultimately diversity of stream fishes increase along the

gradient from headwaters to mouth in a small, non-fragmented stream, due to predictable

increases in the availability and diversity of habitats (Sheldon 1968) and the species pool

of potential colonizers in nearby rivers (Smith and Kraft 2005). These metrics can ebb

and flow upstream and downstream over time in response to discharge and resultant

velocity barriers to potential colonizers from downstream reaches (Grossman et al.,

2010). Ultimately, stream discharge and other physicochemical conditions driven by

landscape changes along this longitudinal gradient lead to patterns of energy and nutrient

flow to which stream ecosystems respond dynamically yet predictably (Vannote et al.,

1980), although interruptions in the continuum evince marked changes in the stream biota

Deleted: These fish

Deleted: It has been shown on the

Deleted: coast

Deleted: that

Deleted: ed

Deleted: likely

Deleted: d

Deleted: presumably

Deleted: ed

Deleted: hips

Deleted: conditions

Deleted: -

4

(Ward and Stanford 1983; Jones 2010). In coastal streams, anadromous fish may

penetrate relatively far into headwaters, and the subsidies of marine-derived nutrients and

energy associated with these migrants provides another gradient to which resident fishes

may respond (Mitchell and Cunjak 2007). If the presence of a dam alters these natural

gradients, then the structure and function of fish communities should shift as well.

Sedgeunkedunk Stream, a small tributary of the Penobscot River below head-of-

tide (Figure 1.1), typifies small streams in Maine impacted by dams. Current restoration

efforts offer an opportunity to assess the resilience of the structure and function of this

fish community when released from habitat fragmentation caused by multiple dams.

Runs of anadromous fishes in Sedgeunkedunk Stream have either disappeared or were

reduced substantially after the building of three dams on the stream. Recent declines in

anadromous fishes in Sedgeunkedunk Stream mirror those in the entire Penobscot

watershed, which contains over 100 known dams; however, because Sedgeunkedunk

Stream flows into the Penobscot River below the lowermost mainstem dam (Veazie

Dam), it is one of only three major tributaries that could receive anadromous fishes

before they encounter a relatively large, mainstem dam. Currently, Atlantic salmon in

Sedgeunkedunk Stream and most of the Penobscot watershed are listed as a federally-

endangered species (74 Fed. Reg. 29344; June 19, 2009), and alewife, sea lamprey, and

rainbow smelt are at historic low levels of abundance throughout Maine watersheds

(Saunders et al. 2006).

As part of a collaborative restoration project, fish passage has been created or

restored in Sedgeunkedunk Stream at the location of two former dam sites between Fields

Pond and the confluence with the Penobscot River (Figure 1.1). The lowermost dam

Deleted:

Deleted: ,

Deleted: , and resilience

Deleted: responding to long-term disturbance by, and then relief from,

Deleted: from

Deleted: Sedgeunkedunk Stream drains a 5400 hectare watershed with numerous wetlands and natural and augmented headwater impoundments.

Deleted: (at 0.6, 5.3 and 8.0 km upstream of the confluence with the Penobscot River; heretoafter referred as “stream km).

Deleted: (PRRT 2009)

Deleted: NOAA 2009

5

(Mill Dam) was removed in August 2009, and the middle dam (Meadow Dam) was

bypassed in August 2008 by a rock-ramp fish-way that allows fish passage while

maintaining current water levels in Fields Pond and adjacent wetlands. Thus habitat

connectivity for migrating anadromous and resident fishes has been restored throughout

most of the Sedgunkedunk watershed.

This study began in 2007, two years prior to removal of the lowermost dam, as a

long term monitoring effort directed toward the effects of this restoration on the fish

community in the Sedgeunkedunk watershed. While the reversion towards a more

historic species composition is desired or expected following dam removal, there is often

an initial decline in abundance of resident species until the geomorphology of the stream

is stabilized (Catalano et al. 2007; Doyle et al. 2005). In small streams the physical

recovery from the immediate disturbance of a dam removal can take as little as a year, if

high water events move the sediment from the impoundment downstream quickly

(Burroughs et al. 2009), but the response of the fish community may be protracted (Hart

et al. 2002), illustrating the importance of long term monitoring to characterize biotic

response. Many small streams are characterized by a great deal of variability at different

scales, both spatially (e.g., longitudinally and within short reaches) and temporally (e.g.,

both annually and seasonally) ,which could make the detection of community changes

difficult (Baldigo and Warren 2008; Jackson et al. 2001, Moyle and Vondracek 1985). In

this study the data collected before the removals are used to quantify the baseline

conditions and variability in this altered system, and a neighboring reference system.

Specifically, our objectives were to: 1) Quantify the baseline distribution and

abundance, and the associated variability, of fish species in Sedgeunkedunk Stream

Deleted: impoundment sediment

Deleted: are

6

before final dam removal; 2) Quantify the distribution and abundance of fish species in

Sedgeunkedunk Stream immediately following removal of the lowermost dam; 3)

Determine if the temporal and spatial variability in fish community metrics will allow us

to detect changes in response to dam removals. This work represents the first step in

assessing the long term research goal of characterizing resilience and recovery in small

coastal systems.

Methods

Study Areas

Sedgeunkedunk Stream is a third-order tributary of the Penobscot River,

Penobscot Co. Maine. The stream flows out of Fields Pond at 44°44’05’’N and

68°45’56’’W and debouches into the Penobscot River (at river km 36, which is

approximately at head-of-tide) at 44°46’08’’N and 68°47’06’’W. The stream drains 5400

hectares including Brewer Lake and Fields Pond (Figure 1.1). The lowermost Mill Dam,

which was removed in August 2009, was located just 600 meters upstream of the

stream’s confluence with the Penobscot

River. The middle Meadow Dam, which

was located at stream km 5.3 upstream

from the Penobscot confluence, was

replaced by a rock-ramp fishway in

August 2008. The removal of these two

barriers provides for presumably

unimpeded access from the Atlantic

Figure 1.1. Study sites on Sedgeunkedunk Stream (S1-S5) and Johnson Brook (J1-J3), Penobscot Co. ME. 2007-2009.

P E N

O B S C

O T

R I V

E R

Swetts Pond

Fields Pond

Brewer Lake

Johnson Brook

Sedgeunkedunk Stream

Mill Dam

Meadow Dam

S1

S2

S3

S4

S5

J1 J2 J3

P E N

O B S C

O T

R I V

E R

Swetts Pond

Fields Pond

Brewer Lake

Johnson Brook

Sedgeunkedunk Stream

Mill Dam

Meadow Dam

S1

S2

S3

S4

S5

J1 J2 J3

Deleted: flowing through the Town of Orrington and the City of Brewer, Maine

Deleted: ,

Deleted: draining

Deleted: and

7

Ocean into Fields Pond, via the lower Penobscot River and then Sedgeunkedunk Stream.

The uppermost Brewer Lake Dam, located between Fields Pond and Brewer Lake, will

remain intact, and continue to block passage further upstream in the watershed.

For future monitoring following the dam removal, our study incorporated a

Before-After-Control-Impact design (BACI; Stewart-Oaten et al. 1986), which accounts

for naturally-occurring temporal and spatial variation so that we can attribute differences

in Sedgeunkedunk Stream fishes to restoration efforts irrespective of background noise.

A BACI design requires that pre-impact conditions and variability are known, which we

established by sampling for one year before the restoration project began, and two years

before the lowermost dam removal took place. We also monitored a reference system,

Johnson Brook, which is a tributary of the Penobscot River located in Orrington, Maine

(Figure 1). The stream flows out of Swetts Pond at 44°42’18’’N and 68°47’10’’W and

debouches into the Penobscot River (at river km 24) at 44°42’08’’N and 68°49’50’’W.

This system includes a natural barrier (Clark’s Falls, at stream km 2.2) and a dam at the

outlet of Swetts Pond (at stream km 5.5); these barriers to passage are analogous to the

dams located in Sedgeunkedunk Stream.

Study Design

Five 100-m sampling locations were selected along Sedgeunkedunk Stream

(Figure 1.1). Site S1 is located at stream km 0.45-0.55, immediately downstream of the

former Mill Dam, and was the only site accessible to anadromous fish. Site S2 is located

immediately upstream of the Mill Dam site at stream km 0.62-0.72. Site S3 is located

approximately halfway between the Mill Dam and Meadow Dam sites at stream km 2.5-

Deleted: O

Deleted: is

Deleted: Our

8

2.6. Site S4 is located immediately downstream of the Meadow Dam / Fishway site at

stream km 5.2-5.3. Site S5 is located in upstream of Brewer Lake; because the Brewer

Lake Dam is not scheduled for removal, this site will remain unaffected by the restoration

project and serve as a reference site within Sedgeunkedunk Stream.

Three 50-m sampling locations were selected along Johnson Brook in order to

compare with sites on Sedgeunkedunk Stream before and after barrier removal (Figure

1). These sites were limited to 50 m due to access problems and the presence of beaver

dams and their associated impoundments, which would have limited our ability to sample

100-m long reaches effectively, but likely did not impede fish passage. Site J1 is located

immediately downstream of Clark’s Falls at stream km 3.62-3.67. Site J2 is located

immediately upstream of Clark’s Falls at stream km 3.71-3.76 Site J3 is located

immediately downstream of the dam at the outlet of Swetts Pond at stream km 7.3-7.8.

