Development of a subarctic peatland linked to slope dynamics at Lac Wiyâshâkimî (Nunavik,Canada)

Abstract

A palaeoecological study of a subarctic minerotrophic peatland was undertaken to reconstruct the formation of the site as an archive of slope geomorphological processes. The study peatland is located about 400 m from Caribou slope (unofficial name) on Lepage Island, Lac Wiyâshâkimî, Nunavik (northern Québec, Canada). The site is close to the lakeshore and receives runoff directly from Caribou slope and its catchment. Gravity processes have been active on Caribou slope since the deglaciation of the region at approximately 6000 cal. yr BP. These processes may be differentiated in terms of Holocene stages of intensity. The objective of our study was to detect evidence of gravity processes in the peatland and to note their frequency since its establishment using loss-on-ignition testing, macrofossil analysis and radiocarbon dating. Our results indicate that peat began to accumulate over the sandy-gravelly sediments at around 4900 cal. yr BP. Larix Laricina, Carex aquatilis and Carex rostrata were present at this time until 4660 cal. yr BP, at which point these taxa were replaced by aquatic taxa such as Hippuris vulgaris and Daphnia (aquatic invertebrates). The percentage of mineral sediments (sand) remained high during this period, which could be linked to slope activity. After 4660 cal. yr BP, sandy sediments diminished while episodes of aquatic conditions and sand inflow occurred on at least three occasions (at 4660, 3905 and 3130 cal. yr BP). The increase in water flow and the introduction of more medium to fine sand into the peatland could be linked to slope movements and the long-distance runout of debris flow that we observed in the field. Given these factors, conditions at the study site remained wet from the earliest phases until the present. Unlike the subarctic permafrost peatlands in northern Québec, permafrost did not become established at the study site.

Keywords

climate change minerotrophic peatland Nunavik permafrost plant-macrofossils slope dynamics

Introduction

Many studies have shown that subarctic peatlands were and still are influenced by allogenic factors such as climate and fire as well as by autogenic processes such as peat accumulation and Sphagnum acidification (e.g. Payette, 1988; Seppälä, 2011). Large amounts of organic carbon are present in permafrost peatlands in the northern hemisphere and, as was shown by the analysis of a large northern peatlands database, Holocene climatic changes such as the Holocene Thermal Maximum and the Neoglacial cooling influenced permafrost peatlands and carbon sequestration rates in these ecosystems (Loisel et al., 2014; Robinson and Moore, 2000). However, the response of northern peatlands to climate is complex and may be hard to detect. This difficulty is likely caused by factors such as autogenic feedback and the geomorphological/geographical location of the peatlands (e.g. Belyea, 2009; Bhiry et al., 2007; Payette, 2009; Waddington et al., 2015). According to a recent study by Bhiry and Lavoie (2018), there have been several periods of permafrost aggradation as well as palsa and peat plateau formation since deglaciation in many sites in Scandinavia, Siberia and Canada. A palsa is a mound of peat rising out of a mire 1–7 m high that contains a permafrost core of peat or silt. The Neoglacial period (which started at about 3500 cal. yr BP) and the Little Ice Age (LIA; 1650–1850 A.D) appear to have triggered permafrost inception in these circumpolar regions (Bhiry and Lavoie 2018) because of both cold conditions and lack of snow precipitation (Payette, 2009).

To our knowledge, few studies have examined the role of local geomorphic factors on peatland succession. Local factors are significant because they may mitigate or exacerbate the impact of changing climatic conditions. For example, Bhiry et al. (2007) underlined the role of geomorphic factors on peatland development at the treeline in northern Québec (northeastern Canada). In that study, the location of the peatland in a depression limited by two small lakes and a large rocky hill favoured occasional inflows of mineral-laden water that reinforced the persistence of shallow water conditions until permafrost inception at about 3670 cal. yr BP. This pattern of peatland evolution differs from the general development of subarctic permafrost peatlands (at least in northern Québec). Typically, development consists of several sub-developmental stages, from shallow open water to fen and occasionally to bog, before reaching the permafrost stage (e.g. Arlen-Pouliot and Bhiry, 2005; Bhiry and Robert, 2006; Couillard and Payette, 1985; Fillion et al., 2014; Heim, 1976; Lavoie and Payette, 1995).

