Spatial and Temporal Trends in Carbon Storage of Peatlands of Continental Western Canada through the Holocene

Abstract

Peatlands of continental western Canada (Alberta, Saskatchewan, and Manitoba) cover 365 157 km2 and store 48.0 Pg of carbon representing 2.1% of the world's terrestrial carbon within 0.25% of the global landbase. Only a small amount, 0.10 Pg (0.2%) of this carbon, is currently stored in the above-ground biomass. Carbon storage in peatlands has changed significantly since deglaciation. Peatlands began to accumulate carbon around 9000 years ago in this region, after an initial deglacial lag. Carbon accumulation was climatically limited throughout much of continental western Canada by early Holocene maximum insolation. After 6000 BP, carbon accumulation increased significantly, with about half of current stores being reached by 4000 BP. Around 3000 BP carbon accumulation in continental western Canada began to slow as permafrost developed throughout the subarctic and boreal region and the current southern limit of peatlands was reached. Peatlands in continental western Canada continue to increase their total carbon storage today by 19.4 g m-2 year-1, indicating that regionally this ecosystem remains a large carbon sink.

Full text

Spatial and temporal trends in carbon storage of
peatlands of continental western Canada through
the Holocene
Dale H. Vitt, Linda A. Halsey, Ilka E. Bauer, and Celina Campbell
Abstract: Peatlands of continental western Canada (Alberta, Saskatchewan, and Manitoba) cover 365 157 km2and
store 48.0 Pg of carbon representing 2.1% of the world’s terrestrial carbon within 0.25% of the global landbase. Only a
small amount, 0.10 Pg (0.2%) of this carbon, is currently stored in the above-ground biomass. Carbon storage in
peatlands has changed significantly since deglaciation. Peatlands began to accumulate carbon around 9000 years ago in
this region, after an initial deglacial lag. Carbon accumulation was climatically limited throughout much of continental
western Canada by early Holocene maximum insolation. After 6000 BP, carbon accumulation increased significantly,
with about half of current stores being reached by 4000 BP. Around 3000 BP carbon accumulation in continental west-
ern Canada began to slow as permafrost developed throughout the subarctic and boreal region and the current southern
limit of peatlands was reached. Peatlands in continental western Canada continue to increase their total carbon storage
today by 19.4 g m–2 year–1, indicating that regionally this ecosystem remains a large carbon sink.
Résumé : Les tourbières de l’Ouest continental canadien (Alberta, Saskatchewan et Manitoba) couvrent plus de
365 157 km2et contiennent 48,0 Pg de carbone, ce qui représente 2,1 % du carbone terrestre dans 0,25 % du territoire
terrestre. Seulement une petite quantité, soit 0,10 Pg (0,2 %) de ce carbone, est présentement entreposée dans la bio-
masse de surface. L’entreposage du carbone dans les tourbières a grandement changé depuis la déglaciation. Les tour-
bières ont commencé à accumuler du carbone dans cette région ilyaenviron 9000 ans, après un premier retard de
déglaciation. L’accumulation de carbone a été limitée par le climat à travers une grande partie de l’Ouest continental
canadien par une insolation maximum à l’Holocène précoce. Après 6000 Av. Pr., l’accumulation de carbone a aug-
menté de façon significative; environ la moitié des réserves présentes a été atteinte vers 4000 Av. Pr. L’accumulation de
carbone dans l’Ouest continental canadien a commencé à ralentir vers 3000 Av. Pr. lorsque le pergélisol s’est déve-
loppé dans les régions boréales et subarctiques et que la limite sud actuelle des tourbières a été atteinte. Les tourbières
de l’Ouest continental canadien augmentent continuellement leur entreposage total de carbone de 19,4 g m–2a–1, indi-
quant ainsi que cet écosystème demeure une grande trappe de carbone.
[Traduit par la Rédaction] Vitt et al. 693
Introduction
Peatlands have long been recognized as large sinks for at-
mospheric CO2, removing an estimated 0.076 Pg (1 Pg =
1015 g) of carbon from the atmosphere annually through the
process of peat accumulation (Gorham 1991). Peat accumu-
lates in wetlands where the rate of biomass production is
greater than the rate of decomposition, and it is most abun-
dant in boreal and subarctic regions of the circumpolar north
where cool and moist climatic conditions favour decreased
rates of decomposition (Gore 1983). Anaerobic decomposi-
tion in northern peatlands produces methane (CH4), with
roughly 0.032 Pg being released to the atmosphere annually
(Frolking 1991). Clearly, peatlands are an important compo-
nent of the terrestrial carbon budget, with northern peatland
storage representing about 455 Pg of carbon, of which
6.8 Pg is living biomass (Gorham 1991). Peatland carbon
dynamics influence atmospheric CO2and CH4concentra-
tions and thus, future changes in peatland carbon storage
have the potential to influence greenhouse gas-induced
warming (Post et al. 1992). How this peatland carbon is dis-
tributed across the landscape and, more importantly, how it
has accumulated through time at a regional scale is largely
unknown, with only vague statements having been applied
to long-term accumulation at the landscape level (Moore et
al. 1998).
Not all peatlands are the same, they have different hydro-
logical, chemical, and biotic gradients. Peatlands are either
ombrogenous and receive their surface water and nutrients
solely from precipitation, or they are geogenous and receive
water not only from precipitation but also from surface wa-
ter and groundwater; the former are termed bogs, while the
latter are fens. Swamps, peat accumulators in eastern Can-
ada, generally do not accumulate >40 cm of organic matter
in continental western Canada (cf. Tarnocai et al. 1995) and,
hence, are not included in this paper.
Fens may be acidic and Sphagnum-dominated (poor fens),
or alkaline, basic to neutral, and dominated by “brown
mosses” (rich fens). Bogs are acidic and are dominated by
Can. J. Earth Sci. 37: 683–693 (2000) © 2000 NRC Canada
683
Received November 23, 1998. Accepted September 13, 1999.
D.H. Vitt,1L.A. Halsey, I.E. Bauer, and C. Campbell.
Department of Biological Sciences, University of Alberta,
Edmonton, AB T6G 2E9, Canada.
1Corresponding author (e-mail: dvitt@gpu.srv.ualberta.ca).