Fish surveys

Each location was sampled in August of 2007. In 2008 and 2009, each location

was sampled twice per year, once in the spring (usually late May) as soon as spring

runoff subsided and again in late summer (usually mid August) during low-flow

conditions. The Meadow Dam was removed after sampling was completed in 2008. The

Mill Dam was removed immediately before the second sampling of 2009. Fish

abundance at each site was estimated by conducting three-pass depletion estimates

(Zippin 1958) using backpack electrofishers, except for spring 2007 when we conducted

2-pass depletions. Electrofisher settings were 30-60Hz, 20% duty cycle, and ~300-500V,

depending on measured ambient conductivity, and were optimized for maximum power

Deleted: ###

Deleted: ,

Deleted: s

Deleted: ###

Deleted: ,

Deleted: and s

Deleted: ###

9

transfer (Reynolds 1996). Each site was closed to immigration and emigration during the

survey using 3mm-mesh blocking nets. All fish caught were identified to species, and the

first 300 individuals of each species at each site were measured (total length; mm) and

weighed (wet mass; 0.1 g). Except for incidental mortality (<2% for all sampling

events), fish were returned to the stream alive. This repeated electrofishing probably did

not influence the fish community measurably (e.g., Latimore and Hayes 2008). In

summer 2008 all sampling was completed before the removal of the Meadow Dam. In

2009 the summer round of sampling took place within 7 days after the removal of the

Brewer Dam. We sampled Site S1 three days after the dam removal, as soon as that

stream had cleared of suspended sediment enough for us electrofish effectively. We note

that although American eel were present and abundant at all sites at all times, our

catchability was extremely low for small elvers (< 100 mm TL), so we limit analysis of

eel data to qualitative, rather than quantitative, measures (i.e., presence / absence).

Abundance Estimates

At each sampling reach, stream width was measured at ten locations and the area

estimated as the product of reach length and mean width. This sampling area was used in

our estimates of total fish density (# of fish / m2), with associated variances, for each

sampling location (with the exception of American eel, as described above). Differences

in density among reaches and over time were assessed by inspecting overlap of 95%

confidence intervals, which is a conservative estimate (Simpson et al. 1961). This

method does not allow us to assume “no significant difference” when there is an overlap,

Deleted: Each location was sampled twice per year, once in the spring (usually late May), as soon as spring runoff subsided and again in late summer (usually mid August) during low-flow conditions.

Deleted: ,

10

however it does allow us to assume a significant difference (at α = 0.05) when there is no

overlap (Payton et al., 2003).

Community Assessment

Temporal and longitudinal changes in the fish assemblages were characterized

using four methods. First, species richness was determined by noting presence/absence

of each species. Comparisons of species richness at among sites, and over time, were

made using a Friedman test and Tukey multiple comparisons (Zar, 1984).

Presence/absence was also used to describe species distribution before barrier removal, in

order to monitor species movement and colonization after removal. Second, we used a

Shannon Wiener species diversity index (Krebs, 1989);

ii PPH ln

for each site over time, where Pi is the proportion of the population belonging to species

i. Third, we used a Sorensen’s Similarity Index (Krebs, 1989);

)/(2 bacQS

where c is the number of common species, and a and b are the total number of species in

the two sites respectively. In this index of similarity the results range from 0 (no

similarity) to 1 (identity).

Differences between mean length and mean mass of species found at all sites

were detected using a two-way ANOVA, with site and sampling time as the main effects.

Lengths of each species were used to construct length frequency histograms, and by

finding natural breaks in the size distribution which separate age classes, we determine

size and age structure, for three common species at each site. A Kolmogorov-Smimov

test was used to find differences between length frequency distributions.

Deleted: e

Deleted: ,

11

Results

Over the course of this study we

encountered 22 species and 26,873

individual fish (Table 1.1). In 2007, we

captured 4,760 fish in Sedgeunkedunk

Stream. In 2008, we captured 7,872 fish

in Sedgeunkedunk Stream and 3,693 fish

in Johnson Brook. In 2009, we captured

8,516 fish in Sedgeunkedunk Steam and

2,032 in Johnson Brook. The dominant

species in both streams was eastern

blacknose dace (Rhinichthys a.

atratulus). In Sedgeunkedunk Stream

species richness, diversity, and the

density of total fish was highest below the Mill Dam and lowest above the dam. These

metrics then increased along the gradient with increasing upstream from the dam, but

then declined in the isolated headwater site (S5).

In Sedgeunkedunk Stream, prior to dam removal, species richness was always

greatest ( p < 0.03) at site S1 (immediately downstream of Mill Dam) (Figure 1.2). Prior

to dam removal, 14 species were encountered at each sampling episode at S1. This

richness decreased to either 5 or 6 species at site S2 (immediately upstream of Mill Dam).

Species richness remained at this lower level at site S3 at all sampling episodes, but at

site S4 had risen to 9 or 11 species. Site S5 had 8 species present in each spring

Species S1 S2 S3 S4 S5 J1 J2 J3Atlantic salmon Blacknose dace Bluntnose minnow Brown bullhead Creek chub Pumpkinseed sunfish Common shiner Eastern brook trout Chain pickerel American eel Fathead minnow Fallfish Finescale dace Golden shiner Northern redbelly dace Ninespine stickleback Redbreast sunfish White sucker Smallmouth bass White perch Black crappie Yellow perch Alewife

Rainbow smelt

Sea Lamprey

Table 1.1. Occurrence of species, before the removal of the Mill Dam, in study sites in Sedgeunkedunk Stream and Johnson Brook, Penobscot Co., Maine. 2007-2009.

density ≥ 0.1 fish/m2 density ≥ 0.01 fish/m2 density < 0.01 fish/m2 no density estimated

Comment [SC1]: Check with the formatting desired by the graduate school, but usually figures and tables are given their own pages, rather than snuggled nest to text. Same comment applies throughout

Deleted: from moving

Deleted: all

12

sampling, and 7 species present in each summer sampling. Immediately after removal of

the Mill Dam the species richness at site S1 decreased from 14 to 10, and the richness at

the sites upstream were similar to that found in previous samplings. Species that were no

longer found at site S1 following the dam removal were: northern redbelly dace

(Phoxinus eos), finescale dace (Phoxinus neogaeus), fathead minnow (Pimephales

promelas), and ninespine stickleback (Pungitius pungitius). In Johnson Brook, the

species richness at site J1

(downstream of Clark’s Falls)

ranged from 8-11. This decreased

to 4-6 species at site J2 (upstream

of Clark’s Falls). Richness was

always highest at site J3

(downstream of Swett’s Pond

Dam, ranging from 10-13 species.

Prior to the removal of the

Mill Dam, Atlantic salmon were

encountered only at site S1. In the fall

of 2008 one Atlantic salmon parr was captured (TL = 152mm), and in the spring of 2009,

one Atlantic salmon fry (TL = 30 mm) was

captured. After the removal of the Mill Dam in

2009, Atlantic salmon parr were captured at site

S2 (5 individuals, mean TL = 83 mm), and 55

were captured at site S3 (55 individuals, mean

0

2

4

6

8

10

12

14

S1 S2 S3 S4 S5

May-08

Jul-08

May-09

Aug-09

0

2

4

6

8

10

12

14

J1 J2 J3

# o

f s

pe

cies

a.Sedgeunkedunk Stream

b.Johnson Brook

0

2

4

6

8

10

12

14

S1 S2 S3 S4 S5

May-08

Jul-08

May-09

Aug-09

0

2

4

6

8

10

12

14

J1 J2 J3

# o

f s

pe

cies

a.Sedgeunkedunk Stream

b.Johnson Brook

Figure 1.2. Species richness at sampling sits along stream gradient in Sedgeunkedunk Stream and Johnson brook, Penobscot Co. Maine 2008-2009.

a. Sedgunkedunk StreamS1 S2 S3 S4 S5

Jul-07 1.18 1.03 0.51 0.82 1.58May-08 1.63 0.66 0.67 1.16 1.67Aug-08 1.22 0.77 0.60 0.65 1.04May-09 1.40 0.43 0.41 0.77 1.32Aug-09 1.44 0.82 0.69 0.56 1.06

b. Johnson BrookJ1 J2 J3

May-08 1.12 1.40 1.88Aug-08 1.06 1.38 1.92May-09 1.36 1.15 1.58Aug-09 1.41 1.24 2.22

Table1.2. Shannon Wiener diversity index values for fish community composition in sites in Sedgeunkedunk Stream and Johnson Brook, Penobscot Co. ME. 2009.

Deleted: In the sampling that followed

Deleted: 5

13

TL = 65 mm).

Species diversity in Sedgeunkedunk Stream was always highest at sites S1 and S5

(Table 1.2). Before the removal of the Mill Dam, sites S1, S4, and S5 had high (i.e., QS

> 0.70) pairwise similarity values, but were less similar to sites S2 and S3; sites S2 which

were highly similar (Table 1.3). In Johnson Brook the diversity was highest at site S3

and lower at site S1 and S2. Total fish density in Sedgeunkedunk Stream was highest at

site S1 and lowest at site S2 during all sampling occasions before the removal of the Mill

Dam (Figure 1.3a). Densities followed a pattern along the stream gradient of increasing

at site S3 and site S4, but decreasing at site S5. Total fish density in Johnson Brook

showed a consistent pattern at all sampling events. Density was higher at site J1, than at

sites J2 and J3 (Figure 1.3b).

The community structure in

Sedgeunkedunk Stream changed following

the removal of the Mill Dam. There was

an increase in diversity at site S2 (Table

1.2). All of the sites in Sedgeunkedunk

stream became more similar (Table 1.3).

