Graduation date: 2008I developed a priori hypotheses and used logistic regression to model Greater Sandhill Crane (Grus canadensis tabida) nest success in relation to weather, habitat and management variables for cranes breeding at Malheur National Wildlife Refuge (MNWR) in southeast Oregon. My primary interest was to investigate the effects of habitat conditions and management practices on nest success as these factors can be influenced by managers. However, I also included variables for weather, water supply, and nest initiation date to provide a context for understanding the relative importance of management actions and habitat conditions. I monitored 506 nests over 9 breeding seasons; mean apparent nest success was 72% ± 4% and varied from 51 to 87%. Nest success varied by habitat management within a field, nest-season temperatures, and water at the nest site. A second analysis which included a subset of nests with nest initiation data revealed that nest success declined with initiation and was higher during years of moderate recipitation. Variable importance analysis indicated initiation date had a greater influence on nest success (0.99) than land use (0.74), water depth (0.50), water supply (0.42), or temperature (0.30). Grazing, haying and predator control had no effect on nest success, while burning had a negative effect, but only during the nesting season following the burn. The lack of a predator control effect may have been due to elimination of individual problem animals before the study commenced, and a protracted effect of the control program. It may also be due to the relatively smaller importance of predator control at higher nest success rates. Nest success at MNWR was higher during my study than in most studies and the low proportion of failed nests may have reduced my ability to detect influences of some variables such as land use practices and predator control. My study is the first to quantify the effect of temperature and moisture conditions on nest success and the results suggest adequate water availability is important for nesting cranes. It affirms the importance of water depth, but not concealment, indicating that the height of cover surrounding a nest is not very important. During the 8-year coyote (Canis latrans) and Common Raven (Corvus corax) control program at MNWR non-target predator populations increased. Such control programs can have unforeseen consequences, such as release of mesopredator populations which can cause additional crane productivity problems. Nest success was relatively high during my study yet productivity was very low, averaging around 3.3 young per 100 pairs from 1995-1998 (GLI, unpublished data). Efforts to further increase nest success at MNWR using predator control are not warranted. The effect of predators on chick survival is likely of more importance and should be quantified to aid in the interpretation of predator control as a management tool. Managers should focus on providing early water to crane territories to facilitate earlier nesting and water levels should be maintained through the nesting season to limit abandonment. Haying and grazing of meadows will likely encourage cranes to nest in deeper water marsh sites. Although burning showed negative effects on nest success during the first season following burns, I advocate continued, but careful use of this important habitat management tool. Lastly, in addition to management to improve nest success, managers should consider efforts to enhance chick and adult survival, as these factors are likely more important than nest success in maintaining populations.