The peatland in this study is a minerotrophic peatland (the Bear peatland, an unofficial name) located about 400 m from a talus slope on Lepage Island, which is located in the western basin of Lac Wiyâshâkimî (formerly known as Lac Wiyasakami and Lac à l’Eau-Claire) (Nunavik). The terrain dips from the slope to the small peatland at a grade of about 10°. The Bear peatland receives runoff directly from the slope and its catchment. A recent study investigated the geomorphic processes responsible for the slope deposits (Decaulne et al., 2018) and found that the slope has been exposed to gravity processes such as rockfall, debris flows, dry avalanches and snow avalanches and ice-crust sliding since the deglaciation of the region at approximately 6000 yr BP. That study also showed that present-day processes remain active (to a limited extent) and that there is a constant redistribution of debris over the talus (Decaulne et al., 2018). The Bear peatland is very different from many sites that have been studied in subarctic Québec on Hudson Bay and the Ungava Bay coast (see Bhiry and Lavoie, 2018). Those sites consist of palsas and peat plateaus, which are periglacial landforms that are commonly found in subarctic peatlands. In the present day, these landforms are melting because of climate warming (note that the collapse of a palsa generates a thermokarst pond). On the nearby island Aux Forreurs, 8 km south-east of the study area, valuable research was conducted on slope dynamics in relation to forest vegetation and fire events (Bégin and Filion, 1985; Marion et al., 1995). Our study aims to detect evidence of gravity processes (e.g. rockfall, debris flows, dry avalanches and snow avalanches and ice-crust sliding) in the peatland and to document the periods of intense geomorphological slope processes that have been recorded in the peatland throughout the Holocene.

Study setting

The Bear peatland is located on Lepage Island (Wiskichanikw in Cree), which is one of the islands that is arranged in a ring shape in the western basin of Lac Wiyâshâkimî (56°20’524”N; 74° 27’167”W). The other large islands are Aux Forreurs (Kamiskutanikaw in Cree) and Atkinson; the Tétard islands are much smaller (Figure 1). Lac Wiyâshâkimî lies inland approximately 150 km from Hudson Bay (Figure 1) and is made up of two circular basins: the western basin is 35 km wide and was created by a meteor impact at about 286 Ma, while the eastern basin is 22 km wide and is older (Schmieder et al., 2015). The region is part of the Canadian Shield, but the islands are formed by felsic lava and granodiorites of Pennsylvanian age (Bostock, 1969) (for the geological synthesis, see Decaulne et al., 2018).

Figure 1.

Location of the study area in (a) northern Québec, (b) Nunavik, (c) on Lepage Island and (d) the northern shore. (e) The Bear peatland is at the mouth of a small catchment dominated by (f) Caribou slope (photos: A. Decaulne 2016 and 2018).

The deglaciation of the Lac Wiyâshâkimî region dates to before 5300 cal. yr BP and the central islands of Lac Wiyâshâkimî were ice-free by 5050 cal. yr BP (Payette, 1984). Discontinuous cover of glacial deposits (till) is visible on the islands and around Lac Wiyâshâkimî. Following deglaciation, a proglacial lake submerged the depression of the lake, as indicated by the presence of several perched lacustrine terraces (Allard and Séguin, 1985).

Lac Wiyâshâkimî is in subarctic Québec and lies in the discontinuous permafrost zone (Allard and Séguin, 1987). According to the SILA weather station of the Centre d’études nordiques (CEN) (situated on the northern shore of the lake), the annual data for 2006, 2007 and for 2011–2015 indicate a mean annual temperature of −3.58°C (data available from the SILA network website: http://www.cen.ulaval.ca/sila.php?xml = cartesila&lt = 72&lg = -79&zm = 3). The Lac Wiyâshâkimî region is part of the forest tundra, a transition area between the boreal forest and arctic tundra (Payette, 1983). The landscape is open tundra with lichen and forest areas dominated by black spruce (Picea mariana) and tamarack (Larix laricina) (Bégin, 1986). Forest patches are restricted to sites sheltered from the wind such as depressions and the lower parts of slopes, while the summit surfaces are devoid of vegetation or covered by discontinuous lichen-heath community cover. We observed many peatland palsas on Lac Wiyâshâkimî Islands that lie on depressions at different levels, from 2 to 100 m above the present lake level (235 m a.s.l). As a consequence of the permafrost thaw in these peatlands, many palsas are collapsing and are being replaced by thermokarst ponds (Langlais, 2016). The Bear peatland is a small minerotrophic peatland colonized primarily by Sphagnum, Carex rostrata, C. calyculata and Eriophurum russeolum. The Bear peatland is close to the lakeshore about 400 m north-east of Caribou slope (30–50 m high). Debris-flow features (levees, channels and lobes) reach about 200 m from the talus slope, towards the Bear peatland; the levees and channels are 2–5 m wide and approximately 2 m high (Figure 1). The site thus receives runoff directly from the slope and its catchment, Caribou slope being the most prominent high point within the catchment and the only one lacking herbaceous plants, shrubs or tree cover. The peatland measures approximately 210 m long and 160 m wide and is surrounded by a black spruce forest that grows on a Sphagnum mat at some locations. Although the peatland is small, its depth varies from between 60 and 200 m (organic-mineral transition).

Methods

A peat core was removed from the deepest location of the peatland (180–200 cm) using a Russian peat corer. The first 60 cm of the core (uppermost part) consists of water and some present-day Sphagnum represented by Sphagnum flexuosum, C. rostrata (fen taxa), E. russeolum and C. calyculata. The rest of the core (140 cm) is composed of sandy organic matter overlain by decomposed peat. Samples were taken from the core for loss on ignition (LOI), plant macrofossil analysis and ¹⁴C dating.

