diapositive 1 - uqam · diapositive 1 author: van bellen created date: 6/13/2011 3:29:16 pm

1
Vegetation, hydrology and fire effects on Holocene carbon accumulation in boreal peatlands of the James Bay region, Quebec 1 INTRODUCTION Besides important long-term net sinks of carbon (C), northern ombrotrophic peatlands form sensitive records of climate-related ecohydrological change, as they depend on precipitation as a source of water and nutrients. Low decomposition rates are primarily the result of cold, anoxic and acidic conditions and decay-resistant biomass. Climatic fluctuations or recurrent fires may change the balance of production and decay, resulting in a temporary net C loss. The interactions between climate, vegetation and fire on millennial timescales are complex, especially considering high spatial variability in peatland vegetation. The aim of the project is to identify the relative influence of vegetation, hydrology and peat fire regimes on C accumulation under Holocene climate variations. Analyses of multiple cores from several peatlands are essential to be able to distinguish regional and local factors. Therefore, we present here the results of nine cores that were extracted from three peatlands. Simon van Bellen 1 , Michelle Garneau 2 and Yves Bergeron 3 1 Chaire DÉCLIQUE / GEOTOP / Institut des Sciences de l’Environnement, Université du Québec à Montréal, C.P. 8888, Succ. Centre-Ville, Montreal, Qc, H3C 3P8, Canada, [email protected] 2 Chaire DÉCLIQUE / GEOTOP / Département de Géographie, Université du Québec à Montréal, C.P. 8888, Succ. Centre-Ville, Montreal, Qc, H3C 3P8, Canada 3 Chaire industrielle CRSNG-UQAT-UQAM en aménagement forestier durable, Université du Québec en Abitibi-Témiscamingue, 445 boul. de l'Université, Rouyn-Noranda, Qc, J9X 5E4, Canada REFERENCES Booth, R.K. et al., 2008. J Quaternary Sci 23, 43-57 Haslett, J. and Parnell, A., 2008. J Roy Stat Soc C-App 57, 399-418 Higuera, P. et al., 2009. Ecol Monogr 79, 201-219 Mansuy, N. et al., 2010. Int J Wildland Fire 19, 1083-1098 Thibault, S. and Payette, S., 2009. Permaforst Periglac 20, 383-389 Turunen, J. et al., 2002. Holocene 12, 69-80 3 MATERIAL AND METHODS Stratigraphic analyses were performed on cores extracted from central/deep and lateral sections of each peatland. Central, deep sections form long and high-resolution records for vegetation and water table reconstructions, whereas lateral sections are more useful for fire reconstructions. Vegetation reconstructions were obtained by plant macrofossil analysis and past water tables were quantified by testate amoeba assemblages from the deep core in each peatland. Past amoeba assemblages were converted to a water table record using a transfer function (Booth, 2008). Macroscopic charcoal fragments (>355 μm) were analyzed for local fire reconstructions from 6 lateral cores (12-132 m from upland forest). Significant charcoal peaks were identified using CharAnalysis (Higuera et al., 2009). Total charcoal accumulation is referred to as CHAR and expressed as fragments cm -2 per unit of time. Chronologies were obtained from a total of 73 radiocarbon dates and piecewise linear interpolation using Bchron (Haslett and Parnell, 2008). Peat C content was quantified by high-resolution bulk density and loss-on-ignition analyses, assuming C represents 50% of the organic matter mass (Turunen et al., 2002). C accumulation rates were defined by the ratio of total amount of C accumulated during a period and the duration of that period, expressed as g m -2 yr -1 . VEGETATION EFFECTS Low accumulation rates between 3000 and 1200 cal BP have been reconstructed in the deep core from LLC (Fig. 5). Forced by fluctuations in water table level, important shifts in Sphagnum sections were reconstructed from the same core after 3000 cal BP (Fig. 6). Whereas the core is generally composed of well-preserved Sphagnum section Acutifolia during the Holocene, the Neoglacial period shows important episodes of the wet-tolerant section Cuspidata. The latter is generally poorly resistant to decomposition, which may partly explain the decrease in C accumulation rates. CARBON ACCUMULATION AND FIRE RECORDS Regional area-weighted mean Holocene C accumulation rate is 16.2 g m -2 yr -1 . The increase in rates towards present-day is the result of incomplete decay in the upper horizons. Eastmain region peatlands show miminum C accumulation rates between 2000 and 1200 cal BP (Fig. 1). During this period, charcoal peak height generally increases, indicating increased biomass burning (Fig. 2). Although fires have likely influenced C accumulation rates, they have not been a driving factor as regression analysis failed to identify a significant negative relationship between C accumulation rates and CHAR (Fig. 3). 4 RESULTS 2 STUDY REGION 5 DISCUSSION AND CONCLUSION Decreasing late-Holocene C accumulation rates are associated to fluctuating water tables, decreasing Sphagnum section Acutifolia presence, and increasing peat fire regimes. Nevertheless, Neoglacial cooling itself may have been the principal cause for declining rates, as cooling generally shortens growing season length while diminishing primary production. Climate- driven changes in temperature and snow cover may have allowed the formation of decadally persistent frozen peat horizons, causing important fluctuations in water table and local vegetation. However, occasional varying trends in accumulation and water table dynamics between cores shows that local, geomorphological or autogenic factors have mediated the climatic influence. ACKNOWLEDGMENTS Thanks to A.A. Ali, H. Asnong, H. Asselin, R.K. Booth, P- L. Dallaire, lab and field assistants and les Tourbeux for discussion and inspiration. Start of peat accumulation Mineral sediments NEOGLACIAL INFLUENCE The combined water table records show important fluctuations since 3000 cal BP (Fig. 4). As this period is associated with cold and dry conditions and the southern limit of sporadic peatland palsas is located ~140 km north of the Eastmain region (Thibault and Payette, 2009), we hypothesize that dry shifts observed may have been amplified by decadal persistence of frozen peat horizons. Individual fires (Fig. 3) may have coincided with regionally recorded dry shifts, however, high-resolution dating is essential for confirmation Fig. 1 Fig. 2 Fig. 3 Fig. 4 Fig. 5 Fig. 6 Sphagnum fuscum stem leaf (×100) Eastmain region is located near 52ºN/75-76ºW. Mean annual precipitation and temperature is 735 mm and -2.1ºC. Mean forest fire frequency is one event every ~90 years (Mansuy et al., 2010). The three peatlands (1.7-2.7 km 2 ) show important hummock-hollow patterns, with Picea mariana and Ericaceae at the transition to the upland forest. Maximum peat thickness attains ~5 m. 10 20 30 0 200 400 600 800 C accumulation rate (g m -2 yr -1 ) CHAR (pieces cm -2 1000 yrs -1 )

