Printer Friendly

High-latitude Yukon boreal-cordilleran grassland plant communities.

ABSTRACT. High-latitude (61.9[degrees]-62.8[degrees] N) graminoid plant communities located near the northern limit of boreal forest occurrence in the Carmacks-Pelly Crossing area of Yukon were sampled and classified, and four sociations were recognized: Calamagrostis purpurascens (purple reedgrass), Hesperostipa comata (needle-and-thread grass), Poa glauca-Artemisia frigida (glaucous bluegrass-pasture sagewort), and Pseudoroegneria spicata-Artemisia frigida (bluebunch wheatgrass-pasture sagewort). These plant communities occurred primarily on south-southwest slopes (180[degrees]-230[degrees]) with 45%-75% gradients. Relative to southern Canada mixedgrass prairie, Yukon grasslands were likely species poor, had less total canopy cover, and produced less than half as much aboveground herb biomass (typically 200-300 kg/ha). The distribution of the sociations appeared limited to ~60[degrees]-65[degrees] N latitude in the Yukon-Alaska region. Calamagrostis was the most abundant and widespread grassland sociation in the study area. The four sociations occupied less than 4% of the Carmacks-Pelly Crossing landscape, although they represented the most extensive high-latitude boreal grassland vegetation in Canada ([greater than or equal to] 2600 ha) and they were among the most northerly in North America.

Key words: Beringia; boreal; ecology; grassland; palaeovegetation; plant community; sociation

RESUME. Les communautes de plantes graminoides en haute latitude (61,9[degrees]-62,8[degrees] N) situees pres de la limite nord du milieu de foret boreale dans la region de Carmacks-Pelly Crossing, au Yukon, ont fait l'objet d'un echantillonnage et d'un classement. Cela a permis de reconnaitre quatre sociations : Calamagrostis purpurascens (calamagrostis pourpre), Hesperostipa comata (stipe comateuse), Poa glauca-Artemisia frigida (paturin glauque--armoise douce) et Pseudoroegneria spicata-Artemisia frigida (agropyre a epi--armoise douce). Ces communautes de plantes se manifestaient surtout sur les versants du sud et du sud-ouest (180[degrees]-230[degrees]) comportant des gradients variant entre 45 % et 75 %. Comparativement a la prairie mixte du sud du Canada, les herbages du Yukon comportaient vraisemblablement peu d'especes et un moins grand couvert au total, et ils produisaient moins de la moitie de biomasse epigee (generalement entre 200 et 300 kg/ha). La repartition des sociations semblait limitee a la latitude ~60[degrees]-65[degrees] N de la region du Yukon-Alaska. Dans l'aire visee par l'etude, Calamagrostis etait la sociation d'herbage la plus abondante et la plus generalisee. Les quatre sociations occupaient moins de 4 % du paysage de Carmacks-Pelly Crossing, meme si elles representaient la vegetation d'herbage boreal la plus vaste en haute latitude au Canada ([greater than or equal to] 2600 ha) et qu'elles figuraient parmi les sociations les plus au nord de l'Amerique du Nord.

Mots cles : Beringie; boreal; ecologie; herbage; paleovegetation; communaute de plantes; sociation

Traduit pour la revue Arctic par Nicole Giguere.


The grassland or Prairie biome of interior North America extends northward to ~54[degrees] N latitude in western Canada (Ecoregions Working Group, 1989). Grass-dominated communities, however, are not restricted to this biome. They also occur among conifer forests on favorable site conditions in the mountains of southwestern Alberta (Stringer, 1973; Holland and Coen, 1982; Grande Cache Coal Company, 2001), northern British Columbia (Seip and Bunnell, 1985; Pojar, 1986), and southwest Yukon (Hoefs et al., 1975; Vetter, 2000; Chambers, 2010). Grass-dominated vegetation was also relatively common among mixedwood forests, as far north as -58[degrees] in the Peace River district of northwestern Alberta until the mid-1900s. Since that time, however, most of these grasslands have been converted to agricultural cropland or replaced by forest vegetation (Moss, 1952; Wilkinson and Johnson, 1983; Strong et al., 2009). Today, grassland remnants are largely confined to steep southerly slopes such as those along the Peace River Valley.

In forested boreal and cordilleran landscapes, grass-dominated vegetation typically occurs as small isolated entities of two distinct ecological types: meadows and grasslands. Meadows occur where early summer flooding results in site moisture that exceeds the needs of the associated terrestrial plants, but conditions become much drier during the remainder of the growing season (Mueller-Dombois and Ellenberg, 1974:481). Calamagrostis canadensis (marsh reedgress), Calamagrostis stricta (northern reedgrass), and Deschampsia cespitosa (tufted hairgrass) are characteristic meadow taxa in forested landscapes. In contrast, grasslands (including dryland Carex spp. or xerophytic sedge stands) are associated with sites that create severe physiological moisture stress for plants. These sites include steep southerly slopes that have rapid lateral drainage and accentuated evaporation, coarse-textured surficial materials with poor water-holding capacity, and saline (Wilkinson and Johnson, 1983) and alkaline soils (Schwarz and Wein, 1997) with adverse soil chemistry. Grasslands that occur on steep slopes are the focus of this study.

