tremuloides, Shepherdia canadensis, Vaccinium vitis-
butions were somewhat variable owing to the complex
idaea, Arctostaphylos uva-ursi, Polytrichum spp., and
nature and depositional environments associated with
Cladina spp. Loamy sites with low and tall scrub pre-
alluvial materials (Table 6d). Dry gravelly sites fre-
sented a problem class, with a mixture of soil proper-
quently included Oxytropis campestris, Dryas
ties and vegetation structures that was difficult to clas-
dr ummondii, Potentilla multifida, Shepherdia
sify and map. Loamy moist forested sites were domi-
canadensis, Elaeagnus commutata, Potentilla fruticosa,
nated by Betula papyrifera, P. tremuloides, Picea
Fragaria virginiana, Populus balsamifera, and
glauca, Picea mariana, Alnus crispa, Rosa acicularis,
Stereocaulon spp. In contrast, moist loamy sites fre-
Calamagrostis canadensis, and Hylocomium splendens.
quently included Betula papyrifera, P. balsamifera,
In the late-successional forested ecotypes, P. glauca,
Picea glauca, Alnus tenuifolia, Rosa acicularis,
Geocaulon lividum, Linnaea borealis, H. splendens
Geocaulon lividum, Linnaea borealis, Ledum
were more important. Upland wet loamy sites were a
groenlandicum, and Hylocomium splendens.
special type restricted to north-facing slopes where per-
Overall, the combination of physiography, soil tex-
mafrost reduces drainage. The Upland Wet Needleleaf
ture (derived from geomorphic units), and vegetation
Forest was similar to Lowland Wet Needleleaf Forest
structure appears to work well at differentiating spe-
with a prevalence of P. mariana, A. crispa, Ledum
cies composition. There are numerous ways to classify
groenlandicum, V. vitis-idaea, H. splendens, and Sph-
vegetation and each has its own advantages and disad-
agnum spp., but had little Betula nana and lacked Ru-
vantages. Vegetation structure is commonly used be-
bus chamaemorus.
cause it can be readily identified by remote sensing and
Lowland areas included a large number of classes
photo interpretation. Unfortunately, structure alone is
whose floristics fell into broad groups associated with
poor at differentiating species associations. Floristic
dry gravelly, moist loamy, and wet, organic-rich soil
analysis arguably provides the best approach to devel-
types (Table 6c). Ecotypes on gravelly lowlands oc-
oping species associations that are closely linked to
curred along a successional sequence from scrub to
environmental properties. Floristic classes are not ame-
needleleaf forests and generally include Picea glauca,
nable to remote sensing, however, because only the
Populus tremuloides, Populus balsamifera, Shepherdia
dominant species in the canopy are visible. In addition,
canadensis, Arctostaphylos uva-ursi, Linnaea borea-
mapping or classification cannot be done until the analy-
lis, and Hylocomium splendens. Ecotypes with some-
ses are completed, and results often change when new
what well-drained to imperfectly drained loamy soils
data are acquired. The ecotype approach used here has
typically supported a successional sequence after fire
advantages from both systems, it relies on structure and
that included Betula papyrifera, Picea mariana, P.
landscape characteristics that can be photointerpreted,
glauca, Alnus crispa, Rosa acicularis, Salix bebbiana,
it separates classes with differing species assemblages,
Calamagrostis canadensis, and Equisetum sylvaticum.
and classification can be done with little or no ground
Ecotypes on wetter, bog soils generally supported vari-
information. The main problem with this approach is
ous combinations that included P. mariana, Betula nana,
that some distinctions are particularly equivocal. Con-
Vaccinium uliginosum, V. vitis-idaea, Ledum
sistent differentiation of physiography, for example, can
groenlandicum, L. decumbens, Empetrum nigrum,
be a problem. While differentiation of some physi-
Eriophorum vaginatum, Rubus chamaemorus,
ographic types is relatively straight-forward, the dis-
Oxycoccus microcarpus, and Sphagnum spp. Ecotypes
tinction between upland and lowland areas can be
on organic soils with minerotrophic groundwater move-
confusing. While the distinction is easy in steep, hilly
ment supported Eriophorum angustifolium, Carex
areas with bedrock control, the differences can be
aquatilis, B. nana, V. uliginosum, and Sphagnum spp.
indistinct when slope changes are subtle or frequent,
Lacustrine ecotypes included ponds with submergent
such as in morainal areas.
vegetation, fens on organic shores, and moist loamy
Environmental properties. A comparison of envi-
meadows in recently drained areas. The ponds had a
unique set of aquatic species dominated by
ronmental properties among ecotypes reveals large
Potamogeton spp., Nuphar polysepalum, and Isotes
differences in elevation, soil texture, permafrost occur-
muricata. The shorelines had emergent species such as
rence, thaw depth, water depth, pH, and electrical con-
Carex aquatilis, C. rostrata, C. canescens, Eriophorum
ductivity (Fig. 11). Ecotypes were grouped by physi-
angustifolium, Equisetum fluviatile, and Hippuris vul-
ography to facilitate comparisons.
garis, while moist well-drained areas were dominated
Elevations of the ground-reference plots ranged from
by Calamagrostis canadensis.
293 to 1535 m. The alpine ecotypes usually were above
Riverine ecotypes fell into two broad groups asso-
900 m, with Alpine Rocky Dry Barrens and Alpine Rocky
ciated with gravelly and loamy soils, although distri-
Dry Dwarf Scrub usually occurring above 1100 m (Fig.
35
Previous Page