LOI was performed at 2-cm intervals to determine the respective percentages of organic material and mineral material. Samples were first dried in a steam room at 100ºC and weighed (initial weight). They were then heated in an oven (Thermolyne 100) to a temperature of 600ºC until the organic matter was completely eliminated (Heiri et al., 2001). The mineral fraction was then weighed again (final weight).

Macrofossil analysis was also performed at 2-cm intervals following the protocol outlined by Bhiry and Filion (2001). The analysis was carried out at the Laboratoire de paléoécologie at the Centre d’études nordiques (CEN–Université Laval). Each 1-cm thick slice consisted of a volume of 8–10 cm³ of sediment with the exception of the two samples at the base of the core (200–190 cm and 190–181 cm), for which 50 and 40 cm³ were analysed, respectively. Sediments were treated with a weak 5% aqueous potassium hydroxide (KOH) solution and boiled for a few minutes to deflocculate. The material was then wet-screened through a 180-mm mesh sieve. In each sample, the frequency of each group of botanical components and sand was estimated as a percentage by volume (10 cm³). The number of tree, shrub and herbaceous macroremains in each sample was counted and converted to counts per 10 cm³. The species of Sphagnum were identified based on a subsample of 100 leaves. Macrofossils were identified under a binocular microscope at 6–20X magnification. References used in identifying plant remains included Rousseau (1974), Montgomery (1977), Porsild and Cody (1980), Marie-Victorin (1995) and Payette (2013, 2015), as well as the collection of CEN. All of the plant macroremains that we identified were incorporated into the CEN collection.

Six organic samples were dated with accelerator mass spectrometry (AMS) at CEN’s laboratory at Université Laval and at the Keck Laboratory, University of California, Irvine (UL-KIU). These samples consisted of needles of Picea or brown moss leaves. The sample from the upper level was composite because the material from one taxon was of insufficient quantity (Table 1). Dates were calibrated using the Calib 7.1 program (Stuiver et al., 2014). For calculation of the age–depth relationship and peat accumulation rates, we used linear interpolation between calibrated dates (the midpoint of the 2-sigma range). In the text, we always use calibrated dates but in Table 1 we included the uncalibrated dates and BC/AD dates to provide a comparison with previous work.

Table 1.

Radiocarbon and calibrated ages of the monolith sampled at Bear peatland, Lac à l’Eau-Claire (LEC), Nunavik.

Sample	14C Age (yr. BP)	Age AD–BC (2ơ)	Interval (cal. yr. BP) (2ơ)	Midpoint calibrated age (cal. yr. BP)	Dated material
60–62 cm	120 ± 20	1682–1936 AD	14–268	113	Macrorests
102–103 cm	2965 ± 20	1256–1117 BC	3067–3209	3128	Moss
122–123 cm	3600 ± 20	2022–1896 BC	3846–3972	3905	Moss
140–141 cm	4125 ± 20	2864–2585 BC	4535–4814	4659	Spruce needle
181–190 cm	4340 ± 25	3017–2902 BC	4852–4967	4903	Moss

Results and interpretation

Peat started to accumulate over gravelly sand in the deepest parts of the peatland (180–200 cm) at around 4905 cal. yr BP. The gravelly sand originates from beach deposits of the lake when it was larger. Between 180 and 140 cm – at a depth where mineral sediment dominates – the material (organic and minerogenic) accumulation rate was high at about 0.184 cm/yr. It decreased thereafter to 0.025 and then to 0.014 cm/yr (Figure 2).

Figure 2.

(a) Material extracted at a depth of between 150 and 200 cm, showing the dominance of sand and gravel. (b) Age–depth relationship illustrating the decreasing material accumulation rate over time.

Five macrofossil zones were identified based on plant assemblages (M-I, M-II, M-III, M-IV and M-V). The percentage of the mineral content is higher than 70% at the lower part of the core, but then it decreases at the expense of organic material. A significant increase in mineral content and a concomitant increase in aquatic species occurred on three occasions: after 4660, 3900 and 3130 cal. yr BP (Figures 3 and 4).

Figure 3.

Summarized plant macrofossil diagram of the Bear peatland core.

Figure 4.

Plant macrofossil diagram of the Bear peatland core.

Zone M-I: 200–181 cm

Zone M-I is dominated by gravel and coarse sand along with sparse moss leaves from Sarmentypnum exannulatum/Warnstorfia fluitans sp.