The most frequently reported indigenous grassland community along the eastern slopes of the Canadian Cordillera is dominated by Leymus innovatus (hairy wild rye), often with Arctostaphylos uva-ursi (bearberry) as a codominant (e.g., Holland and Coen, 1982; Beckingham et al., 1996; Strong, 1996; Grande Cache Coal Company, 2001), and it occurs as far north as 59[degrees] N latitude (Seip and Bunnell, 1985). Such plant communities often represent less than 2% of the regional vegetation (e.g., Strong, 1996; Grande Cache Coal Company, 2001). Several studies have reported the occurrence of boreal and cordilleran grassland communities in northwestern Canada. Hesperostipa spartea-Carex spp.--Artemisia frigida (porcupine grasssedge-pasture sagewort) communities on steep southerly slopes in the boreal forest zone of northern Alberta have been described by Wilkinson and Johnson (1983; ~56[degrees] N, 118[degrees] W) and Redmann and Schwarz (1986; ~59[degrees] N, 112.5[degrees] W). Pojar (1986) documented the occurrence of subalpine Festuca altaica (Altai fescue) and Poa glauca- Carex snpina (glaucous bluegrass-weak arctic sedge) grasslands in north-central British Columbia (~57.6[degrees] N, 129[degrees] W), and Hoefs et al. (1975) recognized Calamagrostis purpurascens (purple reedgrass)--Arctostaphylos uva-ursi, Carex filifolia (threadleaf sedge)--Artemisia frigida, Elymus violaceus (wheatgrass), Elymus x yukonensis (Yukon wheatgrass)--Artemisia frigida, and Festuca altaica types on Sheep Mountain in Kluane National Park, southwestern Yukon (~61[degrees] N, 138.5[degrees] W). In contrast, Vetter (2000) found only Carex filifolia--Artemisia frigida vegetation near Aishihik-Sekulmun Lakes ~85 km to the northeast, whereas Chambers (2010) reported Calamagrostis purpurascens, Carex duriuscula (needleleaf sedge)--Artemisia frigida, and Carex supina-dominated communities in the Takhini Valley ~145 km east-southeast of Sheep Mountain. These examples convey some of the compositional diversity that exists among grasslands when they occur in treedominated landscapes.

In the northernmost portion of the Canadian boreal forest, an atypical concentration of grasslands has been recognized in Yukon from ~10 km south of Braeburn northward to Carmacks (Oswald and Senyk, 1977) and from Braeburn northward to the Stewart River (Smith et al., 2004) (Fig. 1). The latter study described these grasslands as "extensive" and a "notable feature" of the landscape, although they also occur elsewhere in southern Yukon (see Figs. 3 and 4 in Strong, 2013). In addition to creating diversity in forest-dominated landscapes, Yukon grasslands are also known to include rare endemic species such as Nestotus macleanii (Maclean's goldenweed), and prairie floristic elements such as Hesperostipa comata (needle-and-thread grass) and Achnatherum richardsonii (Richardson needlegrass) (Cody, 2000). Scudder (1997:50) considered grasslands to be the "most interesting and notable habitats in the Yukon" with respect to insect diversity. Yukon grasslands also represent an important but limited habitat for grazing ungulates (Strong et al., 2013).

Despite the purportedly common occurrence of grasslands on the central Yukon Plateau, little is known about the plant communities that form the vegetation except that they typically occur on steep southerly slopes. The goals of this study were therefore to (1) sample the botanical composition of the dry upland graminoid-dominated vegetation that occurs in the Carmacks-Pelly Crossing area of Yukon; (2) classify the resulting compositional summaries or releves into distinct plant community types on the basis of species dominance (i.e., sociations); (3) determine the range of slope orientations and gradients that each occupies; and (4) estimate the aboveground productivity of herbaceous plants in each sociation.


Study Area

The study area was limited to grassland polygons located near the Klondike Highway and connecting roads between Pelly Crossing and a point 20 km south of Carmacks, which represents a north-south distance of ~100 km (Fig. 1). The connecting roads included the easternmost portions of the poorly developed Nansen and Free Gold roads that are located immediately west of Carmacks, the western 16 km of the Tatchum Lake Road, and ~18 km of the Robert Campbell Highway along the Yukon River Valley, immediately east of Carmacks. The area included represents the more accessible portions of the region and the overlapping zones of grassland concentration recognized by Oswald and Senyk (1977) and Smith et al. (2004), as well as the core of the central Yukon Plateau.

Rounded mountain ridges with steep lower slopes and broad valleys typify the topography of the Carmacks-Pelly Crossing area. Maximum local relief is ~700 m, with a base elevation of 450 m above sea level. The study area is located within the westernmost portion of Wisconsinan glaciation (Dyke and Prest, 1987). Glacial fluvial deposits on valley bottoms and moraine on lower valley slopes are the most common surficial materials. Bedrock exposures often dominate upper slopes and the highest elevations (Klassen and Morison, 1987; Klassen et al., 1987). All of the study area is within Yukon's Northern Cordilleran Boreal ecoclimatic region, where serai Populus tremuloides (trembling aspen) and successionally more mature P. tremuloides--Picea albertiana ssp. albertiana (western white spruce) vegetation occupy well to moderately well drained sites at elevations below ~1200 m (Strong, 2013). Pinus conforta var. latifolia (lodgepole pine) vegetation forms only a small portion of the study area. Picea albertiana is the dominant late-successional tree on upland sites. The median temperatures are 10.8[degrees]C for summer (May-September) and -18.8[degrees]C for winter (November-March) (Strong, 2013). Most precipitation (177 of 279 mm) falls in summer, but -72 cm of snowfall (~72 mm water equivalent) typically occurs during winter.

Field Sampling and Related Methods

Prior to field sampling, aerial photographs were searched for potential sampling areas, with attention focused primarily on larger grassland polygons. Releve sampling was limited to locations within grassland polygons that had (1) xerophytic graminoids as the dominant plants, (2) a relatively uniform species composition, (3) a relatively uniform slope gradient and orientation, and (4) mineral soils without substantial evidence of active surface erosion, but (5) lacked a notable woody plant component and (6) occurred within a one-hour hike from a road or 20 minutes from another releve. Grassland polygons larger than 2 ha in area were sometimes sampled in more than one location because of the spatial heterogeneity of the vegetation. If more than one releve was sampled, they differed notably in species composition, or were more than 100 m apart, or had a greater than 20[degrees] difference in slope orientation, or more than 10% difference in slope gradient. One hundred releves were sampled at 33 different locations in the study area (Fig. 1).

