Carbon accumulation in peatlands along a boreal to subarctic transect in eastern Canada

Abstract

In this study, we investigated the links between peat carbon accumulation and past ecological and hydrological conditions in three peatlands (Bouleau, Mista, Auassat) which developed along a South-North transect within a watershed encompassing the boreal and subarctic domain in Eastern Canada. Peatland development and long-term apparent rates of carbon accumulation (LORCA) were asynchronous in the watershed, suggesting an influence of both latitude and topography (altitude) on the length of the growing season (GGD₀). Results show that peat initiation within the three peatlands (respectively ca. 9070, 8400, and 6270 cal BP) was delayed after the deglaciation and that LORCA (respectively 35.5, 15.4, and 9.0 g C m⁻² yr⁻¹) decreased from South to North. Peatland development and fen to bog transitions were found to be almost synchronous for the two southernmost sites. The fen to bog transition in the northernmost subarctic site was delayed until the 20th century, owing to the less favorable climatic conditions. This suggests that recent warming has extended the length of the growing season and increased Sphagnum growth enough to potentially influence an ecosystem state-shift as observed in other Subarctic regions of eastern Canada.

Keywords

Boreal and subarctic peatlands carbon dynamics eastern Canada fen to bog transition Holocene latitudinal gradient

Introduction

Peatlands are an important carbon (C) pool owing to the positive balance between net primary production and decomposition (Charman, 2002; Yu et al., 2009). They cover approximately 2.1% of the land surface of the Northern Hemisphere (3.13 million km²; Xu et al., 2018) and form one of the largest terrestrial carbon pools (~547 Gt C; Loisel et al., 2014; Yu et al., 2010). Since the last glaciation, northern peatlands have acted as substantial terrestrial carbon reservoirs (Crowther et al., 2019; Turunen et al., 2002; Yu et al., 2011) and had a net cooling effect on the global climate (Frolking and Roulet, 2007; Gallego-Sala et al., 2018).

Peat initiation is controlled by a combination of allogenic (climate, fires) and autogenic (topography, mineral substrate, vegetation composition, and succession) factors (Charman, 2002). Temperature and a positive moisture balance are the most important drivers for net primary productivity and the apparent peat C accumulation rates tend to decline from south to north (Beilman et al., 2009; Zhang et al., 2018). Garneau et al. (2014) showed that mean summer temperature was one of the most important drivers of peat C accumulation in eastern Canada while Charman et al. (2013) and Gallego-Sala et al. (2018) suggested that photosynthetically active radiation during the growing season and growing degree days above 0°C (GDD₀) were the main factors influencing carbon accumulation in peatlands for the last millennium. Throughout the Holocene, climate variations have impacted biomass productivity by modifying the growing season length and driving shifts in surface wetness and vegetation assemblages (Garneau et al., 2014).

Aside from climate, peat C accumulation is also regulated by several autogenic factors. Beilman et al. (2009) and Loisel et al. (2014) have demonstrated that distinct vegetation assemblages can have an influence on carbon sequestration. For instance, while Sphagnum peat has a lower carbon content than other groups of vegetation, Sphagnum also acidifies the soil water chemistry, which increases the preservation of vegetation over time (Belyea, 2009). Sphagnum assemblages can be more resistant to decomposition than sedges despite the recalcitrance of some plant fragments (e.g. Eriophorum vaginatum) that dominate the peat matrix in some sites (Belyea and Malmer, 2004; Hughes and Dumayne-Peaty, 2002; Yu, 2006).

The biogeographical distribution of peatlands in the boreal and subarctic regions of eastern Canada largely follows that of the ecoclimatic zonation with cool climate conditions and low evapotranspiration (Payette, 2001a; Tarnocai, 2006; Yu et al., 2009). Along the north shore of the Gulf of St Lawrence (GSL), large complexes of maritime peatlands developed on deltaic or fluvioglacial sediments (Magnan et al., 2014; Payette et al., 2013). Peat initiation took place on well-drained podzolic soils following the retreat of the Goldthwait Sea and a rise in atmospheric moisture balance linked to the stabilization of the Atlantic Tropical air masses during the mid-Holocene (Carcaillet and Richard, 2000; Magnan et al., 2014; Payette et al., 2013). Peat initiation in the subarctic region was delayed compared to the boreal region due to the later timing of deglaciation, cooler climatic conditions, and the rugged topography which limited lateral expansion in several regions (Glaser and Janssens, 1986; Morris et al., 2018).

Previous studies of peatlands in eastern Canada (Engstrom and Hansen, 1985; Foster and Glaser, 1986; Glaser and Foster, 1984; Glaser and Janssens, 1986) have focused mainly on surface ecology and patterning, and less on peat stratigraphy and paleoecology. Magnan et al. (2014) and Magnan and Garneau (2014a, 2014b) reconstructed the ecohydrological conditions that influenced carbon accumulation in coastal peatlands, while Foster (1983) and Payette et al. (2013) investigated the fire history that modulated peatland dynamics. No study has yet focused on documenting the effect of a South to North transect encompassing boreal and subarctic biomes in eastern Canada on peatlands carbon dynamics. This study aims to reconstruct the Holocene ecohydrological conditions that influenced carbon accumulation in three peatlands located in the Romaine River watershed (northeastern Canada) along latitudinal and altitudinal gradients.

Study region and sites

Physiographic and ecoclimatic contexts

The studied region is located in northeastern Quebec (eastern Canada), within the Romaine River watershed (14,500 km²) adjacent to the Labrador border. The physiography of the watershed is divided into two sections: the coastal plain and the Laurentian Plateau. The Romaine River flows over 52 km through the coastal plain and 250 km through the Grenville Province of the Canadian Shield on the Laurentian Plateau (Verpaelst et al., 1999). The coastal plain is mainly composed of deltaic sediments up to an elevation ca. 150 m asl (Dubois, 1980). The Laurentian Plateau covers the northern section of the watershed and is divided into two parts: (1) the Highlands located between 150 and 300 m asl which are characterized by steep hills that mark the transition between the coastal and continental environments and (2) the till plains from 300 to 900 m asl that are covered by fluvial and postglacial sediments (Dubois, 1980).

The thawing of the Laurentide Ice Sheet (LIS) on the northern shore of the Gulf of St Lawrence occurred ca. 11,000 cal BP and was followed by the Goldthwait Sea invasion (Dyke et al., 2003). The maximum height of the postglacial Goldthwait Sea was estimated at 128–131 m above present sea level before isostatic rebound influenced land emersion ca. 10,600 cal BP (Dubois, 1980). Substantial deltaic terraces were built by proglacial meltwater following continental uplift at ca. 106, 76–75, 46–45, and 15 m respectively (Bernatchez, 2003, 2005; Dubois, 1980).

Regional vegetation is characterized by the Spruce-moss bioclimatic domain of the closed boreal forest in the Highlands and is replaced by the Spruce-lichen woodland of the taiga in the northernmost sections of the till plains (Payette, 2001b).

Climatic context

The climatic gradient within the watershed shows a transition from a maritime to a continental climate with a slight decrease in mean annual temperature, mean summer temperature, and a slight increase in mean summer precipitation towards the North as shown in Supplemental Figure S1. Mean annual temperature, precipitation and growing degree-days (>0°C) were calculated from gridded adjusted and homogenized climate data (Vincent et al., 2012) for each sampling site using the ANUSPLIN algorithm on a 0.1 arc degree grid (~10 km) (Hopkinson et al., 2011; Hutchinson et al., 2009; McKenney et al., 2011) from 1950 to 2015. Averages from the past 20 years suggest climate warming accelerating faster in the North with summer and autumn temperatures influencing the total increase in GDD₀ values.