Fish density at site S1 showed a significant

decline, while the density at site S2

increased more than five-fold (Figure 1.3a). The density also increased at site S3 and site

S4, but the density at site S5 did not change significantly. There was no change in the

pattern in Johnson Brook over this time period (Figure 1.3b)

a. Sedgeunkedunk StreamS1 S2 S3 S4 S5

S1 - 0.53 0.57 0.80 0.73S2 0.67 - 0.80 0.53 0.57S3 0.78 0.88 - 0.78 0.82S4 0.84 0.71 0.82 - 0.94S5 0.82 0.80 0.93 0.88 -

b. Johnson BrookJ1 J2 J3

J1 - 0.53 0.67J2 0.67 - 0.47J3 0.82 0.59 -

Table 1.3. Sorenson’s Similarity Index values comparing fish community composition between sites in Sedgeunkedunk Stream and Johnson Brook, Penobscot CO. ME. 2009. Values range from 0 (no similarity) and 1 (identity). Values above the dashed line are from before the dam removal. Values below are from after the dam removal.

Deleted: and S3

14

Three species were ubiquitous spatially and temporally in Sedgeunkedunk stream:

eastern blacknose dace, fallfish (Semotilus corporalis), and white sucker (Catostomus

commersoni). Blacknose dace was, often by an order of magnitude, the most abundant

species at all sampling locations and times, and mirrored the pattern seen in overall fish

density. After the removal of the Mill Dam the density of blacknose dace at site S1

decreased significantly, whereas the density at site S2 more than tripled (Figure 1.4a).

The density rose significantly at site S3 and site S4 while the density at site S5 remained

statistically unchanged. Blacknose dace density in Johnson Brook also showed a

S1 S2 S3 S4 S5

Fis

h D

ensi

ty (

#/m

2 )

0

1

2

3

4

5

6

August 2007May 2008July 2008May 2009 August 2009

SITE

J1 J2 J3

0

1

2

3

4

5

6

a. Sedgeunkedunk Stream

b. Johnson Brook

Figure 1.3. Fish density (#/m2) at sites along stream gradient of Sedgeunkedunk Stream and Johnson Brook, Penobscot Co. Maine. 2007-2009. Error bars represent ± 2SE (95% confidence interval).

S1 S2 S3 S4 S5

Fis

h D

ens

ity (

#/m

2 )

0.0

0.5

1.0

1.5

2.0

2.5

3.0

August 2007May 2008July 2008May 2009 August 2009

SITE

J1 J2 J3

0.0

0.5

1.0

1.5

2.0

2.5

3.0

a. Sedgeunkedunk Stream

b. Johnson Brook

Figure 1.4. Blacknose dace density (#/m2) at sites along stream gradient of Sedgeunkedunk Stream and Johnson Brook, Penobscot Co. Maine, 2007-2009. Error bars represent ± 2SE (95% confidence interval).

Deleted: over any other species

Deleted: significantly

15

consistent pattern at all sampling events, with density higher at site J1 than at sites J2 and

J3 (Figure 1.4b).

The mean length and mass of blacknose dace, and all other common species,

differed among all sites at each sampling

occasion (all p < 0.05), but did not show any

consistent spatial or temporal pattern. The

same was true of the size distribution of

blacknose dace, but there was a noticeable

pattern. Before the removal of the Mill

Dam, the size distribution of blacknose dace

lengths was similar among sites S1, S3, and

S4, but skewed towards smaller individuals

at site S2. Immediately following the

removal of the Mill Dam the length

distribution in site S1 was truncated, while the size distribution at site S2 appears to

expand and resemble the distribution in other sites prior to removal (Figure 1.5). The net

losses at site S1 and the net

additions at site S2 of

blacknose dace within size

classes appear to correspond

to each other (Figure 1.6),

especially assuming some

growth in length over the

Figure 1.5. Length frequency distribution of blacknose dace at sites below and above removed dam in Sedgeunkedunk Stream, Penobscot Co. Maine. 2009.

-500 -400 -300 -200 -100 0 100 200 300

25

35

45

55

65

75

Size

Cla

ss

Net Gain in numbers of individuals

S2

S1

Figure 1.6. Net gain of blacknose dace within size classes at site S1 and S2 following dam removal in Sedgeunkedunk Stream, Penobscot Co. Maine. 2009.

Deleted: While our analysis did not show a statistically significant difference between size distributions,

Deleted: ere

Deleted: s

Deleted: and changes

Deleted: S1,

Deleted: if one

Deleted: es

16

time period between samples. There was no such change in Johnson Brook.

Discussion

In the two years of sampling that preceded the removal of the Mill Dam and the

completion of the restoration project in Sedgeunkedunk Stream, we documented

consistent patterns in the distribution and abundance of fishes. These patterns and the

low variability associated with them give us confidence in our ability to detect any long-

term changes that may occur in the system, as does our ability to detect short-term

differences in these patterns that immediately followed the removal of the Mill Dam.

The presence of the Mill Dam in Sedgeunkedunk Stream influenced the

composition, and size structure, of the fish community along the stream gradient. Each

study site along the stream showed a consistent pattern over time, prior to dam removal.

The site directly below the dam (S1), although affected by the hydrological disturbances

associated with a dam, most likely represented the natural habitat condition of the stream

most closely, and was the only site readily accessible to fishes migrating from the

Penobscot River and the Atlantic Ocean. This connectivity to the larger aquatic

ecosystems can allow for the increased species richness and diversity (Lyons and

Schneider 1990; Smith and Kraft 2005); migratory species, like Atlantic salmon and

nine-spine stickleback, as well as resident species from the Penobscot River, could enter

this section of stream. Connectivity also increased habitat availability, due to access to

parts of the Penobscot, and increased thermal variability, due to the presence of a tidal

zone near the mouth of the stream that could create a thermal refuge.

Deleted: .

Deleted: ; migratory

Deleted: variability

17

The study site located directly above the dam (S2) was a straight, slow moving,

impoundment. The riparian zone consisted of grasses that had colonized the sediment

that had been deposited above the dam, replacing the normal forest cover along the

stream. The combination of the exposure, channel straightening, and siltation resulted

seemingly in habitat homogeneity. Because of this, and the lack of connectivity caused

by the dam, this section of stream had fewer fish, and lower species richness and

diversity, than all other sites. The community also contained more large bodied species

like fallfish and white sucker, which could use deeper pools caused by the impoundment

(Petts 1980; Swink and Jacobs 1983). Populations of small-bodied minnows, like

blacknose dace, were dominated by smaller fish. All species found in this site could be

found at other sites as well. While these decreases in abundance and diversity are to be

expected as part of the natural process as one moves up a stream gradient (Sheldon 1968),

the differences between these sites were drastic and occurred over a distance of less than

100 meters.

Species richness, diversity, and the density of total fish increased along the

gradient from moving upstream from the dam. This probably represents a decreasing

impact of the dam on the fish community as it moves upstream from the disturbance

caused by the dam (Kinsolving and Bain 1993; Phillips and Johnston 2004), and shows

that the community at the site directly above the dam represents an interruption of a

natural stream gradient (Sheldon 1968). The fish communities at sites S3and S4 became

more similar to site S1 corresponding to a gradual recover from the interruption caused

by the Mill Dam. The relatively low fish density at site S5, located much further

upstream in the watershed, was probably caused by two factors. Connectivity to

Deleted: utilize

Deleted: ; Vannote et al. 1980),

Deleted: continuum

Deleted: Vannote et al. 1980

Deleted: ,

Deleted: species richness and

18

downstream reaches and the Penobscot River is reduced by another dam and two lakes,

as well as a comparatively long distance from a large pool of potential migrants or

colonizers (e.g., Smith and Kraft 2005). More importantly, this headwaters reach is

located in a steeper, more heavily forested section of the watershed, and is characterized

by lower temperatures and conductivity than in downstream sites. It is also characterized

by shallower pools. Thus, overall production potential of fishes is probably reduced here

(Schlosser 1982). The community at this site is what would be expected in a stream

section higher up in the watershed, and probably is representative of a lightly-impacted

headwaters reach not influenced strongly by anadromous fishes.

The pattern in Johnson Brook was similar to the longitudinal pattern that we

found in Sedgeunkedunk Stream, although the differences above and below the

impassible barrier were not as dramatic, and the distinct changes associated with the Mill

Dam removal was not present here. The site below the impassible falls (J1) showed

higher species richness and diversity than the site directly above the falls (J2), which is

probably a function of habitat disconnectivity, similar to the effect of the Mill Dam in

Sedgeunkedunk Stream. However, there were differences between the site above the

falls in Johnson Brook (J2) and the site above the Mill Dam in Sedgeunkedunk (S2).

While the waterfall acted as a barrier to upstream migration, it was not associated with a

deforested, channelized and sedimented impoundment, like in Sedgeunkedunk Stream, so

it did not harbor a community dominated by larger bodied species. Despite this we still

observed a decline in fish density and species richness above the waterfall in Johnson

Brook analogous to what occurred above the Mill Dam in Sedgeunkedunk. Species that

were found at S1 and J1, and not at S2 or J2 include: fathead minnow, golden shiner

Deleted: Vannote et al., 1980

Deleted: gradient

19

(Notemigonus crysoleucas), northern redbelly dace, and ninespine stickleback. Creek

chub (Semotilus atromaculatus ) was the only species found at site S1, J1 and J2, but not

found at S2. This could imply that it was the impoundment in Sedgeunkedunk that led to

the absence of creek chub, while it was the lack of habitat connectivity that led to the

absence of the previous species. While the farthest upstream site in Johnson Brook (J3)

contained lower fish densities than the site below the falls, it did show an increase in

species richness, most likely due to its location below an old low-head dam located at the

outlet of Swett's Pond. This site often included species normally associated with a lentic

environment that had probably been washed downstream during floods in the spring and

had persisted in a few pools located within the site, associated with a road crossing,

which do not exist in the analogous site in Sedgunkedunk Stream (S4). These species

include brown bullhead Ameiurus nebulosus, chain pickerel Esox niger, and white perch

Morone americana. Predation by these species might explain the very low density of

small-bodied fishes like blacknose dace.

The fact that the fish community in Johnson Brook also showed consistent trends

along the stream gradient over time, similar to those that we found in Sedgeunkedunk

Stream up until dam removal, will allow us to use the BACI design to detect future

changes. During the continued monitoring of the recovery of Sedgeunkedunk Stream,

comparing the changes in the fish community of Sedgeunkedunk Stream to what happens

in Johnson Brook in the context of a BACI design will help isolate changes that are

attributable to restoration from those due to natural variability. This is possible because

the close proximity of the streams should make other regional environmental factors

constant. Over the course of this study longitudinal and temporal patterns in

Formatted: Font: (Default) TimesNew Roman, 12 pt, Italic

Formatted: Font: Not Bold

Formatted: Font: Not Bold, NotItalic

Deleted: e

Deleted: high flow events

Deleted: .

Deleted: Americana

Deleted: background noise

20

Sedgeunkedunk Stream, before the removal of the Mill Dam, and Johnson Brook were

consistent. This consistency allowed us to detect the immediate effects of the disturbance

that took place after the removal of the Mill Dam.

Three days after the removal of the Mill Dam on August 14, 2009, we began post-

removal sampling at site S1. The removal of the dam caused a large amount of sediment

to move through the system; immediately downstream of the former dam site, the stream

was very turbid and the substrate became covered in a layer of silt. The immediate

response to these changes was a decline in species richness and fish density below the

former dam location. Fish mortality can be expected during these types of events of

increased flow and turbidity (Doeg and Koehn 1994). In this case though, some of the

observed decline in density could be caused by movement and not mortality. While there

was not an increase in species richness above the dam site, there was a dramatic increase

in fish density. This is evidence that the species that were common in the stream were

able to move upstream of the disturbance.

The changes in the size structure of blacknose dace show that some of the decline

in fish density below the former dam site could be explained by fish moving upstream.

The length distribution of blacknose dace at site S1 and S2 immediately following the

removal of the Mill Dam show that while site S1 lost a majority of the larger fish during

the disturbance, site S2 showed gains in larger fish. The net gains for each site show that

the losses at site S1 correlate with the gains at site S2, especially if summer growth is

assumed to have taken place. There is still evidence of some mortality associated with

the disturbance, but it is clear that in the larger size-classes of blacknose dace (> 40mm)

many of the fish were able to move upstream. While successful colonization of former

Deleted:

21

dam impoundments by downstream fish species has been shown (Catalanato et al. 2007),

this would have occurred very quickly in our study. Fish densities at sites S3 and S4

reached their highest levels in the study after the dam removal. Small stream fishes like

blacknose dace can move over 1 km in a day (Albanese et al. 2003), which makes the

colonization of site S3 and S4 by fish that previously occupied reaches below the dam

possible. Even if individuals did not migrate all the way up to site S4 it is possible that

there could have been a cascading effect of displacement pushing fish upstream as

densities increased in sites upstream of the former dam. In sites that were not affected by

the removal of the Mill Dam, site S5 in Sedgeunkedunk Stream and all of the sites in

Johnson Brook, we did not detect any deviations from the dominant patterns in

abundance or community structure.

Another possible effect of the dam removal was the colonization of previously

unavailable stream sections by juveniles of anadromous Atlantic salmon. Prior to dam

removal we usually encountered one juvenile Atlantic salmon every time we sampled

below the dam. We were unsure if this was the result of spawning activity in the lower

stream reaches, or if these were individuals who had strayed from elsewhere in the

Penobscot watershed. In the sampling that took place following the dam removal we

captured 5 age 0+ Atlantic salmon parr directly above the former dam site and 55 parr at

site S3. It is possible that adults spawned in Sedgeunkedunk Stream below the Mill

Dam, and that resultant parr moved upstream in response to the disturbance caused by

dam removal. It is also possible that these salmon were juvenile landlocked Atlantic

salmon, but no spawning of landlocked Atlantic salmon has been documented in the

lower Penobscot watershed., Both of these explanations seem unlikely because we

Deleted: .

Deleted: E

Deleted: ,

Deleted: ,

Comment [SC2]: According to Joan, the anadromous explanation is the most likely

Deleted: This seems

22

captured only one fry in the spring of 2009, but otherwise the presence of Atlantic salmon

upstream of the dam immediately after its removal is unexplained and natural

recolonization of formerly inaccessible habitat is the most parsimonious explanation.

In order to detect the response of Sedgeunkedunk Stream to dam removal it is

important to know what the baseline conditions and variability were before the dam

removal. Prior to dam removal we found consistent patterns in space and over time,

which will allow us to detect the changes that take place following the restoration project.

The only deviation from this pattern occurred at the two sites immediately adjacent to the

dam immediately after removal, and our sampling allowed us to detect these changes.

Continued monitoring will allow us to show how the stream fishes below the dam site

recover from the disturbance, and how the entire stream fish community responds to a

more natural hydrology and increased connectivity to large river and marine ecosystems.

Dams have been in place in New England watersheds, for centuries. Every year

more and more dams are being removed in this country in with the goal of river

restoration (Heinz Center 2003). This is often done without a clear idea of what that

restoration means ecologically and without any plan to monitor the response (Bernhardt

et al. 2005). If there is monitoring it is often directed towards a target fish species of

interest (Doyle et al. 2005) rather than a community response.

Dam removals in this country are an expensive undertaking. This money is often

spent without ever really knowing what the results of the efforts are. The preferred

method of implementing a dam removal requires adaptive management. This form of

management involves constant scientific feedback, so that the effects of previous

Formatted: Indent: First line: 0.5"

Deleted: ,

Deleted: ¶

Deleted: Sedgeunkedunk Stream, and in other coastal Maine

Deleted: Their construction and continued presence has caused changes in habitat availability, water quality, and stream connectivity, and ultimately fish abundance and diversity.

Deleted: on

Comment [SC3]: Citation?

23

decisions are known and can inform future actions. This is impossible without effective,

well planned, monitoring (Heinz Center 2003).

Palmer et al. (2005) presented five criteria for successful restoration projects: a

guiding image of the restored system, indicators of ecosystem recovery are improved,

resilience is increased, there is no lasting harm, and that an ecological assessment is

completed. Monitoring is necessary if the project is to be deemed successful.

Sedgeunkedunk Stream provides a model of small coastal stream restoration throughout

Maine and elsewhere. The results of the monitoring here will inform adaptive

management within this system, the Penobscot River watershed, and other restoration

projects in the region.

Formatted: Indent: First line: 0.5"

Deleted: this type of managem

Deleted: which

Comment [SC4]: Here, you should either have an example from Sedgeunkedunk (e.g., monitoring costs vs. removal costs), or a citation. I called Rory and left a message to see if he has those dollar values…

Deleted: inexpensive

Deleted: laid out

Comment [SC5]: Not sure what “improved” means - are they specific?

Comment [SC6]: Technically Johnson IS DAMMED!!!

Deleted: of in the lower Penobscot watershed (includingrecords of ), , and

Deleted: for

Deleted: or not

Deleted: The biotic responses to dam removals usually are not monitored (Bernhardt et al. 2005), or the research is focused on a target fish species (Doyle et al. 2005), rather than a community response, like we have done in this study. Because we do not know what the abundances and size structure of fish species was before the construction of the dams, it is important to know what the baseline conditions and variability were before the restoration. Prior to dam removal we found consistent patterns in space and over time, and low variability, which will allow us to detect the changes that take place following the restoration project. The only deviation from this pattern occurred at the two sites immediately adjacent to the dam immediately after removal, and our sampling allowed us to detect these changes. Continued monitoring will allow us to show how the stream fishes below the dam site recover from the disturbance, and how the entire stream fish community responds to a more natural hydrology and increased connectivity to large river and marine ecosystems. ¶This type of monitoring can be useful in detecting the response of fish communities in dammed streams as dam removals become a more popular restoration effort in Maine and beyond. This restoration project is notable because it takes place downstream of the anticipated Penobscot River Restoration Project, in which two large main-stem dams are to be removed, and fish passage installed or improved at two other dams (PRRT 2009). We view the Penobscot River, structurally and functionally as thousands of Sedgeunkedunk –sized tributaries, and thus this small scale restoration project is the first step towards ... [1]

24

Chapter 2

DISTRIBUTION AND ABUNDANCE OF ANADROMOUS SEA LAMPREY

(Petromyzon marinus) SPAWNERS IN A DAMMED STREAM: CURRENT

STATUS AND POTENTIAL RANGE EXPANSION

FOLLOWING BARRIER REMOVAL

Introduction

Many rivers and ponds in Maine once harbored spawning runs of anadromous fish

species, such as Atlantic salmon (Salmo salar), alewife (Alosa pseudoharengus) sea

lamprey (Petromyzon marinus) and rainbow smelt (Osmerus mordax), (Saunders et al.

2006). These fish likely transported marine derived nutrients and energy (MDNE) into

otherwise oligotrophic freshwater ecosystems, that could stimulate primary productivity

and microbial decomposition of detritus (Brock et al. 2007; Durbin et al. 1979; Wipfli et

al. 1998). MDNE subsidies from Pacific salmon increase productivity and growth of

resident fish species in many Pacific Northwest watersheds (Bilby et al 1996; Scheuerell

et al. 2007; Wipfli et al. 2003), and historically, anadromous fishes in Maine probably

provided similar subsidies (Saunders et al., 20056) . As dams have severed this marine-

freshwater connectivity in Maine and elsewhere, anadromous species have experienced

worldwide decline (Limburg and Waldman 2009) and freshwater systems have become

more oligotrophic (Stockner et al. 2000).

