Ecoclimatic zonation of Yukon (Canada) and ecoclinal variation in vegetation.
ABSTRACT. An ecological climatic zonation of Canada's Yukon Territory (1:1 000 000 scale) was developed from field observations, aerial photographs, ecological literature, forest cover maps, and regression models. The 11 recognized ecoclimatic regions include Arctic (1), Subarctic (1), Alpine (4), Subalpine (1), and Boreal (4) entities. Region differentiation was based on vegetation thought to reflect climate more strongly than soil or topographic conditions. Sites with such vegetation are referred to as "reference sites." The concept of a reference site was used because conventional zonal site criteria are difficult to apply in mountainous terrain and at high latitudes, where permafrost is an integral environmental component. Alpine regions were differentiated from other ecoclimatic regions through regression analysis of tree line elevations (n = 188, > 76% explained variance). An ecocline of vegetation types for each region was developed on the basis of ecological moisture regimes. The climatic distinctiveness of regions was tested by statistical comparison and ordination of monthly temperature and precipitation data (1984-2007) from 26 and 24 meteorological recording locations, respectively. Significant differences (p < 0.001) between regions were found in temperature and most precipitation variables. A latitudinal gradient was evident among ordinated meteorological recording locations. Non-forest vegetation represents at least 70-75% of Yukon. Forests with closed and semi-closed canopies occurred primarily south of 64[degrees] N latitude at low elevations.
Key words: bioclimatic zone, climate, ecoclimatic region, ecology, reference site, vegetation, Yukon, zonation
RESUME. Une zonation dcoclimatique du Territoire du Yukon, au Canada, a ete realisee (moyennant une echelle de 1:1 000000) a partir d'observations sur le terrain, de photographies adriennes, de documentation de nature ecologique, de cartes du couvert forestier et de modeles de regression. Les 11 regions ecoclimatiques se divisent en entites, soit l'entite arctique (1), l'entite subarctique (1), les entites alpines (4), l'entite subalpine (1) et les entites boreales (4). Les regions ont ete differencides en fonction de la vegetation car celle-ci semblait refleter le climat de maniere plus importante que les conditions pedologiques ou topographiques. Les sites dotes d'une telle vegetation sont appeles [much less than] lieux de reference [much greater than]. Le concept du lieu de reference a ate utilise parce qu'il est difficile d'appliquer le critere classique du lieu zonal en terrain montagneux et en latitude elevee, la oil le pergdlisol fait partie integrante de l'environnement. Les regions alpines ont ete differenciees des autres regions ecoclimatiques grace a l'analyse de regression des elevations de la limite forestiere (n= 188, [greater than or equal to] 76 % variance expliquee). L'ecocline des types de vegetation de chaque region a ete prepare en fonction des regimes hygrometriques ecologiques. Le caractere distinctif des regions a ete mis a l'epreuve au moyen de comparaisons statistiques et de l'ordination des donnees portant sur les temperatures et les precipitations mensuelles (1984-2007) depuis 26 et 24 emplacements d'observations meteorologiques, respectivement. Des differences considerables (p < 0,001) ont did relevees entre les regions sur le plan des variables des temperatures et de la plupart des precipitations. Un gradient latitudinal s'averait evident au sein des emplacements d'observations rnereorologiques ayant fait l'objet d'une ordination. La vegetation non forestiere represente au moms 70 % a 75 % du Yukon. Les forets dotees d'un couvert ferme ou semi-ferme se trouvent surtout au sud du 64[degrees] N de latitude, en basse altitude.
Mots des: zone bioclimatique, climat, region ecoclimatique, ecologie, lieu de reference, vegetation, Yukon, zonation
Traduit pour la revue Arctic par Nicole Giguere.
Vegetation in northern North America ranges from boreal forests to Arctic tundra because ambient temperatures decrease with increasing latitude, which creates a distinctive zonation in eastern and central Canada (Ecoregions Working Group, 1989; Ecological Stratification Working Group, 1995). This pattern, however, is less apparent in British Columbia and the Yukon, where mountain ranges disrupt the general global atmospheric circulation pattern. This rugged terrain also complicates the recognition and mapping of climatic zones (Bailey et al., 1985) because changing elevations, slope gradients, and slope orientations cause variability in temperatures and moisture availability that influence vegetation composition. Vegetation on relatively level terrain with deep, well to moderately well drained, medium-textured soil is conventionally considered more reflective of climate (zonal sites) than either soil or terrain conditions (Mueller-Dombois and Ellen-berg, 1974:408). Unfortunately, zonal sites are uncommon in mountainous terrain and where permafrost is an integral part of the landscape, as in northern Yukon. In addition, wildfires and secondary ecological succession create temporal variation in vegetation structure and composition, which further confuses the recognition of climatically representative ecosystems. Despite this terrain and ecological variability, it should be possible to stratify Yukon into climate zones by using vegetation as an indicator, without the necessity of a dense network of long-term meteorological recording stations (Mackey et al., 1996), and stratification may also be a more practical approach in complex terrain (Farr and Hard, 1987). Stratification of landscapes according to climate can have both scientific and applied applications, which include its use as a basis for ecological analyses, land-use planning, and assessment of contemporary climate change.
Four classification schemes have been developed to characterize the regional ecology of the Yukon. These include three national classifications at a scale of 1:7 500 000 (Rowe, 1972; Ecoregions Working Group, 1989; Ecological Stratification Working Group, 1995) and one Yukon-specific classification at 1:2 500 000 (Oswald and Senyk, 1977). As a basis for representing Yukon's climatic zonation, the primary limitation of the Rowe (1972) forest regions and Ecoregions Working Group (1989) ecoclimatic regions classifications is their small scale of mapping, whereas the Oswald and Senyk (1977) and Ecological Stratification Working Group (1995) ecoregion classifications appear more strongly driven by landscape physiography than by climate. In the two latter classifications, multiple climatic/ecological zones were reported in most of the identified landscape units. For instance, Smith et al. (2004) reported 14 occurrences of Alpine, 9 of Subalpine, and 11 of Boreal zones among 23 purportedly unique landscape units recognized by the Ecological Stratification Working Group (1995). Such repetition suggests that considerable biological and climatic diversity occurs within each zone, or alternatively, that a more meaningful approach is needed to adequately portray the ecology and climatic zonation of the territory.
The Yukon has lagged behind other western Canadian jurisdictions with respect to the development of a relatively large-scale climatic zonation (e.g., 1:1 000 000 scale). In comparison, the biogeoclimatic classification of British Columbia was initiated by V.J. Krajina in 1949 (Meidinger and Pojar, 1991:52), whereas Alberta introduced its initial climatic classification in 1981 (Strong, 1992) and was followed shortly thereafter by other Prairie Provinces. Although these and other classifications sometimes used different nomenclature and were occasionally based on different ecological philosophies, their overall intent was to depict climatic zonation. Creation of a similar ecological framework for the Yukon has probably been delayed because of its remote location, sparse population (-0.07 people/km2), minimal ground access, and limited internal research capacity. The growing interest in mineral extraction (gold, silver, copper, lead, and zinc) and related developments that compete with other interests could necessitate a more comprehensive understanding of ecological resources than currently exists in order to manage environmental impacts. The objectives of this analysis were (1) to develop a more comprehensive regional ecological classification of the Yukon than currently exists based on vegetation; (2) to develop an ecological topographic sequence of plant communities (an ecocline) to characterize the vegetation and ecology of each region; (3) to identify the climatically representative vegetation; and (4) to assess temperature and precipitation data to evaluate the climatic distinctiveness of the vegetation-derived regions.
MATERIALS AND METHODS
The Yukon (482 443 [km.sup.2]) is found in northwestern Canada, east of the State of Alaska (USA) and west of the Northwest Territories between 600 and 700 N latitude. The southern two-thirds of the territory is dominated by NNW--SSE trending mountains that range up to 5959 m in elevation, with sea level along the northern boundary. Summers are cool and relatively short, and winters are cold and five or more months long. Whitehorse (60.7[degrees] N) has average May--August temperatures of 11[degrees]C with Decem-ber--March temperatures averaging -13[degrees]C. In contrast, temperatures at Shingle Point (68.9[degrees] N) are about 6[degrees]C colder in May--August and 11[degrees]C colder in December--March (Environment Canada, 2011a). Because of the cold climate, evergreen tree species such as Picea albertiana ssp. albertiana (Strong and Hills, 2006) (western white spruce; equivalent to Picea glauca x engelmannii. Cody, 2000) and Picea mariana (black spruce) are common components of the vegetation.