Study sites

Three peatlands were selected as representative of the Romaine River watershed (Figure 1). Peatlands were characterized using aerial photographs and their representativeness was validated with aerial surveys in 2016. Sites were selected based on their size, latitudinal distribution and ecogeomorphic context within the watershed. The southernmost site, Bouleau (50°31′N, 63°12′W; 108 m asl) is located in a depression at the limit of the coastal plain and the Highlands of the Laurentian Plateau. The withdrawal of the Goldthwait Sea was estimated ca. 10,290 ± 290 cal BP at that elevation (Dubois, 1980). Bouleau is a slightly dome-shaped bog (~1.51 km²) with a clear patterned surface of alternating dry hummocks, wet hollows, and elliptical pools. The vegetation follows the microform humidity gradient. Sphagnum fuscum, S. capillifolium, and Cladonia rangiferina dominate the hummocks while S. magellanicum, S. rubellum, S. cuspidatum, and Trichophorum cespitosum are found on lawns and S. majus and S. pulchrum on wet hollows.

Figure 1.

Location of the three study sites in eastern Canada. White circles show the location of the coring sites.

Mista (50°48′N, 63°20′W, 372 m asl) is located in the Highlands. It is a structured bog (~0.17 km²) with alternating linear hummocks and elliptical pools parallel to the slope. Surface vegetation is composed of S. fuscum, C. rangiferina, ericaceous shrubs, and a few low stands of Picea mariana on hummocks, while S. rubellum and S. majus dominate the lawns and hollows respectively. Both the Bouleau and Mista peatlands belong to the eastern Spruce-moss bioclimatic domain of the closed boreal forest (Payette, 2001b).

Auassat (51°48′N, 63°41′W; 466 m asl) developed on the till plains and northernmost section of the watershed. It is a patterned poor fen (~0.22 km²) located within the Spruce-lichen woodland of the taiga (Payette, 2001b). It is characterized by alternating elongated pools, wet hollows, and ridges with S. fuscum, C. rangiferina, Ericaceae, and krummholz of P. mariana on ridges and S. fallax and Cyperaceae on lawns.

Methods

Fieldwork and sampling

Once representative sites were selected, comprehensive peatland depth surveys were conducted for all sites by systematic manual probing at gridded 50 m intervals. A peat core was retrieved from the deepest section of each peatland from a lawn microform to reconstruct peat accumulation from the earliest phases of peatland development to the present-day. A Box corer (Jeglum et al., 1991) was used for the upper meter and a modified Russian corer (diameter: 7.5 cm) (Jowsey, 1966) was used to collect deeper peat until the mineral contact was reached. Peat cores were wrapped with cellophane and aluminum foil, transported in PVC tubes and stored in a refrigerator at 4°C until further analysis.

In each peatland, vegetation surveys from 1 m² quadrats were carried out on randomly selected microforms using the Braun-Blanquet cover-abundance scale method (Braun-Blanquet, 1932). Testate amoebae surface samples were also taken from the same microforms following the protocol described by Booth et al. (2010).

Loss-on-ignition and C:N ratio

In the laboratory, the three central cores were cut into contiguous 1-cm slices. Peat bulk density was determined after overnight drying of a known volume (1 cm³) at 105°C. The organic matter (OM) density was measured from loss-on-ignition (LOI) for 4 hours at 550°C (Chambers et al., 2011). Carbon density (g cm³) was estimated as 50% of the OM density (Turunen et al., 2002). C:N analyses were also performed on dry subsamples of 0.5 cm³ at 4 cm intervals using a Carlo Erba NC 2500 elemental analyzer (Isotopic stable analysis laboratory, Geotop-UQAM) to quantify peat humification (Kuhry and Vitt, 1996).

Plant macrofossil analyses

Peat samples (5 cm³) were subsampled and analyzed at 4 cm intervals for the Bouleau and Mista cores and at 2 cm intervals for the Auassat core due to its shallower depth. Preparation followed the protocol described by Mauquoy et al. (2010). Samples were gently boiled within a KOH-5% solution to remove fulvic and humic acids, washed through a 125 µm sieve and placed in a gridded petri dish for analysis. A Leitz stereoscopic microscope (10–40×) was used to estimate relative abundance (%) of Sphagnum spp., brown mosses, herbaceous and ligneous remnants, while needles, seeds, and Cenococcum spp. sclerotia were individually counted. When Sphagnum was present, ca. 50 stem or branch leaves were mounted on microscope slides for further identification using a Leitz optical microscope (400–1000×). Lévesque et al. (1988), Mauquoy and van Geel (2007), and the reference collection from the Continental Paleoecology Laboratory at Geotop-UQAM (Garneau, 1995) were used for plant macrofossil identification. The nomenclature of vascular plant taxonomy is based on Marie-Victorin (2005) while Laine et al. (2009) was used for Sphagnum spp., Nilsson and Hjelmqvist (1967) for Cyperaceae and Faubert (2014) for the mosses.

Testate amoeba analyses

Testate amoeba analyses were conducted on the same samples used for plant macrofossil analyses. Using a modified procedure from Booth et al. (2010), 1 cm³ samples were immersed in 100 ml of distilled water and gently boiled for 10 min. One tablet of Lycopodium spores (batch #3862) (Stockmarr, 1971) was incorporated into each sample to calculate test concentrations. Material was then rinsed with distilled water through 355-μm and 15-μm sieves, centrifuged and stored with a drop of glycerin at 4°C. A minimum of 100 tests was counted using a Leitz optical microscope (400×) and identified using the Charman et al. (2000) and Booth (2008) identification keys. Samples with a test concentration <100 were interpreted with caution although assemblages with >50 tests may still provide reliable paleohydrological data (Payne and Mitchell, 2009). Paleohydrological reconstructions were carried out using the transfer function developed by Lamarre et al. (2013) with the Rioja package version 0.1-15.1 (Juggins, 2007) in R. Past water table depths (WTD) were inferred using a weighted average tolerance down-weighted (WA-tol) function with classical deshrinking and evaluated using a bootstrap cross-validation method with 1000 cycles. Diagram zonation was performed visually following the main changes in the assemblages of both proxies.

Chronology

Seventeen samples, mostly Sphagnum or brown mosses, were selected at the mineral-organic contact and the main stratigraphic transitions of each core for radiocarbon dating at the A.E. Lalonde AMS Laboratory (University of Ottawa, Canada) (Supplemental Table S1). Recent chronologies (last ca. 150 years) were constructed using lead-210 (²¹⁰Pb) dating for the upper sections of the three cores (20 cm – Bouleau and Mista; 40 cm – Auassat). Contiguous samples (interval: 2 cm, volume: 2 cm³) were spiked with a ²⁰⁹Po chemical yield tracer and subjected to a sequential HNO₃:HF:H₂O₂:HCl acid digestion to extract and isolate ²¹⁰Po, a daughter product of ²¹⁰Pb decay (Ghaleb, 2009). The ratio of Po isotope activities, used to calculate ²¹⁰Pb concentration, was measured using alpha spectrometry (EGG Ortec 776a) at the Geotop Radiochronology Laboratory (UQAM). The Constant Rate of Supply model (Supplemental Table S2) was used to determine the age of each level (Appleby, 2001; Appleby and Oldfield, 1978). Age-depth models were built by combining ²¹⁰Pb and ¹⁴C results using the rbacon 2.3.9.1 R package (Blaauw and Christen, 2011) in R. The age of the peat surface was set at −66 cal BP, equivalent to the year of coring (i.e. 2016 CE). For consistency with our calibrated chronologies, conventional ages from the literature were also calibrated using the IntCal13 calibration curve (cal BP; Reimer et al., 2013).