some combination of Sphagnum, lichens, and feather mosses
(Belland and Vitt 1995). Unlike domed bogs found in tem-
perate and oceanic areas, continental bogs are relatively flat
across their surfaces (National Wetlands Working Group
1988). Permafrost is an important component of northern
peatlands and is restricted almost exclusively to bogs within
the boreal forest (Vitt et al. 1994). Water table depths are
variable in peatlands, generally in the range of 10 cm above
to 40 cm below the surface for fens (cf. Gignac et al. 1991;
Nicholson et al. 1997), while bogs are drier as a whole, with
water tables 40–60 cm below the surface for nonpermafrost
bogs, and approximately 100 cm below the surface for per-
mafrost bogs (Belland and Vitt 1995). Thus bogs have a rel-
atively thick, upper aerobic zone (acrotelm), while in fens
the acrotelm is shallower; despite the differences, decompo-
sition in both bogs and fens remains less than production.
Peatland distributions have been spatially and temporally
variable throughout the Holocene, developing after an initial
deglacial lag (Halsey et al. 1998). The initiation of peat ac-
cumulation is related to stabilization of seasonal water levels
and restriction of water flow through a wetland (Zoltai and
Vitt 1990) and, in conjunction with leaching of soluble salts
from the mineral substrate, allows the establishment and de-
velopment of a moss layer in hydrologically conducive areas
(Vitt et al. 1993). Thus, climate and geology are important
factors controlling peatland distribution in both space and
time (Halsey et al. 1998). The stabilization of regional water
tables appears to have been an important component in the
successional change from prairie marshes to boreal fens in
the western interior of Canada over the past 10 000 years
(Zoltai and Vitt 1990).
Successional patterns through time have been docu-
mented; bogs can succeed fens (though not all fens become
bogs) with the increased importance of Sphagnum causing a
fundamental change in the functioning of these peatland
ecosystems, leading to acidification and oligotrophication
(Vitt and Kuhry 1992). Permafrost development occurred in
the later part of the Holocene, around 3000 to 4000 BP,
within the subarctic and boreal forest of western Canada,
once Sphagnum accumulation had become significant and
threshold climatic conditions were established (Zoltai 1995).
While net primary production appears to be similar at a
regional scale in all boreal peatland types (Campbell et al.
2000), decomposition rates differ, and result in variable car-
bon accumulation potentials in the acrotelm of a peatland
(Thormann et al. 1999). Permafrost bogs have extremely low
carbon accumulation potentials in the acrotelm at the cen-
tury scale due to the relatively high frequency of fires on dry
peatland surfaces, resulting in near-surface samples with
nonrecent radiocarbon dates (cf. Zoltai 1993). Carbon accu-
mulation potentials at a yearly scale, though probably highly
variable, are unknown for permafrost bogs, while for
nonpermafrost bogs and brown moss peatlands, first year
mass loss estimates range from 14% to 25–61%, respec-
tively, (Thormann et al. 1999). The decomposition of peat in
the deeper anaerobic zone (catotelm) has been approximated
by a simple exponential decay model (Clymo 1984).
This paper examines how carbon is stored in peatlands
within the boreal and subarctic landscape in continental
western Canada. Storage is partitioned into two components:
(1) above-ground storage (including aerial components of
trees, shrubs, and herbs) and (2) surface and below-ground
storage (including living ground layer (nonvascular plants),
below-ground living root biomass (vascular plants), and
dead groundlayer (peat) located in the acrotelm and
catotelm). Carbon storage is examined both spatially and
temporally in 1000 year time slices throughout the Holo-
cene.
Methods
Determination of current and past carbon storage in
peatlands of continental western Canada requires eight types
of data, each derived from different sources:
(1) Inventory of current peatland distribution by peatland
type.
(2) Estimates of current maximum depth distributions.
(3) Calculation of surface and below-ground storage vol-
ume by peatland type using area and maximum depth ad-
justed by basin topography.
(4) Carbon content of organic matter in peat.
(5) Profiles of organic matter density distinguished by
peatland type.
(6) Above-ground carbon content of biomass by peatland
type on a mass–area basis.
(7) Temporal patterns of peatland initiation and expan-
sion.
(8) Long-term catotelm decomposition.
Current peatland distribution
Peatlands and peatland complexes across continental
western Canada were inventoried by type from1:40000to
1 : 60 000 aerial photographs following the classification of
Halsey and Vitt (1997), and the data were transferred to
1 : 250 000 base maps. Peatland types were distinguished on
the basis of hydrology: bog versus fen; permafrost: presence
or absence; patterning: presence or absence; and forest
cover: wooded, shrubby, or open. At the scale of mapping
used, individual peatlands were rarely identified, with most
polygons composed of peatland complexes and the compo-
nents identified to the nearest 10% cover. Peatland dis-
tributions were determined from the 1 : 250 000 base maps
through digitizing onto provincial base maps. Areal extents
were calculated in ARC/INFO for 0.25° latitude and 0.5°
longitude grids by peatland type. Published summaries have
been completed for Alberta (Vitt et al. 1996) and Manitoba
(Halsey et al. 1997), while summary data for Saskatchewan
are available from the authors.
Current maximum depth distribution
Maximum depth values for 818 peatland sites in continen-
tal western Canada were compiled from several sources
(Bannatynne 1980; Zoltai et al. 2000; Vitt published and un-
published data) and contoured through gridding in
MacGridzo for 0.25° latitude and 0.5° longitude grid cells,
with extrapolation in areas with no data along the northeast-
ern margin of Manitoba (Rockware Inc. 1991).
Peatland volume
Below-ground peatland volumes were calculated for 0.25°
latitude and 0.5° longitude grids by multiplying maximum
depth and peatland type areas (maximum volume), adjusted
© 2000 NRC Canada
684 Can. J. Earth Sci. Vol. 37, 2000

by a topography value to account for basin slope. Topogra-
phy by grid cell was determined from numerous surficial ge-
ology studies, soil inventories, and biophysical reports (list
available from authors) and was divided into five types fol-
lowing those established by the Canadian Soil Survey Com-
mittee (1978) for surface expression. These include
(1) level, with 98% of the maximum volume occupied by
peat (1–3% slope); (2) undulating, with 93.5% of the maxi-
mum volume occupied by peat (4–9% slope); (3) rolling,
with 87.5% of the maximum volume occupied by peat (10–
15% slope); (4) hummocky – knob and kettle, with 77% of
the maximum volume occupied by peat (16–30% slope); and
(5) steep–inclined, with 70% of the maximum volume occu-
pied by peat (>31% slope).
Carbon content
Carbon contents of 253 samples were determined from ten
cores recovered from across continental western Canada
(Athabasca: 55°05′N; 113°15′W in an open fen, a wooded
fen, and a nonpermafrost bog; Rainbow Lake: 58°17′N;
119°22′W in a shrubby fen and a permafrost bog; Flintstone
Lake: 50°43′N; 95°18′W in a nonpermafrost bog; Jan Lake:
54°53′N; 102°48′W in a wooded fen and a nonpermafrost
bog; and North Knife Lake: 58°07′N; 97°02′W in two open
fens (Bauer, unpublished data, 1996, 1997).
Samples of known volume were taken from the cores at
10 cm intervals, air dried for 48 h, and weighed for bulk
density. From each of these samples two ground subsamples
were taken. From one of the subsamples, loss on ignition at
550°C was calculated to determine organic matter density
(= ashless bulk density) (Dean 1974). A second subsample
was analyzed for carbon content in a Controlled Equipment
Corporation Model 440 CHN elemental analyzer compared
to a certified standard (Acetanilide). Relating carbon content
to organic matter density is preferable to comparisons of dry
bulk density (Wieder et al. 1994), particularly in continental
western Canada where aeolian activity has been extensive
throughout the Holocene in areas associated with geogenous
fens (Halsey et al. 1990), and volcanic ash deposits are com-
monly found within peat deposits (Zoltai 1989). For this rea-
son, percent carbon is calculated on an ash-free basis.
Organic matter density
Variations in mean organic matter density determined by
stratigraphic horizon (Dean 1974) for 475 cores collected
across western Canada were compared among peatland type,
ecoregion, and whole core depths grouped by 50 cm inter-
vals as main effects within a General Linear Model proce-
dure in SAS (SAS Institute Inc. 1988). All main effects
displayed a normal distribution, thus a Student–Newman–
Keuls post test was utilized to distinguish statistically simi-
lar groupings for all main effects.
Modern peatland carbon storage
The amount of current carbon stored in peatlands was cal-
culated by using the mean organic matter densities for all
main effects groupings distinguished in the post test adjusted
by mean carbon content. Carbon content densities were then
multiplied by peatland type volumes obtained from the 1838
land-based 0.25° latitude and 0.5° longitude grid cells from
across continental western Canada.
Biomass
A literature survey of above-ground biomass was con-
ducted for continental western Canada. Only vascular plants
were placed into above-ground biomass as it is difficult, if
not impossible, to determine the boundary between living
nonvascular plants and dead peat. Pooled, mean biomass val-
ues by peatland type were used to determine total above-
ground biomass for 0.25° latitude and 0.5° longitude grid
cells. When coupled with carbon content, the amount of car-
bon stored in living above-ground vascular plants was deter-
mined. Details are reported in Campbell et al. (2000).
Temporal pattern: calculated past carbon storage
The distribution and extent of peatlands in continental
western Canada has changed throughout the Holocene
(Zoltai and Vitt 1990). Peatland volume at a given time is a
function of the timing of initial peat formation and the rate
of subsequent peatland expansion (paludification) across the
landscape. As peatlands in continental western Canada are
relatively flat the elevation (depth) of the surface at one
point in a peatland will be generally the same across the
peatland. Thus, as a peatland expands through time its eleva-
tion (depth) increases. Following this pattern of lateral
expansion, paludification patterns (depth-calibrated radiocar-
bon date) were established for a permafrost site (Rainbow
Lake, AB: 58°17′N; 119°22′W) and a nonpermafrost site
(Athabasca, AB: 55°03′N; 113°15′W) using curve estima-
tions whose intercepts were forced through zero in SPSS
(SPSS 1995). These relationships were then used to extrapo-
late peatland expansion throughout the Holocene for all grid
cells utilizing timing of peatland initiation determined from
Halsey et al. (1998). Carbon contents in 1000 year incre-
ments were calculated for each grid cell using current stores
with depth adjusted by the paludification patterns derived for
permafrost and nonpermafrost peatlands.
Temporal pattern: modeled past carbon storage
(catotelm decomposition)
Amounts of carbon based on current storage need to be
corrected for long-term catotelm decomposition. Here we
utilize Clymo’s (1984) exponential decay model for catotelm
peat to determine the long-term decay constant in the
catotelm (α) for nine cores located throughout continental
western Canada (Gypsumville: 51°46′N; 98°30′W (Kuhry et
al. 1992); Site 4: 52°51′N; 116°28′W (Zoltai 1989); Site 5:
53°20′N; 117°28′W (Zoltai 1989); Zoltai 81–18A: 54°45′N;
115°51′W (S.C. Zoltai, unpublished data, 1981); Buffalo
Narrows: 55°56′N; 108°34′W (Kuhry 1994); Mariana
Lakes: 55°54′N; 112°04′W (Nicholson and Vitt 1990); Leg-
end Lake: 57°26′N; 112°57′W (Kuhry 1994); Rainbow
Lake: 58°18′N; 119°17′W (Zoltai 1993); and Zama City:
59°07′N; 118°09′W (Zoltai 1993)). Using the mean esti-
mated value of a modeled amount of carbon (Ct) for each
1000 year increment throughout the Holocene was calcu-
lated from the apparent amount of carbon (C′) in each 0.25°
latitude and 0.5° longitude grid cell [1].
[1] Ct/C′=e
–αt
Since acrotelm decay contributes to the actual current stor-
age values, and does not add significantly to long-term de-
composition, no attempt was made to incorporate acrotelm
losses through time.
© 2000 NRC Canada
Vitt et al. 685

Results
Current peatland distribution
Continental western Canada consists of Alberta, Saskatch-
ewan, and Manitoba (Fig. 1) and contains approximately
365 157 km2of peatlands that make up 21% of the landbase
(Table 1). Peatlands are concentrated in northern and north-
eastern Alberta and northeastern Manitoba as well as along
the northeastern shore of Lake Winnipeg (Fig. 1). Sixty-
three percent of these peatlands are fens, 28% are permafrost
bogs, while nonpermafrost bogs represent only 9% of all
peatlands in continental western Canada (Table 1). Perma-
© 2000 NRC Canada
686 Can. J. Earth Sci. Vol. 37, 2000
Fig. 1. Contoured current distribution of peatlands. Contour intervals are in 10% increments of total land surface.
Fig. 2. Contoured maximum depth values for peatlands of continental western Canada. Contour interval is 50 cm.

frost bogs and open nonpatterned fens are most abundant in
the Subarctic Region, while all other peatland types reach
highest abundance in the High Boreal Region. Nearly one-
half of all peatlands are in the High Boreal Region. Sixty-
four percent of the peatlands of continental western Canada
are treed.
Current maximum depth distribution
Contoured maximum depth values for continental western
Canada are presented in Fig. 2. Peatlands are deepest in the
mid-boreal and along the eastern border of Manitoba. For
this reason carbon storage does not reflect peatland distribu-
tion.
Carbon content
Mean carbon content of 253 samples was 47.7 ± 5.0% of
dry bulk density, and when subtraction of ash is included, it
yields a mean of 51.8 ± 4.7% of carbon. This value falls
within the range of carbon contents that can typically be ex-
pected for soils (Nelson and Sommers 1996).
Organic matter density
Analysis of the variation of organic matter density, using
the General Linear Model, shows that all main effects—
peatland type, ecoregion, and whole core depth grouped into
50 cm intervals—are significant in explaining variation in
organic matter density, with only the interaction between
ecoregion and depth being significant (Table 2). Ecoregion
and depth have a significant interaction, as deeper peatlands
(> –350 cm) are found more often within the mid- and low
boreal ecoregions. Of the main effects, only peatland type is
recognized by the Student–Newman–Keuls post test as hav-
ing significantly different variation (Fig. 3). Organic matter
density of wooded and shrubby fens is significantly different
from that of open fens and permafrost and nonpermafrost
bogs, with the former group having a 12% greater whole
core mean organic matter density (Fig. 3).
Current peatland carbon storage
Current peatland carbon pools form two groups:
(1) wooded and shrubby fens with a carbon density of 0.055 ±
0.003 g C@cm–3 and (2) open fens and (permafrost and
nonpermafrost) bogs with a carbon density of 0.049 ±
0.004 g C@cm–3. Carbon storage in continental western Cana-
dian peatlands is concentrated in northern and northeastern
Alberta, northeast of Lake Winnipeg, and within the Hudson
Bay Lowlands of Manitoba (Fig. 4). Currently, 47.9 Pg of
carbon are stored as living groundlayer, below-ground bio-
mass, and peat in continental western Canadian peatlands
(Table 3). Fens contain almost twice as much carbon as
bogs. Bogs contain about 35% of peatland carbon, with per-
mafrost dominated systems having 27% of the total carbon.
Only 9% of the stored carbon is in nonpermafrost bogs. Fens
have 65% of the total carbon, with treed fens having the
most (30%). Treed systems in general contain 65% of the to-
tal carbon. Fifty-one percent of peatland carbon is found in
the High Boreal Region of continental western Canada.
Manitoba contains 57.8% of peatland carbon, followed by
Alberta with 27.9%, and Saskatchewan with 14.3%.
© 2000 NRC Canada
Vitt et al. 687
Peatland type Arctic Subarctic Montane High
Boreal Mid
Boreal Aspen Parkland
and Interlake Total %
total %
area
Permafrost bogs 24 66 159 0 35 749 1 112 0 103 044 28.2 5.9
Nonpermafrost bogs 0 1 749 0 20 884 6 513 1 812 30 958 8.5 1.7
Treed fens 0 10 324 56 68 194 16 698 6 182 101 454 27.8 5.8
Shrubby fens 51 2 986 21 13 638 4 287 3 362 24 345 6.7 1.4
Open nonpatterned fens 202 30 190 26 25 327 10 496 5 724 71 965 19.7 4.1
Open patterned fens 0 4 156 26 15 768 11 445 1 996 33 391 9.1 1.9
Total 277 115 564 129 179 560 50 551 19 076 365 157 100.0 20.8
% 0.1 31.7 0.1 49.2 13.7 5.2 100.0
Note: Regions follow those defined by the Ecological Stratification Working Group (1995).
Table 1. Distribution of peatlands in continental western Canada (Alberta, Saskatchewan, and Manitoba) in square kilometres.
Source Fvalue Pr > F
Ecoregion 3.12 0.0090**
Depth interval 3.06 0.0006**
Peatland type 6.14 0.0001**
Ecoregion*depth interval 1.84 0.0051**
Depth interval*peatland type 0.61 0.8605
Ecoregion*peatland type 0.72 0.8690
Ecoregion*depth interval*peatland type 0.69 0.8928
**Independent variables that are significant at the 95% level.
Table 2. Effects model significance levels of all variables. Fig. 3. Mean organic matter density for different peatland types
obtained from 475 whole cores from across continental western
Canada. Error bars represent the standard deviation of the mean.
Student–Newman–Keuls post test identified three groupings A, B,
and C of which two are statistically distinguishable with means of
0.105 g@cm3representing wooded and shrubby fens (A) and 0.094
g@cm3for open fens and permafrost and nonpermafrost bogs (C).

Biomass
Above-ground vascular plant biomass by peatland type for
continental western Canadian peatlands is presented in Ta-
ble 4. Wooded peatlands have the highest amount of biomass
per unit area, followed by shrubby, and open fens. Of the
wooded peatlands, bogs have the highest above-ground bio-
mass. Carbon contents, coupled with pooled, mean biomass
numbers by peatland type, and area of peatlands results in
© 2000 NRC Canada
688 Can. J. Earth Sci. Vol. 37, 2000
Fig. 4. Contoured current carbon storage in the surface and below-ground component of peatlands across continental western Canada.
Contour interval is 20 kg@m2.
Fig. 5. Contoured current carbon storage of above-ground vascular plant biomass in peatlands of continental western Canada. Contour
interval is 50 g@m2.

0.10 Pg of carbon stored in above-ground vascular peatland
plant biomass, with its distribution following that of
peatland distribution (Fig. 5).
Temporal pattern of past carbon storage
Paludification trends from two bog–fen peatland com-
plexes show that age is linearly related to depth for the
nonpermafrost site, while for the permafrost site age is re-
lated to depth as a power function (Fig. 6). With the devel-
opment of permafrost, peatland development follows a cycle
of aggradation and degradation often related to fire (Zoltai
1993). This cyclic development retards long-term carbon ac-
cumulation once permafrost is initiated. The slope of the
curve fitted to the permafrost site decreases around 4000 BP
(Fig. 6), corresponding to the timing of permafrost expan-
sion into bogs of this area (Zoltai 1993).
Carbon stored in peatlands has increased during the Holo-
cene (Fig. 7; Table 5). Peatlands began to accumulate carbon
around 9000 BP, with half (51.4%) of current stocks present
by 4000 BP. During the last 1000 years about 5.1 Pg
(10.6%) of the total long-term (catotelm) carbon has de-
cayed, and 12.2 Pg (25.5%) of new carbon has accumulated,
with a net gain of 7.1 Pg, or 14.8%. This compares with be-
tween 4000 and 5000 BP when the net carbon gain was
about 67.2%, increasing the stock from 16.0 Pg to 24.7 Pg.
During the early Holocene, between 8000 and 9000 years
ago, the percent net carbon gain was even higher (91.7%),
but the actual increase in stock was minimal (1.1 Pg).
