Soil Texture Gradient Research Articles

Protozoa, nematodes and N-mineralization across a prescribed soil textural gradient

Summary The increase in protozoan and nematode populations following addition of glucose or barley leaf material to five different mixtures of a sandy loam and a silty clay loam was investigated in 2 experiments. Prescribed soil textures (varying in clay content from 15.6% to 28.6%) were incubated at a matric potential of —10 kPa at 15 °C, and the number of protozoa and nematodes and the amount of inorganic nitrogen were estimated after 0, 2 and 5 weeks. In the first experiment, the effect of amendment with glucose was compared with amendment with barley leaves. Numbers of protozoa increased in soil mixtures amended with both glucose and barley leaves, but nematodes only increased in the treatment with barley leaves. There was a large positive effect of the amount of fine-textured soil on the number of protozoa, whereas the nematodes were not affected by soil texture. In the second experiment, the effect of nematodes on protozoa and nitrogen mineralization was examined. Soil mixtures prepared with sterilised soil were amended with barley leaves and either (1) a soil suspension filtered through a 5 μm mesh to remove nematodes, or (2) a filtered soil suspension and a mixture of nematodes extracted from soil. The nematodes that multiplied in the soil mixtures were almost exclusively bacterial-feeding rhabditids. The nematodes had a significantly positive effect on the number of protozoa but an insignificant effect on N-mineralization. Both protozoa and nematodes were affected positively by the proportion of the fine-textured soil in the soil mixtures, but the positive effect on protozoa was larger than the effect on nematodes.

Interpretation of soil enzyme activities in a comparison of tillage practices along a topographic and textural gradient

Assessment of the impacts of soil conservation practices requires measurement of the resulting changes in soil quality at scales inclusive of soil variability comparable to that of typical farms. The objective of this study was to compare changes in six soil enzyme activities (urease, glutaminase, phosphatase, arylsulfatase, β-glucosidase and dehydrogenase) and organic carbon (OC) content resulting from implementation of no-till along a topographic and soil textural gradient. Activities of β-glucosidase, glutaminase, phosphatase and arylsulfatase, and OC content were greater in coarse-textured soils at a lower slope-position than in fine-textured soil at an upper slope-position. Tillage practice influenced activities of urease, glutaminase, β-glucosidase and dehydrogenase but not OC content. The effect of tillage practice on enzyme activities was influenced by sampling depth and slope-position. Phosphatase and arylsulfatase activities of the Ap horizon behaved as indices of soil organic matter (SOM) content along the topographic gradient. Urease and dehydrogenase activities behaved as indices of soil biochemical activity within the Ap horizon. When interpreted in these terms, soil enzyme activities are comparable to other integrative measurements and as such describe system behavior and attributes. Key words: Soil enzyme activity, no-till, topographic position

Distribution patterns of vertebrates in relation to an extensive rainfall gradient and variation in soil texture in the tropical savannas of the Northern Territory, Australia

The abundance, richness and species composition of frog, reptile, bird and mammal faunas varied along an extensive gradient from 470 to 1406 mm annual rainfall, and between three contrasting soil types (clay, loam and sand) in northern Australia. Patterns varied between and within vertebrate classes. In general, the most fertile soils in the highest rainfall sites supported the greatest species richness and abundance, but this association with fertility broke down at lower rainfall sites. Frogs were richest and most abundant at high rainfall sites, especially on clay soils, presumably because these had greatest water availability. Clay soils supported few reptile species, but these were often at relatively high abundance. High rainfall sites supported the richest reptile faunas. On sand and loam soils, bird species richness varied little along the rainfall gradient, but richness declined very substantially on clay soils. This was probably largely due to the far more marked vegetation structural change on clay soils than on other substrates. Few mammals were reported, and no clear trends were associated with either rainfall gradient or soil texture. Turnover in species composition along the rainfall gradient was gradual and limited on sand and loam soils, but far more marked on clay soils. There were few cases of replacements of ecologically comparable species along the gradient. These patterns reflect the disparate history, fragmentation and landscape positioning of clay soil environments, relative to the far more homogeneous eucalypt-dominated vegetation on sand and loam soils. Although comparable studies are lacking on other continents, patterns revealed here may be idiosyncratic due to the virtual monopolisation by eucalypts of the environment across the very extensive rainfall gradient.