LORCA, RERCA, and carbon pool

The LOng-term apparent Rate of Carbon Accumulation (LORCA; g C m⁻² yr⁻¹) was calculated by dividing the total mass of C accumulated by the basal ¹⁴C age (Turunen et al., 2002). REcent apparent Rate of Carbon Accumulation (RERCA) was calculated by dividing the total mass of C for a given period by the age of the ²¹⁰Pb dated horizon and was calculated for the last 50, 100, and 150 years before coring (~1965 CE, ~1915 CE, ~1865 CE). The difference between the two measurements is that LORCA gives an average value of peat accumulation since its initiation and the values are therefore smaller, as part of the organic matter has decomposed over time. The RERCA values provide higher peat accumulation rates because they correspond to the younger and less decomposed surface organic layers of the peatland (Belyea and Clymo, 2001). Carbon accumulation rates (CAR) were calculated by dividing the C density (g cm³) of each centimeter by the deposition time (peat accumulation rate (PAR) in mm yr⁻¹) obtained by the age-depth modelling.

The total C pool (CP) of each peatland was calculated using the equation from Sheng et al. (2004):

C P = A . D . ρ . L . R

where CP is calculated by multiplying (A) peatland surface area (m²), (D) mean peat depth (m), ( $ρ$ ) mean peat density (kg m⁻³), (L) mean organic matter content (%), and (R) relative peat C content calculated for each peatland. An estimation of the sequestration of C per area was obtained by dividing the carbon pool by the area of the peatland. The peatland surface area was calculated from the aerial photos and the high-resolution peat depth values were obtained from the systematic manual probing survey (163, 74, and 56 surveys respectively for Bouleau, Mista, and Auassat), in order to more accurately represent the basin topography. The mean peat density, mean organic matter content and relative peat C content were evaluated by LOI to calculate total peat volume in each peatland. Topo to Raster was used in ArcGIS 10.1 to model peat depths within the three topographic basins. This GIS model was preferred for its clarity and reliability; note that pool surface areas and depths were not subtracted from the calculation.

Results

Chronology and PAR

Bouleau peatland – Peat started to accumulate around 9070 cal BP with a mean rate of 0.65 mm yr⁻¹. Between 8560 and 4560 cal BP, peat accumulation dropped to 0.32 mm yr⁻¹. Following ombrotrophication, PAR increased to 0.81–0.83 mm yr⁻¹ (3690–2500 cal BP) while slower accumulation was recorded between 2500 and 110 cal BP with a mean value of 0.39 mm yr⁻¹. The upper horizons show the highest mean PAR (1.47 mm yr⁻¹) and correspond to the less compacted and decomposed peat in the acrotelm.

Mista peatland – The beginning of peat accumulation, between ca. 8400 and 8130 cal BP, registered a mean apparent rate of 0.48 mm yr⁻¹ followed by a slowdown to 0.14 mm yr⁻¹ between 8130 and 5340 cal BP. The fen to bog transition influenced the rate of peat accumulation which increased to 0.37–0.4 mm yr⁻¹ from 5340 to 1400 cal BP. These high rates slowed down to 0.25 mm yr⁻¹ between 1400 and 110 cal BP. The most recent part of the peatland shows higher accumulation with a mean rate of 1.61 mm yr⁻¹ corresponding to the acrotelm.

Auassat peatland – This peatland initiated with a mean accumulation rate of 0.45 mm yr⁻¹ between 6270 and 5980 cal BP. Between 5980 and 40 cal BP, PAR was reduced to 0.15 mm yr⁻¹. The beginning of the fen to bog transition at 33 cm (40 cal BP; 1910 CE) occurred much more recently than in the two other peatlands with a mean PAR of 4.34 mm yr⁻¹ while the top 10 cm of the acrotelm records as high as 6 mm yr⁻¹ corresponding to the last ca. 20 years of accumulation (2000 CE).

Paleoecohydrological reconstructions

Two main zones were identified at each study site based on their respective nutrient regimes and vegetation assemblages. In the three regions, the deepest sections of peatlands initiated as fens (Bou1, Mis1, Aua1) and shifted into bogs (Bou2, Mis2, Aua2) around 4560 cal BP at Bouleau, 5340 cal BP at Mista, and 40 cal BP at Auassat (Table 1; Figure 2).

Table 1.

Ecohydrological zonation of the three peatlands.

Subzone	Depth (cm)	Age (cal BP)	Dominant plant macrofossil	Dominant testate amoeba	WTD	Dry bulk density (g cm^–3)	C:N ratio	CAR (g C m^–2 yr^–1)
Bou1a	428–396	9070–8560	W. fluitans, Ligneous and Cyperaceae	N.A.	N.A.	0.22–0.51	18–46	36–97
Bou1b	396–276	8560–4560	Cyperaceae	A. wrightianum, A. muscorum	N.A.	0.1–0.27	28–69	16–51
Bou2a	276–204	4560–3690	S. sect. Sphagnum, S. sect. Cuspidata	A. flavum, H. papilio	5	0.07–0.19	51–137	29–94
Bou2b	204–108	3690–2500	S. sect. Acutifolia	D. pulex, H. subflava	20	0.09–0.21	60–170	35–85
Bou2c	108–10	2500 to −10	S. sect. Acutifolia	A. flavum, D. pulex, H. subflava	3—25	0.1–0.27	45–182	19–80
Bou2d	10–0	–10 to −66	S. sect. Acutifolia, Ericaceae	A. flavum, H. elegans, H. papilio	6	0.07–0.11	175–204	29–200
Mis1a	228–215	8400–8130	W. fluitans, Herbaceous	N.A.	N.A.	0.25–0.37	23–27	32–44
Mis1b	215–180	8130–5340	Cyperaceae	A. wrightianum	0	0.11–0.31	17–25	6–19
Mis2a	180–150	5340–4560	S. sect. Sphagnum, S. sect. Cuspidata	A. flavum, A. muscorum	10	0.07–0.16	42–79	7–35
Mis2b	150–46	4560–1400	S. sect. Acutifolia	D. pulex, A. flavum, H. subflava	5–25	0.07–0.18	43–136	6–34
Mis2c	46–18	1400–110	Herbaceous and S. sect. Acutifolia	H. subflava, H. sphagni, T. arcula	20	0.05–0.16	38–61	4–34
Mis2d	18–0	110 to −66	S. sect. Acutifolia	H. elegans, H. papilio, A. muscorum	15	0.07–0.12	78–196	22–149
Aua1a	108–94	6270–5980	S. scorpioides and Cyperaceae	N.A.	N.A.	0.07–0.14	21–34	15–28
Aua1b	94–33	5980–40	Herbaceous and Ligneous	N.A.	N.A.	0.07–0.2	23–34	1–18
Aua2	33–0	40 to −66	S. sect. Acutifolia and Ericaceae	H. papilio, A. muscorum, T. arcula	23	0.04–0.1	80–175	27–392

Figure 2.

Macrofossil abundance (%) and counts (n), testate amoebae (%) and inferred water table depths (WTD) for the Bouleau, Mista, and Auassat peatlands. Inferred WTD with dotted lines correspond to low test concentrations (<100). Peat type (%): Sphagnum (white); herbaceous (light gray); ligneous (dark gray); brown mosses (black).

