Relationships of Site Characteristics to Vegetation in Canyon Grasslands of West Central Idaho and Adjacent Areas

Abstract

The relation of vegetation types to soil and other site characteristics was examined for 57 sample plots representing the Pacific Northwest Bunchgrass Region. Three characterized by Carex spp., Festuca idahoensis, and Agropyron spicatum respectively, and 5 habitat types comprised the vegetation units. These were compared to their associated soil taxa (soil families) and to a group of individual soil and other site characteristics. Relationship to soil taxa was relatively weak, with several soil families associated with each of 4 of the habitat types. Strong relationship of vegetation types to 13 individual soil and site factors was shown by means of stepwise discriminant analysis. Reclassification by these site factors resulted in 92% concurrence with habitat types and even higher agreement with vegetation series. Site factors showing the highest degree of relationship with vegetation units were: elevation, radiation index, color (value), and organic matter of the horizon, and lime depth. This method of relating individual site factors to vegetation provides a powerful tool for testing the validI ty of ecosystems recognized by vegetation, and should be useful also in categorizing sites where plant cover has been disturbed. The grasslands of the Snake and Salmon River valleys in Idaho and adjacent parts of eastern Oregon and Washington constitute a distinctive section of the Pacific Northwest Bunchgrass Region. These grasslands occur mainly on steep canyon slopes and are closely related to the so-called Palouse grasslands of the Columbia Plateau in northeastern Oregon, eastern Washington, and northwestern Idaho. A brief description of this canyon area and a preliminary classification of its grassland has been published (Tisdale 1979). Due to a combination of rugged topography, dry climate, and stony soils, most of the grassland remains uncultivated. Despite the influence of heavy use by livestock over much of the area, many examples of relatively undisturbed vegetation remain. From a study of such relict areas, Tisdale (1979) recognized 2 vegetation and 5 habitat types which constitute most of the grassland vegetation of the area. The Festuca idahoensis includes 3 habitat types: Festuca idahoensisIKoeleria cristata, Festuca idahoensisfAgropyron spicatum, and Festuca idahoensisISymphoricarpos albus. The other group, occupying drier sites, is the Agropyron spicatum which includes Agropyron spicatumI Poa sandbergii and the Agropyron spicatuml Opuntia polyacantha habitat types. Subsequent studies indicated the existence of a third vegetational characterized by the co-dominance of Carex hoodii, Authors are professor emeritus, Dept. of Range Resources, and consultant on range and statistics, Dept. of Plant and Soil Science, Univ. of Idaho, Moscow 83843. The authors are grateful to the Forest, Wildlife and Range Experiment Station, University of Idaho, for use of facilities and for partial financial support for this study. Particular thanks are due to Dr. Minoru Hironaka, professor of range resources, for his continued interest and support, to Dr. M.A. Fosberg, professor of soil science, who identified most of the soils, and to Dr. Kirk Steinhorst, associate professor of agricultural applied statistics, for his advice on the statistical analyses. Manuscript received July 23, 1982. C. geyeri, and other upland Carex species. One habitat type, Carex hoodiil Festuca idahoensis, has been recognized to date. This type is restricted to the highest elevations of the grasslands, and has not been previously described in the literature. In addition to the dominance of Carex, it is characterized by the common occurrence of Bromus carinatus, Danthonia intermedia, Poa nervosa, Antennaria anaphaloides, and Eriogonum flavum, species often associated with subalpine areas and rare or lacking in grasslands of lower elevation. Data from this habitat type are included in the present study. The classification approach used here recognizes the series as a group of communities characterized by a single dominant species. The habitat type is considered to be the aggregate of all areas that support or can support the same primary climax (Daubenmire 1970). The habitat type (h.t.) has relatively uniform biotic and abiotic structure and is the primary unit of ecosystem classification. It is recognized by means of its vegetation, but is characterized also by distinctive habitat features. Following preliminary classification of the vegetation, a second objective was to determine the relationship of habitat factors to the plant communities. We desired to know whether these vegetational groups represented recognizable ecosystems when considered on the basis of site characteristics only.