Asselin Payette Ecoscience 06

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

13 (2): 135-142 (2006)

The forest-tundra biome is a large ecotone between


the boreal forest to the south and the treeless tundra to the
north (Payette, 1983). The basic landscape unit of the for-
est-tundra is a typical vegetation toposequence including a
hilltop covered with shrubs and lichens, mesic lichen wood-
lands midslope, and wet sprucemoss stands (sometimes
peatlands) downslope (Payette, Fortin & Gamache, 2001).
The contact between a forested valley and a tundra-cov-
ered hilltop is called a subarctic tree line (Payette, Fortin &
Gamache, 2001). In northern Qubec, the landscape open-
ing that led to the creation of the forest-tundra can be attrib-
uted to post-fire black spruce (Picea mariana) regeneration
failure under the cool climate of the last 3000 y (Payette
& Gagnon, 1985; Asselin & Payette, 2005a). The sum of
growing degree-days > 5 C in the northern forest-tundra is
usually insufficient for black spruce to produce viable seeds
(Sirois, 2000), and populations are thus maintained by layer-
ing (Laberge, Payette & Pitre, 2001). Hence, black spruce is
presently in disequilibrium with climate in northern Qubec
(Davis, 1986; Arseneault & Payette, 1997).
Temperatures in northern Qubec have been rising
since the end of the Little Ice Age (Payette et al., 1985) and
are expected to continue to rise because of the increased
greenhouse effect caused by anthropogenic overuse of fossil
fuels (IPCC, 2001). Under climate warming, it is assumed
that black spruce krummholz (more or less stunted individu-
als) that are still present on slopes and in valley bottoms of
the forest-tundra will develop tree growth forms (single,
erect stems taller than 2.5 m). This would increase viable
seed production as seeds are born on upper branches (Bgin
& Filion, 1999). Subarctic tree lines would eventually
Origin and long-term dynamics of a
subarctic tree line
1
Hugo ASSELIN
2
& Serge PAYETTE, NSERC Northern Research Chair, Centre
dtudes nordiques, Universit Laval, Qubec, Qubec G1K 7P4, Canada.
Abstract: The basic unit of the forest-tundra landscape is a toposequence extending from a wet, forested valley to a xeric,
deforested hilltop; the contact zone between these two environments being called a subarctic tree line. Dendrochronological
analysis of living, dead, and subfossil black spruce, and radiocarbon dating of peat samples were used to reconstruct the dynamics
of a subarctic tree line since its post-fire origin about 1000 y ago. Fire is not the sole disturbance to have influenced the dynamics
of the toposequence. A regional-scale flooding event ca. 1120 AD killed many black spruce trees, growth of permafrost
during the Little Ice Age, and its subsequent degradation in the 20
th
century, also had major consequences. The climate was
favourable to black spruce growth between ca. 300 and 1100 AD, as evidenced by large growth rings and tree growth forms.
Ring widths then decreased markedly between the 12
th
and 19
th
centuries and trees were replaced by stunted growth forms.
Although climate warming during the 20
th
century resulted in increased ring widths, black spruces have still not produced tree
growth forms, a necessary condition for viable seed production and eventual re-colonization of deforested hilltops.
Keywords: black spruce, climate change, dendrochronology, fire, flooding, permafrost, Picea mariana, subarctic tree line,
toposequence.
Rsum : Lunit de base du paysage de la toundra forestire est une toposquence incluant un bas de pente humide et bois et
un sommet de colline sec et dbois. La zone de contact entre ces deux milieux constitue une limite darbres subarctique. Lanalyse
dendrochronologique dpinettes noires vivantes, mortes et subfossiles et la datation au radiocarbone dchantillons de
tourbe ont t utilises afin de reconstituer la dynamique dune limite darbres subarctique depuis son origine post-incendie il
y a environ 1000 ans. Le feu nest pas la seule perturbation avoir affect la dynamique de la toposquence. Une inondation
dampleur rgionale survenue vers 1120 aprs J.-C. a caus la mort de plusieurs pinettes noires. La formation du perglisol
durant le Petit ge Glaciaire et sa dgradation subsquente au XX
e
sicle ont galement eu des consquences majeures. Le
climat a t favorable la croissance de lpinette noire entre environ 300 et 1100 aprs J.-C., tel que dmontr par les cernes
de croissance larges et les formes de croissance riges. La largeur des cernes de croissance a ensuite diminu de faon
apprciable entre les XII
e
et XIX
e
sicles et les arbres ont t remplacs par des formes de croissance prostres. Mme si le
rchauffement climatique du XX
e
sicle a rsult en une augmentation de la largeur des cernes de croissance, les pinettes
noires nont toujours pas produit de forme de croissance rige, une condition sine qua non la production de graines viables
et la re-colonisation ventuelle des sommets de collines dboiss.
Mots-cls : changement climatique, dendrochronologie, pinette noire, feu, inondation, limite darbres subarctique, perglisol,
Picea mariana, toposquence.
Nomenclature: Crum & Anderson, 1981; Marie-Victorin, 1995; Brodo, Sharnoff & Sharnoff, 2001.
Introduction
1
Rec. 2004-12-09; acc. 2005-10-04.

Guest Editor: Yves Bergeron.