Bouleau peatland

Bou1a: 428–396 cm; 9070–8560 cal BP. The oldest section of Bouleau peatland corresponds to a moderately rich fen dominated by Warnstorfia fluitans, herbaceous species (e.g. Carex spp.) with some Picea and Larix as confirmed by the needles found in the macrofossil assemblages. Low C:N ratios (18–46) suggest some runoff enrichment, as the core was retrieved from the deepest part of the basin. Highly decomposed peat probably resulted in low concentration (⩽5) of testate amoebae (Payne and Mitchell, 2009), which prevented WTD reconstructions.

Bou1b: 396–276; 8560–4560 cal BP. A progressive transition to poorer minerotrophic conditions is recorded with the disappearance of brown mosses through the macrofossil assemblages. Bou1b is characterized by an abundance of herbaceous remnants and Carex spp. seeds along with a few Picea mariana needles. The testate amoeba assemblages are dominated by Amphitrema wrightianum and Assulina muscorum suggesting persistent waterlogged conditions. However, low counts (<20) did not allow the inference of a water table depth.

Bou2a: 276–204 cm; 4560–3690 cal BP. This subzone marks the beginning of ombrotrophication with the presence of lawn-associated species such as Sphagnum sect. Sphagnum, S. sect. Cuspidata, and Andromeda glaucophylla within the macrofossil assemblages. The dominance of Archerella flavum and Hyalosphenia papilio in the testate amoeba assemblages suggests a lowering of the WTD (±5 cm) relative to Bou1b. Peat and related C accumulation increased probably due to the higher recalcitrance of Sphagnum mosses and peat acidification (C:N ratio values from 51 to 137).

Bou2b: 204–108 cm; 3690–2500 cal BP. In this subzone, S. sect. Acutifolia replaced the S. sect. Sphagnum and S. sect. Cuspidata communities found in Bou2a. Along with this vegetation succession, the testate amoeba assemblages shifted from A. flavum to taxa indicators of drier conditions such as Difflugia pulex and Hyalosphenia subflava, suggesting a decrease in the WTD (±20 cm).

Bou2c: 108–10 cm; 2500 to −10 cal BP. Sphagnum sect. Acutifolia also dominates the vegetation in this subzone, but the main difference from Bou2b is the diversity of the testate amoeba species where A. flavum, D. pulex, and H. subflava suggests a period of fluctuating water tables (ca. 3–25 cm). This variation in the testate amoeba composition confirms the greater sensitivity of these taxa to surface moisture changes than to vegetation changes. After 830 cal BP, the vegetation assemblages shifted to a dominance of ligneous species along with a drop in the CAR values (from 80 to 25 g C m⁻² yr⁻¹).

Bou2d: 10–0 cm; −10 to the present-day. At the surface, the vegetation and testate amoeba assemblages suggest a wetter environment than in subzone Bou2c. It corresponds to the recent and contemporary conditions at the surface of the peatland with the presence of S. sect. Acutifolia, Chamaedaphne calyculata, and A. glaucophylla. The testate amoebae A. flavum, Hyalosphenia elegans, and H. papilio suggest an intermediate hydrologic optimum (±6 cm). Dry bulk density (0.07–0.11 g cm⁻³) and a high C:N ratio (175–204) correspond to the recent and partially decomposed acrotelm peat.

Mista peatland

Mis1a: 228–215 cm; 8400–8130 cal BP. As in Bouleau, this subzone corresponds to the beginning of peat accumulation under rich fen conditions where W. fluitans and herbaceous species dominate the vegetation assemblages. A WTD was not inferred for this zone as the number of taxa was too low (<5) in the highly decomposed peat, confirmed by the low C:N ratio values (23–27) (Payne and Mitchell, 2009).

Mis1b: 215–180 cm; 8130–5340 cal BP. The transition between a moderately rich fen (Mis1a) and a poor fen is suggested by the shift from a D. fluitans assemblage to some sedge species (Carex nigra, C. rostrata, C. canescens, C. limosa), few ericaceous shrubs, and Larix laricina. The testate amoeba assemblage is dominated by A. wrightianum and suggests waterlogged conditions. Low C:N ratio values (17–25) confirm the highly decomposed peat and explain the low CAR values (from 6 to 19 g C m⁻² yr⁻¹) in this subzone.

Mis2a: 180–150 cm; 5340–4560 cal BP. This subzone corresponds to the fen to bog transition confirmed by the shift from Cyperaceae towards a dominance of S. sect. Sphagnum and the S. sect. Cuspidata. The testate amoebae taxa A. flavum and A. muscorum suggest a lowering of the water table (±10 cm) associated with a rise in peat accumulation. The higher C:N ratio (42 and 79) in this zone can be associated with the shift in the vegetation assemblage with a dominance of more recalcitrant Sphagnum species and acidic conditions.

Mis2b: 150–46 cm; 4560–1400 cal BP. Ombrotrophic conditions persisted in subzone Mis2b as confirmed by the dominance of S. sect. Acutifolia while D. pulex, A. flavum, and H. subflava suggest conditions with fluctuating water tables (from ca. 5 to 25 cm). Peat C accumulation rates (6–34 g C m⁻² yr⁻¹) and decomposition were variable with C:N ratio values between 43 and 136.

Mis2c: 46–18 cm; 1400–110 cal BP. This subzone is marked by a decrease in Sphagnum mosses and an increase in herbaceous and ligneous remains, P. mariana needles and Cenococcum spp. sclerotia. The reconstruction of the water table suggests low but variable WTD (±20 cm) where Heleopera sphagni, H. subflava, and Trigonopyxis arcula are abundant. Relatively low C:N ratios (38–61) and some low carbon accumulation rates (4–34 g C m⁻² yr⁻¹) suggest conditions of higher peat decomposition than Mis2b at that period.

Mis2d: 18–0 cm; 110 cal BP to the present-day. The vegetation is dominated by S. sect. Acutifolia, which is found at the surface. Surface wetness increased slightly compared to Mis2c (±15 cm) as confirmed by the presence of H. elegans and H. papilio. Low dry bulk density (0.07–0.12 g cm⁻³), high C:N ratio (78–196) and CAR (22–149 g C m⁻² yr⁻¹) correspond to the poorly decomposed acrotelm.

Auassat peatland

Aua1a: 108–94 cm; 6270–5980 cal BP. This subzone marks the beginning of peat accumulation under a rich fen and waterlogged conditions as confirmed by the dominance of Scorpidium scorpioides in the macrofossil assemblages. At the base of the core (108–105 cm), high bulk densities (0.23–0.83 g cm⁻³) correspond to a combination of mineral and organic material. This is followed by low dry bulk density peat (0.07–0.14 g cm⁻³) in which low C:N ratio (21–34) values confirm high decomposition which influenced CAR (from 15 to 28 g C m⁻² yr⁻¹).

Aua1b: 94–33 cm; 5980–40 cal BP. This subzone corresponds to an herbaceous poor fen that lasted several millennia. The WTD could not be inferred due to the poor test preservation as reported in other minerotrophic environments (Payne, 2011; van Bellen et al., 2013). Low dry bulk density (0.07–0.2 g cm⁻³), C:N ratios (23–34), and CAR values (1–18 g C m⁻² yr⁻¹) suggest pronounced decomposition.

Aua2: 33–0 cm; 40 cal BP to present day. This zone marks a recent shift from fen to bog conditions at Auassat peatland. This is shown by a drastic vegetation change from a dominance of sedges to Sphagnum. S. sect. Acutifolia assemblages with remnants of C. calyculata, Kalmia angustifolia, and Vaccinium oxycoccos leaves along with P. mariana and L. laricina needles through the horizons. The testate amoeba assemblages are composed of taxa with different WTD optima from H. papilio to T. arcula. The low dry bulk density (0.04–0.1 g cm⁻³), high C:N ratio (80–175), and high CAR (22–392 g C m⁻² yr⁻¹) values at the coring site correspond to the recent Sphagnum ecosystem state-shift in the poorly decomposed acrotelm.