Zone M-II: 180–142 cm (4905–4659 cal. yr BP)

This zone is characterized by the abundance of aquatic and sub-aquatic species such as Carex rostrata and Hippuris vulgaris and Daphnia sp., a freshwater crustacean. Coniferous remains (seeds and needles from Picea mariana and Larix laricina) are abundant, which indicates that trees were present in the vicinity of the site. Remains from Larix laricina decreased while those from Picea marian reached their peak at a depth of 164–166 cm. Thereafter, remains from coniferous trees become sparse and the macrofossil assemblage becomes less diversified than before. There were only remains from Carex canescens in this zone.

Zone M-III: 141–124 cm

For Zone M-III, sub-aquatic taxa are less frequent than in the previous zone. In fact, H. vulgaris disappeared while Daphnia sp. declined before completely disappearing. Seeds from fen species such as C. rostrata, C. canescens and Juncus filiformis were abundant in the assemblage. Remains from L. laricina are almost absent while those from P. mariana are still present. Charcoal was more abundant at the top of the zone.

Zone M-IV: 124–101 cm

In this zone, L. laricina and S. exannulatum/W. fluitans reestablished themselves while several taxa such as C. canescens and Viola sp. completely disappeared. There was a sporadic appearance of ombrotrophic species such as Chameadaphne calyculata, Vaccinium sp. and Eriophorum russeolum. Some seeds from Menyanthes trifoliata, a minerotrophic species, were identified.

Zone M-V: 100–60 cm

Zone M-V was subdivided into two macrofossil sub-zones: M-Va and M-Vb.

In sub-zone M-Va, remains from L. laricina disappeared, while those from P. marian., Viola sp. and E. russeolum remained. In sub-zone M-Vb, the macrofossil assemblage is much less diverse. P. mariana declined while charcoal is very abundant in this sub-zone, indicating recent fire events.

Discussion

Diverse phases of vegetation succession were interpreted using macrofossil and LOI data of a core taken form the centre of the peatland. The following phases were identified: aquatic phase (pond), filling-in phase and minerotrophic peatland formation.

Aquatic phase (4905–4660 cal. yr BP)

At the study site, the organic matter lies on a mineral deposit consisting of fine-to-coarse sand and gravel. The deposit originates from the beach sediment of Lac Wiyâshâkimî lake when it was larger.

Peat accumulation began at around 4905 cal. yr BP with the establishment of S. exannulatum/W. fluitans, a species of brown moss found in aquatic or humid environments (Crum and Anderson, 1981). The higher percentage of mineral sediments (60–80% fine sand content) in zones M-IIa and M-IIb suggests that the upstream slope was unstable and erosion processes were a significant factor during this period, enabling transport of minerogenic material to the peatland.

The significant inflow of sand would have contributed to the rapid peat accumulation (0.184 cm/yr) during a short period of time (between 4905 and 4660 cal. yr BP). During this first phase, the site contained a pond that was colonized by aquatic plants such as C. rostrata, C. aquatilis, H. vulgaris and Daphinia sp. (a freshwater crustacean). A forest consisting of L. laricina and P. mariana established itself on the edges of the pond, with the later arrival of Betulaceae (probably Betula glandulosa) (Figure 4).

Filling-in and the minerotrophic phase (4660–3905 cal. yr BP)

The second phase of vegetation evolution (M-III) lasted approximately 755 years, during which time 17 cm of material accumulated (at a rate of 0.025/yr). Yet, there was a sudden reduction of the proportion of organic matter from 70.5% (144 cm) to 45.1% (140 cm) at the beginning of this period (4600 cal. yr BP) as shown in Figure 3. This change could reflect the predominance of sand inflow in the peatland, which would have been caused by the dynamics on Caribou slope (Decaulne et al., 2018). During this phase, aquatic species underwent a significant decline and the brown moss S. exannulatum/W. fluitans was replaced by herbaceous species (Figure 4). The disappearance of L. laricina at this time also indicates a lowering of the water table. However, the presence of herbaceous species suggests that the conditions remained relatively humid, since C. rostrata, C. canescens and J. filiformis are commonly found in intermediate minerotrophic peatlands (Garneau, 2001). This filling-in phase at the Bear peatland corresponds to the period of rapid peat formation described by Payette (1988) on the small central island of the lake. According to that study, a humid climate encouraged the expansion of the peatland between 5000 and 3800 in the Lac Wiyâshâkimî region. Thus, the accumulation of peat at the Bear peatland would have been the principal autogenous factor in the succession of vegetation at the study site during this phase.