2
Author for correspondence. Present address: Chaire industrielle CRSNG-UQAT- resent address: Chaire industrielle CRSNG-UQAT-
UQAM en amnagement forestier durable, Universit du Qubec en Abitibi-
Tmiscamingue, 445 boulevard de lUniversit, Rouyn-Noranda, Qubec J9X 5E4,
Canada, e-mail: hugo.asselin@uqat.ca , e-mail: hugo.asselin@uqat.ca e-mail: hugo.asselin@uqat.ca ca
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
ASSELIN & PAYETTE: ORIGIN AND DYNAMICS OF A SUBARCTIC TREE LINE
136 136
move upslope (Gamache & Payette, 2005) and re-colonize
hilltops that shifted from forest to tundra over the last cen-
turies (Arseneault & Payette, 1992). However, black spruce
growth at the tree line and the ability of the species to colo-
nize new habitats are not only influenced by the regional
climate, but also by local factors like permafrost growth
and decay (Camill & Clark, 1998; Osterkamp et al., 2000),
snow depth (Vaganov et al., 1999), wind exposure (Daly &
Shankman, 1985; Scott, Hansell & Erickson, 1993), cata-
strophic floods (Payette & Delwaide, 2004), and seedbed
suitability (Cowles, 1982; Houle & Filion, 2003). Although
subarctic tree lines and their associated toposequences form
the basic unit of the forest-tundra landscape (Payette, Fortin
& Gamache, 2001), little is known about their dynamics
from their post-fire origin to present times.
We used tree-ring analysis and radiocarbon dating of
charcoal and peat samples to reconstruct the dynamics of
a typical forest-tundra toposequence (peatlandmesic slope
with black sprucedeforested hilltop) that developed follow-
ing fire, ca. 1000 y ago. Tree-ring analysis of living, dead,
and subfossil black spruce sampled in the peatland and on
adjacent slopes was used to decipher the respective roles of
and interactions between different climatic and environmen-
tal variables in the secular dynamics of the subarctic tree
line. Emphasis was also put on the links between the differ-
ent constitutive zones of the toposequence.
Methods
STUDY SITE
The study site is located in the Rivire Boniface area of
northern Qubec (57 45' N, 76 20' W), about 30 km east
of Hudson Bay and 10 km south of the latitudinal tree line.
The nearest weather station is located in Inukjuak (58 28' N,
78 04' W), about 125 km northwest of the study area. The
mean annual temperature is -7.0 C, and mean annual
precipitation is ca. 460 mm, of which 40% falls as snow
(Environment Canada, 2004).
The study site consists of three low, elongated hills sur-
rounding a permafrost peatland (Figure 1). Two different
zones can be distinguished in the peatland: an ombrotrophic
zone in the centre, and a minerotrophic zone at the periphery
(Figure 1). The ombrotrophic zone (a thawing palsa pla-
teau) shows a chaotic pattern of hummocks and hollows
with Dicranum and Sphagnum species, ericaceous shrubs,
and stunted black spruce. The minerotrophic zone (a poor
fen) is characterized by the predominance of Eriophorum
and Carex species. Black spruce stands covering the hill-
tops were eliminated by fire between 840 and 1080 AD
(S. Payette, in prep.; Table I). Black spruce is absent from
the hilltops, the species being relegated to the mesic slopes
(high krummholz) and to the hummocks zone of the peat-
land (low krummholz). Only two individuals could be quali-
fied as trees (i.e., single, erect stem taller than 2.5 m) in the
entire study site (ca. 24 ha), and they were growing close
to each other in a well-protected microsite. In summary,
the studied toposequence is composed of the following ele-
ments (all mapped in the field using an infrared theodolite
[Leica T1010, 0.0005-m precision, Leica Geosystem AG,
St. Gallen, Switzerland]): (1) a hummocks zone, (2) a poor
fen, (3) mesic slopes with high spruce krummholz, and
(4) hilltops covered with ericaceous shrubs, dwarf birch
(Betula glandulosa), and lichens (mostly Cladina spp.)
(Figures 1 and 2).
SAMPLING
A 60-m trench was dug across the toposequence in
order to describe soil profiles and to take peat samples for
radiocarbon dating (Figures 1 and 2). Additional peat sam-
ples for radiocarbon dating were taken along two other tran-
sects (Figure 1). Living, dead, and subfossil black spruces
were sampled from the different zones of the toposequence.
Disks were sampled from six living and seven dead black
spruces in the hummocks zone of the peatland and from 10
living and 23 dead individuals on mesic slopes. Many dead
black spruces were found in the northwestern portion of the
fen. A total of 40 disks were sampled from 20 individuals of
the northwestern fen. Subfossil black spruces were sampled
within a 700-m
2
portion of the southeastern fen adjacent
to the soil trench (47 samples) and from the bottom of a
small pond north of hill 101 that was drained with a water
pump prior to sampling (54 samples). Growth form (tree or
krummholz) was described for each sampled black spruce.
FIGURE 1. Schematic view of the study site showing the different zones
of the toposequence. Living, dead, and subfossil black spruce were sampled
in the southwestern pond (SW pond), in the southeastern and northwestern
parts of the fen (SE fen and NW fen), and throughout the hummocks zone
and the mesic slopes. Also shown are radiocarbon dates obtained from
basal/upper sedge peat samples (Table I) and the position of the soil trench
(A-B; Figure 2). The inset shows the location of the Rivire Boniface area
in northern Qubec.