Carbon dynamics and peat properties

Table 2 presents synthesized C data calculated for the three peatlands. In the upper section, mean C values obtained from LOI and the elemental analyzer are quite similar although some differences can be attributed to the vegetation composition within the peat matrix (Figure 3). We chose to present both datasets as the C obtained from the elemental analyzer were not at contiguous intervals. However, the results from the two methods confirm the accuracy of the LOI method of C measurement for multiple cores analysis (Chambers et al., 2011).

Table 2.

Mean values and standard deviations of the carbon measurements and accumulation rates for the three sites.

	Bouleau	Mista	Auassat
C and N values
%C (LOI)	46.7 (6.4)	47.2 (5.8)	47.5 (5.9)
%C (elemental analyzer)	51.9 (8.8)	44.6 (5.2)	46 (8.1)
%N (elemental analyzer)	0.95 (0.5)	1.13 (0.67)	1.65 (0.90)
C:N ratio	84.3 (44.5)	63.7 (38.7)	55.6 (50.6)
Long-term C calculation
LORCA (g C m^–² yr^–1)	35.5	15.4	9.0
Carbon pool (kt C)	200.7 (10.1)	14.7 (1)	9.3 (0.4)
CAR (g C m^–2 yr^–1)	47.2 (21.3)	19.6 (17.3)	12 (6.6)
C per area (kg C m^–²)	133.1 (5.1)	86.1 (3.6)	42.5 (2.5)
Recent C calculation
RERCA (g C m^–² yr^–1)
~1865–1915 CE	67.5	51.9	52.4
~1915–1965 CE	75.9	63.1	93.4
~1965–2016 CE	86.1	84.5	142.7
Carbon pool (kt C)	19.2	1.9	4
C per area (kg C m^–²)	12.8	10.8	18.6

Figure 3.

Characteristics of the Bouleau, Mista, and Auassat peatlands. From left to right: Relative abundance of plant macrofossils; Dry bulk density; %C calculated from LOI; %C and %N from elemental analyzer measurements; C:N ratio; Carbon accumulation rates. Peat type (%) is represented as follows: Sphagnum (white); herbaceous (light gray); ligneous (dark gray); brown mosses (black).

LORCA results are also presented in Table 2: 133.1 (±5.1), 86.1 (±3.6), and 42.5 (±2.5) kg C m⁻² for Bouleau, Mista, and Auassat, respectively. To support the total carbon pool calculation, a peat depth interpolation was created to reconstruct the topographic basin for each peatland (Figure 4). Using the equation in Sheng et al. (2004), the estimated C pools are 200.72 (±10.2), 14.73 (±1), and 9.29 (±0.4) kt C for Bouleau, Mista, and Auassat. These values are slightly overestimated considering that the pools in each peatland were not excluded. On average, mean pool depth in each peatland varies from 85 to 30 cm from South to North while their size varies from 90 to 9000 m^2.

Figure 4.

Peat depth models for the Bouleau, Mista, and Auassat peatlands. The C stock estimates are 200.72, 14.73, and 9.29 kt C, respectively.

The last section of Table 2 presents RERCA results from ca. 1865 CE to the present-day. While apparent C accumulation followed the long-term trend between CE, 1865–1915 on average, results since the beginning of the 20th century show an acceleration in peat accumulation and in C sequestrated in the subarctic Auassat peatland that exceeds the values from Bouleau and Mista peatlands.

Discussion

Carbon accumulation dynamics along a South to North transect within the Romaine River watershed

Overall, the carbon accumulation rates from the three peatlands show values higher than in Loisel et al. (2014) for eastern Canada. Long-term apparent rates of carbon accumulation for the Bouleau and Mista peatlands (35.5 and 15.4 g C m⁻² yr⁻¹) are in line with the range of 19-29 g C m⁻² yr⁻¹ estimated over the Northern Hemisphere from previous studies (Garneau et al., 2014; Gorham, 1991; Turunen et al., 2004). On the Romaine River delta along the coast of the Gulf of Saint-Lawrence, Magnan and Garneau (2014a) reported Holocene C accumulation rates ranging from 16 to 29 g C m⁻² yr⁻¹ within different peatland complexes. Some relatively low values were in part attributed to the harsh climatic conditions with high wind exposure along the coast that limited snow cover accumulation and favored frost penetration and duration within the peat from the late-Holocene cooling onwards.

The results of the present study confirm a decreasing trend in C accumulation rates with increasing latitude and altitude along the Romaine River watershed that encompasses boreal and subarctic conditions. Although peat initiation and ombrotrophication were almost synchronous at the Bouleau (9070 and 4560 cal BP) and Mista (8400 and 5340 cal BP) bogs, the Holocene rates of C accumulation (LORCA) are much higher for Bouleau (35.5 g C m⁻² yr⁻¹) than for Mista (15.4 g C m⁻² yr⁻¹). The GDD₀ and mean summer temperatures seem to be the main variables that influenced peat accumulation (Charman et al., 2013; Garneau et al., 2014; Gallego-Sala et al., 2018). The northward decrease in C accumulation rates is in line with previous studies suggesting that subarctic peatlands are less efficient at storing C over millennial timescales than boreal peatlands (Beilman et al., 2009; Garneau et al., 2014). As a subarctic fen, the Auassat peatland located in the northernmost section of the watershed recorded among the lowest LORCA values (9 g C m⁻² yr⁻¹) reported in northern peatlands throughout the Holocene (Loisel et al., 2014).

Overall, the C accumulation in the three peatlands shows relatively similar although asynchronous trends (Figures 3 and 5). An increase in CAR was registered following the fen to bog shift in both the Bouleau and Mista peatlands which was probably influenced by warmer climate conditions in the mid-Holocene (Yu et al., 2009) combined with a shift to a more recalcitrant Sphagnum communities following ombrotrophication. Conversely, a decrease in peat C accumulation during the Neoglacial cooling was registered in the three sites, as previously documented in Garneau et al. (2014) and Loisel et al. (2014) and was most pronounced at Auassat. Between 3360 and 40 cal BP, Auassat recorded very low carbon accumulation rates (1.4 g C m⁻² yr⁻¹), suggesting periods of very cold climate conditions, very short growing seasons and probable permafrost aggradation (Treat et al., 2016).

Figure 5.

Age-depth models for the three peatland sites modelled using the R package rbacon version 2.3.9.1 (Blaauw and Christen, 2011).

Holocene paleoecological reconstruction

Figure 6 summarizes, for the three peatlands, the changes in carbon accumulation and macrofossil assemblages throughout the Holocene.

Figure 6.

Synthesized reconstructions of the evolution of the three peatlands during the Holocene. Macrofossil assemblages are plotted according to depth and carbon accumulation rates are plotted according to time.

Early to Mid-Holocene (ca. 9000–6000 cal BP)

Following the postglacial uplift and sea retreat, the Bouleau and Mista peatlands developed almost simultaneously while the delayed initiation for Auassat may be explained by the persistence of the LIS at this latitude (Glaser and Janssens, 1986; Morris et al., 2018; Yu et al., 2009).