Classification and Mapping
The fundamental unit of classification (1:1 000 000 scale) was the ecoclimatic region (originally termed a land region, but revised to ecoregion in the late 1970s) and was defined as an area "characterized by a distinctive regional climate as expressed by vegetation" (Lacate, 1969:4). The phrase "eco-climatic region" was selected for use in this study to avoid confusion with those applications of the term "ecoregion" that do not conform to its original definition (cf. Bailey et al., 1985; Ecoregions Working Group, 1989: footnote 1; Ecological Stratification Working Group, 1995:3). Recognition of climatically representative vegetation was based on sites that were neither protected from nor exposed to local climatic extremes" with "neither a lack nor an excess of soil nutrients" (Ecoregions Working Group, 1989:1). Mature seral vegetation was used to differentiate regions because it provided greater potential for separating climatically and ecologically different areas than climatic climax vegetation, which is similar throughout much of the territory. Terrain with gradients of less than 5% and medium-textured soils with submesic to mesic moisture regimes were considered zonal sites, when available. Where only gradients with slopes of more than 5% occurred, vegetation that consistently occurred on multiple aspects, at similar elevations, and with submesic to mesic moisture regimes was considered to be climatically representative. In northern latitudes with Cryosolic soils (nomenclature follows Soil Classification Working Group, 1998), sites on low-gradient terrain with submesic to subhygric midsummer active layers were considered to be climatically representative. Because it was not possible to identify all climatically representative ecosystems using conventional zonal criteria, all climatically representative ecosystems are referred to as reference sites. Such "sites" represent a combination of ecological conditions rather than any specific geographical location (Natural Resources Canada, 2012). Reference sites were not necessarily the most abundant ecosystems within an ecocli-matic region.
To facilitate the systematic summarization of synecological relationships by ecoclimatic region, an ecoclinal sequence was compiled for each region according to a gradient of ecological moisture regimes (Luttmerding et al., 1990:35). These sequences were based primarily on what were perceived as recurring plant communities and secondarily on slope aspect. Sequences were developed from field-observed ecosystems at multiple locations within each region. If a region was not accessible or an ecocline contained gaps with respect to moisture regime classes, sequences were either compiled or supplemented with ecosystems from other ecological studies. The basic unit of plant community description was the vegetation type, which represents plant communities named according to dominant species by stratum (Classification Level VI: National Vegetation Working Group, 1990:7). Vegetation type naming followed the convention of listing dominant species by stratum (i.e., uppermost first), with co-dominants listed according to their relative abundance. Stratum changes were noted by a slash (/) between species names, and a dash (-) was used to indicate co-occurring species within a stratum. Botanical nomenclature was based on Cody (2000), Anderson et al. (1990), and Brodo et al. (2001), unless otherwise indicated.
Differentiation and mapping of ecoclimatic regions was based on location and distribution of reference sites and associated environmental indicators. For example, Arctic and alpine areas were differentiated from boreal, subalpine, and subarctic areas on the basis of whether the tree line was latitudinal or elevational. If a physical boundary such as a mountain ridge was not present, boundaries between subarctic and the most northerly boreal region were somewhat arbitrary because vegetation is similar in these adjoining regions. Subalpine ecosystems between alpine and boreal regions were recognized on the basis of the open structure of the subalpine tree canopy compared to boreal areas and location on mid-elevation slopes in mountainous terrain. Mapping and differentiation of boreal ecoclimatic regions relied on the interpretation of mapped forest cover data in conjunction with contour map information to identify reference sites that indicated a given ecoclimatic region. When reference sites were lacking, ecoclines were used to assign an area to an ecoclimatic region. An initial mapping of eco-climatic region boundaries was reviewed on the ground, where possible, and revised as necessary. Ecoclimatic region nomenclature approximated the system used by the Ecoreaions Working Group (1989).
Mapping information was derived from the synthesis and interpretation of forest cover maps (Yukon Forest Management Branch, 2009), publicly available synecological studies, and aerial photographs of various ages and scales (Yukon Energy, Mines and Resources Library, Whitehorse). Field investigations during the summers of 2010 and 2011 included primarily areas near readily accessible roads; one exception was a hiking trip into the Kotaneelee area in the extreme southeast corner of the territory. The availability of reference materials for mapping was sufficient to cover all of the territory at least once. Satellite imagery was not used because of its small scale, its limited resolution, and the greater difficulty of differentiating vegetation occurrence patterns in the landscape with this imagery compared to aerial photography.
Alpine Tree Line
Distinguishing between elevation-induced climatic or alpine tree lines and geomorphic tree lines that result from steep slopes, substrate factors, and slope instability was often problematic. To provide an objective method for mapping the lower limit of the alpine climatic zone, up to five pairs of north and south tree line elevations were compiled for each 1:250 000 scale Canada National Topographic Series (NTS) map of Yukon areas located south of the Arctic zone. After a stereoscopic review of aerial photographs, those locations considered to be the best available examples were selected for measurement. Whenever possible, measurements were done on opposing slopes that occurred in close proximity. Once identified, each aerial photograph location was compared with the corresponding 1:50000 scale NTS map to determine its elevation.
Models of tree line elevations were constructed using regression analysis based on the average value of each north-south tree line pair. A three-dimensional regression model using quadratic smoothing was developed for areas south of 66[degrees] N, whereas a simple linear model was found adequate for areas north of 64[degrees] (Fig. 1). The reliability of the three-dimensional model was evaluated on the basis of the cdrrelation between observed values and values predicted by the regression. The amount of variance explained by the models was considered equal to the square of the associated coefficient of correlation (r or R) . STA-TISTICA computer software (Statsoft, 1995) was used for these analyses. Mapping of climatic tree lines was based on regression-predicted elevations, plus one standard error (SE), that were specific to individual 1:50000-scale NTS maps.
Climate Data Analysis
Among the available records from Yukon meteorological stations (Environment Canada, 2011b), the longest temperature and precipitation recording period with the most stations and fewest missing monthly values was judged to extend from 1984 to 2007. All 26 included stations had missing monthly data (average~10%), and Environment Canada assigned monthly values that were derived from incomplete sets of daily measurement. It was necessary to accept these assigned values as accurate; otherwise, no analysis (or only a very limited one) would have been possible. Missing values were estimated either by incorporating values from a nearby station (e.g., merger of Ross River YTG and Ross River A station records) or by using a control station. For the latter, Pelly Ranch (station 2100880) was used because it had nearly complete monthly temperature and precipitation records (1 missing value and 19 Environment Canada assigned values, mostly due to a single missing daily measurement within a given month), and because it was more centrally located within the territory (62.817[degrees] N, 137.367[degrees] W) than another comparable station (Mayo Road). Missing values were estimated ([E.sub.ij]) using the algorithm:
[E.sub.ij] = [D.sub.ij] + ([C.sub.j] - [M.sub.j])
where D equals the reported monthly value of the control station, C equals the average monthly value for the control station (1984-2007), M equals the average monthly value for the station with the missing value based on the reported data, i equals year, and j represents the calendar month.
Kruskal-Wallis tests were used to determine whether significant (p < 0.05) differences occurred among ecocli-matic regions with two or more meteorological stations. Comparisons were based on STATISTICA computer software (Statsoft, 1995). Scheffe rank tests ([alpha] = 0.050, Miller, 1966: formula 110) were used to identify which regions differed within significant Kruskal-Wallis tests. Monthly average temperature and total precipitation data were ordinated using detrended correspondence analysis (McCune and Mefford, 1999) to illustrate relative relationships among meteorological recording locations and their associated eco-climatic region. To ordinate the meteorological data, it was necessary to reduce the analysis time frame to 1987-2007 to accommodate software limitations. Explained ordination variance was determined using relative Euclidean distance.
RESULTS AND INTERPRETATION
Eleven ecoclimatic regions were recognized in six broadly defined ecoclimatic provinces (sensu Ecoregions Working Group, 1989). These six provinces, listed from south to north, are Mid-Boreal, Mid-Cordilleran, Northern Pacific Cordilleran, Northern Cordilleran, Subarctic, and Arctic (Fig. 2). Excluding consideration of high-elevation regions, the six ecoclimatic provinces generally formed latitudinal bands. The ecological characteristics of the eleven ecoclimatic regions are summarized below.
