Wetland vegetation of the Blue Mountains, Oregon: Classification, synthesis, and bioassessment

The deep canyon and subalpine vegetation types of the Malheur, Umatilla, and Wallowa-Whitman National Forests were classified based on potential natural vegetation. A plant association is defined as an assemblage of native vegetation in equilibrium with the environment on a specific fluvial surface. The implication is that as the environment (water regime, soils, etc.) changes through time the vegetative potential shifts across that environment space. The classification was designed to overlap with an existing mid-montane northeastern Oregon riparian/wetland classification. Ninety-five vegetation types were classified across the three national forests, including some 45 vegetation types not previously classified for northeastern Oregon subalpine and deep canyon environments.; The local (within plant community) and geo-physiognomic classification (GPC) (a cross-classification of geographic region and vegetation physiognomy) species pools of the deep canyon and subalpine riparian zones of northeastern Oregon (Chapter 2) were compared with local and GPC species pools from adjacent riparian classifications, including central and mid-montane northeastern Oregon, Washington, Idaho, Utah, Nevada, northeastern California, Montana, and Wyoming. Compositional dissimilarity was used as an objective measure of similarity among adjacent regions. Non-metric Multi-Dimensional Scaling (NMDS) was employed as a secondary means of comparison among classifications. The patterns of regional species pool similarities were strongly dependent on the interplay of large-scale geographic and climatic factors, and the development of similar groups of species over millennia.; The reference condition approach to biological monitoring has received considerable attention over the last two decades as a means of detecting impairment in freshwater systems. In chapter 4, the reference condition approach to biological monitoring was applied to riparian zones and wetlands of northeastern Oregon. The reference condition models were associated with high misclassification error rates, suggesting low confidence in an assemblage of riparian vegetation being correctly predicted based on the environment. The poor misclassification rates were largely a factor of the unique character of riparian zones in fostering multiple vegetative potentials, and led to difficulties differentiating true impairment from alternative potential natural vegetation at a site.