Solute Diffusion and Adsorption in Six Soils along a Soil Texture Gradient

The Millington-Quirk (MQ) model is widely used to predict solute diffusivity (ratio of diffusion coefficients in soil and free water) in transport and fate models for nutrients and organic contaminants in soil. Recently, we suggested two soil-type dependent (STD) diffusivity models, a simple power function model and a linear impedance factor model, based on the Campbell soil water retention parameter. In this study, soil water retention and Cl- diffusivity as a function of soil water content were measured in six soils taken along a natural clay gradient (11–46% clay) with the objective of testing the STD and MQ diffusivity models. Both new STD models predicted solute diffusivity well for all six soils, while the MQ model poorly predicted diffusivity for the three most clayey soils. Effective diffusivity of a sorbing solute, NH+4, was evaluated by measuring NH+4 sorption isotherms and calculating retardation factors. The data for the six soils showed an almost linear decrease in Cl- and NH+4 effective diffusivity with increasing clay content at a given soil water content. Numerical calculations, based on the measured diffusivity data, of mean diffusion time from a source of constant strength implied that changes in soil texture (11–46% clay) and soil water content (0.2–0.4 cm3 cm-3) had equally large effects on diffusion times. We recommend that one of the suggested STD diffusivity model should be used instead of the MQ model in order to obtain better solute transport and fate simulations.

Relationships Between Soil Organic Carbon and Site Characteristics in Peninsular Spain

The relationships between soil organic C (SOC) content and soil and site properties, including climate, land use, and other variables, are important for assessing the effects of future land use and climate changes on soil organic C stocks. Data were obtained for 766 soil profiles taken from published soil studies of sites throughout peninsular Spain. Relationships between SOC and mean annual precipitation, mean annual temperature, land use, soil moisture regime, altitude, texture, and slope gradient were statistically analyzed by simple correlation, stepwise multiple regression analysis, and error source estimation. The variables best correlated with SOC were mean annual precipitation and the length of dry summer season. Land use alone explained 33% of the SOC variability. Almost 50% of the SOC variability was explained by all the variables included in the study. Almost 30% of the SOC variability could be attributed to errors in the values assigned to climate variables, SOC measurements, and within-site SOC variability. If the relationships found between SOC and climate and land use hold in the future, a moderate degree of climate change (10% increase in temperature and 10% decrease in precipitation) could lead to a 15% loss of SOC from soils in the most unfavorable case. The magnitude of the effects of land use changes on SOC content, for example, from cropland to scrubland, may equal that of a climate change.

Woody vegetation structure of xeric forest stands under different edaphic site conditions and disturbance histories in the Biosphere Reserve 'Parque Costero del Sur', Argentina

We studied stand structures and soil properties in an old-growth forest and two 30-yr-old second-growth stands. In the old-growth forest, the total density and basal area average 1566 trees > 1.25 m height ha-1 and 46.73 m2 ha-1. The dominant trees are Scutia buxifolia and Celtis tala. Using multivariate techniques we distinguished three stands: PS1, dominated by S. buxifolia, is 1000 m far from the river. Its soil is shallow and loamy. PS2 and PS3, co dominated by S. buxifolia and C. tala, are over 1200 m distant from the river. There the soil is deeper and has thicker texture and higher phosphorus and calcium concentrations than the near-the-river forest soil. Scutia buxifolia shows reverse J-shaped size-distributions and has morphological features of stress-tolerant species. Celtis tala shows oscillating decay size-curves that suggest recruitment pulses related to small gaps and it has morphological features of competitive species. Celtis tala was selectively cut in the past in the second-growth stands SNRD and SRD. The stand SNRD, 1000 m far from the river, is dominated by S. buxifolia. The second species is Schinus polygamus which presents the bell-shaped size-structure of the pioneer species. SNRD does not differ from its old-growth counterpart PS1 in total tree density, basal area and tree branching. The stand SRD, over 1200 m distant from the river, is co-dominated by S. buxifolia and by C. tala trees regenerated from stumps. SRD does not differ from its old-growth counterparts PS2 and PS3 in total tree density and basal area. As to tree branching, it does not differ from PS3, but differs from PS2. All the stands are being invaded by the exotic tree Ligustrum lucidum. The differences between the old-growth stands seem to be related to the gradients of soil texture and nutrient concentrations raising edaphic stress towards the river. In SNRD, the stress, the stress-tolerance of S. buxifolia, and the aptitude of S. polygamus to recruit in disturbed habitats seem to have prevented the post-logging recruitment of C. tala. In SRD, C. tala regenerated possibly due to a better competitive performance in a more favorable site. Under protection the second-growth stands recovered the old-growth quantitative features. We recommend the restoration of the qualitative features and the control of L. lucidum.

Transformation and retention of nitrogen in a coastal forest ecosystem

Transformations and fluxes of N were examined in three forested sites located along a gradient of soil texture in the coastal forests of the Waquoit Bay watershed on Cape Cod. Total N leaching losses to ground water were 0.5 kg ha-1 yr-1 in the loamy sand site and 1.5 kg ha-1 yr-1 in the fine sand site. Leaching loss to groundwater was not measured in the coarse sand site due to the prohibitive depth of the water table but total N leaching loss to 1m depth in the mineral soil was 3.9 kg ha-1 yr-1. DON accounted for most of the leaching losses below the rooting zone (77–89%) and through the soil profile to ground water (60%–80%). Differences in DON retention capacity of the mineral soil in the sites along the soil texture gradient were most likely related to changes in mineral soil particle surface area and percolation rates associated with soil texture. Forests of the watershed functioned as a sink for inorganic N deposited on the surface of the watershed in wet and dry deposition but a source of dissolved organic N to ground water and adjoining coastal ecosystems.

Application of a 15N tracer to simulate and track the fate of atmospherically deposited N in the coastal forests of the Waquoit Bay Watershed, Cape Cod, Massachusetts.

We examined patterns of N retention in the coastal forests of the Waquoit Bay watershed on Cape Cod, Masschusetts using 15N tracer techniques. A solution of 99.6% enriched 15N -NO3-, at a concentration similar to that of background throughfall, was applied to forest plots established along a gradient of soil texture to simulate and track the fate of throughfall NO3- deposition. The tracer solution was applied to replicate plots during both the spring and fall to examine seasonal differences in ecosystem retention. 15N enrichment was subsequently measured in litter, O2 horizon, 0-15 cm mineral soil, fine roots, microbial biomass in the O2 horizon and mineral soil, and lysimeter leachate over a 6 month period following each application. The O2 horizon contained the largest fraction of 15N in all sites immediately following the spring application (19-45%) but was less important following the fall application (10-25%). The mineral soil N pool generally contained the largest fraction of applied 15N (7-28%) in all sites at the end of both 6-month sampling periods. Microbial uptake of applied 15N provided an initial barrier against leaching loss as well as a mechanism for its long-term incorporation into soil organic matter. Microbial processing was less important in the most coarsely textured site, perhaps as a result of lower substrate availability and smaller microbial pool sizes. The highest cumulative leaching losses of applied 15N were observed in the coarse sand site (40, 51%) followed by the fine sand (13, 43%) and loamy sand (4, 19%) sites for the spring and fall applications, respectively. More than 90% of all 15N captured in lysimeters occurred within two days following the applications, and 25-43% of all 15N captured in lysimeters after 2 days was in the form of dissolved organic nitrogen (DON) indicating that it had been assimilated by microbes prior to leaching.

The scaling characteristics of soil parameters: From plot scale heterogeneity to subgrid parameterization

The variation in soil texture, surface moisture or vertical soil moisture gradient in larger scale atmospheric models may lead to significant variations in simulated surface fluxes of water and heat. The parameterization of soil moisture fluxes at spatial scales compatible with the grid size of distributed hydrological models and mesoscale atmospheric models (∼ 100 km 2) faces principal problems which relate to the underlying microscopic or field scale heterogeneity in soil characteristics. The most widely used parameterization in soil hydrology, the Darcy-Richards (DR) equation, is gaining increasing importance in mesoscale and climate modelling. This is mainly due to the need to introduce plant-interactive soil water depletion and stomatal conductance parameterizations and to improve the calculation of deep percolation and runoff. Covering a grid of several hundreds of square kilometres, the DR parameterization in soil-vegetation-atmosphere-transfer schemes (SVATs) is assumed to be scale-invariant. The parameters describing the non-linear, area-average soil hydraulic functions in this scale-invariant DR-equation should be treated as calibration-parameters, which do not necessarily have a physical meaning. The saturated hydraulic conductivity is one of the soil parameters to which the models show very high sensitivity. It is shown that saturated hydraulic conductivity can be scaled in both vertical and horizontal directions for large flow domains. In this paper, a distinction is made between effective and aggregated soil parameters. Effective parameters are defined as area-average values or distributions over a domain with a single, distinct textural soil type. They can be obtained by scaling or inverse modelling. Aggregated soil parameters represent grid-domains with several textural soil types. In soil science dimensional methods have been developed to scale up soil hydraulic characteristics. With some specific assumptions, these techniques can be extrapolated from classical field-scale problems in soil heterogeneity to larger domains, compatible with the grid-size of large scale models. Particularly promising is the estimation of effective soil hydraulic parameters from area averaging measurements through inverse modelling of the unsaturated flow. Techniques to scale and aggregate the soil characteristics presented in this paper qualify for direct or indirect use in large scale meteorological models. One of the interesting results is the effective behaviour of the reference curve, which can be obtained from similar media scaling. If the conclusions of this paper survive further studies, a relatively simple method will become available to parameterize soil variability at large scales. The inverse technique is found to provide effective soil parameters which perform well in predicting both the area-average evaporation and the area-average soil moisture fluxes, such as subsurface runoff. This is not the case for aggregated soil parameters. Obtained from regression relationships between soil textural composition and hydraulic characteristics, these aggregated parameters predict evaporation fluxes well, but fail to predict water balance terms such as percolation and runoff. This is a serious drawback which could eventually hamper the improvement of the representation of the hydrological cycle in mesoscale atmospheric models and in GCMs.

Relationships between spatial pattern of the endogeic earthworm Polypheretima elongata and soil heterogeneity

Spatial distribution of the endogeic earthworm Polypheretima elongata, soil texture, carbon and nitrogen contents were investigated in a 10-y-old pasture established on a vertisol, by measuring each variable at 57 points randomly located in a 25 × 60 m plot. Soil variables were measured at three depths: 0–10, 10–20 and 20–30 cm. Principal component analysis (PCA) of the data set showed the absence of relationships between density of P. elongata and environmental variables. The first principal component expressed a marked gradient of soil texture and soil organic matter (SOM). Sampling point coordinates on the first principal component axis were used as a composite variable assumed to account for the major source of soil heterogeneity at this site. The spatial pattern of this variable and three age classes of P. elongata were investigated using geostatistics. There was no evidence of any relationship between soil heterogeneity and earthworm distribution. Population of P. elongata exhibited a marked patchy distribution with opposite patterns between adults and other age classes (juveniles and cocoons). These results suggest that at the study site processes responsible for the spatial distribution of P. elongata population are of demographic rather than environmental origin.

Vegetation and its relation to soil nutrient and salinity in the Calabar mangrove swamp, Nigeria

The study examines vegetation – environment relationships. Vegetation measurements included species frequency, density, diameter and tree height, while environmental measurements were soil particle size distribution, acid properties (pH, Al, SO4), nutrient cations (Ca, Na, Mg, K), organic carbon, nitrogen, phosphorus and chloride content. Nypa fruticans was the dominant species in the A stratum (> 3 m tall) while Rhizophora mangle was dominant in the B stratum (1–3 m tall). The C stratum (< 1 m tall) was dominated by mangrove, Nypa and Raphia saplings. Silt was dominant and the most variable particle size fraction. A principal components analysis of the soil data indicated the first three dominant components influencing the vegetation were salinity, nutrient and soil texture. Tree height and density correlated highly with the salinity and soil texture gradients (P < 0.01), while basal area correlated with salinity and nutrient gradients (P < 0.01). While Avicannia africana in the A stratum was influenced largely by the salinity and soil texture gradients. Nypa fruticans in the B stratum was influenced by salinity and nutrients.

Long‐Term Patterns in Coastal California Grasslands in Relation to Cultivation, Gophers, and Grazing

Grasslands at the Hastings Natural History Reservation (HNHR) and in adjacent Santa Lucia coastal range of Monterey County, California were sampled from 1971 to 1991. Grasslands on HNHR showed two distinct and stable associations: stands with and without historical cultivation (≈ 1865‐1937$). Relict stands dominated by native, perennial grasses (e.g., Nassella pulchra, Poa secunda) are limited to uncultivated, steeper stands, often where soils have more clay. Abandoned agricultural fields have stable compositions dominated by Avena fatua, Bromus mollis, B. diandrus, Erodium spp., Hypochaeris glabra, Vulpia spp., Eremocarpus setigerus, and Amsinckia spp. Patterns in species composition were associated with gradients in soil texture, gopher abundance, and slope. Gophers provide a significant and continuous source of soil disturbance and may slow successional processes in old fields. Where gophers are excluded, aboveground biomass accumulates. Germination and establishment of native perennial grasses (compared to introduced, annual grasses) are reduced on gopher tailings in old fields. Species composition patterns reflecting past cultivation on both grazed and ungrazed stands are apparent. Relict (uncultivated) stands of native grasses persist under many historical levels of grazing. Effects of grazing are often only seen on old fields, and not on relict grasslands. Compared to stands where grazing was removed in 1937, stands currently or recently grazed by cattle show higher soil nitrogen, but reductions in cover of gopher tailings, species diversity, soil phosphate, and sulphate.

Flow Regulation, Geomorphology, and Colorado River Marsh Development in the Grand Canyon, Arizona

New, productive fluvial marshes may develop along regulated canyon rivers through reduction in flood frequency, thereby increasing diversity, production, and wildlife habitat availability. Few fluvial marshes occurred along the eddy‐dominated Colorado River in the Grand Canyon prior to construction of Glen Canyon Dam in 1963. Reduction of flooding after 1963 permitted widespread marsh development. Fluvial marshes exhibited low stability but high resilience, quickly redeveloping after scouring by high flows between 1983 and 1986. In 1991, 253 fluvial wet marshes (cattail/reed and horseweed/Bermudagrass) and 850 dry marshes (horsetail/willow) occupied 25.0 ha (1%) of the 363 km mainstream riparian corridor between Lees Ferry and Diamond Creek, Arizona. Fluvial marsh development and composition varied in relation to local and reach‐based geomorphology, and microsite gradients in inundation frequency and soil texture. Colorado River marsh density (number/km2) increased with distance downstream, and marshes were larger and more abundant in wide reaches. Wet marsh cattail/reed stands developed on silty loam soils in low velocity depositional environments that were inundated 54% of the days from 1986 to 1991, whereas dry horsetail/willow marshes occupied less frequently inundated sites with sandy soils. Mean marsh standing mass (641 g C/m2) was comparable with values from regulated alluvial river marshes, but litter retention appeared limited by flow variability in both regulated and unregulated fluvial marshes. We discuss implications of flow management on the four marsh assemblages, and the need for consensus on priorities for management of regulated fluvial wetlands.

Contrasting patterns of resource limitation in tree seedlings across a gradient in soil texture

Resource competition theory for terrestrial plant communities explains gradients in species composition among various habitats as the outcome of differential supply and demand for limiting resources. To examine one group of assumptions of this theory, for forest vegetation in southeastern Texas, we grew seedlings of three broadly distributed tree species, Acerrubrum L. (red maple), Liquidambarstyraciflua L. (sweetgum), and Pinustaeda L. (loblolly pine), in intact soil cores collected from three sites located along a gradient in soil texture. We tested resource limitation in a temperature-controlled greenhouse with nutrient fertilization, light, and moisture treatments applied in a resource-minus or screening design. This design provided most resources at levels not limiting for growth while simulating (potentially limiting) field conditions for a single resource or group of related resources. Light was most limiting with clay-rich floodplain soil cores under simulated floodplain forest conditions. Soil nutrients were most limiting in sandy upland soil cores under upper slope forest conditions. Both light and soil nutrients were strongly limiting with loamy intermediate soil cores under simulated mesic forest conditions. The most limiting soil resource was phosphorus. The pattern of reciprocal limitation by aboveground or belowground resources in extreme habitats and joint limitation in intermediate habitats was consistent with requirements of the theory.

Spatial variation and temporal persistence of grapevine response to a soil texture gradient

Studying the water transport in the soil-plant system requires information on the spatio-temporal variability of both subsystems and the ability to assess the impact of the soil heterogeneity and of the biological responses on the coupling between vegetation and its substrate. This study was conducted for 2 years in a vineyard in the Aude Valley, France, by measuring the particle size distribution of the topsoil, the instantaneous isotopic ratios ( 18 O 16 O , 2 H 1 H ) of leaf water, annual shoot biomass production, and interannual persistence of this biomass along a 360 m transect. The resultant spatial series were analysed for their correlations and converted to spectra. Changes in the isotopic ratios along the transect reflect the soil texture gradient, suggesting that the vines root deeper on the gravel layers than elsewhere. This could provide a mechanism for the partial decoupling between soil and vegetation, and thus explain (1) the strong temporal persistence of the vegetation pattern, (2) the low overall correlation between biomass production and soil texture. The spectra show that this correlation concentrates at specific scales which correspond to a minimum variability in the shoot biomass. In this case, therefore, soil texture plays only a minor role in determining the spatial heterogeneity of shoot biomass in grapevine.

Edaphic factors and competition as determinants of pattern in South African karoo vegetation

The roles of edaphic factors and competition in explaining species distribution of Pteronia (family Asteraceae) dominated plant communities, were investigated. Soil characteristics, species plant covers and competitive interactions were determined at three sites (one on a colluvial plain and two on tillite slopes) on the farm Tierberg in the semi-arid Karoo, South Africa. These sites represented areas where Pteronia species replace each other in apparently distinct habitats. Direct (Canonical Correspondence Analysis, CCA) and indirect (Detrended Correspondence Analysis, DCA) gradient analyses gave similar eigen values, indicating that edaphic variables accounted for much of the variation in species data. The first axis of the CCA-biplot separated species on the colluvial plain from those on tillite slopes along a soil texture gradient. The separation of species in the second CCA axis indicated that soil phosphorous, pH, magnesium and calcium levels were also important in determining species distribution patterns. Together, these axes accounted for 50.1% of the variation in the species data. Nearest-neighbour analyses indicated that there were strong inter- and intraspecific competitive interactions at Pteronia community boundaries. If edaphic specialization alone was responsible for Pteronia distribution patterns, one would expect there to be reduced competition at the boundaries of species overlap. It is concluded that Pteronia species distribution patterns in the Karoo result from the combined effects of soil physico-chemical factors and competition.

Inferences for life history strategies of Artemisia tridentata subspecies

AbstractCommunities dominated by Artemisia tridentata ssp. tridentata, A. tridentata ssp. wyomingensis, and A. tridentata ssp. vaseyana, found in the Piceance Basin of western Colorado, were evaluated for life history strategy. Species cover data were analyzed using Grime's (1984) triangular model. Then a canonical analysis was conducted to obtain an ordination of relative species cover. Results from these two analyses were then used to infer strategy differences among life histories of subspecies, based on the respective plant community associates. These differences were found to be consistent with the divergent evolution assumed to have occurred in this species. The primary ordination axis was interpreted as an elevation‐moisture gradient. Further analysis of soil data by factor analysis also separated the three Artemisia subspecies along a soil texture gradient.

Characterisation of small-scale stand types in a mixed hardwood forest

Gradients in soil moisture, texture, nutrient state and spacing patterns enable us to divide stands into separate substands, which have their own tree species composition and structures. Nevertheless, such a separation is just the beginning of a set of conclusions and observations. It is based upon differences in flora and chemical characteristics. As a matter of fact, every separate sub stand has its own microclimate in which a well defined complex of environmental variables regulates regeneration, flora presence and leaf decomposition. The results of such an intensive research should lead to the improvement of the stand tending.

A Site-Growth Study ofPinus radiatain the Southern Cape

SYNOPSIS An intensive study of the site requirements of Pinus radiata yielded rather disappointing results for the prediction of site index (R2 = 0,47 Root MSE = 3,01m). This is ascribed to the adoption of a design dependent sampling strategy. Significant factors identified by the multiple regression analysis were complex, difficult to interpret and reflected regional gradients in soil texture, temperature and water-balance variables. Pinus radiata responds to a variety of environmental gradients in a complex fashion. The relatively poor performance of soil morphological and chemical factors in the prediction of site index was unexpected.

Patterns of Community Structure and Morphology of Bryophytes and Lichens Relative to Edaphic Gradients in the Subarctic Forest-Tundra of Northwestern Canada

Patterns of Community Structure and Morphology of Bryophytes and Lichens Relative to Edaphic Gradients in the Subarctic Forest-Tundra of Northwestern Canada