Mid-Boreal Subhumid (MBs): The MBs region occurs in the extreme southeast corner of the Yukon (Fig. 2) and is the smallest region in the territory (Table 1). The reference vegetation is forest stands that are dominated by 20-30 m tall, closed-canopied, Populzis trenmloides (aspen) and also include a 1-2 m tall shrub stratum of Viburnum edzde (low-bush cranberry) and Rosa acicularis (prickly rose) (Fig. 3). Herbs such as Aralia nudicaulis (wild sarsaparilla) and Rubus pubescens (dwarf raspberry) commonly occur in the understory vegetation (Jeffrey, 1964:27), as well as Calamagrostis canadensis (marsh reedgrass). The under-story vegetation of drier P. tremuloides stands is dominated by R. acicularis. Subhygric sites develop Populus balsamif: era (balsam poplar) stands with Alnus incana (river alder) and Salix spp. (willows) that are often more than 3 m tall, and V edule. These sites sometimes include such Yukon rarities as Maianthemum canadensis (wild lily-of-the-valley) (Cody, 2000:616) and Matteuccia struthiopteris (ostrich fern). Betula papyrifrra (paper birch) trees are common in this ecoclimatic region. Upland forests are rapidly succeeded by Picea albertiana, which at maturity can reach heights of 25-35 m. Besides having deciduous rather than coniferous forest reference vegetation (cf. MCb), the MBs region is structurally more diverse, botanically richer, and more productive in terms of phytomass than other regions in the Yukon. Gray Luvisolic soils occur on reference sites, which is also atypical of the territory. Picea albertianal Hylocomium splendens (stairstep moss) is the potential climatic climax vegetation.
TABLE 1. Estimated percent areal extent of ecoelimatic regions in Yukon based on Figure 2. Ecoclimatic region Area (%) MBs Mid-Boreal Subhum id 0.1 MCb Mid-Cordilleran Boreal 8.9 MCs Mid-Cord illeran Subalpine 12.8 MCa Mid-Cordilleran Alpine 3.0 NPg Northern Pacific Cordilleran Glacierized 3.4 NCb Northern Cordilleran Boreal 2.9 NCh Northern Cordilleran High Boreal 32.8 NCa Northern Cordilleran Alpine 9.2 HS High Subarctic 20.3 LA Low Arctic 3.6 LAa Low Arctic Alpine 3.0
Mid-Cordilleran Boreal (MCb): This region occurs primarily south of 61[degrees] N at low elevations and in valley bottoms (Fig. 2). Its reference sites have closed and semiclosed canopied stands of Pinus contorta ssp. latifolia (lodgepole pine). Mature P. contorta seldom exceed 20 m in height and are typically shorter. In the western portion of the region, the understory vegetation is characterized by Shepherdia canadensis (buffaloberry), whereas Ledum groenlandicum (Labrador tea) is more prominent on reference sites in the eastern portion (Fig. 3). Associated under-story species commonly include Vaccinium vitis-idaea (bog cranberry), Cornus canadensis (bunchberry), Hylo-comium splendens, and Pleurozium schreberi (Schreber's moss) (Strong, 2002). Calamagrostis purpurascens (purple reedgrass) also occurs in association with S. canadensis. Brunisolic soils are characteristic of MCb reference sites, although Gray Luvisolic soils may commonly occur in the eastern extreme of the region (White et al., 1992). The lack of P. contorta at low elevations in the far western portion of the region (Champagne to Haines Junction area) is likely the result of high soil NaC1 and pH levels (Day, 1962). Picea alhertiana, Picea mariana, and sometime Abies lasiocarpa with Hylocomium splendens are the potential climatic climax species on submesic to hygric sites.
Stands on submesic coarse-textured soils develop a ground cover of Arctostaphylos uva-ursi (bearberry), whereas drier sites have reduced P. contorta cover and an abundance of Cladina spp. (reindeer lichens) (Fig. 3). In contrast, warm south-facing slopes develop Populus trem-uloides1Rosa acicularis and P. tremuloidesIA. uva-ursi stands, with the latter occupying the driest sites. C. purpu-raseens--A. uva-ursi vegetation occurs on subxeric steep southfacing slopes. Wetlands of Salix spp. and Carex aqua-ti/is (water sedge) develop where early summer flooding and water pooling occur, whereas P. mariana-dominated vegetation develops where near-surface soils are wet, but not continuously flooded with water, and often have a poor nutrient status (Fig. 3).
Mid-Cordilleran Subalpine (MCs): The MCs ecocli-matic region occurs primarily west of 1290 W and south of 620 N, although unmapped examples may extend into the Dawson area (Fig. 2). Topographically, this region occurs at mid-elevations in mountainous terrain above boreal regions (MCb, NCb, and NCh) and below alpine regions (MCa and NCa areas < 64.5[degrees] N). The MCs region has an elevation range of less than 400 m: Scattered, stout conifer trees 13 m or more in height occur on reference sites within a matrix of tall (1.5-3 m), relatively dense Betula glandulosa (dwarf birch) and Salix spp. (Danby and Hik, 2007: Fig. 2b). Picea albertiana is the most common tree, but Abies lasiocarpa (subalpine fir), Pinus contorta var. yukonensis (Yukon pine; Strong, 2010), Pinus contorta spp. latifolia, and Picea mar-iana also occur. Picea mariana is more frequent near the NCh zone than near other ecoclimatic regions, whereas A. lasiocarpa is most common at higher elevations. Tree and shrub heights decrease with increasing elevation. In the core of its geographical range, ground vegetation on reference sites is often modest in abundance and composed of scattered forbs, graminoids, ericaceous shrubs, and lichens (e.g., Cladonia spp.). High-elevation sites, north aspects, and more northern locations (e.g., in NCh) have less herb and greater moss cover (Ayotte, 2002:84). Steep north aspects at lower elevations can develop tall (5-6 m) Salix stands that include scattered conifer trees. Grass patches and Populus tremuloides stands occasionally occur on southwest slopes, which are warmer and drier than reference sites (Fig. 3). Eutric Brunisolic soils appear to be the most abundant soil great group, but Dystric Brunisols also occur. Cryosolic soils are more frequent in proximity to the NCh on steep north aspects.
Mid-Cordilleran Alpine (MCa): Mid-Cordilleran Alpine has a discontinuous distribution in southern Yukon because it occurs only at high elevations. Because it was difficult to confirm the extent of the MCa region, the working model arbitrarily limited its occurrences to southern Yukon areas with MCs vegetation (Fig. 2). This arbitrary limit was necessary because few alpine vegetation studies have been conducted in Yukon and ground access to the region's more than 200 map polygons was very limited. Many small alpine areas were unmappable at a 1:1 000 000 scale. The lower elevation limit of the MCa increases from--1100 m in the southwest portion of the territory to--1480 m in the eastern portion of its range (Fig. la). Tree lines are higher on south aspects than on north aspects, with a differential of 110-155 m (based on second and third quartile data values; n = 188). This differential did not vary with latitude (based on correlation p> 0.050). Vegetation in the MCa region is highly variable in composition because of macro- and microtopographic varigbility and the associated microclimatic differences. Vegetation cover and stature decrease with increasing elevation.
Reference sites are dominated by moderately dense stands of Betula glandulosa and Salix spp. with a height of less than 1 m (Fig. 3). Ground species in this vegetation include Vaccinium uliginosum (bog bilberry), Empetrum nigrum (crowberry), Cladonia spp. (lichens), Flavocetraria cucullata (curled snow lichen), and Polytrichum spp. (hair-cap mosses). Sites that are exposed to wind or with coarse-textured soils often have Dryas integrijblia (smooth-leaved mountain-aven) as the dominant species, whereas open-growing Betula glandulosa stands with Stereocaulon spp. (coral lichens) occur on thin dry soils. Dryas vegetation is typically under 15 cm tall. Steep southerly aspects with warm and dry conditions develop Festuca altaica--Arte-misia norvegica (northern rough fescue--wormwood) vegetation. Kobresia myosuroides (kobresia) vegetation occurs on sites with subxeric conditions and limited winter snow cover (Douglas et al., 1980). On moist sites, open-growing tall (~1 m tall) and low-growing (< 10 cm tall) Salix-dominant vegetations develop in response to surface water flow and melting snowbeds, respectively (Fig. 3). The former vegetation often includes a diverse mixture of mesophytic herbs such as Polemonium acutillorum (Jacob's-ladder) and Senecio triangularis (groundsel). Cas-siope tetragona (white heather) stands are usually associated with cool northern aspect. Soil profile development is often limited by the cold climate and shallow surficial materials.
North Pacific Cordilleran Glacierized (NPg): The majority of this region occurs in the St. Elias Mountains (southwest Yukon) within or near Kluane National Park. Small disjunct examples also occur between--62.5[degrees] and 64.5[degrees] N at high elevations along the east side of the territory (Fig. 2). The NPg occurs above the MCa and NCa regions and is differentiated from them by the occurrence of icefields. Areas not occupied by glaciers are likely unvegetated, or at best, have a spare, discontinuous cover of low-growing plants (e.g., mosses, lichens, or very cold-hardy vascular plants). Where unconsolidated surficial materials occur, Cryosolic (Douglas et al., 1980:189) and Regosolic soils are common.
Northern Cordilleran Boreal (NCb): The NCb region occurs in the central portion of southern Yukon (Fig. 2) in a topographic depression between the east and west flanks of the Yukon Plateau (Bostock, 1948). The region is surrounded by the NCh region on three sides and abuts the MCb region at its southern limit. Its upper elevation limit is formed by the MCs region.
Mature seral vegetation on reference sites is dominated by closed-canopied Populus tremuloides trees (< 15 in tall), often with a large proportion of Picea albertiana seedlings and trees. Such stands are usually of fire origin. Medium height (< 50 cm tall) Shepherdia canadensis, Rosa aciczi-laris, and Calamagrostis purpurascens form the under-story vegetation, with a scattered cover of 2-5 m tall Salix (Strong, 2009). Populus tremuloides stands tend to contain only 10-20 vascular and nonvascular plant species, with a combined foliar cover of 50% or less. Such stands are floristically poor compared to those of southern boreal forests (Strong and Redburn, 2009). Picea albertiana establish early in the successional development of reference stands and reach heights of 20-25 m. As a result, P. albertiana can replace P. tremuloides as the dominant trees within 100 years after stand initiation (Strong, 2009). Climax under-story vegetation is dominated by Hylocomium splendens in conjunction with species such as Peltigera aphthosa and P. malacea (dog-tongue lichens). Eutric Brunisolic soils are characteristic of upland sites.
Dry south-facing slopes develop P. tremuloideslArcto-staphylos uva-ursi stands, whereas Calamagrostis purpu-rascens--Artemisia frigida (pasture sage) vegetation is a common landscape component on steep southerly aspects. Sanborn (2010) suggests that some of the Brunisolic-like soils that occur in association with C. purpurascens could technically be considered Dark Brown Chernozems. Pinus contorta ssp. latifolia stands also occur in the region, but appear to be limited to coarse-textured soils and steep south-facing slopes. Salix spp. and Carex aquatilis are the principal species of wetlands (Fig. 4).
Northern Cordilleran High Boreal (NCh): NCh occurs north of the MCb region to about 64.5' N (Fig. 2) and represents the most northern component of the boreal forest. The region is ecologically complex because of its diverse terrain, which ranges from undulating and rolling topography to steep mountains slopes. MCs and Northern Cordilleran Alpine (NCa) form the upper elevation limit of this region. The boundary between the NCh and. High Subarctic (HS) regions is somewhat arbitrary because both include the same tree species. Occurrence of the NCh is more strongly determined by elevation than by latitude, compared to HS. Tree densities and heights are greater on reference sites in the NCh than in the HS, although they decrease with increasing elevation. The NCh comprises one-third of the Yukon Territory (Table 1).
Reference sites have moderately dense (< 10 m tall) Picea mariana stands that often include Picea albertiana. The ground vegetation is dominated by Ledum groenlandi-cum and Ilylocomium splendens. In the eastern half of the region, Betula glandulosa is a common tall shrub in this vegetation (Fig. 4). The soils are primarily Cryosols, but Dystric Brunisols sometime occur. Non-permafrost, mesic (not south-facing) sites within areas dominated by Picea mariana sometimes develop Betula papyri/era stands. In contrast, south-facing slopes with Populus tremuloides stands will be replaced by P. albertiana. Picea albertiana trees on southerly slopes (a common occurrence) are often taller (15-20 m) and larger than those of reference sites. Steeply sloping sites develop graminoid vegetation dominated by Calamagrostis purpurascens, Festuca altaica, or sometimes Elymus spp. (wheatgrasses). Artemisia spp. (sage or wormwood) are an abundant secondary component of these grass stands. Coarse-textured soils can develop open-growing P. albertiana stands, with a relatively continuous stratum of ground lichens such as Cladina rangife-rina (grey reindeer), Cladina stellaris (star-tipped reindeer), and Flavocetraria cucullata (Kojima, 1996: Fig. 3b). Plant communities intermediate to the P. albertiana/Cladina and reference vegetation likely occur, but were not observed or found in the literature searched. Subhygric to subhy-dric sites, steep north-facing slopes, and depressions are typically vegetated by open-growing stunted (< 5 m tall) P. mariana, with a thick moss stratum that insulates the underlying permafrost. Saliv spp. and Carex aquatilis stands occur on subhydric and hydric sites, respectively.
Northern Cordilleran Alpine (NCa): Examples of the NCa ecoclimatic region occur throughout the NCh and HS in the Yukon, with the greatest concentrations in the Ogilvie and Selwyn Mountains (Bostock, 1948), northeast of Kluane National Park, and along the eastern edge of the territory south of 63[degrees] N (Fig. 2). This region occurs at elevations ranging from 1480 m in the southeast to--665 m in the north (Fig. 1), which represents a 175 m decrease per degree of latitude increase from 64[degrees] to 68[degrees] N (Fig. lb). A 110-155 m differential in tree line elevation occurs between north and south slopes. The NCa is the most extensive alpine ecoclimatic region in the Yukon (Table 1).
Stands of Betula glandulosa and Salix spp. less than 1 m tall, with a ground stratum of lichens and feather-mosses, form the reference vegetation (Fig. 4). Graminoids and forbs are normally absent or have very limited cover. Cryosolic soils are associated with reference sites. Ilylocomium spkndens is the dominant ground species on low-gradient north aspects, whereas Stereocaulon spp, are associated with drier sites. Low-growing Dryas spp. and Salix spp. vegetation develops on southerly slopes with submesic to subxeric moisture regimes that lack permafrost. Physiognomically similar vegetation occurs on north aspects, but includes Carex misandra (sedge) and Cassiope tetragona with Dryas octopetala white mountain-aven) (Fig. 4).
High Subarctic (HS): The HS ecoclimatic region occurs geographically and ecologically between the Low Arctic and the NCh, and spans 4.5[degrees] of latitude in the northern one-third of the territory (Fig. 2). The latitudinal tree line forms the northern limit of the region (slightly modified from Wiken et al., 1981). Open-growing and stunted evergreen forests dominate the landscape, with tree heights and densities decreasing with increasing latitude. Permafrost is common in the landscape. HS is the second largest ecoclimatic region in the Yukon (Table 1).
Reference sites have open-growing, stunted (5-10 m tall) Picea mariana trees. Spaces between trees are occupied by medium-height (1-1.5 m tall) Betula glandulosa, sometimes with Salix spp. The ground vegetation includes cold-hardy species such as Ledum groenlandicum, Vaccin-ium vitis-idaea, Hylocomium splendens, and Cladonia spp. (Stanek et al., 1981:16-19; Boggs and Sturdy, 2005:57-60). Picea mariana stands with Picea albertiana occur on coarse-textured soils that develop a lichen ground stratum, whereas open-growing P. albertiana stands occur on steep southerly aspects. Open-growing 3-5 m tall P. mariana in association with Eriophorum spp. (cottongrasses) and Sphagnum spp. (peat mosses) develop on subhygric to hygric sites. Similar stands dominated by Larix lariciana (tamarack) occur on hygric sites. Reference and wetter sites have Cryosolic soils, with Brunisols occurring on drier sites.
Low Arctic (LA): The Low Arctic region occupies a--120 km wide band south of the Arctic Ocean and extends southward along the lower slopes of the Richardson Mountain (Fig. 2). The southern boundary is formed by the latitudinal tree line. LA is predicted to occur below 600 and 470 m elevation (south and north, respectively) in the western portions of its range, and below 1000 m around the southern Richardson Mountains on the basis of Fig. lb. Permafrost is relatively continuous in this ecoclimatic region.