Holocene Climatic Optimum (ca. 6000–4000 cal BP)

From ca. 6000 cal BP, with the disappearance of the LIS, a major change in atmospheric circulation and precipitation patterns occurred due to the stabilization of the Atlantic Maritime Tropical air masses over eastern Canada, increasing summer humidity (Carcaillet and Richard, 2000; Payette et al., 2013; Viau et al., 2006; Yu et al., 2009). This change, which is also marked by an increase in summer temperature, may have favored the fen to bog transition both at Bouleau and Mista, while peatland accumulation had just initiated at Auassat. The fen to bog transition is one of the most important internal feedbacks in terms of carbon storage in peatlands. Peat and C accumulation increased possibly due a decrease in the water table and isolation from the groundwater influenced by warmer climate conditions and longer growing season length that influenced Sphagnum growth in the two ombrotrophic systems (Hughes, 2000; Hughes and Barber, 2003).

The fen to bog transition also registered an increase in C accumulation during this period, and this increase took place chronologically from South to North (ca. 5800, 5080, 40 cal BP) with rates for this period varying from 54.8 g C m⁻² yr⁻¹ at Bouleau, to 21.4 and 12.1 g C m⁻² yr⁻¹ at Mista and Auassat respectively (Figure 3).

Neoglacial: Little Ice Age (ca. 4000–100 cal BP)

A decrease in solar insolation caused a general cooling in temperatures over the Northern Hemisphere after ca. 4000 cal BP (Kutzbach, 1981; Viau et al., 2002, 2006). The northernmost peatland, Auassat, currently located at the ecotonal limit between ombrotrophic and subarctic peatland distribution was likely more sensitive to this cooling than the boreal sites. A decrease in peat C accumulation was recorded from North to South over time (respectively ca. 3360, 2700, and 2330 cal BP). After 3000 cal BP, the Bouleau and Mista peatlands registered conditions with fluctuating water tables, as registered in coastal peatlands by Magnan and Garneau (2014b), while peat C accumulation for Auassat decreased drastically (1.4 g C m⁻² yr⁻¹). We hypothesize that Auassat may have been affected by very short growing seasons as already recorded in eastern Canadian subarctic peatlands by Lamarre et al. (2012), Robitaille et al. (2021), and van Bellen et al. (2013). The brief Medieval Climate Anomaly was not recorded in any core probably due to the absence of more precise radiocarbon dating in the corresponding horizons. However, the potential effects of Little Ice Age cooling were registered both in Bouleau and Mista with a shift from Sphagnum mosses to a dominance of ligneous remains along with lower water tables as registered in other peatlands of eastern Canada (Beaulieu-Audy et al., 2009) while, in Auassat, peat accumulation was almost lacking.

Recent period (ca. 100–present day)

The climatic data series from 1950 to 2015 shows a general warming trend which is more pronounced in the subarctic (Auassat) than in the boreal domain (Bouleau and Mista). Since 2000 CE, mean summer temperatures are now higher in the subarctic Auassat peatland than in the Mista peatland, which is located further South, and they both record comparable GDD₀ despite their difference in latitude. The average autumn temperatures for Bouleau have exceeded the threshold of 0°C, extending the growing season length through September. On the other hand, precipitation does not show any clear pattern. The most important rates of C accumulation in the northernmost part of the watershed induced a recent fen to bog transition (~40 BP) on the lawn and hummock microforms. This ecosystem state-shift in Auassat is probably related to an increase in vegetation growth that also influenced an apparent lowering of the water table depth.

Subarctic regions currently record higher RERCA in northwestern Québec (147.1 g C m⁻² yr⁻¹; Lamarre et al., 2012), in north central Québec (141.4 g C m⁻² yr⁻¹; Robitaille et al., 2021), and in northeastern Québec (142.7 g C m⁻² yr⁻¹; Auassat) than boreal peatlands (83.5 g C m⁻² yr⁻¹; Loisel and Garneau, 2010) and in this study (86.1 and 84.47 g C m⁻² yr⁻¹; Bouleau and Mista respectively). The accelerated warming recorded in northernmost regions of Canada during the past decades (Zhang et al., 2019) may explain the higher RERCA values registered in the subarctic peatland of the Romaine watershed due to an increase in mean summer temperature and GDD₀ (Charman et al., 2013; Garneau et al., 2014).

Conclusion

This study brings a comprehensive analysis of peatlands C dynamics along a South to North transect within a watershed encompassing boreal and subarctic biomes in eastern Canada. The results show that the onset of peat accumulation was asynchronous and influenced both by latitude and altitude. The consequences of climate variations on peatlands were not uniform across the studied sites and suggest a nonlinear peatland response to climate forcing (Belyea, 2009). Peat initiation occurred rapidly following ice sheet and postglacial sea retreat in the South but was delayed in the northernmost section of the watershed. During the Holocene Climatic Optimum, peat C accumulation was slower in the northern minerotrophic site compared to the southernmost peatlands. The Neoglacial cooling may have induced an overall slowdown in peat accumulation and a shift in vegetation communities in peatlands from the boreal domain while the subarctic peatland recorded shorter growing seasons and possible periods of permafrost aggradation. These results confirm the sensitivity of subarctic peatlands to climate variations. They seem to warm more slowly but also react more rapidly to cooling than boreal peatlands, suggesting that variables affecting productivity, in this case, mean summer temperature and GDD₀, are the main drivers of peat accumulation (Charman et al., 2013; Garneau et al., 2014). Furthermore, our results show that recent rates of accumulation are higher towards the North. The recent warming of the 20th century favored the ombrotrophication of some sections of the subarctic peatland as registered in other regions of eastern Canada with the increase in summer and autumn temperatures that influenced the vegetation productivity and lengthened the growing season duration. The results of this study also suggest that the climate acceleration toward the North may potentially influence the northward migration of the ombrotrophic peatland distribution within the next decades.

Supplemental Material

sj-pdf-1-hol-10.1177_0959683620988031 – Supplemental material for Carbon accumulation in peatlands along a boreal to subarctic transect in eastern Canada

Supplemental material, sj-pdf-1-hol-10.1177_0959683620988031 for Carbon accumulation in peatlands along a boreal to subarctic transect in eastern Canada by Guillaume Primeau and Michelle Garneau in The Holocene

Footnotes

Acknowledgements

We would like to thank Charles-Élie Dubé-Poirier, Nolann Chaumont, Pierre Grondin and Steve Pratte for field and laboratory assistance and Guillaume Dueymes and Philippe Gachon (Centre ESCER-UQAM) for providing the ANUSPLIN climate data. Thanks to Mylène Robitaille, Joannie Beaulne, Dr. Simon van Bellen, and all the members of Les Tourbeux for their constant help and support. The important contribution from Nicole Sanderson for the development of the age-depth models and editing of the text was greatly appreciated. We also thank Dr. Gabriel Magnan (UQAM) and Professor Julie Talbot (UdeM) as well as two anonymous reviewers for providing thoughtful comments on earlier versions of the manuscript.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Natural Sciences and Engineering Research Council of Canada to Michelle Garneau (RDCPJ 514218-17) and the Mitacs Acceleration program.

ORCID iD

Guillaume Primeau

Supplemental material

Supplemental material for this article is available online.

References

Appleby

(2001) Chronostratigraphic techniques in recent sediments. In: Last

Smol

(eds) Tracking Environmental Change Using Lake Sediments vol. 1 Basin Analysis, Coring, and Chronological Techniques. Dordrecht: Kluwer Academic Publishers, pp.171–203.

Appleby

Oldfield

(1978) The calculation of lead-210 dates assuming a constant rate of supply of unsupported 210Pb to the sediment. CATENA 5(1): 1–8.

Beaulieu-Audy

Garneau

Richard

PJH

, et al. (2009) Holocene palaeoecological reconstruction of three boreal peatlands in the La Grande Rivière region, Québec, Canada. The Holocene 19(3): 459–476.

Beilman

MacDonald

Smith

, et al. (2009) Carbon accumulation in peatlands of West Siberia over the last 2000 years. Global Biogeochemical Cycles 23(1): 1–12.

Belyea

Clymo

(2001) Feedback control of the rate of peat formation. Proceedings of the Royal Society B: Biological Sciences 268(1473): 1315–1321.

Belyea

(2009) Nonlinear dynamics of peatlands and potential feedbacks on the climate system. Carbon Cycling in Northern Peatlands 184: 5–18.

Belyea

Malmer

(2004) Carbon sequestration in peatland: Patterns and mechanisms of response to climate change. Global Change Biology 10(7): 1043–1052.

Bernatchez

(2003) Évolution littorale holocène et actuelle des complexes deltaïques de Betsiamites et de Manicouagan-Outardes: Synthèse, processus, causes et perspectives. Université Laval. Available at: https://corpus.ulaval.ca/jspui/bitstream/20.500.11794/33004/1/20745.pdf.

Bernatchez

(2005) Déglaciation et variations du niveau marin de la Mer de Goldthwait sur la Côte-Nord. Centre Boréal du Saint-Laurent.

10.

Blaauw

Christen

(2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6(3): 457–474.

11.

Booth

(2008) Testate amoebae as proxies for mean annual water-table depth in Sphagnum-dominated peatlands of North America. Journal of Quaternary Science: Published for the Quaternary Research Association 23(1): 43–57.

12.

Booth

Lamentowicz

Charman

(2010) Preparation and analysis of testate amoebae in peatland palaeoenvironmental studies. Mires and Peat 7(2): 1–7.

13.

Braun-Blanquet

(1932) Plant sociology. The study of plant communities, 1st edn. New York and London: McGraw-Hill Book Co., Inc.

14.

Carcaillet

Richard

PJH

(2000) Holocene changes in seasonal precipitation highlighted by fire incidence in eastern Canada. Climate Dynamics 16(7): 549–559.

15.

Chambers

Beilman

(2011) Methods for determining peat humification and for quantifying peat bulk density, organic matter and carbon content for palaeostudies of climate and peatland carbon dynamics. Mires and Peat 7(7): 1–10.

16.

Charman

(2002) Peatlands and Environmental Change. Chichester, UK: Wiley and sons.

17.

Charman

Beilman

Blaauw

, et al. (2013) Climate-related changes in peatland carbon accumulation during the last millennium. Biogeosciences 10(2): 929–944.

18.

Charman

Hendon

Woodland

(2000) The Identification of Testate Amoebae (Protozoa: Rhizopoda) in Peats: QRA Technical Guide No. 9. London: Quaternary Research Association.

19.

Crowther

van den Hoogen

Wan

, et al. (2019) The global soil community and its influence on biogeochemistry. Science 365(6455): eaav0550.

20.

Dubois

J-MM

(1980) Environnements quaternaires et évolution postglaciaire d’une zone côtière en émersion en bordure sud du Bouclier canadien : la Moyenne Côte Nord du Saint-Laurent, Québec. Thèse de doctorat, département de géographie, Université d’Ottawa.

21.

Dyke

Moore

Robertson

(2003) Deglaciation of North America. Ottawa, ON: Geological Survey of Canada.

22.

Engstrom

Hansen

BCS

(1985) Postglacial vegetational change and soil development in southeastern Labrador as inferred from pollen and chemical stratigraphy. Canadian Journal of Botany 63: 543–561.

23.

Faubert

(2014) Flore des bryophytes du Québec-Labrador. Saint-Valérien, Québec: Société québécoise de bryologie, viii + 465 p., illus.

24.

Foster

(1983) The history and pattern of fire in the boreal forest of southeastern Labrador. Canadian Journal of Botany 61(9): 2459–2471.

25.

Foster

Glaser

(1986) The raised bogs of south-eastern Labrador, Canada: Classification, distribution, vegetation and recent dynamics. Journal of Ecology 74(1): 47–72.

26.

Frolking

Roulet

(2007) Holocene radiative forcing impact of northern peatland carbon accumulation and methane emissions. Global Change Biology 13(5): 1079–1088.

27.

Gallego-Sala

Charman

Brewer

, et al. (2018) Latitudinal limits to the predicted increase of the peatland carbon sink with warming. Nature Climate Change 8(10): 907–913.

28.

Garneau

(1995) Collection de référence de graines et autres macrofossiles végétaux de taxons provenant du Québec méridional et boréal/Reference collection of seeds and other botanical macrofossils from meridional and boreal Quebec. Commission géologique du Canada, Dossier public 3048, p.38.

29.

Garneau

van Bellen

Magnan

, et al. (2014) Holocene carbon dynamics of boreal and subarctic peatlands from Québec, Canada. The Holocene 24(9): 1043–1053.

30.

Ghaleb

(2009) Overview of the methods for the measurement and interpretation of short-lived radioisotopes and their limits. IOP Conference Series: Earth and Environmental Science 5: 012007.

31.

Glaser

Foster

(1984) The vascular flora of raised bogs in south-eastern Labrador and its phytogeographic significance. Canadian Journal of Botany 62(7): 1361–1364.

32.

Glaser

Janssens

(1986) Raised bogs in eastern North America: Transitions in landforms and gross stratigraphy. Canadian Journal of Botany 64(2): 395–415.

33.

Gorham

(1991) Northern peatlands : Role in the carbon cycle and probable responses to climatic warming. Ecological Applications 1(2): 182–195.

34.

Hopkinson

McKenney

Milewska

, et al. (2011) Impact of aligning climatological day on gridding daily maximum–minimum temperature and precipitation over Canada. Journal of Applied Meteorology and Climatology 50: 1654–1665.

35.

Hughes

PDM

(2000) A reappraisal of the mechanisms leading to ombrotrophy in British raised mires. Ecology Letters 3(1): 7–9.

36.

Hughes

PDM

Barber

(2003) Mire development across the fen-bog transition on the Teifi floodplain at Tregaron Bog, Ceredigion, Wales, and a comparison with 13 other raised bogs. Journal of Ecology 91(2): 253–264.

37.

Hughes

PDM

Dumayne-Peaty

(2002) Testing theories of mire development using multiple successions at Crymlyn Bog, West Glamorgan, South Wales, UK. Journal of Ecology 90(3): 456–471.

38.

Hutchinson

McKenney

Lawrence

, et al. (2009) Development and testing of Canada-wide interpolated spatial models of daily minimum/maximum temperature and precipitation for 1961-2003. Journal of Applied Meteorology and Climatology 48:, 725–741.

39.

Jeglum

Rothwell

Berry

, et al. (1991) New volumetric sampler increases speed and accuracy of peat surveys. Frontline - Forestry Research Applications 9: 1–4.

40.

Jowsey

(1966) An improved peat sampler. New Phytologist 65(2): 245–248.

41.

Juggins

(2007) C2 Version 1.5: Software for ecological and palaeoecological data analysis and visualisation. Newcastle University, Newcastle upon Tyne.

42.

Kuhry

Vitt

(1996) Fossil Carbon/Nitrogen ratios as a measure of peat decomposition. Ecology 77(1): 271–275.

43.

Kutzbach

(1981) Monsoon climate of the early Holocene: Climate experiment with the earth’s orbital parameters for 9000 years ago. Science 214(4516): 59–61.

44.

Laine

Harju

Timonen

, et al. (2009) The Intricate Beauty of Sphagnum Mosses-a Finnish Guide to Identification. Helsinki, FI: University of Helsinki.

45.

Lamarre

Garneau

Asnong

(2012) Holocene paleohydrological reconstruction and carbon accumulation of a permafrost peatland using testate amoeba and macrofossil analyses, Kuujjuarapik, subarctic Québec, Canada. Review of Palaeobotany and Palynology 186: 131–141.