Minerotrophic peatland and a more humid phase (3905–3130 cal. yr BP)

According to Payette (1988), the Lac Wiyâshâkimî region was subjected to colder temperature and an increase in atmospheric humidity at around 3500 cal. yr BP; these conditions induced an increase in precipitation inputs to the site. Similar conditions were reported in several studies conducted in northern Quebec and in Nunavik. These studies detected a higher water table between 4200 and 3000 cal. yr BP (Bhiry and Robert, 2006; Lavoie and Payette, 1995; Miousse et al., 2003; Payette and Fillion, 1993; Tremblay et al., 2014). At the study site, this transition to a more humid climate brought about the return of the brown moss S. exannulatum/W. fluitans and L. laricina in zone M-IV. We also found seeds from Menyanthes trifoliata dating to this period, which is a minerotrophic and semi-aquatic species (Marie-Victorin, 1995; Porsild and Cody, 1980). Similar findings were highlighted for the period between 4350 and 3500 cal. yr BP by Langlais (2016) in a palsa peatland in the vicinity (about 7 km north). Towards the beginning of this phase at 3905 cal. yr BP, we also found a sudden decrease in the proportion of organic mater, which dropped from 69.5% (126 cm) to 55.6% (124 cm). This corresponds to an input of minerogenic material potentially caused by hillslope erosion processes, transportation and subsequent sedimentation within the Bear peatland. The rate of accumulation was around 0.025 cm/yr between 4660 and 3130 cal. yr BP.

Relative decrease in the water table phase (3130–1595 cal. yr BP)

The start of this phase coincides with the climate cooling that occurred during the Neoglacial period (at approximately 3000 cal. yr BP). At the study site, the climatic change caused a reduction in the rate of peat accumulation (0.014 cm/yr). During this period, the brown moss S. exannulatum/W. fluitans was again replaced by herbaceous species, while L. laricina was replaced by a forest dominated by P. mariana. This succession indicates that the conditions became less humid and coincided with a decrease in the water table. Cooler temperatures and a shorter growing season may have also caused this plant succession. This finding is also suggested by Langlais (2016) for the period between 3500 and 2210 cal. yr BP, during which time a wide variety of trees and shrubs populated the palsa peatland. Despite the fact that some charcoal fragments were found in zone M-Va, fires appear to have been less common during this period. The results of the LOI testing revealed that the Bear peatland was significantly affected by the presence of sand (between 96 and 90 cm), which was likely due to the erosion of the neighbouring hillslope.

Minerotrophic peatland and frequent fires phase (1595–115 cal. yr BP)

In the Lac Wiyâshâkimî region, conditions became cold and dry (with less snow) between 1700 and 1100 cal. yr BP (Payette, 1988). This climate was likely responsible for the progressive decline of spruce trees and it corresponded to the significant decrease in P. mariana macroremains in the end of the M-Va zone (before 1595 cal. yr BP). The abundance of charcoal found in the M-Vb zone indicates the occurrence of several fire events, which supports the hypothesis that conditions were drier. The rate of accumulation remained constant (0.014 cm/yr), but there was very little (almost no) addition of sand during this period. However, even though there are only a few macrofossil remains, the presence of C. canescens in the peatland suggests that local conditions remained humid during this period (Marie-Victorin, 1995).

Minerotrophic to sphagnum peatland phase (155 cal. yr BP–present)

In the present day, the site is still very humid. Sphagnum flexuosum, a species that forms a mat on the surface of poor-to-moderate rich fens (Faubert, 2013), dominated this site, while the minerotrophic species C. rostrata and two ombrotrophic species (E. russeolum and C. calyculata) were also identified.

Absence of permafrost

The study site is located in the discontinuous permafrost zone. The specific combination of climate factors (e.g. low precipitation and thin snow cover, annual temperature below 0°C) makes this zone conducive to the development of palsa peatlands and palsa plateaus (Bhiry et al., 2007; Kuhry, 2008; Sollid and Sørbel, 1998; Thibault and Payette, 2009; Zoltai, 1993, 1995; Zoltai and Tarnocai, 1971). Although it is possible that the studied peatland had been affected by permafrost at some point during its development, the macrofossil analysis did not reveal any signs of a pre-existing palsa at this site.

When a palsa forms, the gradual uplifting of the peat surface that is triggered by the ice lens formation in the peat causes the surface to become drier, which leads to the establishment of species that thrive in well-drained environments such as lichen, Empetrum nigrum and Betula glandulosa (Bhiry et al., 2007; Kuhry, 2008; Langlais, 2016; Seppälä, 2011; Sollid and Sørbel, 1998; Thibault and Payette, 2009; Zoltai, 1993, 1995). However, the macroremains identified at the study site are dominated by hygrophilous taxa and species indicating moist conditions, which means that the peatland has remained humid since the formation of the minerotrophic peatland (4905 cal.yr BP).

Several conditions are necessary to form a palsa and numerous factors could explain the absence of permafrost at the study site. The thickness of the snow cover is frequently cited as one of the most important factors. The insulating effect of a thick cover of snow reduces the penetration of cold into the peat and thus prevents the establishment of permafrost (Gurney, 2001; Nicholson, 1979; Pissart, 2013; Seppälä, 1988, 2003; Zoltai, 1993). It is thus possible that the Bear peatland was affected by a great accumulation of snow due to the topography and orientation of the site. After melting, a significant amount of water could also slow the development of permafrost by causing ice lenses to melt (Seppälä, 2011). However, the peatland appears to have received a significant amount of water in addition to mineral sediments.