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
RADIOCARBON DATING
Charcoal fragments were sampled below the lichen car-
pet on the three hilltops and radiocarbon-dated in order to
determine the time of fire occurrence (S. Payette, in prep.).
Peat inception dates from different points in the peatland
were inferred by radiocarbon-dating basal sedge peat,
enabling us to reconstruct the lateral expansion of the peat-
land. Dates of the permafrost aggradation that caused the
rising of the palsa plateau were based on radiocarbon-dating
of the uppermost hygrophilous sedge peat, at the contact
with Sphagnum peat (Seppl, 1988; Allard & Sguin,
1987a,b; Vasilchuk & Vasilchuk, 1998; Payette, 2001a).
All radiocarbon dates were calibrated using version 4.4 of
the CALIB program (Stuiver & Reimer, 1993) (Table I).
To facilitate comparisons with tree-ring dates, radiocarbon
dates are given in years BC/AD.
DENDROCHRONOLOGICAL ANALYSIS
All black spruce wood samples were air-dried and sand-
ed for growth ring dating and measurement. Cross-dating of
the samples was done at 40X under a dissecting microscope,
using diagnostic light rings, i.e., growth rings almost exclu-
sively composed of earlywood cells, with very few layers of
latewood cells (Filion et al., 1986). Light rings are caused
by low summer temperatures induced by delayed springs,
cool summers, or early-ending autumns (Yamaguchi, Filion
& Savage, 1993; Wang, Payette & Bgin, 2000). Two light-
ring chronologies were used: a solid chronology covering
690 AD to present and a floating chronology, radiocarbon
dated between 178 BC and 785 AD (Lavoie & Payette,
1997; Arseneault & Payette, 1997; 1998). Samples from
the present study allowed us to fill the gap between the
chronologies, yielding a 2184-y chronology of black spruce
growth at tree line (Asselin et al., in prep.). Tree ring widths
were measured at 40X under a dissecting microscope, using
a Velmex micrometer ( 0.002 mm) linked to a computer
for data recording.
It is commonplace in dendrochronological analysis
to first remove age-related biological growth trends by
detrending (or standardization) before pooling the series
in a common chronology (Schweingruber, 1996). This
ensures that the remaining variation can be mostly attrib-
uted to climate. However, it has been shown that detrending
techniques remove most of the low-frequency (long-term)
variation when mean segment length is short relative to
total chronology length (Cook et al., 1995). The regional
curve standardization (RCS) technique has been suggested
to preserve low-frequency variability (Briffa et al., 1992;
Esper, Cook & Schweingruber, 2002). This method works
by removing the same (regional) growth trend from all
series, instead of computing a different growth trend for
each series. However, the RCS technique requires a large
COSCIENCE, VOL. 13 (2), 2006
137
TABLE I. List of radiocarbon dates used in this study. Dates from soil charcoal fragments sampled below the lichen carpet on the three hill-
tops (hill identifcation numbers as in Figure 1) give a maximum age of fre occurrence. Dates from basal and upper sedge peat samples
(order of dates as in Figure 1 starting from upper left corner and going clockwise) give an estimation of the time of peat inception and palsa
formation, respectively.
Site Lab. No. Dated material Measured Conventional date
2
Median probability
3
Date in
radiocarbon age
1
(y BP 1) (cal. y BP) years BC/AD
HILLS
101 Beta-35507 charcoal na 970 60 870 1080 AD
136 Beta-35515 charcoal na 1130 50 1040 910 AD
140 Beta-35519 charcoal na 1110 50 1020 930 AD
140 UL-2113 wood 1170 60 1190 70 1110 840 AD
PEATLAND
UL-2338 basal peat 250 60 220 80 210 1740 AD
UL-2337 basal peat 4490 100 4460 110 5100 3150 BC
UL-2306 upper peat 360 60 330 80 380 1570 AD
UL-2318 basal peat 2050 70 2020 90 1980 30 BC
UL-2299 basal peat 2310 60 2280 80 2260 310 BC
UL-2305 basal peat 900 90 870 100 800 1150 AD
UL-2322 basal peat 1660 70 1630 90 1520 430 AD
UL-2319 basal peat 2370 70 2340 90 2380 430 BC
UL-2301 basal peat 2140 90 2110 100 2090 140 BC
UL-2307 upper peat 640 60 610 80 600 1350 AD
UL-2303 basal peat 3600 70 3570 90 3860 1910 BC
UL-2298 basal peat 980 60 950 80 850 1100 AD
UL-2281 basal peat 1220 90 1190 100 1110 840 AD
UL-2292 basal peat 2030 70 2000 90 1950 0 AD
UL-2293 upper peat 710 60 680 80 630 1320 AD
UL-2285 basal peat 5510 100 5480 110 6250 4300 BC
UL-2287 upper peat 650 60 620 80 600 1350 AD
UL-2288 basal peat 5200 100 5170 110 5920 3970 BC
UL-2282 basal peat 3550 100 3520 110 3800 1850 BC
UL-2289 basal peat 4260 100 4230 110 4740 2790 BC
1
Measured radiocarbon age is given only for Laboratoire de datation au
14
C de lUniversit Laval (UL) which does not normalize dates to account for isotopic
fractionation.
