A comparison of two survey methods for assessing bird species richness and abundance in tropical farmlands

Capsule Timed species counts could provide a more valuable bird survey tool in the humid tropics than previously suggested. Aims To compare the relative performance of two bird survey methods that differ in complexity - point counts with distance sampling; and timed species counts - for estimating the species richness and abundance of bird species in tropical farmlands. Methods Twenty-six sites were surveyed in banana - coffee farmland habitats north of Lake Victoria, central Uganda. Point counts and timed species counts, each lasting ten minutes, were conducted at ten stations in each site. Results The two methods provided similar estimates of species richness in farmland and a significant positive correlation between measures of frequency of occurrence of bird species from the two methods was found. There was also a significant positive correlation between density from point counts with distance sampling and frequency of occurrence of bird species from the ten-minute counts. Ten-minute counts tended to accumulate bird species faster than point counts but this difference was not significant. Conclusions In these relatively open farmland habitats species richness estimates from both methods were not significantly different. Although density can only be derived from point counts these densities were highly correlated with frequencies from the ten-minute counts. The frequency of occurrence method is much quicker and simpler, in terms of both field methods and analysis, because it does not involve distance estimation. The results presented here suggest that, at least in relatively open habitats, frequency of occurrence provides a valuable index of density of birds.