The thermal properties of the peat and the type of mineral substratum underlying it are additional factors that affect the formation of palsas (Bhiry and Robert, 2006; Seppälä, 2011). The insulating property of sphagnum peat on the surface of a subarctic peatland is essential to the preservation of ice cores during the warm season (Bhiry and Robert, 2006; Kujala et al., 2008; Zoltai, 1993, 1995). However, the fact that sphagnum was only present at the Bear peatland for a relatively short time (about 115 years) could also explain the absence of permafrost. Finally, according to Allard and Rousseau (1999), silt and clay sediments provide greater support for the formation of segregated ice than is provided by the sandy sediments found in the studied peatland.

An archive of slope processes

During most of the post-glacial period, the peatland reveals the presence of vegetation. A few species of plants were established soon after deglaciation and a significant number of trees were also present. Therefore, the regolith had not been significantly exposed and runoff under ‘normal’ conditions cannot be the source of the minerogenic material found in such episodic phases in the Bear peatland. Results from LOI measurements suggest that minerogenic input was dominant until c. 4500 cal. yr. BP (Figure 5: Synthesis of results and significance regarding slope processes). Organic material developed after that period. This initial appearance of coarse sand and gravel may be attributed to the reworking of glacial material that was left in the area by post-glacial processes, including heavy flows that were a part of the paraglacial processes (Ballantyne, 2002). Today, Caribou slope is the only active talus slope in the catchment. Its development is the result of multiple originating processes such as rockfalls, avalanches and debris flows (Decaulne et al., 2018). The levees of the debris flow have the longest runout distance in the direction of the peatland. This geomorphological process must have occurred between 4000 and 3000 cal. yr. BP, as it requires a large supply of rock debris. Furthermore, the debris flow landforms are visible at the foot of the talus, but they cannot be tracked to the precise source area due to the filling-in by subsequent rockfall and debris redistribution. Since 3000 cal. yr. BP, there has been only a small amount of sediment inflow into the Bear peatland, indicating that no major event occurred on the slope (at least none with such a long runout distance). By comparing activities on the slope through the inflow of minerogenic material into the peat area, combined with the occurrence of fires as indicated by the presence of charcoal within the core extracted from the peatland, it appears that fire periods do not favour the inflow of minerogenic material into the peatland. Slope processes (rockfall, debris flow, etc.) are thus not controlled primarily by climatic conditions, since runoff was not recorded as being higher after fire periods. Instead, they are controlled by meteorological conditions in the form of extreme precipitation events and high-volume spring melt following high snowfall amounts.

Figure 5.

Synthesis diagram comparing the contribution from mineral erosion due to slope processes to the development of the Bear peatland and vegetation cover within the catchment.

Conclusion

Following site emersion, organic matter began to accumulate at approximately 4900 cal. yr BP. The percentage of mineral sediments (sand) remained high between 4900 and 4660 cal. cal. yr BP, which may be linked to paraglacial activity on Caribou slope, predominantly in the form of rockfalls. As evidenced by the continuous presence of hygrophilous taxa, the study site was dominated by minerotrophic conditions right up to the present day. Unlike the subarctic permafrost peatlands in northern Québec, there was no permafrost establishment at the site. Since the peatland is located 400 m downstream of a debris-covered talus slope, we attribute this unusual peatland succession to its geomorphic setting. The inflow of a greater amount of water and more sand into the peatland at 4660, 3905 and 3130 cal. yr BP may be linked to the slope movements. Decaulne et al. (2018) conclude that the formation of Caribou slope was predominantly linked to paraglacial rockfall activity following deglaciation. The presence of displaced parts of the rockfall on the slope indicates that events of greater magnitude than scree formation occurred in the past. Our palaeoecological investigation identified at least three significant phases associated to slope movements at 4660, 3905 and 3130 cal. yr BP.

Footnotes

Acknowledgements

Thanks are extended to L. Qungisiruk, C. Nowra, J. Lebrun and S. Veilleux for field assistance, to S. Auger for laboratory assistance and to L. Burns, A. B. Beaudoin (reviewer) and the anonymous reviewer for their insightful comments on the manuscript. We are also grateful to Annie Novalinga and Véronique Nadeau from Tursujuq National Park and Denis Sarrazin, the Centre d’études nordiques and to the community of Umiujaq for their support and fieldwork assistance.

Funding

Funding was provided by NSERC, Hudsonie21 (CEN), and by LabEx DRIIHM and OHMi NUNAVIK (TUKISIG) through the MOVE grant.

ORCID iD

Najat Bhiry

References

Allard

Rousseau

(1999) The international structure of a pala and a peat plateau in the Rivière Boniface region, Québec: Interferences on the formation of ice segregation mounds. Géographie physique et Quaternaire 53: 373–387.

Allard

Séguin

(1987) Le pergélisol au Québec nordique: bilan et perspectives. Géographie physique et Quaternaire 41: 141–152.