2
Dates normalized for isotopic fractionation. Values of 27.0 3 and 24.0 2 were used to normalize UL dates obtained from peat and from charred
subfossil wood respectively (following Stuiver & Pollach, 1977). Dates from Beta Analytics (Beta) are conventional dates as provided by the laboratory.
3
Calibrated median probability dates were obtained using CALIB 4.4 (Stuiver & Reimer, 1993) with the IntCal98 calibration dataset of Stuiver et al. (1998).
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
ASSELIN & PAYETTE: ORIGIN AND DYNAMICS OF A SUBARCTIC TREE LINE
138
sample set to be efficient (Briffa et al., 1992). Furthermore,
krummholz do not show clear age-related growth trends and
are sometimes standardized using a horizontal line (slope
= 0). This is only recommendable when short time periods
are studied or when all samples have grown concurrently.
For these reasons, the chronologies discussed in the present
study are presented as raw ring widths.
Results
DYNAMICS OF THE TOPOSEQUENCE
The radiocarbon dates obtained from soil charcoal frag-
ments sampled below the lichen carpet on the three hilltops
yielded ages of 910, 930, and 1080 AD for hilltops 136,
140, and 101, respectively (S. Payette, in prep.; Table I).
A date of 840 AD was also obtained from a charred black
spruce stem sampled in the southeastern portion of the fen,
at the contact with the hilltop (S. Payette, in prep.; Table I).
Characterization and measurement of hilltop soil horizons
showed that the soils were poorly developed podzols,
with 9 5 cm of organic matter, a thin eluviated (Ae)
horizon (3 2 cm), and a lightly coloured illuviated (B)
horizon (15 12 cm) (soil nomenclature following Expert
Committee on Soil Survey, 1987). The bedrock was some-
times exposed or only covered with a thin layer of organic
matter without mineral horizons. According to vegeta-
tion surveys conducted on hilltop 140 (Asselin & Payette,
2005b), it is presently dominated by dwarf birch (Betula
glandulosa), ericaceous shrubs (Vaccinium vitis-idaea and
Ledum decumbens), lichens (mostly Cladina spp.), and
sedges (mostly Carex bigelowii); these taxa accounted for
more than 97% of the vegetation cover on the hilltop.
The mesic slopes marking the transition between the
hilltops and the peatland were characterized by high black
spruce krummholz and had more developed podzols than
those present on the hilltops. The organic, Ae, and B hori-
zons were 7 5, 4 3, and 24 11 cm thick, respectively.
Furthermore, the eluviated horizon was paler, and the illuvi-
ated horizon darker, than in hilltop soils.
According to radiocarbon dating of basal peat samples
from the deepest parts of the peatland, peat started accumu-
lating in 2790 BC in the southwestern pond (at that time a
peatland) and in 4300 and 3150 BC in the southern and north-
ern parts of the peatland, respectively (Figure 1; Table I).
Other basal peat dates indicate that the peatland reached
half its present size by ca. 30 BC and that maximum lateral
expansion occurred between 430 and 1150 AD on the east-
ern side and more recently (1740 AD) on the western side
(Figure 1; Table I). Radiocarbon dating of the uppermost
sedge peat beneath the Sphagnum carpet suggests that the
formation of the palsa plateau (present hummock zone)
occurred between 1320 and 1570 AD (Figure 1; Table I).
DENDROECOLOGICAL ANALYSES
The oldest black spruce individuals preserved in the
peat established shortly before 300 AD in the southeastern
fen and shortly after 400 AD in the southwestern pond area
(Figure 3a,b). Ring width values for the period 2781096
AD are almost always above the long-term average, and
the growth curves obtained for the southeastern fen and
the southwestern pond area show similar patterns for this
period, except from 900 to 1096 AD (Figure 3a,b). The
percentage of samples showing tree growth forms for the
early portion of the record (2781096) is always above
50%, except for five brief episodes: 548573 (no sample),
641693, 833865, 961993, and 10651096 (Figure 3g).
No samples were found for the period 10971133 AD.
Only krummholz lived in the southeastern fen and in the
southwestern pond area between 1134 and 1535 (Figure 3g),
and ring width values for this period are mostly below the
long-term average (Figure 3a,b). From 1536 to present, no
black spruce established in the minerotrophic part of the
peatland or on the palsa that had formed in the southwestern
pond area. However, black spruce krummholz colonized
the newly formed palsa plateau starting around 1525 AD.
Ring widths of these individuals reached the lowest values
on the record in the first half of the 17
th
century (Figure 3c).
Growth remained minimal and below the long-term aver-
age until the 20
th
century, although an increasing trend
is noticeable. Black spruce krummholz also established
at the periphery of the fen (northwestern section) in the
early 17
th
century. Although currently located in the fen,
these spruces were rooted in mineral soil and produced
adventitious roots. Ring width values were low during the
17
th
and 18
th
centuries, higher during the 19
th
century, and
decreased again in the early 20
th
century (Figure 3d). These
spruces were all dead by 1961. The oldest samples from the
mesic slopes (all krummholz) date from 1580. Their growth
was below the long-term average until ca. 1900 and has
increased markedly since then (Figure 3e).
