Winter Bird Census 



Table 14 summarizes the results of the winter bird census for the 

 five bird census plots. Information on use of the plots by mammals is 

 included here as well. A more complete account of the 1980 winter bird 

 census is given in appendix A, and plot descriptions can be found in 

 appendix F of the WBS. 



Roadside Wildlife Survey 



Data obtained during the fifteen 1979 roadside survey runs are presented 

 in appendix B. The cumulative numbers of species for all May-July runs 

 (1977-1979) are presented in table 15. Since the roadside wildlife survey 

 appears to be a particularly appropriate and sensitive monitoring technique, 

 some features of the analysis will be described in detail below. 



Table 16 summarizes 1978-1979 changes in May-July samole abundances 

 of selected species, based, on data obtained during runs of the five road- 

 side survey routes. The eastern kingbird, yellow warbler, common yellow- 

 throat, and lark bunting showed significant decreases, while Sprague's 

 pipit showed a significant increase. Figure 5, which shows year-to-year 

 changes in sample abundances of ring-necked pheasants, is an example of 

 the graphs which were prepared to monitor individual species. 



A computer program was developed which calculates a number of diversity 

 parameters for the community samples obtained by the roadside counts. 

 Factor analysis was used to determine which of these community parameters 

 were most sensitive to ecological differences among the five routes, and 

 which were redundant (that is, which are measuring essentially the same 

 community attribute). The first factor, which accounted for most of the 

 variation among routes and replicates (65%), was closely related to measures 

 which are sensitive to the variety of species in the sample, such as 

 species number (S), species richness (D), and the height of the mode of 

 the lognormal species-abundance curve (Sg). The parameters with the highest 

 loading on this factor were species richness and species number, which may 

 be considered the best measures of species variety for this data set. The 

 second factor accounted for an additional 24% of the variation, and was 

 closely related to measures sensitive to the equitability of distribution 

 of individuals among species. Showing high loadings on this factor were 

 the logarithmic dispersion factor (a), the lognormal standard deviation 

 (sigma), and evenness (e) (see WBS for definitions of parameters). The 

 Shannon-Weaver diversity index H', as well as the theoretical total num- 

 ber of species S^, are "hybrid" parameters with moderately high loadings 

 on both factors. Figures 20 and 21 show year-to-year changes in D and 

 sigma, the two parameters loading most heavily on Factors I and II, 

 respectively. Changes in species number (excluding visitors, migrants, 

 and water birds) are shown in figure 22. Note that the relative positions 

 of the routes remain fairly constant from year to year. This analysis 

 also showed that June data alone provide nearly as much discriminating 

 power among routes as do pooled May-July data or even pooled 1977-1979 

 data. 



57 