Allard

Séguin

(1985) La déglaciation d’une partie du versant hudsonien québécois: bassins des rivières Nastapoca, Sheldrake et à l’Eau Claire. Géographie physique et Quaternaire 39(1): 13–24.

Arlen-Pouliot

Bhiry

(2005) Palaeoecology of a palsa and a filled thermokarst pond in a permafrost peatland, subarctic Québec, Canada. The Holocene 15: 408–419.

Ballantyne

(2002) Paraglacial geomorphology. Quaternary Science Reviews 21: 1935–2017.

Bégin

Filion

(1985) Analyse dendrochronologique d’un glissement de terrain de la région du Lac à l’Eau Claire (Québec nordique). Canadian Journal of Earth Science 22: 175–182.

Bégin

(1986) Dynamique de la végétation riveraine du lac à l’Eau-Claire, Québec subarctique. PhD Thesis, Department of Geography, Université Laval, Quebec City.

Belyea

(2009) Non-linear dynamics of peatlands and potential feedbacks on the climate system. In: Baird

Belyea

Comas

, et al. (eds) Northern Peatlands and Carbon Cycling (American Geophysical Union Monograph Series). Washington, DC: American Geophysical Union, pp: 5–18.

Bhiry

Filion

(2001) Analyse des macrorestes végétaux. In: Payette

Rochefort

(eds) Écologie des tourbières du Québec-Labrador. Quebec City: Presses de l’Université Laval, pp. 259–272.

10.

Bhiry

Lavoie

(2018) Timeline of permafrost peatland formation in subarctic Quebec (Canada). In: Deline

Bodin

Ravanel

, et al. (eds) 5th European Conference on Permafrost Book of Abstracts, Chamonix, France, 23 June–1 July 2018, pp. 687–688. Le Bourget-du-Lac: Laboratoire EDYTEM.

11.

Bhiry

Robert

(2006) Reconstruction of changes in vegetation and trophic conditions of a palsa in a permafrost peatland, subarctic Québec, Canada. Écoscience 13: 56–65.

12.

Bhiry

Payette

Robert

(2007) Peatland development at the arctic tree line (Québec, Canada) influenced by flooding and permafrost. Quaternary Research 67: 426–437.

13.

Bostock

(1969) The Clearwater complex, New Quebec. Geological Survey Canadian Bulletin, 178: 63.

14.

Couillard

Payette

(1985) Évolution holocène d’une tourbière à pergélisol (Québec nordique). Canadian Journal of Botany 63: 1104–1121.

15.

Crum

Anderson

(1981) Mosses of Eastern North America. New York: Columbia University Press.

16.

Decaulne

Bhiry

Lebrun

, et al. (2018) Geomorphic evidence of slope dynamics on the Canadian Shield throughout the Holocene–A study from Lac à l’Eau-Claire, western Nunavik. Ecoscience 25: 343–357.

17.

Faubert

(2013) Flore des bryophytes du Québec-Labrador (Volume 2: Mousses, Première Partie). Saint-Valérien: Société Québécoise de Bryologie, p. 402. Available at: http://www.societequebecoisedebryologie.org/documents/Flyer_FloreVol2_130429.pdf.

18.

Fillion

MÈ

Bhiry

Touazi

(2014) Differential development of two palsa fields in a peatland located near Whapmagoostui-Kuujjuarapik, Northern Québec, Canada. Arctic, Antarctic, and Alpine Research 46(1): 40–54.

19.

Garneau

(2001) Les taxons préférentiels des tourbières (Annexe 1). In: Payette

Rochefort

(eds) Écologie des tourbières du Québec-Labrador. Québec City: Les Presses de l’Université Laval, pp. 523–532.

20.

Gurney

(2001) Aspects of the genesis, geomorphology and terminology of palsas: perennial cryogenic mounds. Progress in Physical Geography 25: 249–260.

21.

Heim

(1976) Étude palynologique d’une palse de la région du golfe de Richmond (Nouveau-Québec, Canada). Cahiers de Géographie de Québec 20: 221–238.

22.

Heiri

Lotter

Lemcke

(2001) Loss on ignition as a method for estimating organic and carbonate content in sediments: Reproducibility and comparability of results. Journal of Paleolimnology 25: 101–110.

23.

Kuhry

(2008) Palsa and peat plateau development in the Hudson Bay Lowlands, Canada: Timing, pathways and causes. Boreas 37: 316–327.

24.

Kujala

Seppälä

Holappa

(2008) Physical properties of peat and palsa formation. Cold Regions Science and Technology 52: 408–414.

25.

Langlais

(2016)Dynamique holocène d’une tourbière à palses de la région du lac à l’Eau-Claire (Nunavik). MSc Thesis, Department of Geography, Université Laval, Québec City.

26.

Lavoie

Payette

(1995) Analyse macrofossile d’une palse subarctique (Québec nordique). Canadian Journal of Botany 73: 527–537.