Species composition sampling and canopy cover estimation for each releve were based on five 100 cm x 100 cm quadrats located at 2.5 m intervals along a 15 m transect. The lower left corner of the first quadrat was located at the 2.5 m mark on each transect. For each releve, a transect was placed in the center of a vegetation stand that by visual survey was judged to have a relatively consistent structure and species composition. Transects were aligned parallel to the terrain contour, with the quadrat frame held horizontal during sampling (cf. Beers, 1969; Abella et al., 2004). Ocular estimation was used to gauge the percent canopy cover of individual species (sensu Daubenmire, 1968). The larger-than-normal quadrat size for grassland sampling (cf. Wilkinson and Johnson, 1983; Redmann and Schwarz, 1986) was used to accommodate the tussock habit and large size (e.g., up to 70 cm tall foliage; Cody, 2000) of locally common Calamagrostis purpurascens. A short transect was used to minimize the topographic variability that could occur within a releve, so a specific slope orientation and gradient might be linked to each plant community. Furthermore, the short transects helped minimize the amount of physical site damage caused by sampling. Slope orientation and gradient were determined for each releve using a compass and clinometer, and location and elevation, with a hand-held global positioning system device. June 22 solar insolation (kJ/[m.sup.2]) input was estimated by regression analysis of total daily values (Buffo et al., 1972), with adjustments for the location, slope orientation, and slope gradient for each releve. Plant nomenclature follows ITIS Partners (2013) for vascular plants, except for Picea spp. (Strong and Hills, 2006) and lichens (Esslinger, 2014).

The vegetation at selected releves was sampled to estimate aboveground plant biomass availability. To ensure at least five biomass estimates were available for each plant community type in the final vegetation classification, releves were stratified into five tentative plant community types on the basis of general species compositions. Ten releves were selected at random from the two more common plant community types and five from each of the other three types, for a total of 35 samples. The tentative releve classification was discarded after the biomass sampling was completed. Biomass sampling followed the same design used for releve composition sampling. The current summer's growth of herbaceous plants was clipped within 3 cm of the ground surface and separated into graminoids and forbs. The collected material was air-dried in paper bags for approximately three weeks in a warm environment and then oven-dried at 60[degrees]C for 24 hours prior to weighing. Vegetation composition sampling was conducted from 24 June to 9 July 2013, and biomass sampling from 5 to 11 August. The latter was completed before the anticipated first late-summer frost, which typically occurs in mid to late August, as indicated by Carmacks meteorological data for 1981-2012 (Environment Canada, 2014).

An estimate of grassland areal extent in the Carmacks-Pelly Crossing study area was determined for a corridor 5 km wide centered on the three main access roads (a total area of ~840 [km.sup.2]). Mapping of individual grassland polygons was done through stereoscopic interpretation of aerial photographs and the plotting of polygons at a map scale of 1:50000. The mapped polygons included all sampled grassland polygons, as well as some judged unsuitable for sampling for whatever reason and those that were not readily accessible (e.g., across a river). The areal extent of the individual grassland polygons and the study area was determined by planimeter measurement.

Data Analysis

A cluster analysis based on Ward's method and relative Euclidean distance was used to aggregate releves according to similarities in species composition and abundance (McCune and Mefford, 1999). The groups recognized in the resulting dendrogram were scrutinized for consistency in composition and then classified according to dominant species, i.e., sociations sensu G.E. Du Rietz (Mueller-Dombois and Ellenberg, 1974). The closest North American approximations to this classification approach are represented by the proposed Canadian System of Vegetation Classification (Strong et al., 1990) and the system used in Alaska (Viereck et al., 1992). Two releves that were ambiguous with respect to sociation membership were excluded from classification. Detrended correspondence analysis (McCune and Mefford, 1999) was used to illustrate the compositional relationship among individual releves and the distinctiveness of each sociation. The degree of ordination explanation was based on relative Euclidean distance. Constancy was defined as the percent occurrence of a taxon among releves. Dominance concentration of taxa in individual releves was estimated using the algorithm presented by Strong (2002).

Prior to statistical analysis, data were tested for normality using the Kolmogorov-Smirnov one-sample test. Because most variables were not normal distributed, Kruskal-Wallis tests were used to identify compositional and site condition differences among sociations. Nonparametric Scheffe rank tests at the [alpha] = 0.05 level (Miller, 1966: formula 110) were used to identify differences between individual sociations. Paired groups with normally distributed values were compared using t-tests. All statistical analyses were based on STATISTICA software (StatSoft, 1995) except Scheffe rank tests, which were assessed manually.


Grasslands represented at least 2600 ha or 3%-4% of the terrestrial vegetation within the 5 km wide study corridor along the main access roads in the Carmacks-Pelly Crossing area. In total, 100 grassland polygons were mapped, with upper quartile values ([Q.sub.3]-maximum) of 27 to 345 ha. Both the size and the frequency of grassland polygons appeared to diminish with increasing distance from the highways. The sampled grasslands were located between 61.9[degrees] and 62.8[degrees] N latitude. Releves were distributed throughout the study area, with one notable exception: the five most northern releves were sampled near Pelly Crossing, which was ~19 km northeast of the next nearest location (Fig. 1). The next potentially available grassland polygon to the north was located ca. 70 km away, immediately north of Stewart Crossing. Most of the polygons visited were dominated by a single type of grassland community, but at least one other community was often present because of variation in site conditions and topographic position (e.g., toe versus middle and upper slopes). Variable site conditions also tended to create differences in plant densities and composition within communities. Most sampled polygons had mineral matter (sandy loam to loamy sand texture) in the uppermost portion of their soil profile, with bedrock often within 1 m of the ground surface. No evidence of permafrost was observed. Stunted and scattered individuals and small clones of Populus tremuloides sometimes occurred in conjunction with grassland areas, but well-developed forest vegetation occurred only at the margins.