Sedgeunkedunk Stream, a small tributary of the Penobscot River below head-of-

tide (Figure 2.1), typifies small streams in Maine impacted by dams. Current restoration

efforts offer an opportunity to assess how the structure, function, and resilience of the fish

Deleted: likely

Deleted: ing

Deleted: rates

Deleted: Presumably, such

Deleted: ed

Deleted: subsidized

Comment [SC7]: Alaska IS the PNW

Deleted: and Alaskan

25

community responds to long-term disturbance by, and then relief from, habitat

fragmentation from multiple dams. Runs of anadromous fishes in Sedgeunkedunk

Stream either disappeared or were reduced substantially after the building of three dams

on the stream (at 0.6, 5.3 and 8.0 km upstream of the confluence with the Penobscot

River; “stream km”). The downstream dam has been there, in some form, for over two

centuries (S. Shepard, Aquatic Science Associates, personal communication). Declines

in anadromous fishes in Sedgeunkedunk Stream mirror those in the entire Penobscot

watershed, which contains over 100 known dams; because Sedgeunkedunk Stream is one

of only three major tributaries that could receive anadromous fishes before they

encounter a relatively large, mainstem dam on the Penobscot River (i.e., Veazie Dam),

restoration of anadromous fishes here should precede that in tributaries farther upstream

in the watershed when the mainstem dams are removed or bypassed. Currently, Atlantic

salmon in Sedgeunkedunk Stream and most of the Penobscot watershed are listed as a

federally-endangered species (74 Fed. Reg. 29344; June 19, 2009), and alewife, sea

lamprey, and rainbow smelt are at historic low levels of abundance (Saunders et al.

2006).

As part of a collaborative restoration project, fish passage has been created or

restored in Sedgeunkedunk Stream at the location of two former dam sites between Fields

Pond and the confluence with the Penobscot River (Figure 2.1). The lowermost dam

(Mill Dam) was removed in August 2009, and the middle dam (Meadow Dam) was

bypassed in August 2008 by a rock-ramp fish-way that allows fish passage while

maintaining current water levels in Fields Pond and adjacent wetlands. Thus habitat

Deleted: Recent d

Deleted: (PRRT 2009)

Deleted: be accelerated relative to

Deleted: NOAA 2009

26

connectivity for migrating anadromous and resident fishes has been restored throughout

most of the Sedgunkedunk watershed.

Currently, Sedgeunkedunk Stream receives a small run of anadromous sea

lamprey, the only anadromous fish known to spawn in the stream reliably. Adult sea

lampreys are parasitic in the ocean and return to freshwater after two to seven years to

spawn (Beamish 1980). Unlike many anadromous species, sea lamprey do not home to

natal streams (Bergstedt and Seeyle 1995; Waldman et al. 2008), and may select

spawning streams based on flow, temperature (Andrade et al. 2007), or the presence of

chemical compounds released by both ammocoetes and sexually-mature males (Li et al.

2002; Vrieze et al. 2010). This should allow for faster expansion and colonization into

new habitats than for anadromous species with strong homing tendencies, like Atlantic

salmon. The rapid expansion of sea lamprey throughout the upper Great Lakes is strong

evidence of their ability to colonize suitable and accessible streams (Smith and Tibbles

1980). Their capacity for rapid range expansion makes sea lamprey likely to benefit

immediately from restoration efforts in Sedgeunkedunk Stream.

In North America, research on sea lamprey is dominated by Great Lakes issues,

where the species is invasive upstream of Niagara Falls but probably native throughout

the watersheds of Lakes Ontario and Champlain (Waldman et al. 2004, 2006; but see

Eshenmeyer 2009). Sea lamprey invasion has been devastating for economically-

valuable recreational and commercial fisheries (Christie 1973). Because of their highly-

visible parasitism on popular sport fish, sea lamprey have acquired a negative public

image not only in the Great Lakes region, but also within their native range.

Deleted: documented

Deleted: Sea lamprey, u

Deleted: es

Deleted: certainly

Deleted: , and

27

Recolonization and recovery of sea lamprey may be a critical step in the

restoration of native anadromous fish assemblages in Maine. Sea lamprey are an

ecologically significant member of stream ecosystems, both as spawning adults and in the

larval form. They are selemparous, dying in early summer after spawning, making the

adults potential sources of MDNE in otherwise oligotrophic streams – a potentially

important combination of time and place for nutrient subsidies in Maine (Nislow and

Kynard 2009). Adults build nests in shallow riffles or the tails of pools, by moving

cobbles and pebbles into mounds located downstream of a pit. Spawning occurs in the

pit and includes tail and body flexions and vibrations on or near the substrate. Both nest

building and spawning activities appear to clear away fine sediments from the coarse

substrate on a local scale (Kircheis 2004), perhaps reducing substrate armoring and

embeddedness, similar to the effects seen from Pacific salmon aggregations

(Montgomery et al. 1996). Thus it has been suggested that sea lamprey nest building

activities may “condition” potential spawning habitat for Atlantic salmon (Kircheis 2004;

Saunders 2006). Finally, ammocoetes are a potential prey base for other resident species

and act as filter feeders actively trapping nutrients from the water column (Applegate

1950).

This study provides a baseline assessment of the status of sea lamprey in

Sedgeunkedunk Stream, where restoration efforts should make the watershed more

accessible to anadromous fishes. Specifically, our objectives are to: 1) quantify the

abundance and movements of sea lamprey in Sedgeunkedunk Stream before the removal

of the Mill Dam 2) quantify the distribution and abundance sea lamprey nests in

Sedgeunkedunk Stream before the removal of the Mill Dam 3) Assess the likelihood of

Deleted: Sea lamprey are of particular interest in this system not only because they are the lone anadromous species that has been documented to spawn reliably, but because they likely play an important ecological role as a member of a historic assemblage of native anadromous fishes.

Deleted: Q

Deleted: Q

28

colonization of sea lamprey into new habitats made accessible following the removal of

the Mill Dam. This work represents the first step in a long term monitoring of the

response of sea lamprey to this restoration, and the influence of their presence on the

Sedgeunkedunk Stream ecosystem.

Methods

Study Area

Sedgeunkedunk Stream is a third-order tributary of the Penobscot River, flowing

through the Town of Orrington and the City of Brewer, Maine. (Figure 2.1). The

lowermost Mill Dam, which was removed in August 2009, was located just 700 meters

upstream of head of tide, near the stream’s confluence with the Penobscot River, and

Figure 2.1. Location of Sedgeunkedunk Stream, Fields Pond, and the dams that were removed during the Sedgeunkedunk Stream restoration. Penobscot Co. Maine, USA.

P E N

O B S C

O T

R

I V E R

Fields Pond

Sedgeunkedunk Stream

Mill Dam

Meadow Dam

Comment [SC8]: Text in this figure way too small. Same comments about embedding figure in textbody

Deleted: <sp>

29

limited access for anadromous species farther upstream. The removal of this barrier

provides for presumably unimpeded access from the Atlantic Ocean into Sedgeunkedunk

Stream, via the lower Penobscot River. The uppermost Brewer Lake Dam, located

between Fields Pond and Brewer Lake, will remain intact, and continue to block passage

further upstream in the watershed.

Sea lamprey population estimate and behavioral evaluation

We captured sea lamprey in an Indiana-style trap-net that spanned the entire width

of the Sedgeunkedunk Stream, 90 m upstream of the confluence with the Penobscot River

(river km 36), approximately at head-of-tide. Net dimensions included a 3 mm square

mesh, 1.3m x 1.6m rectangular mouth, 1m diameter circular cod end, 2.5 m length, and

wings that extended the width of the stream (4.0 m at time of deployment). We deployed

the net from 15 May to 1 July, 2008. At deployment the thalweg depth was 0.8m, but

water depth varied with stream discharge and tidal cycle. During the first five sampling

days (18-22 June) high water rendered this net ineffective. During this time period we

captured sea lamprey easily with hand nets during stream surveys. The lack of deep

pools and instream structure, combined with low turbidity, facilitated capture of sea

lamprey in the stream. We tagged each captured sea lamprey with both an internal

Passive Integrated Transponder (PIT) tag and an external floy tag. Mass, length, and sex

were recorded for each sea lamprey before release.

During the sea lamprey run, stream surveys were conducted daily. We surveyed

the 610-m reach from the trap net to the Mill Dam, and 2 km above the Mill Dam, and

identified observed fish using a portable PIT tag reader, thus obviating the need to handle

Deleted: 1 inch

Deleted: made us confident in out ability to spot sea la

Deleted: ,

Deleted: was

Deleted: walked

Deleted: -

Deleted: pack

30

fish repeatedly (Hill et al. 2006). We captured any non-tagged sea lamprey with a hand

net and processed them accordingly. At each encounter with a sea lamprey the following

were recorded: whether it was alive or dead, the location in the stream, whether or not it

was found on a nest, and if so which nest and what other individuals were in the nest.

We estimated abundance of adult sea lamprey in Sedgeunkedunk Stream using a POPAN

Jolly-Seber open population mode (Arnason and Schwarz 1995) in the program MARK

(White, G.C. 2008).

Nest surveys and abundance estimation

Daily nest surveys were also conducted throughout the duration of the sea

lamprey spawning run. Each nest location recorded and marked on the stream bank with

flagging. Maximum length, width, and depth were recorded for both the upstream

depression and the downstream mound of each nest, and any sea lamprey occupying nests

were identified. Abundance of nests was estimated using a Cormack-Jolly-Seber mark-

recapture method in MARK. The 610 meters of accessible stream below the Mill Dam

was divided into 61 10-meter sections and lamprey nest abundance was recorded for each

section. We also recorded substrate size along a random transect, running perpendicular

to the stream bank, in each

stream section. Substrate was

sampled by walking heel to toe

along the transect and measuring

the size of the substrate

component at each step.

0

0.2

0.4

0.6

0.8

1

6/18 6/19 6/20 6/21 6/22 6/23 6/24 6/25 6/26 6/27

Date

Prop

orti

on e

nter

ed

Figure 2.2 The cumulative proportion of sea lamprey entering Sedgeunkedunk Stream, Penobscot Co. Maine. 2008.

31

Results

In 2008, the spawning run of

sea lamprey in Sedgeunkedunk

Stream began on 18 June and ended

on 27 June. During that time period

a total of 47 sea lamprey (21 female

and 26 male) was captured and

tagged. The mark-recapture model estimated total abundance (N ± 2SE) of 47 ± 0. Of the

47 lamprey caught only 16 were captured using the trap net. Of those sea lamprey that

evaded the trap net, the mean point of capture was 345 ± 52 m upstream of the net, and 5

sea lamprey were captured initially in the pool directly downstream of the Mill Dam. No

lamprey were observed above the Mill Dam. Males and females were similar in size

(males mean length was 625 ± 23cm and mean mass was 750 ± 100g, females mean

length was 612 ± 22cm and mean mass was 700 ± 100g). There was not a detectable

pattern in size or sex of a sea lamprey and the arrival of it to the stream. Sea lamprey

entered the stream at a

relatively steady rate

throughout the period (Figure

2.2).

Individual sea

lamprey were active in the

stream for an average of 2.5

± 0.5 days (range 1 – 6 days),

Date entered Total Males Females Water temp. (°C) Days active Nests used6/18 5 1 4 16.9 4.4±1.0 1.8±1.26/19 9 5 4 16.5 3.3±1.0 2.0±0.76/20 3 3 0 16.9 4.0±2.0 3.3±1.36/21 7 4 3 18.8 2.1±0.5 1.4±0.76/22 4 2 2 19.5 3.0±2.2 2.8±1.76/23 5 2 3 19.5 1.4±0.8 0.8±0.66/24 3 1 2 20.9 2.0±2.0 0.3±0.36/25 5 4 1 20.9 1.2±0.4 0.2±0.26/26 3 2 1 20.8 1.0 0.3±0.36/27 1 1 0 21.9 1.0 0.0N/A 2 1 1 - - -Total 47 26 21 - 2.5±0.5 1.4±0.4

Table 2.1. Number of live sea lamprey captured for the first time on each day of the sampling period, the mean water temperature of that day and the mean number of days observed active in the river, and mean number of total nests sea lamprey that entered on that day used before they died, Sedgeunkedunk Stream, Penobscot

Co., ME 2008. Means are presented with ±2SE.

02468

1012

0 1 2 3 4 5

# of nests used

# of

lam

prey

Male

Female

Figure 2.3. Frequency of the number of nests individual male, and female sea lamprey were observed using during the duration of their activity in Sedgeunkedunk Stream, Penobscot, Co., Maine. 2008

Deleted: captured

32

and were observed on average 2.3 ± 0.4 times (range 1 – 5 observations). Average linear

distance traveled, based on successive observations of individuals, was 103 ± 48 m (not

including individuals observed only once); minimum and maximum daily means were 89

± 55 m, and 138 ± 50 m. Daily distances ranged from 0 m (i.e. individuals found on the

same nest on consecutive days, n = 3 ) to 255 m. Every lamprey was observed alive; of

these 47 live fish captured, we observed the carcasses of 17 individuals. All observed

carcasses were observed in the wetted stream channel. Carcasses were found, on average,

116 ± 71 m downstream of the point of last live observation. We did not observe any

additional downstream movement of carcasses after first sighting.

While sea lamprey entered the stream everyday from June 18-June 27, we did not

observe any live sea lamprey after June 27. Mean water temperature increased everyday

during the run starting at 16.9° C and reaching 21.9°C on the last day. Sea lamprey that

arrived in the stream earlier, regardless of sex, were able to spend more days in the

stream and, on average, were associated with more total nests before death (Table 2.1).

Females were more likely to be observed on only one or two nests than were males,

whereas males were less likely to be observed associated with a nest or be seen on 3 or

more nests, than were

females (Figure 2.3).

During stream

surveys, 31 nests were

identified. There were

no nests identified

above the Mill Dam.

0

1

2

3

4

5

6

0100 200 300 400 500 600 700

Stream meter

# of

lam

prey

nes

ts

00.10.20.30.40.50.60.70.80.91

% fi

nes

Nests

Fines

Mill Dam

NO ACCESS

Figure 2.4. Distribution sea lamprey nests, and percentage of substrate made up fines (smaller than 2mm) along the stream gradient of Sedgeunkedunk Stream, Penobscot Co., Maine. 2008.

Deleted: .

Deleted: all had died by

Deleted: 8

33

The mark-recapture model estimated total nest abundance (N ± 2SE) of 31 ± 0. Sea

lamprey nests were present in 39% (24 of 61 stream sections) of the stream available to

them. These nests were distributed throughout the stream upstream of head-of-tide to

immediately downstream of the Mill Dam, where the substrate had a low percentage of

fine sediment (<2mm) (Figure 2.4). Sea lamprey nests were present in 87% of the stream

sections characterized by less than 20% fine sediment. No nests were found below head-

of-tide or above the dam. Mean nest dimensions included a pit that was 75cm long, 56cm

wide and 20cm deep, and a mound that was 62cm long, 57cm wide

and 8cm deep (Table 2.2). Nests as attendance ranged from 0 to 6 sea lamprey at one

time, never with more than 2 males or 4 females on any one nest, with a mean number

1.7±0.4 individuals. Females on nests were associated with, on average, 0.9 ± 0.3 other

males and 0.8 ± 0.4 other females. Males, on average, occupied nests with 0.2 ± 0.2

other males, and 0.7 ± 0.4 other females. Females shared nests with, on average, more

male partners than did males (two sample paired means t-test p < 0.0004, df = 15).

Discussion

In 2008, lamprey arrived in Sedgeunkedunk Stream later than in most streams in

the lower Penobscot watershed (Oliver Cox, Maine Department of Marine Resources,

personal communication). In Maine, sea lamprey spawning occurs in late May and early

June when the water temperature reaches 17-19° C (Kircheis 2004). By the time the sea

lamprey arrived the mean daily water temperature in Sedgeunkedunk Stream had

Pit length Pit width Pit depth Pile length Pile width Pile depth Max. # of sea lampreySmallest 4 2 40 23 37 35 11 0Largest 126 101 23 128 101 9 6M ean 7 5±1 1 56±9 20±2 62±14 57±8 8±1 1.8±0.4

Table 2.2. Smallest, largest and mean sizes of sea lamprey nests, and the maximum number of sea lamprey observed on them in Sedgeunkedunk Stream ,Penobscot Co. ME. 2008. Measurements are in cm. Mean of the maximum number of sea lamprey for all nests is shown with ±2SE. Smallest and largest nests determines by area.

Comment [SC9]: Very awkward. “Females shared nests with, on average, more male partners than did males” or something like that

Deleted: There was significantly different between males

Deleted: and females in the number of other males occupying the dame nest nest

34

exceeded 20° C for most of early June. The sea lamprey run might have been triggered

by the temperatures dropping below 17° C on June 17, and then increasing back up to

20°C over 9 days (Binder and McDonald 2008). This decrease in temperature coincided

with an increase in discharge, which has been shown to initiate migration (Almeida et al.

2002). This period of lower temperatures was short, so individuals who arrived early had

more opportunity to take advantage of a brief period of favorable spawning conditions.

We saw no evidence of nest fidelity in either male or female sea lamprey. In fact

we observed that sea lamprey moved often in the system and used multiple nests, with

individuals using as many as 5 nests and travelling as much as 255m between nest sites in

consecutive days. This is consistent with what has been observed in other lamprey

species (Moser et al. 2007). Thus, a single sea lamprey may modify the substrate in

several reaches of the same stream, and multiple sea lampreys may build and expand

upon on a communal nest. Males initiate nest building (Kircheis 2004), and typically

were associated with multiple nests. Males were more likely to be observed away from a

nest than were females, which could be caused by their tendency to move among multiple

nests, possibly in order to avoid sharing nests with other males.

Despite the relatively short reach of Sedgeunkedunk Stream accessible to

anadromous fishes, sea lampreys did spawn there in modest numbers. Our methods of

capture and survey for both sea lamprey and their nests were successful, and our models

of abundance indicated that we were able to capture every sea lamprey in the system and

record every nest location. The use of PIT packs was also very successful in identifying

individuals without any observable disruption of their behavior and we were able to

record the location and activity of individuals on a daily basis.

Deleted: was

35

Historic abundances of sea lamprey in Maine are unknown (Kircheis 2004).

Therefore, we do not know whether 47 sea lamprey in Sedgeunkedunk Stream represents

a run of spawners that is persistent in the system, despite the limited habitat available to

them, or one that is in decline. During 20 year study on the Fort River (which is similar in

size to Sedgeunkedunk Stream), Massachusetts, there was a mean of 80 spawners per

year entering the stream (Nislow and Kynard 2009). Tributaries of Lake Ontario with

less than 1km of available stream below a dam support runs of up to 800 sea lamprey

(Binder et al. 2010). Now that the dams on Sedgeunkedunk have been removed, what

happens next will help us to determine the viability of the sea lamprey spawning run in

the stream. In Portugal, sea lamprey have been shown to use previously-inaccessible

habitat after connectivity is restored (Almeida et al. 2002). It is likely that there will be

an immediate expansion of sea lamprey spawning range in Sedgeunkedunk Stream now

that the Mill Dam has been removed. Many sea lamprey swam to the Mill Dam on the

first day they entered the stream before moving downstream to other nesting habitats.

Also. sea lamprey nests were distributed throughout the available stream below the dam.

We found nests in almost all of the reaches that were not dominated by fine sediment.

This makes us believe that the only factor preventing spawning by sea lamprey further

upstream was the Mill Dam.

Sea lamprey potentially provide important ecological functions in the stream

communities, particularly ones that have lost other anadromous components. Their

semelparous life history make them a potential source of MDNE. We only observed 17

carcasses in the stream, which suggests that the other sea lamprey left before death, the

carcasses were transported somewhere that we could not see it, either by the current or

Deleted:

Deleted: danger of disappearing

Deleted: currently

Deleted: i

Deleted: all the way upstream either to the riffle, or the pool, below the

Deleted: as

36

terrestrial scavengers, or the carcass was washed out of the system. Most sea lamprey

carcasses decompose within 1-2 weeks (Nislow and Kynard 2009), such that MDNE

addition occurs in Sedgeunkedunk Stream before high discharge events wash the

carcasses into the Penobscot River. We did not observe downstream movements of intact

carcasses, but did witness rapid disintegration following carcass deposition.

The nest building behavior of sea lamprey has the potential to act as a substrate

conditioner for Atlantic salmon, especially in a system that does not currently support

reliable salmon spawning. The mean length of sea lamprey nests in the stream was

1.37m, and even the smallest nests that we observed were nearly a meter long; nests of

this size could have a potentially important impact on the substrate of a stream that is

often less than 4 meters wide. If the abundance and distribution of the spawning run

increases with restoration, as we expect, this creates a large potential for sea lamprey to

affect the substrate of a large portion of Sedgeunkedunk Stream. The synergism between

a suite of anadromous species could be an important factor in each species’ recovery

(Saunders et al. 2006). If the numbers of spawning sea lamprey using Sedgeunkedunk

Stream increase following dam removals it could potentially impact the ecology of other

anadromous species through nutrient addition and substrate modification.

Deleted: t

37

REFERENCES

Albanese, B., P.L. Angermeier and C. Gowan. 2003. Designing mark-recapture

studies to reduce the effects of distance weighting on movement distance

distributions of stream fishes. Transactions of the American Fisheries Society

132: 925-939.

Almeida, P.R., B.R. Quintella and N.M. Dias. 2002. Movement of radio-tagged

anadromous sea lamprey during spawning migration in the River Mondego

(Portugal). Hydrobiologia 483:1-8.

Andrade, N.O., B.R. Quintella, J. Ferreira, S. Pinela, I. Póvoa, S. Pedro and P.R.

Almeida. 2007. Sea lamprey spawning migration in the Vouga river basin

(Portugal): poaching impact, preferential resting sites and spawning grounds.

Hydrobiologia 582:121-132.

Applegate, V.C. 1950. Natural history of the sea lamprey in Michigan. U.S. Fish Wild.

Serv., Spec. Sci. Rep. Fish. 55.

Arnason, A. N. and C.J. Schwarz. 1995. POPAN-4: Enhancements to a system for

the analysis of mark-recapture data from open populations. Journal of Applied

Statistics 22:785-800

Baldigo, B.P. and D.R. Warren. 2008. Detecting the response of fish assemblages to

stream restoration: effects of different sampling designs. North American Journal

of Fisheries Management 28: 919-934.

Beamish, F.W.H. 1980. Biology of the North American anadromous sea lamprey.

Canadian Journal of Fisheries and Aquatic Sciences 37:1924-1943.

Benke, A.C. 1990. A perspective on America’s vanishing streams. Journal of the North

38

American Benthological Society. 9:77-88.

Bernhardt, E.S., M.A. Palmer, J.D. Allan, G. Alexander, K. Barnas, S. Brooks, J.Carr,

S. Clayton, C. Dahm, J. Follstad-Shah, D. Galat, S. Gloss, P. Goodwin, D. Hart,

B. Hassett, R. Jenkinson, S. Katz, G.M. Kondolf, P.S. Lake, R. Lave, J.L. Meyer,

T.K. O’Donnell, L. Pagano, B. Powell, E. Suddeth. 2005. Sythesizing U.S. River

Restoration Efforts. Ecology 308: 636-637.

Bergstedt, R.A., J.G. Seelye. 1995. Evidence for lack of homing by sea lampreys.

Trans. Am. Fish. Soc. 124, 235–239.

Bilby, R.E., B.R. Fransen, P.A. Bisson. 1996. Incorporation of nitrogen and carbon from

spawning coho salmon into the trophic system of small streams: Evidence from

stable isotopes. Canadian Journal of Fisheries and Aquatic Sciences 53: 164-173.

Binder, T.R., and D.G. McDonald. 2008. The role of temperature in controlling diel

activity in upstream migrant sea lampreys (Petromyzon marinus). Can. J. Fish.

Aquat. Sci. 65:1113-1121.

Binder, T.R., R.L. McLaughlin and D.G. McDonald. 2010. Relative importance of water

temperature, water level, and lunar cycle to migratory activity in spawning-phase

sea lampreys in Lake Ontario. Transactions of the American Fisheries Society

139:700-712.

Brock, C.S., P.R. Leavitt, D.E. Schindler and P.D. Quay. 2007. Variable effects of marine

derived nutrients on algal production in salmon nursery lakes of Alaska during the

past 300 years. Limnology and Oceanography 52:1588-1598.

39

Burroughs, B.A., D.B. Hayes, K.D. Klomp, J.F. Hansen and J. Mistak. 2009. Effects of

Stronach Dam removal on fluvial geomorphology in the Pine River, Michigan,

United States. Geomorphology 110: 96-107.

Catalano, M.J., M.A. Bozek and T.D. Pellett. 2007. Effects of dam removal on fish

assemblage structure and spatial distributions in the Baraboo River, Wisconsin.

North American Journal of Fisheries Management 27:519-530.

Christie, W.J. 1973. A review of the changes in fish species composition of Lake Ontario.

Great Lakes Fishery Commission, Technical Report 23, Ann Arbor, Michigan.

Doeg, T.J. and J.D. Koehn. 1994. Effects of draining and desilting a small weir on

downstream fish and macroinvertebrates. Regul River 9:263-277.

Doyle, M.W., E.H. Stanley, C.H. Orr, A.R. Selle, S.A. Sethi and J.M. Harbor. 2005.

Stream ecosystem response to small dam removal: lessons from the Heartland.

Geomorphology 71:227-244.

Durbin, A.G., S.W. Nixon and C.A. Oviatt. 1979. Effects of spawning migration of the

alewife on freshwater ecosystems. Ecology 60:8-17.

Ehrlich, P.R. and J.P. Holden. 1971. Impact of population growth. Science 171:1212-

1217.

Elmqvist, T., C. Folke, M. Nystrom, G. Peterson, J. Bengtsson, B. Walker and J.

Norberg. 2003. Response diversity, ecosystem change, and resilience. Frontiers in

Ecology and the Environment 1:488-494.

Grossman, G.D., R.E. Ratajczak, C.M. Wagner and J.T. Petty. 2010. Dynamics and

regulation of the southern brook trout (Salvelinusfontinalis) population in an

Appalachian stream. Freshwater Biology 55: 1494-1508.

40

Gunderson, L.H. 2000. Ecological resilience – in theory and application. Annual Review

of Ecology and Systematics. 31:425-439.

The Heinz Center. 2003. Dam Removal: Science and Decision Making.

Hall, C.A.S. 1972. Migration and Metabolism in a Temperate Stream Ecosystem. Ecology

53:585-604.

Hart, D.D., T.E. Johnson, K.L. Bushaw-Newton, R.J. Horwitz, A.T. Bednarek, D.F.

Charles, D.A. Kreeger and D.J. Velinsky. 2002. Dam removal: Challenges and

opportunities for ecological research and river restoration. Bioscience 52:669-681.

Hill, M.S., G.B. Zydlewski, J.D. Zydlewski, and J.M. Gasvoda. 2006. Development and

evaluation of portable PIT tag detection units: PIT packs. Fisheries Research

77:102-109.

Holling, C.S. 2004. From complex regions to complex worlds. Ecology and Society 9:11.

Jackson, D.A., P.R. Peres-Neto and J.D. Olden. 2001. What controls who is where in

Freshwater fish communities – the roles of biotic, abiotic, and spatial factors.

Canadian Journal of Fisheries and Aquatic Science 58: 157-170.

Jones, N.E. 2010. Incorporating lakes within the river discontinuum: longitudinal changes

in ecological characteristics in stream-lake networks. Canadian Journal of

Fisheries and Aquatic Sciences. 67:1350-1362.

Koster, D., J. Lichter, P.D. Lea, and A. Nurse. 2007. Histirical Eutrophication in a River-

Estuary Complex in Mid-Coast Maine. Ecological Applications. 17:765-778.

Kinsolving, A. D. and M. B. Bain 1993. Fish assemblage recovery along a riverine

disturbance gradient. Ecological Applications 3:531-544.

Kircheis, F.W. 2004. Sea lamprey. F.W. Kircheis L.L.C., Carmel, Maine.

Krebs, C. 1989. Ecological Methodology. HarperCollins, New York.

41

Latimore, J.A. and D.B. Hayes. 2008. The vigor-organization-resilience concept of

ecological health: lessons from temperate warmwater stream fish communities.

Pp. 1807 1813 In J. Nielsen, J. Dodson, K. Friedland, T. Hamon, N. Hughes, J.