27.

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.

28.

Marie-Victorin

(1995) Flore Laurentienne. 3rd Edition. Montréal: Les Presses de l’Université de Montréal, p. 1093.

29.

Marion

Filion

Hétu

(1995) The Holocene development of a debris slope in subarctic Québec, Canada. The Holocene 5(4): 409–419.

30.

Miousse

Bhiry

Lavoie

(2003) Isolation and water-level fluctuations of lake Kachishayoot, Northern Québec, Canada. Quaternary Research 60: 149–161.

31.

Montgomery

(1977) Seeds and Fruits of Plants of Eastern Canada and Northeastern United States. Toronto: University of Toronto Press.

32.

Nicholson

(1979) Permafrost spatial and temporal variations near Schefferville, Nouveau-Québec. Géographie physique et Quaternaire 33: 265–277.

33.

Payette

(1983) The forest tundra and present tree-line of the northern Québec-Labrador Peninsula. Nordicana 47: 3–23.

34.

Payette

(1984) Peat inception and climatic change in Northern Quebec. In Marner

Karlèn

(eds) Climatic Change on a Yearly to Millennial Basis. Boston, MA: Reidel, pp. 173–179.

35.

Payette

(1988) Late-Holocene development of subarctic ombrotrophic peatlands: Allogenic and autogenic succession. Ecology 69: 516–531.

36.

Payette

(2009) Rise and fall of palsas and peat plateaus in eastern Canada. In: Fronzek

Johansson

Christensen

, et al. (eds) Climate Change Impacts on Sub-arctic Palsa Mires and Greenhouse Gas Feedbacks. Helsinki: Finnish Environment Institute 3, pp. 56–59.

37.

Payette

(2013) Flore nordique du Québec et du Labrador. Tome 1. Quebec City: Presses de l’Université Laval, p. 550.

38.

Payette

(2015) Flore nordique du Québec et du Labrador. Tome 1. Quebec City: Presses de l’Université Laval, p. 710.

39.

Payette

Fillion

(1993) Holocene water-level fluctuations of a subarctic lake at the tree line in northern Québec. Boreas 22: 7–14.

40.

Pissart

(2013) Palsas and lithalsas. In: Shroder

Giardino

Harbor

(eds) Treatise on Geomorphology, vol. 8. San Diego, CA: Academic Press, pp. 223–237.

41.

Porsild

Cody

(1980) Vascular plants of continental northwest territories, Canada. Ottawa: National Museum of Sciences, National Museum of Canada, p. 667.

42.

Robinson

Moore

(2000) The influence of permafrost and fire upon carbon accumulation in high boreal peatlands, Northwest Territories, Canada. Arctic, Antarctic and Alpine Research 32: 155–166.

43.

Rousseau

(1974) Géographie floristique du Québec/Labrador. Québec: Presses de l’Université Laval, p. 789.

44.

Schmieder

Schwarz

Trieloff

, et al. (2015) New 40Ar/39Ar dating of the Clearwater Lake impact structures (Québec, Canada) – not the binary asteroid impact it seems? Geochim Cosmochim Acta 148: 304–324.

45.

Seppälä

(1988) Palsas and related forms. In: Clark

(ed.) Advances in Periglacial Geomorphology. Chichester: John Wiley, pp. 247–278.

46.

Seppälä

(2003) Surface abrasion of palsas by wind action in Finnish Lapland. Geomorphology 52: 141–148.

47.

Seppälä

(2011) Synthesis of studies of palsa formation underlining the importance of local environmental and physical characteristics. Quaternary Research 75: 366–370.

48.

Sollid

Sørbel

(1998) Palsa bogs as a climate indicator: Examples from Dovrefjell, southern Norway. Ambio 27: 287–291.

49.

Stuiver

Reimer

(2014) CALIB Rev 7.1. Available at: http://calib.org.

50.

Thibault

Payette

(2009) Recent permafrost degradation in bogs of the James Bay area, northern Quebec, Canada. Permafrost and Periglacial Processes 20: 383–389.

51.

Tremblay

Bhiry

Lavoie

(2014) Long-term dynamics of a palsa in the sporadic permafrost zone of northwestern Quebec (Canada). Canadian Journal of Earth Sciences 51: 500–509.

52.

Waddington

Morris

Kettridge

, et al. (2015) Hydrological feedbacks in northern peatlands. Ecohydrology 8: 113–127.

53.

Zoltai

(1993) Cyclic development of permafrost in the peatlands of northwestern Alberta, Canada. Arctic and Alpine Research 25: 240–246.

54.

Zoltai

(1995) Permafrost distribution in peatlands of west-central Canada during the Holocene warm period 6000 years BP. Géographie physique et Quaternaire 49: 45–54.

55.

Zoltai

Tarnocai

(1971) Properties of a wooded palsa in northern Manitoba. Arctic and Alpine Research 3: 115–129.