Discussion
HOLOCENE DYNAMICS OF THE TOPOSEQUENCE
According to the radiocarbon dates obtained from
black spruce wood and soil charcoal fragments (S. Payette,
FIGURE 2. Schematic view of the soil trench. Shown are radiocarbon
dates obtained from basal and upper sedge peat samples (same dates as in
Figure 1 and Table I). Vertical enhancement factor = 10.
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
COSCIENCE, VOL. 13 (2), 2006
139
in prep.; Table I), two fires probably occurred in the study
area, one around 840930 AD (hilltops 136 and 140) and
the other around 1080 AD (hilltop 101). Taking into account
the error associated with radiocarbon dating, it is impossible
to determine if a single fire or two different events burned
hilltops 136 and 140. The 840 AD date obtained from a
charred black spruce stem is direct evidence that the fire
that burned hilltop 140 reached the margin of the peatland
in some areas. This is further evidenced by the presence of
a charcoal layer between the organic and Ae soil horizons
found beneath the high krummholz of the mesic slopes.
Therefore, the fire margin was re-colonized to some extent
from surviving individuals (located in portions of the mesic
slopes where no charcoal was found between the organic
and Ae soil horizons).
The Holocene dynamics of the studied peatland closely
follow what is known from other northern Qubec sites
(Payette, 2001a,b). Peat started accumulating in the deepest
parts of the peatland around 4300 BC, i.e., shortly after the
retreat of the postglacial Tyrrell Sea (Lauriol & Gray, 1987).
Similar dates of peat inception were obtained for other
parts of northern Qubec (Payette, 2001b). Accumulation
continued for several hundred years, with peat expanding
both vertically and laterally. By ca. 1150 AD, the peatland
was almost at its present size (a small portion in the western
area was absent). At that time the peatland was a treed fen,
with no evidence of permafrost. Then, between 1320 and
1570 AD, permafrost developed in the peatland, forming a
palsa plateau. The age range for permafrost aggradation in
the studied peatland is comparable to those reported in other
studies conducted in northern Qubec (Payette & Sguin,
1979; Couillard & Payette, 1985; Allard & Sguin, 1987a,b;
Gah, Allard & Sguin, 1987; Payette & Delwaide, 2004)
and elsewhere in the northern hemisphere (Van Vliet-Lano,
Bourgeois & Dauteuil, 1998; Vasilchuk & Vasilchuk,
1998). Starting ca. 1740 AD, wetter conditions favoured
peat expansion on the western side of the peatland. Greater
humidity can be attributed to increased winter precipita-
tion from 1750 AD to present (Payette & Delwaide, 2004).
Climate warming starting in the mid-1800s induced thawing
of the palsa plateau, the four ponds present in the peatland
being direct evidence of recent permafrost degradation. This
phenomenon is widespread in northern Qubec (Allard &
Sguin, 1987b; Allard, Sguin & Lvesque, 1987; Payette,
2001a; Payette et al., 2004) and elsewhere in the northern
hemisphere (Osterkamp et al., 2000; Luoto & Seppl,
2003; Agafonov, Strunk & Nuber, 2004).
DENDROECOLOGICAL RECONSTRUCTION OF THE TOPOSEQUENCE
DYNAMICS
Trees grew in the peatland (at that time a poor fen)
between 278 and 1096 AD (Figure 3a,b). No sample (either
tree or krummholz) was found for the period 548573 AD.
It is not possible to determine if this is due to a mass-mor-
tality event or a sampling artifact, as samples are less abun-
dant in the earliest portion of the chronology. Nevertheless,
climatic conditions were favourable to tree growth between
278 and 1096 AD, as shown by high ring width values both
in the southeastern fen and in the southwestern pond area.
Tree-ring curves for the southeastern fen and the south-
western pond area suddenly diverged between 900 and 1096
AD (Figure 3a,b). This can be explained by the fact that hill
140 burned earlier than hill 101 (Table I) and is a further
confirmation that these were separate fire events. Post-fire
regeneration failure on hilltop 140 caused excess snow to
accumulate in the peatland, as it was no longer trapped by
black spruce on the hilltop. It is possible that this excess
snow took a longer time to melt in the spring, keeping the
soil frozen and thus causing a delayed growing season and
reduced tree growth in the southeastern fen.
No sample was found in the peatland or in the pond for
the period 10971133 AD. We hypothesize that this may be
attributed to a mass-mortality event caused by a regional
flooding episode that is known to have occurred between
1120 and 1155 AD in other sites of the Rivire Boniface area
FIGURE 3. Black spruce growth curves (yearly values and 49-y low
pass filter to show long-term trends) for the southwestern pond (a), the
southeastern fen (b), the hummocks zone (c), the northwestern fen (d),
and the mesic slopes (e). The long-term average of ring width values for
all chronologies combined (0.237 mm) is indicated as a reference for each
chronology (horizontal black line). Also shown is the number of samples
included in each curve (f) and the percentage of samples showing a tree
growth form (the remaining being krummholz) for each year of the record
(g). The shaded areas represent the flooding event (10971133 AD) and the
time of permafrost aggradation (13201570 AD). The arrows indicate the
fires that burned hilltop 140 (930 AD) and 101 (1080 AD).
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
ASSELIN & PAYETTE: ORIGIN AND DYNAMICS OF A SUBARCTIC TREE LINE
140
(Payette & Delwaide, 2004). Black spruce re-colonized the
southeastern fen and the southwestern pond area at the begin-
ning of the 12
th
century (Figure 3a,b), probably meaning that
the effects of the flood were no longer apparent. Krummholz
grew in the southeastern fen and in the southwestern pond
area between 1134 and 1535 AD. However, ring widths for
this period are markedly thinner than during the preceding
period (2781096 AD). Another episode of massive death
(after 1535 AD) can be attributed to the formation of the
palsa plateau at the center of the peatland and a palsa in the
southwestern pond area. According to the peat stratigraphy,
permafrost growth started in the peatland between 1320 and
1570 AD (no date available for the pond area), favoured
by the cold conditions prevailing during the first half of the
Little Ice Age. The palsa plateau reached its full extension
during the first half of the 16
th
century, drastically altering the
drainage. Water accumulated in the surrounding (remaining)
fen, killing black spruce krummholz growing there. Soil up-
thrusting due to permafrost aggradation probably caused up-
rooting of some individuals, a phenomenon likely to explain
most of the mortality in the southwestern pond, where no
spruce was found in growth position.
Soon after its formation, the palsa plateau was dry
enough to be colonized by black spruce, which estab-
lished in the 16
th
century and still grow in the present day
(Figure 3c). Although ring widths have remained below
the long-term average, a slight increasing trend is clearly
visible and may be due in part to increased temperatures
since the end of the Little Ice Age. Nevertheless, ring width
values for the 20
th
century are far from being comparable
to those of the period from 278 to 1096 AD (Figure 3).
The palsa that once occupied the present location of the
southwestern pond was not colonized by black spruce dur-
ing recent centuries. Although it is difficult to pinpoint a
specific explanation for the absence of colonization, it may
be due to submersion of the palsa by rainwater. Before the
formation of the palsa, rainwater coming from hilltop 101
fed the small peatland, and excess water drained to the
river. It is likely that the palsa blocked the normal course
of the water between hilltop 101 and the river, and water
accumulated, rapidly covering the palsa. This hypothesis is
supported by the absence of a Sphagnum peat cover on top
of the sedge peat in the pond stratigraphy, as opposed to the
palsa plateau stratigraphy.
The climate warming of the 20
th
century was accom-
panied by wetter conditions. Increased wetness, which
started ca. 1750 AD, is attributable to increased winter
precipitation (Payette & Delwaide, 2004). Humid condi-
tions favoured the lateral expansion of the western part of
the peatland into the surrounding forest, gradually killing
the spruces growing there and rooted in the mineral soil
(Figure 3d). Interestingly, the long-term growth trends of
spruce of the northwestern fen are the opposite of those of
spruce growing in the hummocks zone (Figure 3c,d). This
can be explained by the differential response of these indi-
viduals to moisture fluctuations: increased moisture was
detrimental to spruce growing in the peatland but beneficial
to those growing at the periphery.
The black spruce growing on mesic slopes responded
most strongly to the 20
th
century climate warming (Figure 3e),
although their ring width values do not yet exceed those
characteristic of the period from 278 to 1096 AD. Moreover,
only two trees (not sampled) are present in the study site
today, while the majority of the samples of the period from
278 to 1096 AD were trees (Figure 3g). As black spruce
krummholz can adopt a tree growth form in a few decades
under warm climate (Payette et al., 1989), this confirms that
conditions at the study site are not yet as favourable as they
were between 278 and 1096 AD.
If climate warming continues, trees may eventually
come back on the hilltops, a phenomenon already appar-
ent in the southern part of the forest-tundra (Gamache &
Payette, 2005). However, as thawing of permafrost will
soon transform the peatland into a treeless poor fen, hilltop
re-colonization will rely on tree-like spruce (high krumm-
holz) presently growing on mesic slopes. These individuals
must first adopt vertical (tree) growth forms and produce
viable seeds (Sirois, 2000) for invasion of hilltops to pro-
ceed. Seeds must then fall on suitable sites, i.e., mudboils
or cracks in the lichen mat where mineral soil is surfacing
(Cowles, 1982; Sirois, 1993; Houle & Filion, 2003).
Acknowledgements
Many thanks to M. Beauchemin, S. Champagne, and S. Valle
for field assistance. Help from A. Delwaide and M.-C. Martel
during lab work was greatly appreciated. This study was finan-
cially supported by the Natural Sciences and Engineering Research
Council of Canada, the Canadian Department of Indian Affairs and
Northern Development, and the Ministre de la Recherche, de la
Science et de la Technologie du Qubec (FCAR program).
Literature cited
Agafonov, L., H. Strunk & T. Nuber, 2004. Thermokarst dynam-
ics in Western Siberia: Insights from dendrochronological
research. Palaeogeography, Palaeoclimatology, Palaeoecology,
209: 183196.
Allard, M. & M. K. Sguin, 1987a. Le perglisol au Qubec
nordique : bilan et perspectives. Gographie physique et
Quaternaire, 41: 141152.
Allard, M. & M. K. Sguin, 1987b. The Holocene evolution of
permafrost near the tree line, on the eastern coast of Hudson
Bay (northern Qubec). Canadian Journal of Earth Sciences,
24: 22062222.
Allard, M., M. K. Sguin & R. Lvesque, 1987. Palsas and mineral
permafrost mounds in northern Qubec. Pages 285309 in V.
Gardiner (ed.). International Geomorphology 1986, Part II.
John Wiley & Sons, London.
Arseneault, D. & S. Payette, 1992. A postfire shift from lichen
spruce to lichentundra vegetation at tree line. Ecology, 73:
10671081.
Arseneault, D. & S. Payette, 1997. Reconstruction of millennial
forest dynamics from tree remains in a subarctic tree line peat-
land. Ecology, 78: 18731883.
Arseneault, D. & S. Payette, 1998. Chronologie des cernes ples
de lpinette noire (Picea mariana [Mill.] BSP.) au Qubec
subarctique : de 706 1675 ap. J.-C. Gographie physique et
Quaternaire, 52: 219226.
Asselin, H. & S. Payette, 2005a. Late Holocene opening of the
forest tundra landscape in northern Qubec, Canada. Global
Ecology and Biogeography, 14: 307313.
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
COSCIENCE, VOL. 13 (2), 2006
141
Asselin, H. & S. Payette, 2005b. Late Holocene deforestation of a
tree line site: Estimation of pre-fire vegetation composition and
black spruce cover using soil charcoal. Ecography, 28: 801-805.
Bgin, C. & L. Filion, 1999. Black spruce (Picea mariana) archi-
tecture. Canadian Journal of Botany, 77: 664672.
Briffa, K. R., P. D. Jones, T. S. Bartholin, D. Eckstein, F. H.
Schweingruber, W. Karln, P. Zetterberg & M. Eronen, 1992.
Fennoscandian summers from AD 500: Temperature changes
on short and long timescales. Climate Dynamics, 7: 111119.
Brodo, I. M., S. D. Sharnoff & S. Sharnoff, 2001. Lichens of North
America. Yale University Press, New Haven, Connecticut.
Camill, P. & J. S. Clark, 1998. Long-term perspectives on lagged
ecosystem responses to climate change: Permafrost in boreal
peatlands and the grassland/woodland boundary. Ecosystems,
3: 534544.
Cook, E. R., K. R. Briffa, D. M. Meko, D. A. Graybill & G.
Funkhouser, 1995. The segment length curse in long tree-ring
chronology development for palaeoclimatic studies. Holocene,
5: 229237.
Couillard, L. & S. Payette, 1985. volution holocne dune tour-
bire perglisol (Qubec nordique). Canadian Journal of
Botany, 63: 11041121.
Cowles, S., 1982. Preliminary results investigating the effect of
lichen ground cover on the growth of black spruce. Naturaliste
Canadien, 109: 573581.
Crum, H. A. & L. E. Anderson, 1981. Mosses of Eastern North
America. Columbia University Press, New York, New York.
Daly, C. & D. Shankman, 1985. Seedling establishment by coni-
fers above tree limit on Niwot Ridge, Front Range, Colorado,
USA. Arctic and Alpine Research, 17: 389400.
Davis, M. B., 1986. Climatic instability, time lags, and community
disequilibrium. Pages 269284 in J. Diamond & T. J. Case (eds.).
Community Ecology. Harper & Row, New York, New York.
Environment Canada, 2004. Canadian Climate Normals or
Averages 19712000. [Online] URL: http://www.climate.
weatheroffice.ec.gc.ca/climate_normals
Esper, J., E. R. Cook & F. H. Schweingruber, 2002. Low-frequen-
cy signals in long tree-ring chronologies for reconstructing past
temperature variability. Science, 295: 22502253.
Expert Committee on Soil Survey, 1987. The Canadian System of
Soil Classification. 2
nd
Edition. Agriculture Canada publication
1646, Ottawa, Ontario.
Filion, L., S. Payette, L. Gauthier & Y. Boutin, 1986. Light rings
in subarctic conifers as a dendrochronological tool. Quaternary
Research, 26: 272279.
Gah, ., M. Allard & M. K. Sguin, 1987. Gophysique et
dynamique holocne de plateaux palsiques Kangiqsualujjuaq,
Qubec nordique. Gographie physique et Quaternaire, 41:
3346.
Gamache, I. & S. Payette, 2005. Latitudinal response of subarctic
tree lines to recent climate change in eastern Canada. Journal of
Biogeography, 32: 849862.
Houle, G. & L. Filion, 2003. The effects of lichens on white spruce
seedling establishment and juvenile growth in a spruce-lichen
woodland of subarctic Qubec. coscience, 10: 8084.
IPCC, 2001. Climate Change 2001: The Scientific Basis.
Contribution of Working Group I to the Third Assessment
Report of the Intergovernmental Panel on Climate Change.
Cambridge University Press, Cambridge.
Laberge, M.-J., S. Payette & N. Pitre, 2001. Development of
stunted black spruce (Picea mariana) clones in the subarctic
environment: A dendroarchitectural analysis. coscience, 8:
489498.
Lauriol, B. & J. T. Gray, 1987. The decay and disappearance of the
Late Wisconsin ice sheet in the Ungava Peninsula, Northern
Qubec, Canada. Arctic and Alpine Research, 19: 109126.
Lavoie, C. & S. Payette, 1997. Late-Holocene light-ring chronolo-
gies from subfossil black spruce stems in mires of subarctic
Qubec. Holocene, 7: 129137.
Luoto, M. & M. Seppl, 2003. Thermokarst ponds as indicators of
the former distribution of palsas in Finnish Lapland. Permafrost
and Periglacial Processes, 14: 1927.
Marie-Victorin, F..C., 1995. Flore laurentienne. 3
e
dition.
Presses de l'Universit de Montral, Montral, Qubec.
Osterkamp, T. E., L. Viereck, Y. Shur, M. T. Jorgenson, C. Racine,
A. Doyle & R. D. Boone, 2000. Observations of thermo-
karst and its impact on boreal forests in Alaska, USA. Arctic,
Antarctic, and Alpine Research, 32: 303315.
Payette, S., 1983. The forest tundra and present tree-lines of the
northern QubecLabrador Peninsula. Nordicana, 47: 323.
Payette, S., 2001a. Les processus et les formes priglaciaires.
Pages 199239 in S. Payette & L. Rochefort (eds.). cologie
des tourbires du QubecLabrador. Presses de lUniversit
Laval, Qubec.
Payette, S., 2001b. La stratigraphie. Pages 345370 in S. Payette
& L. Rochefort (eds.). cologie des tourbires du Qubec
Labrador. Presses de lUniversit Laval, Qubec.
Payette, S. & A. Delwaide, 2004. Dynamics of subarctic wetland
forests over the past 1500 years. Ecological Monographs, 74:
373391.
Payette, S. & R. Gagnon, 1985. Late Holocene deforestation and
tree regeneration in the forest-tundra of Qubec. Nature, 313:
570572.
Payette, S., M.-J. Fortin & I. Gamache, 2001. The subarctic for-
est-tundra: The structure of a biome in a changing climate.
BioScience, 51: 709718.
Payette, S. & M. K. Sguin, 1979. Les buttes minrales cryognes
dans les basses terres de la Rivire-aux-Feuilles, Nouveau-
Qubec. Gographie physique et Quaternaire, 33: 339358.
Payette, S., L. Filion, L. Gauthier & Y. Boutin, 1985. Secular
climate change in old-growth tree-line vegetation of northern
Qubec. Nature, 315: 135138.
Payette, S., L. Filion, A. Delwaide & C. Bgin, 1989.
Reconstruction of tree-line vegetation response to long-term
climate change. Nature, 341: 429432.
Payette, S., A. Delwaide, M. Caccianiga & M. Beauchemin, 2004.
Accelerated thawing of subarctic peatland permafrost over the
last 50 years. Geophysical Research Letters, 31: doi:10.1029/
2004GL020358.
Schweingruber, F. H., 1996. Tree Rings and Environment.
Dendroecology. Swiss Federal Institute for Forest, Snow and
Landscape Research. Paul Haupt, Birmensdorf and Berne.
Scott, P. A., R. I. C. Hansell & W. R. Erickson, 1993. Influences of
wind and snow on northern tree-line environments at Churchill,
Manitoba, Canada. Arctic, 46: 316323.
Seppl, M., 1988. Palsas and related forms. Pages 247278 in M.
J. Clark (ed.). Advances in Periglacial Geomorphology. John
Wiley & Sons, Chichester.
Sirois, L., 1993. Impact of fire on Picea mariana and Pinus
banksiana seedlings in subarctic lichen woodlands. Journal of
Vegetation Science, 4: 795802.
Sirois, L., 2000. Spatiotemporal variation in black spruce cone and
seed crops along a boreal foresttree line transect. Canadian
Journal of Forest Research, 30: 900909.
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.
ASSELIN & PAYETTE: ORIGIN AND DYNAMICS OF A SUBARCTIC TREE LINE
142
Stuiver, M. & H. A. Polach, 1977. Discussion: Reporting of
14
C
data. Radiocarbon, 19: 355363.
Stuiver, M. & P. J. Reimer, 1993. Extended
14
C database and
revised CALIB radiocarbon calibration program. Radiocarbon,
35: 215230.
Stuiver, M., P. J. Reimer, E. Bard, J. W. Beck, G. S. Burr, K. A.
Hughen, B. Kromer, F. G. McCormac, J. van der Plicht & M.
Spurk, 1998. INTCAL98 Radiocarbon age calibration 24,0000
cal BP. Radiocarbon, 40: 10411083.
Vaganov, E. A., M. K. Hughes, A. V. Kirdyanov, F. H. Schweingru-
ber & P. P. Silkin, 1999. Influence of snowfall and melt timing
on tree growth in subarctic Eurasia. Nature, 400: 149151.
Van Vliet-Lano, B., O. Bourgeois & O. Dauteuil, 1998. Thufur
formation in northern Iceland and its relation to Holocene cli-
mate change. Permafrost and Periglacial Processes, 9: 347365.
Vasilchuk, Y. K. & A. C. Vasilchuk, 1998. The
14
C age of palsas
in northern Eurasia. Radiocarbon, 40: 895904.
Wang, L., S. Payette & Y. Bgin, 2000. A quantitative definition of
light rings in black spruce (Picea mariana) at the arctic treeline
in Northern Qubec, Canada. Arctic, Antarctic, and Alpine
Research, 32: 324330.
Yamaguchi, D. K., L. Filion & M. Savage, 1993. Relationship of
temperature and light ring formation at subarctic treeline and
implications for climate reconstruction. Quaternary Research,
39: 256262.
coscience
Droit de visualisation personnel seulement. Ne pas reproduire ou redistribuer de faon lectronique.
For personnal viewing purpose only. Do not copy or electronically redistribute this article.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy