The pitfalls of pitfalls: a comparison of pitfall trap catches and absolute density estimates of epigeal invertebrate predators in arable land

In an agroecosystem, catches of epigeal invertebrate predators obtained by pitfall traps were compared to absolute population densities estimated by g round photoeclectors in two different habitats, a field and an adjacent set -aside land. In general, abundance of Carabidae and Lycosidae were overesti mated by pitfalls, while Staphylinidae and Linyphiidae were underestimated, and beetle larvae showed no obvious trend. The overestimation of Carabidae and Lycosidae by pitfalls had its maximum in spring and summer. Numbers co llected by pitfalls did represent actual population densities better in the field than in the set-aside. This is probably caused by the higher vegetat ion diversity and complexity of the set-aside land which additionally influ enced the catches. Pitfalls recovered higher numbers of animals and species , and species composition differed significantly between pitfalls and photo eclectors, in the field as well as in the set-aside. Body size seemed tb be the main factor in determining the catch, the relatively larger species be ing more frequently caught by pitfall traps. In sum, the relationship betwe en pitfall tray catches and actual population densities appeared to be eith er absent, weak or highly variable among taxa, habitat and time of the seas on. Thus numbers caught in pitfall traps cannot be considered as reliable i ndices of "real" abundance nor do they reflect the relative abundance withi n a given predator community correctly. Therefore, in most cases absolute d ensity estimates should be preferred for the study of epigeal invertebrates in arable land. However, I suggest a combination of both pitfall traps and standardised area samples, because the two approaches possibly provide inf ormation of different qualities equally important for the evaluation of epi geal predator species, namely searching activity and population density.