Four grass-dominated sociations were recognized among 98 releves in the final classification (Table 1): Calamagrostis purpurascens (purple reedgrass, Fig. 2), Pseudoroegneria spicata--Artemisia frigida (bluebunch wheatgrass-pasture sagewort), Hesperostipa comata (needle-and-thread grass), and Poa glauca--Artemisia frigida (glaucous bluegrass--pasture sagewort). The cluster analysis groupings that represented the first three sociations formed in the dendrogram at the 13% or less amalgamation error level, whereas the Poa sociation formed at the 23% error level. Cluster amalgamation was discontinued after formation of the latter group. Three percent chaining occurred within the dendrogram. Calamagrostis was the most frequently sampled sociation (49%, Table 1), but field observations suggested that it probably comprised at least two-thirds of the grassland vegetation. Hesperostipa vegetation appeared the least abundant.

The Calamagrostis, Pseudoroegneria, and Hesperostipa sociations were easily differentiated on the basis of a single high-constancy species that had 30%-50% canopy cover (Table 1; cf. Fig. 3). On a relative basis, these dominant species individually represented at least two-thirds of the total plant cover in their respective sociations. The Poa sociation was compositionally more heterogeneous, with four taxa having more than 86% constancy and individual average canopy cover values of 10%-18% (Table 1). This greater compositional diversity was reflected in the greater dispersion of Poa releves in ordination space compared to the other sociations (Fig. 3). The Calamagrostis, Hesperostipa, and Pseudoroegneria sociations also had more than one high-constancy taxon, but the other taxa had much less cover than the dominant one. In addition to greater Poa glauca canopy cover than the other communities, the Poa sociation was also distinguishable by its greater proportion of Flavocetraria cucullata (curled snow lichen) and Cladonia (fruticose lichens) squamule cover. The two releves excluded from the classification had botanical composition that was intermediate between the Pseudoroegneria and Poa sociations judging by their position within the detrended correspondence analysis ordination (Fig. 3). Neither excluded releve contained a grass species with more than 5% canopy cover.

Plant heights within the four sociations seldom exceeded 40-45 cm, except that flowering stalks sometimes reached 1 m tall. Average total plant canopy cover of sociations ranged from 48% to 73%, with the Poa sociation having the greatest cover (Table 1). The Calamagrostis and Poa sociations contained on average ~9.5 taxa per releve, whereas the Pseudoroegneria sociation averaged 6.4 taxa, which was significantly fewer (p < 0.001). According to Scheffe rank tests, the Hesperostipa sociation was intermediate in total species richness, although the average number of taxa was essentially the same as that of the Pseudoroegneria sociation (Table 1). This inconsistency likely stems from the small Hesperostipa sample. Dominance concentration among sociations was moderate, but greatest in the Calamagrostis sociation. The least amount of bare ground was associated with the Poa sociation, in part because it included abundant ground lichens (Table 1).

Distribution and Site Conditions

Only the Hesperostipa sociation was not distributed throughout most of the study area, with seven of its eight releves located east of Carmacks on the north side of the Yukon River Valley. South of Carmacks, only Calamagrostis vegetation was observed. Releves occurred at elevations ranging from 497 to 872 m, with no significant difference among sociations (Table 1). Grasslands occurred on southerly aspects ranging from 90[degrees] to 270[degrees], but 72% (n = 71) occurred on south-southwest orientations (158[degrees]-247[degrees]), with quartile values of 180[degrees] and 230[degrees]. No significant difference existed among grassland sociations with respect to slope orientation. Calamagrostis and Pseudoroegneria releves occurred on the steepest slopes (Table 1), with 89% of all releves having gradients of 45% to 75%. Poa releves had the lowest slope gradients, with quartile values ranging from 45% to 56%.

Most grassland releves had at least xeric moisture regimes due to the associated steep slopes and rapid lateral drainage. Hesperostipa vegetation, however, was typically associated with the toe of slopes, and therefore, it likely had somewhat moister (e.g., subxeric to submesic) regimes than upslope grasslands. On the basis of slope orientation and gradient values, Pseudoroegneria releves were associated with the least amount of June 22 solar insolation; as a result, they likely had the coolest grassland environments. Poa vegetation occurred on the sites with highest insolation (Table 1). Observation suggested that Pseudoroegneria grassland communities were also typically associated with convex slopes, whereas Calamagrostis vegetation occupied concave slopes.

Herbaceous Biomass

No sampled releve had a standing crop of herbaceous biomass greater than 485 kg/ha (excluding dead material from previous years) at the end of the 2013 growing season. Most releves (n = 18 of 34) produced 200-300 kg/ha of total herbaceous biomass and 150-250 kg/ha of graminoid biomass. Graminoids comprised the bulk of the standing crop, although forbs constituted almost half in the Poa sociation (Table 1). Total graminoid biomass was greatest in the Calamagrostis and Hesperostipa sociations.


