Assessing ant assemblages: pitfall trapping versus nest counting (Hymenoptera, Formicidae)

Pitfall trapping and nest counting are the most common census methods for ant assemblages. We examined the concordance between pitfall catches and nest counts on dry grassland. Spearman rank correlations and non-metric multidimensional scaling of the Bray Curtis similarity index revealed moderate concordance between the data collated by the two methods, but overall method-related differences were considerable. The dissimilarity was influenced by the type of land management, but not by trapping period or plot shape. Trapping success depended on nest density, ground vegetation cover and species-specific traits (inhabited stratum, colony size, foraging distance). Even when these factors were taken into account, the convertibility of pitfall trap and nest density values was unsatisfactory: the census method proved to be crucial in designing ant-ecological studies and interpreting literature data. Received 11 November 2005; revised 9 February 2006; accepted 1 March 2006.