46.

Lamarre

Magnan

Garneau

, et al. (2013) A testate amoeba-based transfer function for paleohydrological reconstruction from boreal and subarctic peatlands in northeastern Canada. Quaternary International 306: 88–96.

47.

Lévesque

PEM

Dinel

Larouche

(1988) Guide illustré des macrofossiles végétaux des tourbières du Canada. Ottawa, ON: Direction générale de la recherche, Agriculture Canada.

48.

Loisel

Garneau

(2010) Late-Holocene paleoecohydrology and carbon accumulation estimates from two boreal peat bogs in eastern Canada: Potential and limits of multi-proxy archives. Palaeogeography, Palaeoclimatology, Palaeoecology 291(3–4): 493–533.

49.

Loisel

Beilman

, et al. (2014) A database and synthesis of northern peatland soil properties and Holocene carbon and nitrogen accumulation. The Holocene 24(9): 1028–1042.

50.

Magnan

Garneau

(2014a) Climatic and autogenic control on Holocene carbon sequestration in ombrotrophic peatlands of maritime Quebec, eastern Canada. The Holocene 24(9): 1054–1062.

51.

Magnan

Garneau

(2014b) Evaluating long-term regional climate variability in the maritime region of the St. Lawrence North Shore (eastern Canada) using a multi-site comparison of peat-based paleohydrological records. Journal of Quaternary Science 29(3): 209–220.

52.

Magnan

Garneau

Payette

(2014) Holocene development of maritime ombrotrophic peatlands of the St. Lawrence North Shore in eastern Canada. Quaternary Research 82(1): 96–106.

53.

Marie-Victorin

(2005) Flore Laurentienne. Montréal: Les Presses de l’Université de Montréal.

54.

Mauquoy

Hughes

PDM

van Geel

(2010) A protocol for plant macrofossil analysis of peat deposits. Mires and Peat 7(6): 1–5.

55.

Mauquoy

van Geel

(2007) Plant macrofossil methods and studies: Mire and peat macros. In: Elias

(ed.) Encyclopedia of Quaternary Science. Amsterdam: Elsevier, pp.2315–2336.

56.

McKenney

Hutchinson

Papadopol

, et al. (2011) Customized spatial climate models for North America. Bulletin of the American Meteorological Society 92(12): 1611–1622.

57.

Morris

Swindles

Valdes

, et al. (2018) Global peatland initiation driven by regionally asynchronous warming. Proceedings of the National Academy of Sciences of the United States of America 115(19): 4851–4856.

58.

Nilsson

Hjelmqvist

(1967) Studies on nutlet structure of South Scandinavian species of carex. Lund: Greerup.

59.

Payette

(2001a) Les principaux types de tourbières. In: Payette

Rochefort

(eds) Écologie des tourbières du Québec-Labrador. Québec: Presses de l’Université Laval, pp.39–89.

60.

Payette

(2001b) Le contexte physique et biogéographique. In: Payette

Rochefort

(eds) Écologie des tourbières du Québec-Labrador. Québec: Presses de l’Université Laval, pp.9–37.

61.

Payette

Garneau

Delwaide

, et al. (2013) Forest soil paludification and mid-Holocene retreat of jack pine in easternmost North America: Evidence for a climatic shift from fire-prone to peat-prone conditions. The Holocene 23(4): 494–503.

62.

Payne

(2011) Can testate amoeba-based palaeohydrology be extended to fens? Journal of Quaternary Science 26(1): 15–27.

63.

Payne

Mitchell

EAD

(2009) How many is enough? Determining optimal count totals for ecological and palaeoecological studies of testate amoebae. Journal of Paleolimnology 42(4): 483–495.

64.

Reimer

Bard

Bayliss

, et al. (2013) IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon 55(4): 1869–1887.

65.

Robitaille

Garneau

van Bellen

, et al. (2021) Long-term and recent ecohydrological dynamics of patterned peatlands in north-central Quebec (Canada). The Holocene. Epub ahead of print 18 January 2021. DOI: 10.1177/0959683620988051.

66.

Sheng

Smith

MacDonald

, et al. (2004) A high-resolution GIS-based inventory of the west Siberian peat carbon pool. Global Biogeochemical Cycles 18(3).

67.

Stockmarr

(1971) Tablets with spores used in absolute pollen analysis. Pollen et Spores XIII(January 1971): 615–621.

68.

Tarnocai

(2006) The effect of climate change on carbon in Canadian peatlands. Global and Planetary Change 53(4): 222–232.

69.

Treat

Jones

Camill

, et al. (2016) Effects of permafrost aggradation on peat properties as determined from a pan-Arctic synthesis of plant macrofossils. Journal of Geophysical Research: Biogeosciences 121(1): 78–94.

70.

Turunen

Roulet

Moore

, et al. (2004) Nitrogen deposition and increased carbon accumulation in ombrotrophic peatlands in eastern Canada. Global Biogeochemical Cycles 18(3): 1–12.

71.

Turunen

Tolonen

Tomppo

, et al. (2002) Estimating carbon accumulation rates of undrained mires in Finland - Application to boreal and subarctic regions. The Holocene 12(1): 69–80.

72.

van Bellen

Garneau

Ali

, et al. (2013) Poor fen succession over ombrotrophic peat related to late-Holocene increased surface wetness in subarctic Quebec, Canada. Journal of Quaternary Science 28(8): 748–760.

73.

Verpaelst

Madore

Brisebois

, et al. (1999) Géologie de la région du lac Jérôme, du lac Sanson et de la petite rivière de l’Abbé-Huard (SNRC 12L/13, 12L/14, 12M/03). Québec: Ministère des Ressources naturelles, RG 98-02, p.38.

74.

Viau

Gajewski

Fines

, et al. (2002) Widespread evidence of 1500 yr climate variability in North America during the past 14 000 yr. Geology 30(5): 455–458.

75.

Viau

Gajewski

Sawada

, et al. (2006) Millennial-scale temperature variations in North America during the Holocene. Journal of Geophysical Research: Atmospheres 111(9): 1–12.

76.

Vincent

Wang

Milewska

, et al. (2012) A second generation of homogenized Canadian monthly surface air temperature for climate trend analysis. Journal of Geophysical Research: Atmospheres 117: D18110

77.

Morris

Liu

, et al. (2018) PEATMAP: Refining estimates of global peatland distribution based on a meta-analysis. Catena 160: 134–140.

78.

(2006) Holocene carbon accumulation of fen peatlands in boreal Western Canada: A complex ecosystem response to climate variation and disturbance. Ecosystems 9: 1278–1288.

79.

Beilman

Frolking

, et al. (2011) Peatlands and their role in the global carbon cycle. Eos, Transactions American Geophysical Union 92(12): 97–108.

80.

Beilman

Jones

(2009) Sensitivity of northern peatland carbon dynamics to Holocene climate change. Carbon Cycling in Northern Peatlands 184: 55–69.

81.

Loisel

Brosseau

, et al. (2010) Global peatland dynamics since the Last Glacial Maximum. Geophysical Research Letters 37(13): 1–5.

82.

Zhang

Piilo

Amesbury

, et al. (2018) The role of climate change in regulating Arctic permafrost peatland hydrological and vegetation change over the last millennium. Quaternary Science Reviews 182: 121–130.

83.

Zhang

Flato

Kirchmeier-Young

, et al. (2019) Changes in temperature and precipitation across Canada. In: Bush

Lemmen

(eds) Canada’s Changing Climate Report. Ottawa, ON: Government of Canada, pp.112–193.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.60 MB