High-Latitude Boreal Grasslands

Eastern and central Canada lack high-latitude (> 60[degrees] N) boreal grasslands because the boreal biome extends northward only to ~53[degrees] N in eastern Canada and ~57[degrees] N in central Canada (Ecoregions Working Group, 1989). Grassland areas occur farther west in the boreal forests of the Northwest Territories, but they are rare, e.g., Elymus trachycaulus--Muhlenbergia richardsonis or slender wheatgrass-mat muhly (Thieret, 1959; Ecosystem Classification Group, 2010). North of 62[degrees] N, or at about the southern limit of the current study area, Pseudoroegneria spicata vegetation has been described as "abundant" on "widely separated" steep south-facing slopes in the Tanana River Valley near Fairbanks (64.7[degrees] N, 147.5[degrees] W) in central Alaska (Hansen, 1951). Kassler (1979) also reported occurrences of P. spicata, Poa glauca, and Calamagrostis purpurpascens communities up to 440 km farther west in the Galena area. In contrast, Leymus innovatus and Festuca altaica dominated grassland communities ~120 km southwest of Fairbanks (Hansen, 1951). Unfortunately, no data were available to determine the areal extent of grasslands for any of the areas. Immediately west of the Yukon boundary in eastcentral Alaska, however, Calamagrostis purpurascens and Festuca altaica communities have been reported in the Ogilvie Mountains (65.2[degrees] N, 141.8[degrees] W), where they comprised about two-thirds of the 468 ha of boreal cordilleran grassland in the ~4120 [km.sup.2] Yukon-Charley Rivers National Preserve (Boggs and Sturdy, 2005; also see Edwards and Armbruster, 1989). Other grassland occurrences have been reported in Alaska, but they occur either south of 62[degrees] N or in nonboreal environments (e.g., alpine or Arctic ecosystems). Available information suggests that the Carmacks-Pelly Crossing boreal grasslands represent the most extensive occurrence in Canada, with an areal extent greater than 2600 ha, and they are among the most northerly grasslands in North America.

The concentration of grasslands in the Carmacks-Pelly Crossing area likely stems from the occurrence of northwest-southeast trending mountain ridges and the similarly aligned Yukon River Valley, in conjunction with relatively low elevations, and from the area's location in the lee of the Coast Mountains. The lee position creates a regional rain shadow, but lower elevations allow greater potential for adiabatic warming than at higher elevations; hence, there is a greater potential for evaporation. The effect of these circumstances was represented in a regional potential evaporation analysis conducted by Jatzold (2000), who found that a severe April-June moisture deficit occurred in and around the Yukon River Valley and the lower reaches of its main tributaries. Jatzold (2000) estimated that precipitation compensated for less than 25% of the potential early summer climatic moisture deficit, although this percentage varies with slope orientation and steepness. He equated the severity of deficits on steep south-southwest slopes to that of semi-desert environments in low latitudes. Similar moisture deficits were also associated with the Yukon-Charley Rivers National Preserve and the Tanana River Valley in central Alaska (see Fig. 2 in Jatzold, 2000).

Regional Plant Community Differences

The Carmacks-Pelly Crossing Calamagrostis, Poa, and Pseudoroegneria sociations appear taxonomically and ecologically analogous to the communities that occur in central Alaska, although Hansen (1951) described only one releve per plant community type. There are no Hesperostipa comata plant communities in Alaska because the species is absent from the state's flora. All four Carmacks sociations appeared to be limited to locations between ~60[degrees] and 65[degrees] N latitude. The Calamagrostis sociation is the more widespread (Fig. 2), but probably represents less than 2% of the vegetation in southern Yukon. Leymus innovatus-dominant plant communities appeared to replace Calamagrostis purpurascens vegetation south of the sociation's distribution (cf. Hansen, 1951; Seip and Bunnell, 1985).

The Carmacks-Pelly Crossing sociations differed from other grasslands in various respects. In the Peace River district of northwest Alberta, for example, Hesperostipa vegetation was dominated by H. spartea rather than by H. comata (Wilkinson and Johnson, 1983). In addition, the H. spartea-dominant vegetation included large proportions of A. frigida (5%-11%) and presumably dryland Carex spp. (7%-21%), whereas H. comata comprised ~78% of the total plant cover in the Yukon sociation. Similarly, in the Canadian Prairie biome, H. comata mixedgrass vegetation typically includes 3-4 codominant graminoid taxa, each with 10%-15% canopy cover (Adams et ah, 2004, 2005). Mixedgrass prairie also has greater total vascular plant cover than grasslands in the Carmacks-Pelly Crossing area (i.e., 79%-94% versus 44%-59%), as well as much greater annual aboveground herb biomass production (typically > 1000 kg/ha versus < 300 kg/ha, cf. Coupland, 1961; Adams et ah, 2004, 2005). No prairie releve data based on 5 [m.sup.2] of sampling area were available for a comparison of floristic richness. However, releves based on 2 [m.sup.2] of sampled area from mixedgrass prairie in the Suffield National Wildlife Area of southeastern Alberta (Adams et ah, 1997) did not differ from those in the Carmacks-Pelly Crossing area with respect to vascular plant richness (7.9 taxa per 2 [m.sup.2] of sampling area versus 7.1 taxa per 5 [m.sup.2] of sampling area, respectively, t-test, p = 0.063). If this 2.5 x differential in sampling area could be rectified, it is very likely that the Carmacks grasslands would be less floristically diverse than their mixedgrass prairie physiognomic counterparts. These botanical differences likely reflect the ecological constraints imposed by a cold high-latitude climate compared to a mid-latitude climate. Vetter (2000:172) concluded appropriately that "Yukon and Alaska grasslands are more similar to each other than either [is] to the Great Plains [prairie] grasslands" in terms of both floristics and phytosociology, although both the Great Plains and the Carmacks-Pelly Crossing grassland soils are subject to similar Chernozemic soil development processes (Sanborn, 2010).


Scott Gilbert (Yukon College, Whitehorse) provided access to an analytical scale for weighing the biomass samples. Dr. Mary Vetter (Luther College, University of Regina) provided a comprehensive and constructive review that improved the manuscript.


Abella, S.R., Gering, L.R., and Shelburne, V.B. 2004. Slope correction of plot dimensions for vegetation sampling in mountainous terrain. Natural Areas Journal 24(4):358-360.

Adams, B.W., Poulin-Klein, L., Moisey, D., and McNeil, R.L. 2005. Rangeland plant communities and range health assessment guidelines for the dry mixedgrass natural subregion of Alberta. Publication T/040. Lethbridge: Alberta Sustainable Resource Development.