Discussion
Peatlands of continental western Canada contain a signifi-
cant amount of carbon—47.9 Pg with an additional 0.10 Pg
© 2000 NRC Canada
Vitt et al. 689
Peatland type Arctic Subarctic Montane High Boreal Mid Boreal Aspen Parkland
and Interlake Total %
Permafrost bogs 1.77 × 1012 8.34 × 1015 0 4.17×10
15 1.44 × 1014 0 1.27×10
16 26.5
Nonpermafrost bogs 0 2.96 × 1014 0 2.84×10
15 8.05 × 1014 2.01 × 1014 4.14 × 1015 8.6
Treed fens 0 1.36 × 1015 3.78 × 1012 9.72 × 1015 2.25 × 1015 7.47 × 1014 1.41 × 1016 29.4
Shrubby fens 3.95 × 1012 4.04 × 1014 1.40 × 1012 1.77 × 1015 3.89 × 1014 2.65 × 1014 2.83 × 1015 5.9
Open nonpatterned fens 1.59 × 1013 4.09 × 1015 1.71 × 1012 3.23 × 1015 9.55 × 1014 4.51 × 1014 8.74 × 1015 18.3
Open patterned fens 0 7.09 × 1014 1.83 × 1012 2.63 × 1015 1.59 × 1015 4.48 × 1014 5.38 × 1015 11.2
Total 2.16 × 1013 1.52 × 1016 8.72 × 1012 2.44 × 1016 6.13 × 1015 2.11 × 1015 4.79 × 1016 100
% 0.1 31.7 0.0 50.9 12.8 4.4 100
Note: Regions follow those defined by the Ecological Stratification Working Group (1995).
Table 3. Distribution of carbon in peatlands of continental western Canada (Alberta, Saskatchewan, and Manitoba) in grams.
Site Above-ground
biomass (g m–2) Reference Pooled mean above-
ground biomass (g m–2)
Wooded bogs
Nonpermafrost bog 901.3 Reader and Stewart 1972 662 775
Nonpermafrost bog 423.3 Reader and Stewart 1972
Nonpermafrost bog (1991A) 823.2aSzumigalski 1995 887
Nonpermafrost bog (1992A) 854.8 Szumigalski 1995
Nonpermafrost bog (1994A) 984.0aThormann 1995
Wooded fens
Rich fen (1991B) 887.3aSzumigalski 1995 750
Rich fen (1992B) 613.3 Szumigalski 1995
Shrubby fens
Poor fen (1991C) 364.1 Szumigalski 1995 395 275
Poor fen (1992C) 424.8 Szumigalski 1995
Rich fen (1991D) 224.6 Szumigalski 1995 201
Rich fen (1992D) 139.8 Szumigalski 1995
Rich fen (1993D) 110.0 Thormann 1995
Rich fen (1994D) 328.0 Thormann 1995
Rich fen (1991E) 190.0 Szumigalski 1995 230
Rich fen (1992E) 269.0 Szumigalski 1995
Open fens
Rich fen (1991F) 105.0 Szumigalski 1995 99 254
Rich fen (1992F) 92.7 Szumigalski 1995
Rich fen (1993G) 339.0 Thormann 1995 410
Rich fen (1994G) 480.0 Thormann 1995
Note: Swamp and marsh sites are not included. Year of collection as well as a site designator (A through G) are given for sites measured over multiple years.
aTree layer biomass component is based on 1992 modeled values from Szumigalski (1995).
Table 4. Total above-ground biomass for all continental western Canadian peatland sites.

in above-ground biomass. Together this represents roughly
2.1% of the world’s terrestrial carbon within 0.25% of the
global terrestrial surface area. Other workers have calculated
the amount of carbon stored in organic soils (peats) in Can-
ada (Tarnocai 1998), with an estimated 38.1 Pg (–20.5% de-
viation when compared to the amount of carbon we
estimate) within Alberta, Saskatchewan, and Manitoba (C.
Tarnocai and B. Lacelle, personal communication, 1998).
This value is considered low by these workers, with the dis-
crepancy attributed to lack of information on peat depths (C.
Tarnocai and B. Lacelle, personal communication, 1998).
Gorham (1991) examined current carbon storage on a
general circumboreal level. Comparing our more detailed ap-
proach over a much smaller area, variation in the contribut-
ing components is as follows: peatland area for continental
western Canada +14.4% (Gorham’s (1991) value is higher);
+11.7% for mean carbon density (Gorham’s (1991) value is
higher); and –10.4% for depth (calculation used here provid-
ing higher contribution). Thus, while depth is a major source
of error in calculating carbon storage (cf. Botch et al. 1995),
carbon densities, and peatland distributions are also variable.
Detailed inventories of peatland distributions, such as those
we present here, will reduce error in carbon storage esti-
mates substantially.
Adding together the estimated variation in this study for
peatland distribution (±4%), standard error of the mean for
carbon content (±0.3%), standard error the mean for organic
matter density for treed and shrubby fens (±0.4%) that com-
prise 35% of peatlands, and for bogs and open fens (±0.5%)
representing 65% of peatlands. When a ±10.4% variation
derived from comparing differing depth measurement meth-
ods (cf. Gorham 1991) is included, the root-mean-square
error is 11.2%, with carbon storage in peatlands of continen-
tal western Canada estimated to represent 48.0 ± 5.4 Pg.
Carbon storage in peatlands has changed considerably
over the Holocene (Fig. 7). Peatlands began to accumulate
carbon around 9000 BP in this region, after an initial
deglacial lag. Carbon accumulation was climatically limited,
however, as the amount of land with suitable climatic condi-
tions for peatlands was restricted by early Holocene maxi-
mum summer insolation (Halsey et al. 1998). As summer
insolation decreased, more land became climatically avail-
able for peatland formation, with the peatland climatic
threshold transgressing southward (Halsey et al. 1998).
Over a span of 3000 years during the mid-Holocene
(6000–3000 BP), about half of current stocks accumulated.
Since the mid-Holocene, peatland carbon stocks have contin-
ued to increase with a further one-third accumulating over
the last 3000 years. The net increase in carbon stocks over
the last 3000 years has declined relative to the mid-
Holocene. This decline corresponds temporally to the expan-
sion of permafrost into continental western Canada (Zoltai
© 2000 NRC Canada
690 Can. J. Earth Sci. Vol. 37, 2000
Site Location Radiocarbon dates
(years BP) Depth (cm) Reference Decay rate
(year–1)
Gypsumville 51°46′N and 98°30′W 1790±90 (AECV-1082C) 146.0 Kuhry et al. 1992 1.9 × 10–4
2710±100 (AECV-1081C) 186.0
4230±100 (AECV-1031C) 235.5
Site 4 52°51′N and 116°28′W 4460±170 (BGS-771) 205.0 Zoltai 1989 4.3 × 10–5
6170±140 (BGS-770) 241.0
8600±250 (BGS-772) 354.5
Site 5 53°20′N and 117°28 ′W 2800±300 (BGS-773) 190.0 Zoltai 1989 2.3 × 10–4
6140±200 (BGS-774) 332.0
8400±270 (BGS-775) 506.0
Zoltai 81–18A 54°45′N and 115°51′W 2820±220 (BGS-776) 243.5 Zoltai unpublished 1.4 × 10–4
6170±180 (BGS-777) 442.5
8940±240 (BGS-554) 551.0
Buffalo Narrows 55°56 ′N and 108°34′W 1480±100 (AECV-1092C) 86.0 Kuhry 1994 1.3 × 10–4
5230±90 (AECV-1739C) 120.0
7870±130 (AECV-1091C) 163.0
Mariana Lakes 55°54′N and 112°04′W 3170±80 (AECV-262C) 107.0 Nicholson and Vitt 1992 1.8 × 10–5
5270±90 (AECV-263C) 160.0
6740±100 (AECV-212C) 189.0
Legend Lake 57°26′N and 112°57′W 1180±80 (AECV-1645C) 60.5 Kuhry 1994 2.6 × 10–4
4390±91 (AECV-1738C) 90.5
7950±100 (AECV-1900C) 112.0
Rainbow Lake 58°18′N and 119°17′W 1070±90 (AECV-989C) 46.5 Zoltai 1993 8.9 × 10–5
3710±100 (AECV-990C) 84.5
7620±120 (AECV-991C) 187.0
Zama City 59°7′N and 118°9′W 1420±90 (AECV-984C) 52.0 Zoltai 1993 1.7 × 10–4
2410±120 (AECV-985C) 111.0
5840±100 (AECV-986C) 258.0
Mean 1.41 × 10–4
Standard deviation 8.13 × 10–5
Table 5. Sites and associated radiocarbon dates used to calculate catotelm decay rates.

1995), and to the establishment of the current southern limit
of peatlands (Halsey et al. 1998).
Since the accumulation of peatland carbon in continental
western Canada was initially limited by global maximum
early Holocene insolation (Halsey et al. 1998), accumulation
patterns similar to those presented here would be expected
throughout the circumboreal. High-resolution records of at-
mospheric methane in Greenland show an increase in con-
centration after a mid-Holocene minimum (Chappellaz et al.
1997). This increase in methane concentration corresponds
temporally to the highest peat accumulation rates in our
data.
With the development of permafrost in continental west-
ern Canada 3000–4000 BP, in conjunction with the southern
limit of peatlands being reached about 3000 BP, the rate of
increase in overall carbon storage in peatlands of continental
western Canada declined. This decrease relates to the de-
cline in the Greenland/Antarctic methane concentration ra-
tio, suggesting that the contribution of methane from
northern sources became less important globally around this
time (Chappellaz et al. 1997). While there was more carbon
stored in continental western Canada 3000 years ago than at
any other time previously in the Holocene, the widespread
development of bogs and also permafrost probably resulted
in decreased methane fluxes. Bogs, with their thick
acrotelms, are known to produce very low methane fluxes
relative to other peatland types, with permafrost bogs pro-
ducing the least (Klinger et al. 1994; Bubier et al. 1995).
This suggests that overall methane flux from continental
western Canadian peatlands may have decreased after the
widespread development of permafrost around 3000 BP.
Although the rate of peatland carbon sequestration is less
today than 3000 years ago, this regional study documents
that peatlands in continental western Canada continue to
function as a carbon sink. This finding follows a similar ob-
servation made by Kuhry and Vitt (1996) at the site specific
level. Our estimate suggests that in continental western
Canada 7.1 × 1012 gC@year–1 (19.4 g C@m–2@year–1) have
been sequestered during the past 1000 years. This is lower
than Gorham’s (1991) estimate for northern peatlands of
28.1 g C@m–2@year–1, but comparable to the 21 g C@m–2@year–1
for very poorly drained soils (peatlands) estimated for boreal
Manitoba where accumulation–loss of carbon from the soil
profile was determined (Raphalee et al. 1998), and to those
of Finnish mires summarized by Mäkilä (1997) that range
from 14.1 to 22.5 g C@m–2@year–1 The carbon stock increase
of 10.6% over the past 1000 years indicates that current
stocks are increasing at a rate of 0.011% annually. Thus,
while regional storage is beginning to level off due to in-
creased total catotelm decay, as predicted by Clymo (1984),
more important to this decline in net carbon storage is the
development of permafrost, and the establishment of the cur-
rent southern limit of peatlands.
Conclusions
Peatlands cover 365 157 km2of continental western Can-
ada and store 48.0 Pg of carbon, representing 2.1% of global
terrestrial carbon over 0.25% of the landbase. Peatland car-
bon stores have been highly variable through the Holocene.
© 2000 NRC Canada
Vitt et al. 691
Fig. 6. Relationship between peatland depth and calibrated, basal
radiocarbon date (I. Bauer, unpublished data). Basal dates from
the permafrost site at Rainbow Lake, AB are represented as open
squares, while dates from the nonpermafrost site at Athabasca,
AB are represented by closed circles. Regression lines were de-
termined from curve estimation in SPSS (SPSS 1995). The
nonpermafrost site has a linear relationship between peatland
depth and calibrated radiocarbon date (depth = 0.052 × date, r2=
0.76), while the permafrost site displays a power relationship
(depth = 4.47 × 10–7 × date2.23,r
2= 0.66).
Fig. 7. Changes in carbon storage through the Holocene for
peatlands in continental western Canada. Solid bar segments are
estimated carbon present at each time interval; stippled bar seg-
ments represent the modeled amount of carbon decayed since de-
position (1 Pg=1×10
15 g).

Accumulation began after an initial deglacial lag, and in-
creased around 6000 BP as more land area passed through a
climatic threshold that was previously limited in extent by
early Holocene maximum insolation. Contemporaneous in-
creases in methane concentrations in Greenland ice cores
(Chappellaz et al. 1997), correlate well with rates of peat ac-
cumulation, not with the total carbon pool. About half of
current peatland carbon was present by 4000 BP. The rate of
increase in carbon storage in continental western Canadian
peatlands began to decline around 3000 BP, responding to
widespread permafrost development and the establishment
of the current southern peatland limit. Declines in the ratio
of Greenland/Antarctic methane concentration after 3000 BP
suggest a declining boreal source (Chappellaz et al. 1997)
that appears to be related to the rate of carbon accumulation.
Northern peatlands have played an important role in atmo-
spheric carbon budgets through the Holocene, and currently
sequester about 19.4 g C@m–2@year–1. The changes in carbon
storage presented here are based on limited data on the
depth/age relationship of peatland expansion and should be
viewed only in the context of the overall trends. Collection
of more data on peatland expansion spanning all climatic
and physiographic regions is required to reach a better un-
derstanding of how peatlands, and the carbon they store, re-
spond to changing climates.
Acknowledgments
Funding for this project was provided by a Climate Sys-
tem History and Dynamics, National Science and Engi-
neering Research Council Research Network Grant and a
Network of Centres of Excellence in Sustainable Forest
Management Grant to Dale Vitt. R. Kelman Wieder im-
proved an earlier draft of this manuscript, for which we are
thankful. Dr. Bob Vance and the Geological Survey of Can-
ada provided the radiocarbon dates for the Rainbow Lake
Core, for which we are grateful. In addition thanks are also
extended to Nigel Roulet and an anonymous reviewer,
whose thoughtful and thorough reviews greatly improved the
manuscript. Graphics were produced by Laureen Snook and
Sandi Vitt. We dedicate this manuscript to Stephen Zoltai,
our dear friend and always enthusiastic colleague, whose ex-
tensive work in the peatlands of western Canada contributed
significantly to this manuscript.
References
Bannatyne, B.B. 1980. Sphagnum bogs in southern Manitoba and
their identification by remote sensing. Manitoba Department of
Energy and Mines, Economic Geology Report ER79-7.
Belland, R.J., and Vitt, D.H. 1995. Bryophyte vegetation patterns
along environmental gradients in continental bogs. Ecoscience,
2: 395–407.
Botch, M.S., Kobak, K.I., Vinson, T.S., and Kolchugina, T.P. 1995.
Carbon pools and accumulation in peatlands of the former So-
viet Union. Global Biogeochemical Cycles, 9: 37–46.
Brown, R.J.E., and Williams, G.P. 1972. The freezing of peatlands.
National Research Council of Canada, Division of Building Re-
search, Ottawa, Ontario, Technical Paper 381.
Bubier, J.L., Moore, T.R., Bellisario, L., and Comer, N.T. 1995.
Ecological controls on methane emissions from a northern
peatland complex in the zone of discontinuous permafrost, Man-
itoba, Canada. Global Biogeochemical Cycles, 9: 455–470.
Campbell. C., Vitt, D.H., Halsey, L.A., Campbell, I.D., Thorman,
M.N., and Bayley, S.E. 2000. Spatial distribution of net primary
production in the wetlands of continental western Canada.
NOR-X-369. Northern Forestry Centre, Canadian Forest Ser-
vice, Edmonton, Alta.
Canadian Soil Survey Committee 1978. The Canadian system of
soil classification. Canada Department of Agriculture, Research
Branch, Ottawa, Ont.
Chappellaz, J., Blunier, T., Kints, S., Dällenbach, A., Barnola, J.-
M., Schwander, J., Raynaud, D., and Stauffer, B. 1997. Changes
in the atmospheric CH4gradient between Greenland and
Antarctica during the Holocene. Journal of Geophysical Re-
search, 102: 15 987 – 15 997.
Clymo, R.S. 1984. The limits to peat bog growth. Philosophical
Transactions of the Royal Society of London, B303: 605–654.
Dean, W.E., Jr. 1974. Determination of carbonate and organic mat-
ter in calcareous sediments and sedimentary rocks by loss on ig-
nition: comparison with other methods. Journal of Sedimentary
Petrology, 44: 242–248.
Ecological Stratification Working Group 1995. Terrestrial
ecozones, ecoregions, and ecodistricts of Canada: Provinces of
Alberta, Saskatchewan, and Manitoba. A national ecological
framework for Canada. Agriculture and Agri-Foods Canada, Re-
search Branch Centre for Land and Biological Resources Re-
search and Environment Canada, State of the Environment
Directorate, Ecozone Analysis Branch, Ottawa–Hull, Canada.
Frolking, S. 1991. Methane from northern peatlands and climate
change. In Carbon cycling in boreal forest and sub-arctic eco-
systems: biospheric responses and feedbacks to global climate
change. Edited by T.S. Vinson and T.P. Kolchungina. United
States Environmental Protection Agency, Publication 600R-
93/084, Corvallis, Oreg.
Gignac, L.D., Vitt, D.H., Zoltai, S.C., and Bayley, S.C. 1991.
Bryophyte response surfaces along climatic, chemical, and phys-
ical gradients in peatlands of western Canada. Nova Hedwigia,
53: 27–71.
Gore, A.J.P. 1983. Ecosystems of the World 4B. Mire: swamp,
bog, fen, and moor. regional studies. Elsevier Scientific Pub-
lishing Co., Amsterdam, The Netherlands.
Gorham, E. 1991. Northern peatlands: role in the carbon cycle and
probable responses to climatic warming. Ecological Applica-
tions, 1: 182–195.
Halsey, L.A., and Vitt, D.H. 1997. Alberta Wetland Inventory Stan-
dards. In Alberta Vegetation Inventory Standards Manual, Ver-
sion 2.2. Compiled by R. Nesby. Alberta Environmental
Protection, Edmonton, Alta.
Halsey, L.A., Catto, N.R., and Rutter, N.W. 1990. Sedimentology
and development of parabolic dunes, Grande Prairie dune field,
Alberta. Canadian Journal of Earth Sciences, 27: 1762–1772.
Halsey, L.A., Vitt, D.H., and Zoltai, S.C. 1995. Disequilibrium re-
sponse of permafrost in boreal continental western Canada to
climatic change. Climatic Change, 30: 57–73.
Halsey, L.A., Vitt, D.H., and Zoltai, S.C. 1997. Climatic and phys-
iographic controls on wetland type and distribution in Manitoba,
Canada. Wetlands, 17: 243–262.
Halsey, L.A., Vitt, D.H., and Bauer, I.E. 1998. Peatland initiation
during the Holocene in continental western Canada. Climatic
Change, 40: 315–342.
Klinger, L.F., Zimmerman, P.R., Greenberg, J.P., Heidt, L.E., and
Guenther, A.B. 1994. Carbon trace gas fluxes along a
successional gradient in the Hudson Bay lowland. Journal of
Geophysical Research, 99: 1469–1494.
© 2000 NRC Canada
692 Can. J. Earth Sci. Vol. 37, 2000

© 2000 NRC Canada
Vitt et al. 693
Kuhry, P. 1994. The role of fire in the development of Sphagnum-
dominated peatlands in western boreal Canada. Journal of Ecol-
ogy, 82: 899–910.
Kuhry, P., and Vitt, D.H. 1996. Fossil carbon/nitrogen rations as a
measure of peat decomposition. Ecology, 77: 271–275.
Kuhry, P., Halsey, L.A., Bayley, S.E., and Vitt, D.H. 1992. Peatland
development in relation to Holocene climatic change in Mani-
toba and Saskatchewan (Canada). Canadian Journal of Earth
Sciences, 29: 1070–1090.
Mäkilä, M. 1997. Holocene lateral expansion, peat growth and car-
bon accumulation on Haukkasuo, a raised bog in southeastern
Finland. Boreas, 26: 1–14.
Moore, T.R., Roulet, N.T., and Waddington, J.M. 1998. Uncer-
tainty in predicting the effect of climatic change on the carbon
cycling of Canadian peatlands. Climatic Change, 40: 229–245.
National Wetlands Working Group. 1988. Wetlands of Canada.
Ecological Land Classification Series, No. 24. Sustainable De-
velopment Branch, Environment Canada, Ottawa, Ont. and
Polyscience Publications Inc., Montreal, Que.
Nelson, D.W., and Sommers, L.E. 1996. Total carbon, organic carbon,
and organic matter. In Methods of Soil Analysis Part 3. Chemical
Methods. Soil Science Society of America. Madison, Wis.
Nicholson, B.J., and Vitt, D.H. 1990. The paleoecology of a
peatland complex in western Canada. Canadian Journal of Bot-
any, 68: 121–138.
Nicholson, B.J., Gignac, L.D., Bayley, S.E., and Vitt, D.H. 1997.
Vegetation response to global warming: interactions between bo-
real forest, wetlands, and regional hydrology. In Mackenzie Ba-
sin Impact Study Final Report. Edited by S. Cohen.
Environment Canada, Ottawa, Ont.
Post, W.M., Chavez, F., Mulholland, P.J., Pastor, J., Peng, T.-H.,
Prentice, K., and Webb, III, T. 1992. Climatic feedbacks in the
global carbon cycle. In The science of global change: the im-
pacts of human activities on the environment. Edited by D.A.
Dunnette and R.J. O’Brien. American Chemical Society, Wash-
ington, D.C.
Rapalee, G., Trumbore, S.E., Davidson, E.A., Harden, J.W., and
Veldhuis, H. 1998. Soil Carbon stocks and their rates of accu-
mulation and loss in a boreal forest landscape. Global
Biogeochemical Cycles, 12: 687–701.
Reader, R.J., and Stewart, J.M. 1972. The relationship between net
primary production and accumulation for peatlands in southeast-
ern Manitoba. Ecology, 53 1024–1037.
Rockware Inc. 1991. Macgridzo: the contouring mapping program
for the Macintosh Version 3.3. RockWare, Wheat Ridge, Colo.
SAS Institute Inc. 1988. SAS/Stat User’s Guide Release 6.03 Edi-
tion, Cary, N.C.
SPSS. 1995. SPSS Version 6.1.1, SPSS Inc., Chicago, Ill.
Szumigalski, A.R. 1995. Production and decomposition of vegeta-
tion along a wetland gradient in central Alberta. M.Sc. thesis.
Department of Biological Sciences, University of Alberta, Ed-
monton, Alta.
Thormann, M.N. 1995. Production and decomposition in Alberta
wetlands. M.Sc. thesis. Department of Biological Sciences, Uni-
versity of Alberta, Edmonton, Alta.
Thormann, M.N, Szumigalski, A., and Bayley, S.E. 1999. Above
ground peat and carbon accumulation potentials along a bog-
fen-marsh wetland gradient in southern boreal Alberta, Canada.
Wetlands, 19: 305–317.
Tarnocai, C. 1998. The amount of organic carbon in various soil
orders and ecological provinces of Canada. In Soil processes
and the carbon cycle. Edited by R. Lal, J.M. Kimble, R.F.
Follett, and B.A. Stewart. CRC Press, Boca Raton, Fla.
Tarnocai, C., Kettles, I.M., and Ballard, M. 1995. Peatlands of Can-
ada preliminary. Geological Survey of Canada Open File 3152.
Vitt, D.H., and Kuhry, P. 1992. Changes in moss-dominated wet-
land ecosystems. In Bryophytes and lichens in a changing envi-
ronment. Edited by J.W. Bates and A.M. Farmer. Clarendon
Press, London, U.K.
Vitt, D.H., Van Wirdum, G., Halsey, L.A., and Zoltai, S.C. 1993.
The effects of water chemistry on the growth of Scorpidium
scorpioides in Canada and The Netherlands. The Bryologist, 96:
106–111.
Vitt, D.H., Halsey, L.A., and Zoltai, S.C. 1994. The bog landforms
of continental Canada in relation to climate and permafrost pat-
terns. Arctic and Alpine Research, 26: 1–13.
Vitt, D.H., Halsey, L.A., Thormann, M.N., and Martin, R. 1996.
Peatland inventory of Alberta. Alberta Peat Task Force, Network
of Centres of Excellence in Sustainable Forest Management
Grant, University of Alberta, Edmonton, Alta.
Wieder, R.K., Novak, M., Schell, W., and Rhodes, T. 1994. Rates
of peat accumulation over the past 200 years in five Sphagnum-
dominated peatlands in the United States. Journal of
Paleolimnology, 12: 35–47.
Zoltai, S.C. 1989. Late Quaternary volcanic ash in the peatlands of
central Alberta. Canadian Journal of Earth Sciences, 26: 207–214.
Zoltai, S.C. 1993. Cyclic development of permafrost in the
peatlands of northwestern Alberta. Arctic and Alpine Research,
25: 240–246.
Zoltai, S.C. 1995. Permafrost distribution in peatlands of west-
central Canada during the Holocene warm period 6000 years BP.
Géographie Physique et Quaternaire, 49: 45–54.
Zoltai, S.C., and Vitt, D.H. 1990. Holocene climatic change and
the distribution of peatlands in western interior Canada. Quater-
nary Research, 33: 231–240.
Zoltai, S.C., Siltanen, R.M., and Johnson, J.D. In press. A wetland
environmental data base. NOX Report, Northern Forestry Cen-
tre, Canadian Forest Service, Edmonton, Alta.