the southern portion of the study area migrated northward to 

 areas of greater relief with steep south-facing slopes. Those 

 areas were characteristically poor in forage resources, 

 regardless of deer numbers. The most preferred winter forage, 

 rubber rabbitbrush, was covered by deep snow and was usually 

 absent on the steeper areas where deer are forced to winter. 



Paradoxically, effectively long winters (periods of 

 negative energy balance), were often the result of hot dry 

 springs, summers and autumns rather than deep snow and cold 

 temperature during the calendar months of winter. Dry 

 conditions during spring through autumn resulted in earlier 

 than normal and heavy use of rubber rabbitbrush and snowberry 

 which led to unseasonably early dependence on big sagebrush 

 and Rocky Mountain juniper. 



The nutritional quality of the mule deer diet was 

 influenced at least as much by the density-independent factor 

 of climate as by the number of deer competing for forage. 

 Weather certainly acted with more regularity than 

 intraspecif ic competition on the nutritional quality of the 

 deer ' s diet . 



General Patterns of Habitat Use 



Earlier studies (Mackie 1970, Knowles 1975, Komberec 

 1976) defined general patterns of habitat use by mule deer. 

 Therefore, data obtained during 1976-1984 were primarily used 

 to evaluate and refine previous conclusions about habitat 

 selection and habitat factors influencing population ecology 

 of deer in riverbreaks habitat. Aerial surveys ensuring 

 complete coverage of the study area during July, September or 

 October, December or January, and March or April provided the 

 most complete data for analysis. Locations from individual 

 surveys, recorded as the mid-point of 3.2 ha cells, were 

 compiled seasonally and interpreted to represent habitat use 

 and selection during summer, autumn, winter, and early spring. 



A "block analysis" (Porter and Church 1987, Wood 1987) 

 was employed. This involved (1) compiling numbers of deer 

 observed during aerial surveys within 28.8-ha blocks to 

 determine relative seasonal density or intensity of use of 

 each of the 953 blocks on the area and (2) relating those data 

 to various habitat attributes measured within each block. The 

 latter included the kind and amount of each vegetation cover 

 type, the number of different cover types, the number of 

 different patches of cover, a topographic relief index, and 

 distance to free water during wet and dry years. 



Habitat parameters were measured from a vegetation cover 

 type map developed from aerial photographs and ground-truthing 



245 