Adams, B.W., Richman, J., Poulin-Klein, L., France, K., and McNeil, R.L. 2013. Rangeland plant communities and range health assessment guidelines for the mixedgrass natural subregion of Alberta (2nd approximation). Publication T/039. Lethbridge: Alberta Environment and Sustainable Resource Development.

Adams, G.D., Trottier, G.C., Strong, W.L, MacDonald, I.D., Barry, S.J., Gregoire, P.G., Babish, G.W., and Weiss, G. 1997. Canadian Forces Base Suffield, National Wildlife Area, wildlife inventory: Vegetation component report. Edmonton, Alberta: Canadian Wildlife Service.

Beckingham, J.D., Corns, I.G.W., and Archibald, J.H. 1996. Field guide to ecosites of west-central Alberta. Special Report 9. Edmonton, Alberta: Canadian Forest Service.

Beers, T.W. 1969. Slope correction in horizontal point sampling. Journal of Forestry 67(3): 188-192.

Boggs, K., and Sturdy, M. 2005. Plant associations and post-fire vegetation succession in Yukon-Charley Rivers National Preserve. Anchorage: Alaska Natural Heritage Program, Environment and Natural Resources Institute, University of Alaska Anchorage.

Buffo, J., Fritschen, L.J., and Murphy, J.L. 1972. Direct solar radiation on various slopes from 0 to 60 degrees north latitude. Research Paper PNW-142. Portland, Oregon: USDA Forest Service.

Chambers, J.H.S. 2010. Habitat use and ecologically sustainable carrying capacity for elk (Cervus elaphus) in the Takhini Valley, Yukon. MEDes thesis, University of Calgary, Calgary, Alberta.

Cody, W.J. 2000. Flora of the Yukon Territory, 2nd ed. Ottawa: NRC Research Press.

Coupland, R.T. 1961. A reconsideration of grassland classification in the northern Great Plains of North America. Journal of Ecology 49(1):135-167.

Daubenmire, R. 1968. Plant communities. New York: Harper & Row Publishers.

Dyke, A.S., and Prest, V.K. 1987. Late Wisconsinan and Holocene retreat of the Laurentide ice sheet (scale 1:5000000). Map 1702A. Ottawa: Geological Survey of Canada.

Ecoregions Working Group. 1989. Ecoclimatic regions of Canada: First approximation. Ecological Land Classification Series No. 23. Ottawa: Canadian Wildlife Service.

Ecosystem Classification Group. 2010. Ecological regions of the Northwest Territories--Cordillera. Yellowknife: Northwest Territories Department of Environment and Natural Resources.

Edwards, M.E., and Armbruster, W.S. 1989. A tundra-steppe transition on Kathul Mountain, Alaska, U.S.A. Arctic and Alpine Research 21(3):296-304.

Environment Canada. 2014. Historical climate data.

Esslinger, T.L. 2014. A cumulative checklist for the lichen-forming, lichenicolous and allied fungi of the continental United States and Canada (version 19).

Grande Cache Coal Company. 2001. Vegetation (Vol. 4, Part 2). In: Application for approval and environmental impact assessment for Grande Cache Coal project. Calgary: Grande Cache Coal Company. Available at Government of Alberta, Great West Life Site, 9920-108 St., 6th floor, Edmonton, Alberta T5K 2M4.

Hansen, H.C. 1951. Characteristics of some grassland, marsh, and other plant communities in western Alaska. Ecological Monographs 21(4):317-378.

Hoefs, M., McTaggart-Cowan, I., and Krajina, V.J. 1975. Phytosociological analysis and synthesis of Sheep Mountain, southwest Yukon Territory, Canada. Syesis 8(Suppl. 1): 125-228.

Holland, W.D., and Coen, G.M. 1982. Ecological (biophysical) land classification of Banff and Jasper National Parks. Publication SS-82-44. Edmonton: Alberta Institute of Pedology.

ITIS Partners. 2013. Integrated taxonomic information system.

Jatzold, R. 2000. Semi-arid regions of the boreal zone as demonstrated in the Yukon basin. Erdkunde 54(1):1-19.

Kassler, K.C. 1979. Relicts of the late Pleistocene Arctic-steppe: Investigations of certain south-facing slopes in interior Alaska. BA thesis, Middlebury College, Middlebury, Vermont.

Klassen, R.W., and Morison, S.R.. 1987. Surficial geology, Laberge, Yukon Territory (1:250000-scale). Map 8-1985. Ottawa: Geological Survey of Canada.

Klassen, R.W., Morison, S.R., and Duk-Rodkin, A. 1987. Surficial Geology, Carmacks, Yukon Territory (1:250000-scale). Map 9-1985. Ottawa: Geological Survey of Canada.

McCune, B., and Mefford, M.J. 1999. PC-ORD for Windows: Multivariate analysis of ecological data. Gleneden Beach, Oregon: MjM Software.

Miller, R.G., Jr. 1966. Simultaneous statistical inference, 1st ed. New York: McGraw-Hill Book Company.

Moss, E.H. 1952. Grassland of the Peace River region, western Canada. Canadian Journal of Botany 30(1):98-124.

Mueller-Dombois, D., and Ellenberg, H. 1974. Aims and methods of vegetation ecology, 1st ed. New York: John Wiley & Sons.

Oswald, E.T., and Senyk, J.P. 1977. Ecoregions of Yukon Territory. Victoria, British Columbia: Canadian Forest Service, Pacific Forestry Centre.

Pojar, J. 1986. Vegetation and ungulate habitat in the Gladys Lake Ecological Reserve, northern British Columbia. Research Report RR85009-PR. Victoria: British Columbia Ministry of Forests.

Redmann, R.E., and Schwarz, A.G. 1986. Dry grassland plant communities in Wood Buffalo National Park, Alberta. Canadian Field-Naturalist 100(4):526-532.

