Preliminary evaluation of sampling strategies to estimate the species richness of diurnal, terrestrial birds using Monte Carlo simulation

Various environmental stratifications for estimating the total bird species richness across the south east region of Australia were evaluated using Monte Carlo simulation techniques. This was possible because of the availability of gee-referenced point data for the birds observed in the region that could be generally associated with a set of primary environmental attribute data. The bird data set consisted of a total of 173 species recorded from 1,075 sites. Several stratification options were assessed including: random sampling; sampling within gradsects positioned to span some of the region's major environmental gradients; sampling within stratifications based on climate attributes; and sampling within a stratification based on climate and substrate variables. Variation in Sample size was the most important factor affecting estimates of bird species richness. Several limitations associated with the origins of the bird data set and the distribution of the bird sites across the region dictated what could be achieved by the simulation study. We discuss some of the problems and limitations associated with the use of existing data sets to investigate biological issues at a regional scale. (C) 1997 Elsevier Science B.V.