Upload: others

Post on 03-Jul-2020

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Diapositive 1 - UQAM · Diapositive 1 Author: van Bellen Created Date: 6/13/2011 3:29:16 PM

Vegetation, hydrology and fire effects on Holocene carbon accumulation

in boreal peatlands of the James Bay region, Quebec

1 INTRODUCTION

Besides important long-term net sinks of carbon (C), northern ombrotrophic

peatlands form sensitive records of climate-related ecohydrological change, as

they depend on precipitation as a source of water and nutrients. Low

decomposition rates are primarily the result of cold, anoxic and acidic conditions

and decay-resistant biomass. Climatic fluctuations or recurrent fires may change

the balance of production and decay, resulting in a temporary net C loss. The

interactions between climate, vegetation and fire on millennial timescales are

complex, especially considering high spatial variability in peatland vegetation.

The aim of the project is to identify the relative influence of vegetation, hydrology

and peat fire regimes on C accumulation under Holocene climate variations.

Analyses of multiple cores from several peatlands are essential to be able to

distinguish regional and local factors. Therefore, we present here the results of

nine cores that were extracted from three peatlands.

Simon van Bellen1, Michelle Garneau2 and Yves Bergeron3

1Chaire DÉCLIQUE / GEOTOP / Institut des Sciences de l’Environnement, Université du Québec à Montréal, C.P. 8888, Succ. Centre-Ville, Montreal, Qc, H3C 3P8, Canada,

[email protected] 2Chaire DÉCLIQUE / GEOTOP / Département de Géographie, Université du Québec à Montréal, C.P. 8888, Succ. Centre-Ville, Montreal, Qc, H3C 3P8, Canada

3Chaire industrielle CRSNG-UQAT-UQAM en aménagement forestier durable, Université du Québec en Abitibi-Témiscamingue, 445 boul. de l'Université, Rouyn-Noranda, Qc, J9X 5E4, Canada

REFERENCES

Booth, R.K. et al., 2008. J Quaternary Sci 23, 43-57

Haslett, J. and Parnell, A., 2008. J Roy Stat Soc C-App 57, 399-418

Higuera, P. et al., 2009. Ecol Monogr 79, 201-219

Mansuy, N. et al., 2010. Int J Wildland Fire 19, 1083-1098

Thibault, S. and Payette, S., 2009. Permaforst Periglac 20, 383-389

Turunen, J. et al., 2002. Holocene 12, 69-80

3 MATERIAL AND METHODS

Stratigraphic analyses were performed on cores

extracted from central/deep and lateral sections of

each peatland. Central, deep sections form long and

high-resolution records for vegetation and water

table reconstructions, whereas lateral sections are

more useful for fire reconstructions.

Vegetation reconstructions were obtained by plant

macrofossil analysis and past water tables were

quantified by testate amoeba assemblages from the

deep core in each peatland. Past amoeba

assemblages were converted to a water table

record using a transfer function (Booth, 2008).

Macroscopic charcoal fragments (>355 µm) were

analyzed for local fire reconstructions from 6 lateral

cores (12-132 m from upland forest). Significant

charcoal peaks were identified using CharAnalysis

(Higuera et al., 2009). Total charcoal accumulation

is referred to as CHAR and expressed as fragments

cm-2 per unit of time.

Chronologies were obtained from a total of 73

radiocarbon dates and piecewise linear interpolation

using Bchron (Haslett and Parnell, 2008). Peat C

content was quantified by high-resolution bulk

density and loss-on-ignition analyses, assuming C

represents 50% of the organic matter mass

(Turunen et al., 2002). C accumulation rates were

defined by the ratio of total amount of C

accumulated during a period and the duration of that

period, expressed as g m-2 yr-1.

VEGETATION EFFECTS

Low accumulation rates between 3000

and 1200 cal BP have been

reconstructed in the deep core from LLC

(Fig. 5).

Forced by fluctuations in water table

level, important shifts in Sphagnum

sections were reconstructed from the

same core after 3000 cal BP (Fig. 6).

Whereas the core is generally

composed of well-preserved Sphagnum

section Acutifolia during the Holocene,

the Neoglacial period shows important

episodes of the wet-tolerant section

Cuspidata. The latter is generally poorly

resistant to decomposition, which may

partly explain the decrease in C

accumulation rates.

CARBON ACCUMULATION AND FIRE

RECORDS

Regional area-weighted mean Holocene

C accumulation rate is 16.2 g m-2 yr-1. The

increase in rates towards present-day is

the result of incomplete decay in the

upper horizons.

Eastmain region peatlands show

miminum C accumulation rates between

2000 and 1200 cal BP (Fig. 1). During this

period, charcoal peak height generally

increases, indicating increased biomass

burning (Fig. 2). Although fires have likely

influenced C accumulation rates, they

have not been a driving factor as

regression analysis failed to identify a

significant negative relationship between

C accumulation rates and CHAR (Fig. 3).

4 RESULTS

2 STUDY REGION

5 DISCUSSION AND CONCLUSION

Decreasing late-Holocene C accumulation rates are associated to fluctuating water tables, decreasing Sphagnum section

Acutifolia presence, and increasing peat fire regimes. Nevertheless, Neoglacial cooling itself may have been the principal

cause for declining rates, as cooling generally shortens growing season length while diminishing primary production. Climate-

driven changes in temperature and snow cover may have allowed the formation of decadally persistent frozen peat horizons,

causing important fluctuations in water table and local vegetation. However, occasional varying trends in accumulation and

water table dynamics between cores shows that local, geomorphological or autogenic factors have mediated the climatic

influence.

ACKNOWLEDGMENTS

Thanks to A.A. Ali, H. Asnong, H. Asselin, R.K. Booth, P-

L. Dallaire, lab and field assistants and les Tourbeux for

discussion and inspiration.

Start of peat

accumulation

Mineral

sediments

NEOGLACIAL INFLUENCE

The combined water table records show important fluctuations

since 3000 cal BP (Fig. 4). As this period is associated with cold

and dry conditions and the southern limit of sporadic peatland

palsas is located ~140 km north of the Eastmain region (Thibault

and Payette, 2009), we hypothesize that dry shifts observed may

have been amplified by decadal persistence of frozen peat

horizons.

Individual fires (Fig. 3) may have coincided with regionally

recorded dry shifts, however, high-resolution dating is essential

for confirmation

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Sphagnum fuscum stem leaf (×100)

Eastmain region is located near

52ºN/75-76ºW. Mean annual

precipitation and temperature is

735 mm and -2.1ºC. Mean forest

fire frequency is one event every

~90 years (Mansuy et al., 2010).

The three peatlands (1.7-2.7 km2)

show important hummock-hollow

patterns, with Picea mariana and

Ericaceae at the transition to the

upland forest. Maximum peat

thickness attains ~5 m.

10

20

30

0 200 400 600 800

C a

ccum

ula

tion r

ate

(g

m-2

yr-1

)

CHAR (pieces cm-2

1000 yrs-1

)