Graduation date: 2007Thirteen different habitat parameters were measured either quantitatively or qualitatively around seven different Salix (willow) species on a defined reach of stream on the southwestern slope of Steens Mountain in southeastern Oregon in order to both broaden the current information base of riparian ecology and inform future willow plantings aimed at improving riparian habitat. Willow species included in the research were: S. lemmonii (Lemmon willow), S. scouleriana (Scouler willow), S. boothii (Booth willow), S. geyeriana (Geyer willow), S. exigua (coyote willow), S. lasiandra (whiplash willow), and S. lutea (yellow willow). The habitat parameters measured were: height and distance of willow individual from stream channel bankfull elevation, stream channel slope, entrenchment ratio, width-to-depth ratio, watershed elevation, valley azimuth, valley width, soil rock size distribution, soil texture, and winter air temperature. Logistic regression models were built to describe probabilities of occurrence for five out of the seven willow species based on the habitat data collected. Multi-response permutation procedures were performed to describe differences between respective willow habitats. Booth and Geyer willow habitats were best described by watershed elevation as these species dominated the research stream channel’s upper reaches. Whiplash willow habitat was best described by stream channel entrenchment and width-to-depth ratio as this species occurred, on average, in habitats with high stream channel entrenchment and width-to-depth ratios. Specifically, the odds of whiplash willow occurrence are expected to increase by 17% with a unit increase in width-to-depth ratio. Yellow willow habitat was best described by watershed elevation and stream channel width-to-depth ratio as this species occurred only on the lower reaches of the research stream channel. Lemmon willow habitat was best described by interactions between watershed elevation and valley azimuth and between stream channel slope and width-to-depth ratio as the first of these interactions worked to create air temperatures favorable for species occurrence. Coyote willow habitat was best described by stream channel slope, entrenchment, and width-todepth ratio as this species occurred, on average, in habitats with combinations of stream channel slopes, entrenchment, and width-to-depth ratios able to promote frequent streambank rejuvenation. Scouler willow habitat was best described by stream channel entrenchment ratio, as well as interactions between watershed elevation and valley azimuth and between stream channel slope and width-to-depth ratio. Specifically, the odds of Scouler willow occurrence are expected to decrease by 89% with a unit increase in entrenchment ratio. Scouler willow, on average, occurs in habitats with air temperature dynamics similar to those described above for Lemmon willow, and on hydraulically efficient stream channels with low entrenchment and width-to-depth ratios.