Musick, and E. Verspoor, eds. Proceedings of the Fourth World Fisheries

Congress: Reconciling Fisheries with Conservation. American Fisheries Society,

Symposium 49, Bethesda, MD.

Li, W., A.P. Scott, M.J. Siefkes, H. Yan, Q. Lui, S.-S. Yun and D.A. Gage. 2002. Bile

acid secreted by male sea lamprey that acts as a sex pheromone. Science 296:138-

141.

Ligon, F.K. , W.E. Dietrich and W.J. Trush. 1995. Downstream ecological effects of

dams. Bioscience 45:183-192.

Limburg, K.E., and J. R. Waldman. 2009. Dramatic declines in North Atlantic

diadromous fishes. BioScience 59: 955-965.

Lyons, J. and D.W. Schneider. 1990. Factors influencing fish distribution and community

structure in a small coastal river in southwestern Costa Rica. Hydrobiologia

203:1-14.

Martinez, P.J.. T.E. Chart, M.A. Trammel, J.G. Wullschleger, and E.P. Bergerson. 1994.

Fish species composition before and after construction of a main stem reservoir

on the White River, Colorado. Environmental Biology of Fishes 40:199-232.

Mitchell, S.C. and R.A. Cunjak. 2007. Stream flow, salmon and beaver dams: roles in the

structuring of stream fish communities within an anadromous salmon dominated

stream. Journal of Animal Ecology 76: 1062-1074.

42

Montgomery, D.R.. J.M. Buffington, N.P. Peterson, D. Schuett-Hames and T.P. Quinn.

1996. Stream-bed scour, egg burial depths, and the influence of salmonid

spawning on bed surface mobility and embryo survival. Canadian Journal of

Fisheries and Aquatic Sciences 53:1061-1070.

Moser, M.L., J.M. Butzerin and D.G. Dey. 2007. Capture and collection of lampreys: the

state of the science. Rev. Fish Biol. Fisheries 17:45-46.

Moyle, P.B. and B. Vondracek. 1985. Persistence and structure of the fish assemblage in

a small California stream. Ecology. 66: 1-13.

Nislow, K.H. and B.E. Kynard. 2009. The role of anadromous sea lamprey in nutrient

and material transport between marine and freshwater environments. American

Fisheries Society Symposium 69: 485-494.

Palmer, M.A., E.S. Bernhardt, J.D. Allan, P.S. Lake, G. Alexander, S. Brooks, J. Carr, S.

Clayton, C.N. Dahm, J. Follstad Shah, D.L. Galat, S.G. Loss, P. Goodwin, D.D.

Hart, B. Hassett, R. Jenkinson, G.M. Kondolf, R. Lave, J.L. Meyer, T.K.

O’Donnell, L. Pagano and E Sudduth. 2005. Standards for ecologically successful

river restoration. Journal of Applied Ecology 42:208-217.

Payton, M.E., M.H. Greenstone and N. Schenker. 2003. Overlapping confidence intervals

or standard error intervals: what do they mean in terms of statistical significance?

Journal of Insect Science 3:34.

Petts, G.E. 1980. Long-term consequences of upstream impoundment. Environmental

Conservation. 7: 325-332.

Phillips, B.W. and C.E. Johnston. 2004. Fish assemblage recovery and persistence.

Ecology of Freshwater Fish. 13: 145-153.

43

Quinn, J.W. and T.J. Kwak. 2003. Fish assemblage changes in an Ozark River after

impoundment: a long-term perspective. Transaction of the American Fisheries

Society 132: 110-119.

Reynolds, J. B. 1996. Electrofishing. Pages 221–253 in B. R. Murphy and D. W. Willis,

editors. Fisheries Techniques. American Society, Bethesda,Maryland.

Saunders, R., M.A. Hachey and C.W. Fay. 2006. Maine’s anadromous fish community:

past, present, and implications for Atlantic salmon recovery. Fisheries 31:537-

547.

Scheuerell, M.D., J.W. Moore, D.E. Schindler and C.J. Harvey. 2007. Varying effects of

anadromous sockeye salmon on the trophic ecology of two species of resident

salmonids in southwest Alaska. Freshwater Biology 52:1944-1956.

Schlosser, I.J. 1982. Fish community structure and function along two habitat gradients in

a headwater stream. Ecological Monographs 52:395-414.

Schmetterling, D.A. and S.B. Adams. 2004. Summer movements within the fish

community of a small montane stream. North American Journal of Fisheries

Management 24:1163-1172.

Sheldon, A.L. 1968. Diversity and Longitudinal Succession in Stream Fishes. Ecological

Society of America 49:194-198.

Smith, T. and C.E. Kraft. 2005. Stream fish assemblage in relation to landscape position

and local habitat variables. Transactions of the American Fisheries Society 134:

430-440.

44

Smith, B. R., and J. J. Tibbies. 1980. Sea lamprey (Petromyzon marinus) in Lakes Huron,

Michigan, and Superior: history of invasion and control, 1936-78. Canadian

Journal of Fisheries and Aquatic Sciences 37:1780-1801.

Stewart-Oaten, A., W.W. Murdoch and K.R. Parker. 1986. Environmental impact

assessment: “pseudoreplication” in time? Ecology 67: 929-940

Stocker, J.G., E. Rydin and P. Hyenstrand. 2000. Cultural oligotrophication: causes and

consequences for fisheries resources. Fisheries 25:7-14.

Swink, W. D. and K. E. Jacobs. 1983. Influence of a Kentucky flood-control reservoir on

the tailwater and headwater fish populations. North American Journal of

Fisheries Management 3:197-203.

Vitousek, P.M., H.A. Mooney, J. Lubchenco and J.M. Melillo. 1997. Human of Earth’s

ecosystems. Science 277:494-499.

Vrieze, L.A., R. Bjerselius and P.W. Sorenson. 2010. Importance of the olfactory sense to

migratory sea lampreys seeking riverine spawning habitat. Journal of Fish

Biology 76:949-964.

Waldman, J.R., C. Grunwald, N.K. Roy and I.I. Wirgin. 2004. Mitochondrial DNA

analysis indicates sea lampreys are indigenous to Lake Ontario. Transactions of

the American Fisheries Society 133:950-960.

Waldman, J.R., C. Grunwald and I. Wirgin. 2006. Evaluation of the native status of sea

lampreys in Lake Champlain based on mitochondrial DNA sequencing analysis.

Transactions of the American Fisheries Society. 135:1076-1085.

Waldman, J.R., C. Grunwald and I. Wirgin. 2008. Sea lamprey Petromyzon marinus: an

exception to the rule of homing in anadromous fishes. Biology Letters 4:659-662.

Deleted: Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell and C.E. Cushing. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Science 37:130-137¶

45

Ward, J. V., and J. A. Stanford. 1983. The serial discontinuity concept of lotic

ecosystems, Dynamics of Lotic Ecosystems. In: The Dynamics of Lotic

Ecosystems, T.D. Fontaine,III and S.M. Bartell (eds.), pp. 29-42 edition. Ann

Arbor Science, Ann Arbor, MI.

Winston, M.R., C.M. Taylor and J. Pigg. 1991. Upstream extirpation of four minnow

species due to the damming of a prairie stream. Transactions of the American

Fisheries Society 120:98-105.

Wipfli, M.S., J. Hudson and J. Caoutte. 1998. Influence of salmon carcasses on stream

productivity: response of biofilm and benthic macroinvertebrates in southeastern

Alaska, U.S.A. Canadian Journal of Fisheries and Aquatic Science 55: 1503-

1511.

Wipfli, M.S., J.P. Hudson, J.P. Caoutte, and D.T. Chaloner. 2003. Marine subsidies in

freshwater ecosystems: salmon carcasses increase the growth rates of stream-

resident salmonids. Transactions of the American Fisheries Society 132: 371-381.

Zar, J.H. 1999. Biostatistical Analysis. Prentice Hall. Upper Sadle River, NJ.

Zippen, C. 1958. The removal method of population estimation. The Journal of Wildlife

Management. 22:82-90.

46

BIOGRAPHY OF THE AUTHOR

Cory Gardner was born in Bangor, Maine on September 13. 1978. He was raised

in Orono, Maine and graduated from Orono High School in 1997. He attended the

University of Maine and graduated in 2007 with a Bachelor’s degree in Wildlife Ecology.

Cory is a candidate for the Master of Science degree in Wildlife Ecology from The

University of Maine in December, 2010.

Page 23: [1] Deleted Nutt234-01-08 10/25/2010 11:43:00 AM

The biotic responses to dam removals usually are not monitored (Bernhardt et al.

2005), or the research is focused on a target fish species (Doyle et al. 2005), rather than a

community response, like we have done in this study. Because we do not know what the

abundances and size structure of fish species was before the construction of the dams, it

is important to know what the baseline conditions and variability were before the

restoration. Prior to dam removal we found consistent patterns in space and over time,

and low variability, which will allow us to detect the changes that take place following

the restoration project. The only deviation from this pattern occurred at the two sites

immediately adjacent to the dam immediately after removal, and our sampling allowed us

to detect these changes. Continued monitoring will allow us to show how the stream

fishes below the dam site recover from the disturbance, and how the entire stream fish

community responds to a more natural hydrology and increased connectivity to large

river and marine ecosystems.

This type of monitoring can be useful in detecting the response of fish communities in

dammed streams as dam removals become a more popular restoration effort in Maine and

beyond. This restoration project is notable because it takes place downstream of the

anticipated Penobscot River Restoration Project, in which two large main-stem dams are

to be removed, and fish passage installed or improved at two other dams (PRRT 2009).

We view the Penobscot River, structurally and functionally as thousands of

Sedgeunkedunk –sized tributaries, and thus this small scale restoration project is the first

step towards rehabilitation of a major river system.