Reference sites have tussocky tundra dominated by Erio-phorum vaginatum--Ledum decumbens--Betula glandulosa (sheathed cottongrass--northern Labrador tea--dwarf birch) vegetation (<30 cm tall) (Fig. 5). Low-growing (< 10 cm) Salix arctica--and S. reticulata--Dryas integrifolia (arctic willow--and net-veined willow--smooth-leaved mountain-aven) vegetation is associated with submesic sites, whereas--1 m tall Salix alaxensis (felt-leaved willow), B. glandu-losa, and Alnus crispa (green alder) stands can occur on drier sites (Zoltai and Pettapiece, 1973; Wiken et al., 1981). Scattered and stunted Picea albertiana are sometimes present on steep southerly aspects at low elevations, where the topography enhances the local thermal regime. Sites wetter than the reference sites are vegetated by graminoids with hydrophilic mosses. Submesic and wetter sites have Cryo-solic soils, with the active layer increasing in depth with better site drainage (Zoltai and Pettapiece, 1973:19). On Herschel Island, Smith et al. (1989) reported active layers of 45-50 cm on reference sites.
Low Arctic Alpine (LAa): Recognition of the Low Arctic Alpine ecoclimatic region is somewhat theoretical, based on the assumption that high-elevation sites have different climatic regimes compared to low elevation sites because of adiabatic cooling, although this region was also recognized by Simpson et al. (2002:1179) in Alaska and by the CAVM Team (2003). Delineation of this region was based on the northward projection of the lower limit of alpine vegetation using the regression model presented in Fig. lb (see LA summary for lower elevation limits). This ecoclimatic region has a discontinuous distribution at mid to upper elevations in the British and Richardson Mountains (Fig. 2). It also appears to occur at high elevations along the north face of the Ogilvie Mountains in the vicinity of Tombstone Territorial Park in central Yukon. Much of the landscape consists of colluvium, stone stripes, frost polygons, and associated cryoturbation features. Unvegetated and sparsely vegetated areas are common (Wiken et al., 1981).
This ecoclimatic region has tundra vegetation that is dominated by low-growing shrubs (Wiken et al., 1981:18) with graminoids, ground lichens, and mosses. Reference sites appear to be a mixture of Betula glandulosa and Ledum decumbens with Eriophorum vaginatum. Dryas octopetala and low-growing shrubs such as Salix reticulata with SaWraga oppositifolia (purple saxifrage) occur on submesic sites. The reference vegetation of the LA region occurs on subhygric sites in the LAa (Fig. 5). Wet sites develop Carex aquatills-dominated stands. Salix spp. and open-growing Picea albertiana stands occasionally occur on dry and warm sites, respectively. Turbic Cryosolic soils predominate on submesic and subhygric upland sites, with a 35-75 cm deep active layer (Wiken et al., 1981).
The warmest monthly temperatures (15[degrees]C) within the territory occurred in the NCb and NCh regions during July (Table 2). In June and August, temperatures were similar within the MCb and HS regions, but distinctly colder in the LA. Although data were available for only one MCa meteorological station (Blanchard River near British Columbia border and east of Kluane National Park), summer temperatures appeared cooler than in other regions (Table 2) except the LA. In contrast, temperatures in the NPg were likely much colder than in MCa or LA areas (Marcus and LaBelle, 1970:110). January temperatures were 32[degrees]-40[degrees]C colder than July values. MCb areas had the warmest winter (November--March) values (-8[degrees] to -18[degrees]C) in the territory. Overall winter temperatures were 10[degrees]C colder in northern than in southern Yukon (Table 2). Winter MCa temperatures appeared to be similar to or warmer than those in the MCb region. On an annual basis, a gradient of decreasing annual temperature with increasing latitude was apparent among regions, with temperatures ranging from -1.1[degrees] to -9.9[degrees]C (Table 2). Despite similar monthly and seasonal temperatures in the NCb and NCh regions, annual temperatures were--.1.1[degrees]C warmer in the NCb.
Monthly summer precipitation was greatest in the HS and NCh regions. Peak precipitation occurred in July, with 24-33% less precipitation during June and August. For the one MCa station, maximum precipitation occurred in September. Total summer precipitation was similar in the HS and NCh (-205 mm), whereas the NCb and MCb regions received about 173 mm. Although NCb precipitation totals were only about 16% less than NCh totals, the NCb region occurred within an area considered by Jatzold (2000:6) to have the potential for severe evaporation deficits in April--June. Winter precipitation was 22 mm or less (water equivalent) per month in subarctic and boreal regions, with total winter precipitation of 72-92 mm. Northern regions received--2.4 times more summer than winter precipitation, but the differential was less in Mid-Cordilleran regions (Table 2). Although significant differences occurred among regions, the range in annual total precipitation was not great (-300 [+ or -] 21 mm or 7%).
TABLE 2. Median monthly temperatures and total precipitation of Yukon ecoclimatic regions based on 1984-2007 meteorological data, with comparisons using Kruskal-Wallis tests. (1) Summer Ecoclimatic Number of Apr May June July Aug Region stations Median Temperature ([degrees] C) Low Arctic 2 - 16a - 4a 6a 10a 8a (3) High Subarctic 3 - 9a 3b 12b 14b 10b Northern 8 0b 7c 13bc 15c 12c Cordilleran High Boreal Northern 3 lb 7c 13c 15c 12c Cordilleran Boreal Mid - 9 lb 7c 12b 14b l2c Cordilleran Boreal Mid - 1 p 4 9 11 9 Cordilleran Alpine (4) Total P <0.00l O.001 <0.00l <0.001 <0.001 Precipitation (mm) High Subarctic 3 8b 20a 43ab 52bc 53c Northern 8 8ab 25b 42b 61c 4lbc Cordilleran High Boreal Northern 3 5a 22ah 31a 46ab 34ab Cordilleran Boreal Mid - 9 5a 14a 32a 42a 38a Cordilleran Boreal Mid - 1 p 16 36 45 45 Cordilleran Alpine (4) 0.001 <0.00l <0.001 <0.0Ol <0.001 Winter Ecoclimatic Sept Oct Nov Dec Jan Feb Mar Region Median Temperature ([degrees] C) Low Arctic 2a - 8a - 19ab - 22ab - 24a - 25a - 24a b High Subarctic 4a - 9a - 22a - 23a - 26a - 24a - 20a Northern 6b - 4b - 16bc - 20bc - 23b - 18b - 12b Cordilleran High Boreal Northern 6bc - 2c - 15c - 18c - 22b - 16b - l0bc Cordilleran Boreal Mid - 7c 0d - l1d - I4d - 18c - 13c - 8c Cordilleran Boreal Mid - 5 - 1 - 10 - 10 - 14 - 11 - 9 Cordilleran Alpine (4) Total <0.00l <0.001 <0.001 <0.00l <0.001 <0.001 <0.001 Precipitation (mm) High Subarctic 32a 30b 16a 20ab 13a 14a 11b Northern 30a 24ab 20a 21b 16ab 12a 10ab Cordilleran High Boreal Northern 29a 20a 18a 16a [4a 10a 8a Cordilleran Boreal Mid - 32a 25ab 21a 20b 22b 14a 9ab Cordilleran Boreal Mid - 73 54 49 29 46 31 22 Cordilleran Alpine (4) 0.810 0.001 0.008 0.011 <0.001 0.021 0.004 Season (2) Ecoclimatic Summer Winter Year Region Median Temperature ([degrees] C) Low Arctic 4a - 23a - 9.9a High Subarctic 8b - 23 a - 7.9a Northern Kid - 17b - 3.4b Cordilleran High Boreal Northern lid - 16b - 2.3c Cordilleran Boreal Mid - 10c - 13c - 1.1d Cordilleran Boreal Mid - 8 - 10 - 1.5 Cordilleran Alpine (4) Total <0.001 < 0.001 <0.001 Precipitation (mm) High Subarctic 203 b 84b 322b Northern 208b 86b 324b Cordilleran High Boreal Northern 177a 72a 279a Cordilleran Boreal Mid - 169a 92b 306a Cordilleran Boreal Mid - 208 224 522 Cordilleran Alpine (4) 0.001 <0.001 <0.001 (1.) Only ecoclimatic regions with meteorological data were included in the summary. (2.) Summer is defined as May through September, and winter, as November through March. Summer included those months with median temperatures above 1[degrees]C except in the Low Arctic, where temperatures were below 1[degrees]C in May. April and October were considered seasonal transitional months. (3.) Ecoclimatic regions by time period followed by the same letter do not differ at [alpha] = 0.050 based on Scheffe rank tests. (4.) Excluded from Kruskal-Wallis test.
Climate Data Ordination
Figure 6a illustrates the relative similarity of assessed meteorological recording locations based on ordinated 1984-2007 monthly temperatures. The horizontal axis appeared to represent a gradient of increasing latitude from right to left and decreasing temperature from bottom to top. The ordination explained 89% of the variance in temperatures, with the horizontal axis accounting for three-fourths of the total. Meteorological recording locations within the same ecoclimatic region generally occurred in closer proximity than those of different regions (Fig. 6a). The location of individual regions approximated their relative spatial sequence within the Yukon landscape based on an upward arc from MCb through NCb and NCh to LA. The position of meteorological recording locations in the ordination suggested that NCb temperatures were more closely associated with the NCh regime than with the MCb regime. Of the 26 assessed locations, only Watson Lake was inconsistently located with respect to its ecoclimatic region classification (Fig. 6a). This anomaly could be due to its location adjacent to a lake, or its greater exposure to winter Arctic fronts than other MCb locations. The Klondike (NCh) and Eagle Plains (HS) sites were disjunct from other members of their region, but also represented the most northern and highest-elevation recording sites, when adjusted for latitude, within their respective regions.
The ordination of monthly total precipitation in Figure 6b, although graphically somewhat different from the temperature data in Figure 6a, suggested similar ecological trends. MCb sites occurred in the lower portion of the ordination, with NCb and HS sites progressively higher along the vertical axis, suggesting a latitudinal gradient. NCb sites occurred between the NCh and MCb regions. No precipitation data were available for the LA. The horizontal axis appeared to represent a complex gradient based on temporal variability in precipitation regimes. The Figure 6b ordination explained 59% of the variance in data (horizontal axis = 32%). Of the 24 sites assessed, Swift River represented the only inconsistency with respect to classification. Swift River was closely associated with three NCh sites (Fig. 6b), although it is classified as MCb and occurred along the Yukon--British Columbia border. Extensive areas of the MCs and MCa regions occurred close to the Swift River location; therefore, it might be subject to somewhat greater precipitation than other MCb areas.
The original intent of this study was to develop an ecologically based climatic zonation of Yukon based on vegetation, independent of any existing classification, using modified zonal site criteria. Once this zonation was compiled, however, a strong similarity was apparent with the Ecoregions Working Group (1989) classification based on the location and vegetation characteristics of the mapped polygons. The Rowe (1972) classification had broad similarities to Figure 2 at the forest region level. A weak resemblance to the Oswald and Senyk (1977) and the Ecological Stratification Working Group (1995) classifications occurred in part because physiography was excluded as a variable in the current classification. The effects of physiography and topographic variation were indirectly reflected in the ecological response of vegetation to changing local and regional elevation. Although these latter classifications recognized the existence of definable elevation-induced climate zones within "ecoregions," few were formally classified or mapped, nor were quantitative assessments done to identify climatic differences. Several differences occurred between Figure 2 and the Ecoregions Working Group (1989) classification. Beside greater mapping detail, these differences included (1) reclassification of the Alpine Northern Subarctic Cordilleran or NSCa as LAa; (2) subdivision and reclassification of the original NCb as NCb and NCh; (3) northward expansion of the MCb from northern British Columbia; and (4) reduction in area of the MBs. Most importantly, quantifiable temperature and precipitation differences were identified among tested ecoclimatic regions.
Despite the occurrence of NNW--SSE trending mountains, latitudinal climatic zones were apparent in Figure 2, if elevation-induced ecoclimatic regions are overlooked: MCb [right arrow]NCh[right arrow]HS[right arrow]LA (south to north). The LA and HS regions were comparable to those recognized by the Ecore-gions Working Group (1989) in central and eastern Canada. The NCh region was compatible with the characteristics of the High Boreal (HB) ecoclimatic province on the basis of prominent Picea mariana with ericaceous shrubs and feathermosses on reference sites (Ecoregions Working Group, 1989). Although the MCb region occurred as a latitudinal band across southern Yukon (Fig. 2), it represented the northern limit of an extensive longitudinal transition zone between boreal and cordilleran flora along the eastern slopes of the Rocky Mountains from southwestern Alberta to the Yukon (Strong, 1992, 2002). It is also noteworthy that some elevation-induced regions occurred in more than one ecoclimatic province, usually as a northward extension beyond their primary area of concentration (e.g., MCs occurring in the Northern Cordilleran ecoclimatic province). The ecoclimatic diversity in the Yukon, based on amount of area per region, was greater than in Alberta (43 858 km2 versus 50 860 km2/region; Strong, 1992), but slightly less than in British Columbia (39525 km2/region; Ecoregions Working Group, 1989). British Columbia also had twice as many regions as the Yukon.
The recognized ecoclimatic regions (Fig. 2) are compatible with the vegetation biomes recognized in Alaska (Simpson et al., 2002: Fig. 11b, 2007:344) and subsequently supported by discriminant analysis modeling of climate and environment variables (Simpson et al., 2007:360). The only major difference was subdivision of the Alaskan boreal vegetation biome (also referred to as Subarctic; Simpson et al., 2007:343) into HS and NCh ecoclimatic regions in the Yukon. In some respects, the HS does represent a northward latitudinal continuation of the NCh, when based on vegetation composition, without regard to the physiognomy of the reference vegetation (i.e., stunted open-growing trees versus taller semi-closed tree stands). In support of this subdivision, some of the climatic parameters modeled by Simpson et al. (2002, 2007:347-349), particularly temperatures, tended to point toward a difference between the north and south portions of the Alaskan boreal vegetation biome along the Yukon border. Along the southern Yukon border, three biogeoclimatic zones have been recognized in British Columbia (Meidinger and Pojar, 1991): Alpine Tundra, Spruce--Willow--Birch, and Boreal White and Black Spruce. The Alpine Tundra zone encompassed the NPg, MCa, and NCa ecoclimatic regions, whereas the Spruce--Willow--Birch zone was equivalent to the MCs. The MCb and MBs regions were part of the British Columbia Boreal White and Black Spruce zone and would be recognized at a subzone level of classification (Meidinger and Pojar, 1991:241). Along the Yukon/Northwest Territories border, the majority of adjoining land south of 65.5[degrees] N consisted of alpine ecosystems (NWT Ecosystem Classification Group, 2010:4, 147). In the far south, the NCh region had vegetation similar to the NWT High Boreal and Mid-Boreal ecoregions (NWT Ecosystem Classification Group, 2010:134, 144). Land adjacent to the southeastern tip of the Yukon was not included in the 2010 NWT ecoregion classification. HS was recognized in the north on both sides of the border, except that the southern limit was considered part of the Low Subarctic region in the NWT. It was difficult to match the LA region in the Yukon with its equivalent in the NWT because of the north--south trending Richardson Mountains along the territorial boundary, but it appeared similar to the NWT Southern Arctic ecoregion in location and ecological characteristics (NWT Ecosystem Classification Group, 2010:4). In conclusion, the proposed Yukon ecoclimatic regions were comparable to the ecological units recognized in Alaska and British Columbia and similar to those in the NWT classification. These similarities occurred despite the use of different approaches to classification.
The appropriate construct of an ecological landscape classification system has been debated in North America since at least the late 1960s (e.g., the basis for Lacate, 1969; Orians, 1993; Bailey et al., 1985). At broad scales of classification, a dichotomy of approaches has developed in Canada. One approach is based on climate/vegetation zonation as the initial basis of landscape subdivision (e.g., British Columbia: Meidinger and Pojar, 1991; Alberta: Strong, 1992; Ontario: Mackey et al., 1996; Quebec: Ministere des Ressources Naturelles, de la Faune et des Parcs, 2003), followed by subdivisions based on physiography and landscape patterns down to individual synecological types at increasingly larger scales of analysis. The two uppermost levels of classification are similar in concept to ecoclimatic province and ecoclimatic region. The other (aggregative) approach relies essentially on the simultaneous classification of very broadly defined climate zones and physiography features, referred to as ecozones and ecoregions (Oswald and Senyk, 1977; Ecological Stratification Working Group, 1995). Superficially, the latter approach might seem more comprehensive and ecologically holistic than just using climate as the initial criterion for landscape classification, but climatic zones are major ecosystems with different biological potentials that have developed in response to global circulation patterns and topographic variability. With respect to topographic variability, the critical factor is elevation change, rather than geologic origin or configuration. The aggregative approach may be viable in areas where the terrain is relatively simple, but from a hierarchical classification perspective, it results in the grouping of different climatic and ecological types (e.g., alpine + subalpine + boreal), and the taxonomic segregation of like ecosystems in mountainous areas (e.g., subalpine climatic zones with similar vegetation occur in both Yukon's Pelly Mountain and Southern Lakes ecoregions; Ecological Stratification Working Group, 1995). Even the inclusion of a new classification stratum above the ecoregion level (Marshall et al., 1999) does not appear to resolve these pragmatic concerns. The aggregative approach may be a convenient national and international administrative tool (Marshall et al., 1999; Hirvonen, 2001), but as currently designed, it will lead to the duplication of ecological landscape types at increasingly larger scales of analysis. Solving these problems might prove expensive if users must reconfigure major geographical information system databases to eliminate such taxonomic and ecological inconsistencies. As applied in Yukon, the aggregative approach (e.g., Oswald and Senyk, 1977; Ecological Stratification Working Group, 1995) provides a less ecologically meaningful perspective than a system that uses climate alone as the initial basis for landscape classification.
Successive approximation (sensu Poore, 1962) is a concept that is applicable to the development of ecoclimatic region classifications. It applies because the slow, sporadic, and spatially uneven accumulation of synecological data within a large geographical area often requires multiple iterations to develop a comprehensive classification. It would therefore be unrealistic to consider Figure 2 the final classification for the Yukon; rather, it is the next step. Future research might (1) challenge what was considered reference vegetation in an ecoclimatic region, (2) expand the ecocline sequences, (3) refine boundary locations, (4) subdivide regions (e.g., upper and lower HS and NCh), and (5) evaluate latitude-driven botanical and phytosociologi-cal differences within elevation-induced regions (such as the MCa and NCa) that occur in more than one ecoclimatic province. Without a regional context as a starting point for landscape analysis, little systematic progress will be made toward a comprehensive understanding and management of Yukon's ecological resources.
Yukon Forest Management Branch provided forest cover maps, which proved of great value in the mapping of ecoclimatic regions. Dr. L.V. Hills (University of Calgary), C.E. Kennedy (Yukon Departmebt of Environment), Dr. M. Raynolds (University of Alaska Fairbanks), and another reviewer for the journal provided comments that improved the clarity and completeness of the manuscript. V. Loewen (Yukon Department of Environment), Dr. S.S. Talbot (U.S. Fish and Wildlife Service), and J.F. Bisaillon (Ivvavik National Park) provided background materials. Hope Ventures (Fort Liard) assisted with access into the Kota-neelee area of southeast Yukon. The efforts and patience of the Yukon Energy, Mines and Resources Library staff are also gratefully acknowledged.
[c] The Arctic Institute of North America
Anderson, L.E., Crum, H.A., and Buck, W.R. 1990. List of the mosses of North America north of Mexico. The Bryologist 93(4):448--499.
Ayotte, N. 2002. White spruce dynamics in the forest-tundra ecotone, the southwest Yukon Territory. MSc thesis, University of Ottawa, Ottawa, Ontario.
Bailey, R.G., Zoltai, S.C., and Wiken, E.B. 1985. Ecological regionalization in Canada and the United States. Geoforum 16(3):265-275.
Boggs, K., and Sturdy, M. 2005. Plant associations and post-fire vegetation succession in Yukon--Charley Rivers National Preserve. Anchorage: Natural Heritage Program, Environment and Natural Resources Institute, University of Alaska Anchorage.
Bostock, H.S. 1948. Physiography of the Canadian cordillera, with special reference to the area north of the fifty-fifth parallel. Geological Survey Memoir 247. Ottawa: Canada Department of Mines and Resources.
Boyd, C.N. 1984. An analysis of the natural vegetation of the MacMillan Pass/Sheldon Lake area of the east-central Yukon Territory. MSc thesis, University of Alberta, Edmonton, Alberta.
Brodo, I.M., Sharnoff, S.D., and Sharnoff, S. 2001. Lichens of North America. New Haven: Yale University Press.
CAVM Team. 2003. Circumpolar Arctic vegetation map (1:7,500,000 scale). Conservation of Arctic Flora and Fauna Map No. 1. Anchorage: U.S. Fish and Wildlife Service.
Cody, W.J. 2000. Flora of the Yukon Territory. Ottawa: NRC Research Press.
Danby, R.K., and Hik, D.S. 2007. Evidence of recent treeline dynamics in southwest Yukon from aerial photographs. Arctic 60(4):411--420.
Day, J.H. 1962. Reconnaissance soil survey of the Takhini and Desadeash Valleys in the Yukon Territory. Ottawa: Canada Department of Agriculture.
Douglas, G.W., Ballard, T.M., and Otchere-Boateng, J. 1980. Biophysical inventory studies of Kluane National Park. Unpubl. ms by Douglas Ecological Consultants Ltd., Victoria, British Columbia. Available at Yukon Energy, Mines and Resources Library, 335--300 Main Street, PO Box 2703 (K335), Whitehorse, Yukon YlA 2C6.
Drew, J.V., and Shanks, R.E. 1965. Landscape relationships of soils and vegetation in the forest-tundra ecotone, Upper Firth River Valley, Alaska-Canada. Ecological Monographs 35(3):285--306.
Ecological Stratification Working Group. 1995. A national ecological framework for Canada. Ottawa: Agriculture and Agri-Food Canada and Environment Canada.
Ecoregions Working Group. 1989. Ecoclimatic regions of Canada: first approximation. Ecological Land Classification Series No. 23. Ottawa: Canadian Wildlife Service, Environment Canada.
Environment Canada. 2011a. Canadian climate normals or averages 1971-2000. National Climate Data and Information Archive. <http://www.climate.weatheroffice.gc.ca/climate_normals/index_e.html.
--. 2011b. Climate data online. National climate data and information archive. http://climate.weatheroffice.gc.ca/climateData/canada_e.html.
Farr, W.A., and Hard, J.S. 1987. Multivariate analysis of climate along the southern coast of Alaska: Some forestry implications. Research Paper PNW-RP-372. Portland, Oregon: Pacific Northwest Research Station, USDA Forest Service.
Felix, N.A., Raynolds, M.K., Jorgenson, J.C., and DuBois, K.E. 1992. Resistance and resilience of tundra plant communities to disturbance by winter seismic vehicles. Arctic and Alpine Research 24(1):69--77.
Grods, J. 2006. Ecological assessment of devil's club (Oplopanax horridus) in the LaBiche and Beaver River drainages. Unpubl. ms by Makonis Consulting Ltd., Kelowna, British Columbia. Available at Yukon Energy, Mines and Resources Library, 335--300 Main Street, PO Box 2703 (K-335), Whitehorse, Yukon YlA 2C6.
Hettinger, L., Janz, A., and Wein, R.W. 1973. Vegetation of the northern Yukon Territory. Arctic Gas Biological Report Series Vol. 1. Calgary, Alberta: Northern Engineering Services Company Ltd.
Hirvonen, H. 2001. Canada's National Ecological Framework: An asset to reporting on the health of Canadian forests. The Forestry Chronicle 77(1):111--115.
Hoefs, M., McTaggart-Cowan, 1., and Krajina, V.J. 1975. Phytosociological analysis and synthesis of Sheep Mountain, southwest Yukon Territory, Canada. Syesis 8(Suppl.):125-228.
Jatzold, R. 2000. Semi-arid regions of the boreal zone as demonstrated
in the Yukon Basin. Erdkunde 54(1):1--19.
Jeffrey, W.W. 1964. Forest types along lower Liard River, Northwest Territories. Publication 1035. Ottawa: Canada Department of Forestry.
Kennedy, C.E., and Smith, C.A.S. 1999. Vegetation, terrain and natural features in the Tombstone area, Yukon Territory. Whitehorse: Yukon Department of Natural Resources and Agriculture and Agri-Food Canada.
Kojima, S. 1996. Ecosystem types of boreal forest in the North Klondike River valley, Yukon Territory, Canada, and their productivity potentials. Environmental Monitoring and Assessment 39(1--3):265-281.
Lacate, D.S. 1969. Guidelines for biophysical land classification. Forest Service Publication 1264. Ottawa: Canada Department of Fisheries and Forestry.
Lambert, J.D.H. 1968. The ecology and successional trends of tundra plant communities in the low arctic subalpine zone of the Richardson and British Mountains of the Canadian western Arctic. PhD thesis, University of British Columbia, Vancouver, British Columbia.
Lipovsky, P.S., and McKenna, K. 2005. Local-scale biophysical mapping for integrated resource management, Watson * Lake area (NTS 105A/2), Yukon. YGS Open File 2005-6. Whitehorse: Yukon Geological Survey.
Luttmerding, H.A., Demarchi, D.A., Lea, E.C., Meidinger, D.V., and Vold, T. 1990. Describing ecosystems in the field. MOE Manual 11. Victoria: British Columbia Ministry of Environment, Lands and Parks and Ministry of Forests.
Mackey, B.G., McKenney, D.W., Yang, D.Q., McMahon, J.P., and Hutchinson, M.F. 1996. Site regions revisited: A climatic analysis of Hills' site regions for the province of Ontario using a parametric method. Canadian Journal of Forest Research 26(3):333--354.
Marcus, M.G., and LaBelle, J.R. 1970. Summer climatic observations at the 5,360 meter level, Mt. Logan, 1968-1969. Arctic and Alpine Research 2(2):103--114.
Marshall, I.B., Schut, P.H., and Ballard, M. 1999. National Ecological Framework for Canada: Attribute data report. Ottawa/Hull: Agriculture and Agri-Food Canada, and Environment Canada. http://sis.agr.gc.ca/cansis/nsdb/ecostrat/1999report/intro.html.
McCune, B., and Mefford, M.J. 1999. PC-ORD for Windows, multivariate analysis of ecological data, version 4.25. Gleneden Beach, Oregon: MjM Software.
Meidinger, D., and Pojar, J. 1991. Ecosystems of British Columbia. Special Report Series 6. Victoria: B.C. Ministry of Forests.
Miller, R.G., Jr. 1966. Simultaneous statistical inference. New York: McGraw-Hill Book Company.
Ministere des Ressources Naturelles de la Faune et des Pares. 2003. The ecological land classification hierarchy. Quebec: Gouvernement du Quebec. http://www.mrn.gouv.qc.ca/english/publications/forest/publications/ecological.pdf.
Mueller-Dombois, D., and Ellenberg, H. 1974. Aims and methods of vegetation ecology. New York: John Wiley & Sons.
National Vegetation Working Group. 1990. The Canadian vegetation classification system. Edited by W.L Strong, E.T. Oswald, and D.J. Downing. Ecological Land Classification Series No. 25. Ottawa: Environment Canada.
Natural Resources Canada. 2012. Silvicultural terms in Canada. Ottawa: National Forestry Database, Canadian Forest Service. http://nfdp.ccfm.org/terms/terms_e.php.
NWT Ecosystem Classification Group. 2010. Ecological regions of the Northwest Territories--Cordillera. Yellowknife: Northwest Territories Department of Environment and Natural Resources.
Orians, G.H. 1993. Endangered at what level? Ecological Applications 3(2):206--208.
Oswald, E.T., and Brown, B.N. 1986. Forest communities in Lake Laberge ecoregion, Yukon Territory. Information Report BC-X-282. Victoria: Canadian Forest Service.
Oswald, E.T., and Senyk, J.P. 1977. Ecoreg ions of Yukon Territory. Victoria: Canadian Forest Service.
Poore, M.E.D. 1962.. The method of successive approximation in descriptive ecology. Advances in Ecological Research 1(1):35--68.
Rowe, J.S. 1972. Forest regions of Canada. Publication 1300. Ottawa: Canadian Forestry Service.
Sanborn, P. 2010. Topographically controlled grassland soils in the Boreal Cordillera ecozone, northwestern Canada. Canadian Journal of Soil Science 90(1):89--101.
See, M.G., and Bliss, L.C. 1980. Alpine lichen-dominated communities in Alberta and the Yukon. Canadian Journal of Botany 58(20):2148-2170.
Simpson, J.J., Hufford, G.L., Fleming, M.D., Berg, IS., and Ashton, J.B. 2002. Long-term climate patterns in Alaskan surface temperature and precipitation and their biological consequences. IEEE Transactions on Geoscience and Remote Sensing 40(5):1164-1184.
Simpson, J.J., Stuart, M.C., and Daly, C. 2007. A discriminant analysis model of Alaskan biomes on spatial climatic and environmental data. Arctic 60(4):341-369.
Smith, C.A.S., Kennedy, C.E., Hargrave, A.E., and McKenna, K.M. 1989. Soil and vegetation of Herschel Island. Yukon Soil Survey Report I. Whitehorse: Land Resource Research Centre, Agriculture Canada.
Smith, C.A.S., Meikle, J.C., and Roots, C.F. 2004. Ecoregions of the Yukon Territory: Biophysical properties of Yukon landscapes. PARC Technical Bulletin No. 04-01. Summerland: Agriculture and Agri-Food Canada.
Soil Classification Working Group. 1998. The Canadian system of soil classification. Agriculture and Agri-Food Canada Publication 1646. Ottawa: NRC Research Press.
Stanek, W., Alexander, K., and Simmons, C.S. 1981. Reconnaissance of vegetation and soils along the Dempster Highway, Yukon Territory: 1. Vegetation types. Information Report BC-X-217. Victoria, British Columbia: Canadian Forestry Service.
Statsoft. 1995. STATISTICA for Windows, Version 5.0. Tulsa, Oklahoma: Statsoft Inc.
Strong, W.L 1992. Ecoregions and ecodistricts of Alberta. Publication T/244. Edmonton: Alberta Forestry, Lands and Wildlife.
--. 2002. Lodgepole pine/Labrador tea type communities of western Canada. Canadian Journal of Botany 80(2):151-165.
--. 2009. Populus tremuloides Michx. postfire stand dynamics in the northern boreal-cordilleran ecoclimatic region of central Yukon Territory, Canada. Forest Ecology and Management 258(7):1110-1120.
--. 2010. Pinus contorta var. yukonensis var. nov. (Pinaceae) from south-central Yukon, Canada. Nordic Journal of Botany 28(4):448-452.
Strong, W.L, and Hills, L.V. 2006. Taxonomy and origin of present-day morphometric variation in Picea glauca (xengelmannii) seed-cone scales in North America. Canadian Journal of Botany 84(7):1129-1141.
Strong, W.L, and Redburn, M.J. 2009. Latitude-related variation in understory vegetation of boreal Populus tremuloides stands in Alberta, Canada. Community Ecology 10(1):35-44.
Vitt, D.H., Marsh, J.E., and Bovey, R.B. 1988. Mosses, lichens and ferns of northwest North America. Edmonton, Alberta: Lone Pine Publishing.
White, M.P., Smith, C.A.S., Kroetsch, D., and McKenna, K. 1992. Soil landscapes of Canada--Yukon Territory. Contribution No. 89-05. Ottawa: Agriculture Canada.
Wiken, E.B., Welch, D.M., Ironside, G.R., and Taylor, D.G. 1981. The northern Yukon: An ecological land survey. Ecological Land Classification Series 6. Vancouver: Environment Canada, Lands Directorate.
Yukon Forest Management Branch. 2009. Yukon forest cover mapsheets (416 sheets), 1:50000 scale. Whitehorse: Yukon Energy, Mines and Resources.
Zoladeski, C.A., Cowell, D.W., and Ecosystem Classification Advisory Committee. 1996. Ecosystem classification for the southeast Yukon, field guide. Whitehorse: Yukon Renewable Resources, Canadian Forest Service, and Canada Department of Indian Affairs and Northern Development.
Zoltai, S.C., and Pettapiece, W.W. 1973. Studies of vegetation, landform and permafrost in the Mackenzie Valley: Terrain, vegetation and permafrost relationships in the northern part of the Mackenzie Valley and northern Yukon. Environmental-Social Committee Northern Pipelines, Task Force on Northern Oil Development Report 73-4. Ottawa: Information Canada.
(1.) Arctic Institute of North America, 2500 University Drive NW, Calgary, Alberta T2N 1N4, Canada; mailing address: PO Box 40186 Station Main, Whitehorse, Yukon YlA 6M9, Canada; firstname.lastname@example.org
WAYNE L STRONG (1)
|Printer friendly Cite/link Email Feedback|
|Author:||Strong, Wayne L.|
|Date:||Mar 1, 2013|
|Previous Article:||Effects of climate-induced changes in parasitism, predation and predator-predator interactions on reproduction and survival of an Arctic Marine bird.|
|Next Article:||Ways we respect caribou: Teetl'it Gwich'in rules.|