1. Ring-necked pheasant Sample Abundance (SA)= 16.49-. 15(tinie)**- .04(CU)+. 21 

 (trees and tall shrubs) (21%) 



2. Horned lark SA= 9.78+.28(CU)+.28(GR)-.88(HJ+BW) (30%) 



3. Magpie SA=-2.49+.02(GR)+.39(trees and tall shrubs)** (39%) 



4. Yellow warbler SA= 2.52-.03(GR)**+.06(trees and tall shrubs)** (35%) 



5. Rufous-sided towhee SA= 1 .25-.01 (GR)+.10(trees and tall shrubs)** (23%) 



5. Vesper sparrow SA=-1 . 55+.01 (CU)-. 16(trees and tall shrubs)*+.06 (big 

 sagebrush badlands)** (20%) 



7. Brewer's sparrow SA- 0.93-.01 (CU)-.04(trees and tall shrubs)*+.04 (big 

 sagebrush badlands)** (25%) 



Note the similarity of the equations for the three species preferring trees 

 and/or tall shrubs (equations 3 through 5) and for the two species characteristic 

 of big sagebrush badlands (equations 5 and 7). Sample abundances of the species 

 within these two groups showed highly significant (p<.001) positive correlations. 



Multiple regression analysis also revealed habitat effects on the number of 

 species encountered at each stop during May-July runs (NS), as shown in the fol- 

 lowing equation: 



NS= 20.9-.12(time)**+.42(trees and tall shrub)** (36%) 



Thus, time in relation to sunrise and the presence of higher foliage layers 

 (trees and shrubs) contribute substantially to the number of species encountered. 



Community Patterns . Correlation coefficients among sample abundances of 

 individual species revealed a number of highly significant correlations (a sum- 

 mary of these correlations is on file with DNRC). While these data are somewhat 

 difficult to interpret, it is evident that at least three distinct groups of 

 species exist (tree- and shrub-loving species, species of open grassland, and 

 species of big sagebrush badlands); sample abundances of species within each 

 group are positively correlated, and species abundances of species of different 

 groups are negatively correlated. Principal component analysis was used to create 

 an ordination by means of which these community patterns may be more easily visual- 

 ized (figure 50). In this figure, the horizontal axis represents the first prin- 

 cipal component and the vertical axis the second principal component; individual 

 species are plotted with respect to their relative positions on these two axes. 

 Only 20.2 percent of variation among variables is explained by the first two 

 principal components; nine principal components are necessary to account for 

 50% of the variation, and 54 are required to explain 100%. 



The first principal component appears to reflect primarily differences in 

 foliage height diversity, progressing from an open grassland on the extreme left 

 through low shrubland, taller shrubs, and finally cottonwood forest with a tall 

 shrub understory on the extreme right. The second principal component is not 

 as clearly defined, but appears to progress from steep, heavily dissected top- 

 ography with little grass cover at the bottom of this graph to flat or rolling 

 areas with high grass cover near the top. 



171 