Sanborn, P. 2010. Topographically controlled grassland soils in the Boreal Cordillera ecozone, northwestern Canada. Canadian Journal of Soil Science 90(1):89-101.

Schwarz, A.G., and Wein, R.W. 1997. Threatened dry grasslands in the continental boreal forests of Wood Buffalo National Park. Canadian Journal of Botany 75(8): 1363-1370.

Scudder, G.G.E. 1997. Environment of the Yukon. In: Danks, H.V., and Downes, J.A., eds. Insects of the Yukon. Monograph Series No. 2. Ottawa: Biological Survey of Canada. 13-57.

Seip, D.R., and Bunnell, F.L. 1985. Species composition and herbage production of mountain rangelands in northern British Columbia. Canadian Journal of Botany 63(11):2077-2080.

Smith, C.A.S., Meikle, J.C., and Roots, C.F., eds. 2004. Ecoregions of the Yukon Territory: Biophysical properties of Yukon landscapes. PARC Technical Bulletin 04-01. Summerland, British Columbia: Agriculture and Agri-Food Canada. 313 p.

StatSoft. 1995. STATISTICA for Windows, Version 5.0. Tulsa, Oklahoma: StatSoft Inc.

Stringer, P.W. 1973. An ecological study of grasslands in Banff, Jasper, and Waterton Lakes National Parks. Canadian Journal of Botany 51(2):383-411.

Strong, W.L. 1996. Vegetation resource survey and impact assessment of the proposed Cheviot Mine and associated areas (Appendix 27). In: Cheviot Mine project application. Hinton, Alberta: Cardinal River Coals Ltd. Available at University of Calgary, High Density Library, Calgary, Alberta T2N 1N4.

--. 2002. Assessing species abundance unevenness within and between plant communities. Community Ecology 3(2):237-246. http://dx.doi.Org/10.1556/comec.3.2002.2.9

--, 2013. Ecoclimatic zonation of Yukon (Canada) and ecoclinal variation in vegetation. Arctic 66(1):52-67.

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, Oswald, E.T., and Downing, D.J., eds. 1990. The Canadian vegetation classification system: First approximation. Ecological Land Classification Series No. 25. Ottawa: Environment Canada.

Strong, W.L, Redburn, M.J., and Gates, C.C. 2009. Climate and vegetation change during the twentieth century in the lower Peace River district, northern Alberta, Canada. The Holocene 19(2): 199-207.

Strong, W.L, Chambers, J.H.S., and Jung, T.S. 2013. Range constraints for introduced elk in southwest Yukon, Canada. Arctic 66(4):470-482.

Thieret, J.W. 1959. Grassland vegetation near Fort Providence, Northwest Territories. Canadian Field-Naturalist 73(3):161-167.

Vetter, M.A. 2000. Grasslands of the Aishihik-Sekulmun Lakes area, Yukon Territory, Canada. Arctic 53(2):165-173.

Viereck, L.A., Dyrness, C.T., Batten, A.R., and Wenzlick, K.J. 1992. The Alaska vegetation classification. PNW-GTR-286. Portland, Oregon: USDA Forest Service.

Wilkinson, K., and Johnson, E.A. 1983. Distribution of prairies and solonetzic soils in the Peace River district, Alberta. Canadian Journal of Botany 61(7):1851-1860.

(Received 13 January 2014; accepted in revised form 10 June 2014)

Wayne L Strong (1)

(1) Arctic Institute of North America, 2500 University Drive NW, Calgary, Alberta T2N 1N4, Canada; present address: Yukon Research Centre, Yukon College, 500 College Drive, PO Box 2799, Whitehorse, Yukon Y1A 5K4, Canada;

TABLE 1. Species composition, site characteristics, and net primary
productivity of boreal grassland sociations in the Carmacks-Pelly
Crossing area in Yukon, Canada. Group comparisons are based on
Kruskal-Wallis tests. See Table 2 for list of other occurring taxa.


                                Calamagrostis    Pseudoroegneria
                                purpurascens         spicata

Taxonomic Composition (1)         Average percent cover [SD]
                                constancy, Scheffe test (2,3)
Androsace septentrionales        +[+]6a (4)          +[+]3a
Anemone patens                     2[4]5b            +[+]0a
Artemisia frigida                  4[3]9a            8[6]9ab
Calamagrostis purpurascens        44[ll]10b          +[1]1a
Carex spp. (5)                     3[5]6a            +[1]2a
Carex supina                       2[5]4ab            +[2]a
Hesperostipa comata                  0a                0a
Penstemon gormanii                  1[1]6             2[2]7
Poa glauca                         +[1]2a            +[1]2a
Potentilla pensylvanica            1[1]7a            1[1]6a
Pseudoroegneria spicata            +[1]2a           30[14]10b
Cladonia spp. (as squamules)       1[2]5ab           +[1]2a
Flavocetraria cucullata            2[2]9a            4[6]8a

Botanical Characteristics      Average [SD] Scheffe test
Total plant cover (%)              63[14]b           48[16]a
Bare ground (%)                    36[20]b           50[18]b
Taxa richness per releve          9.5[3.l]b         6.4[2.6]a
Vascular plant richness           8.1[3.0]b         5.2[2.0]a
  per releve
Dominance concentration (6)      0.63[0.08]b       0.52[0.12]a

Site Characteristics           Average [SD] Scheffe test
Elevation (m)                      618[92]           565[42]
Slope (%)                          57[13]b           67[12]b
Aspect ([degrees])                 205[13]           213[32]
June 22 solar insolation         3.0[0.3]ab         2.7[0.4]a
Number of releves                    49                25

Net Primary Productivity       Average [SD] Scheffe test
Graminoids                        298[88]b          168[32] a
Forbs                              48[27]            61[51]
Total                             346[91]b          229[38]a
Number of samples (7)                10                 9


                                Hesperostipa      Poa glauca
                                   comata         Artemisia

Taxonomic Composition (1)         Average percent cover [SD]
                                constancy, Scheffe test (2,3)
Androsace septentrionales          +[+]1a           +[+]7a
Anemone patens                       0a             1[2]5a
Artemisia frigida                  3[4]7a          12[11]9b
Calamagrostis purpurascens         +[+]2a           +[+]1a
Carex spp. (5)                     +[1]5a         13[11]10b
Carex supina                      +[+]2ab          8[11]7b
Hesperostipa comata              49[20]10b          +[+]1a
Penstemon gormanii                 1[1]4            1[1]6
Poa glauca                           0a           18[14]10b
Potentilla pensylvanica            +[1]2a           2[3]7a
Pseudoroegneria spicata              0a             2[3]a
Cladonia spp. (as squamules)      +[1]5ab          10[18]6b
Flavocetraria cucullata            4[5]9a         11[10]10b

Botanical Characteristics      Average [SD] Scheffe test
Total plant cover (%)             63[18]b          73[19]b
Bare ground (%)                   31[17]ab         17[15]a
Taxa richness per releve         6.5[2.7]ab       9.6[2.5]b
Vascular plant richness          5.0[1.0]a        8.1[4.0]b
  per releve
Dominance concentration (6)     0.61[0.17]ab     0.49[0.08]a

Site Characteristics           Average [SD] Scheffe test
Elevation (m)                     579[51]          591[54]
Slope (%)                         58[7]ab          49[10]a
Aspect ([degrees])                182[24]          195[10]
June 22 solar insolation         3.0[0.2]ab       3.2[0.2]b
Number of releves                    8                16

Net Primary Productivity       Average [SD] Scheffe test
Graminoids                       248[71]ab         130[70]a
Forbs                              42[51]          109[61]
Total                            290[117]ab       239[59]ab
Number of samples (7)                5                10


Taxonomic Composition (1)

Androsace septentrionales       0.003
Anemone patens                 < 0.001
Artemisia frigida              < 0.001
Calamagrostis purpurascens     < 0.001
Carex spp. (5)                 < 0.001
Carex supina                   < 0.001
Hesperostipa comata            < 0.001
Penstemon gormanii              0.245
Poa glauca                      <0.001
Potentilla pensylvanica         0.047
Pseudoroegneria spicata         <0.001
Cladonia spp. (as squamules)    0.023
Flavocetraria cucullata         0.002

Botanical Characteristics
Total plant cover (%)          < 0.001
Bare ground (%)                 <0.001
Taxa richness per releve        <0.001
Vascular plant richness         <0.001
  per releve
Dominance concentration (6)    < 0.001

Site Characteristics
Elevation (m)                   0.111
Slope (%)                      < 0.001
Aspect ([degrees])              0.075
June 22 solar insolation       < 0.001
Number of releves

Net Primary Productivity
Graminoids                      <0.001
Forbs                           0.058
Total                           0.016
Number of samples (7)

(1) Includes taxa that occurred in 50% or more of the releves in one
or more sociations.

(2) Constancy classes: 0[0-5%]; 1[6-15%]; 2[16-25%]; 3[26-35%];
4[36-45%]; 5[46-55%]; 6[56-65%]; 7[66-75%]; 8[76-85%]; 9[86-95%]; and

(3) Sociations with the same Scheffe rank test letter do not differ at
the a = 0.05 level.

(4) A "+" represents a value of less than 0.56.

(5) Includes C. supina, C. duriuscula, C.filifolia, C. obtusata, and
C. aurea.

(6) Based on a scale of 0 to 1, in which 1 represents maximum species
dominance concentration (Strong, 2002).

(7) Two samples were excluded from analysis because they were
unclassified with respect to sociation membership (Fig. 3).

TABLE 2. List of flora encountered in the Carmacks-Pelly
Crossing grassland releves that do not appear in Table 1.

Taxa                                    % constancy

Achillea millefolium                         2
Arnica angustifolia                          2
Artemisia campestris                         3
Artemisia laciniata                          1
Aster alpinus ssp. vierhapperi               4
Astragalus laxmannii var. tananicus          1
Astragalus agrestis                          8
Astragalus australis                         15
Boechera retrofracta                         25
Bupleurum americanum                         23
Castilleja yukonis                           6
Cerastium arvense                            1
Chamaerhodos erecta                          17
Chamerion angustifolium                      3
Elymus lanceolatus ssp. psammophilus         3
Erigeron glabellus spp. pubescens            4
Erysimum coarctatum                          6
Festuca saximontana var. saximontana         2
Galium boreale                               13
Juniperus horizontalis                       1
Koeleria macrantha                           6
Linum lewisii                                4
Minuartia yukonensis                         4
Nestotus macleanii                           1
Oxytropis campestris ssp. varions            13
Phlox hoodii                                 4
Plantago canescens                           6
Polemonium pulcherrimum                      2
Potentilla nivea                             6
Rosa acicularis                              3
Saxifraga reflexa                            3
Saxifraga tricuspidata                       2
Solidago simplex                             11
Stellaria longipes                           1
Townsendia hookeri                           1
Anticlea elegans                             1
Cladonia amaurocraea                         1
Flavocetraria nivalis                        1
Peltigera rufescens                          2
Psora decipiens                              1
Xanthoparmelia wyomingica                    2
Unidentified mosses                          5
COPYRIGHT 2015 Arctic Institute of North America of the University of Calgary
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Strong, Wayne L.
Article Type:Report
Geographic Code:1CYUK
Date:Mar 1, 2015
Previous Article:Wolf (Canis lupus) winter density and territory size in a low biomass moose (Alces alces) system.
Next Article:Micrometeorological and thermal control of frost flower growth and decay on young sea ice.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |