Incomplete species lists produced by pitfall trapping: How many carabid species and which functional traits are missing?

Pitfall trapping is a widely used sampling technique that is standardly employed to collect various data on ground-dwelling arthropods. However, the validity of this method as a quantitative tool has been repeatedly questioned as the representation of species is influenced by their relative activity and some species are only rarely collected using pitfall traps. In this study, we compared species lists of carabid beetles recorded by pitfall trapping and hand searching techniques at 41 sites covering a range of typical habitats present in the Czech Republic. The number of species and their conservation values, the mean species trait values and phylogenetic composition of assemblages were compared between the methods. In general, hand searching performed by an experienced entomologist can record almost twice as many carabid species compared to pitfall trapping using an intermediate sampling effort (i.e., 5 to 12 pitfall traps per site exposed for an entire vegetation season and 50 to 120 min of hand searching per month). In particular, red-listed species were only rarely recorded in pitfall trap samples. On the other hand, the efficiency of the hand searching technique was significantly affected by the experience of the collector and novices were less efficient in recording carabid species than pitfall trapping. Sampling method also significantly affected the species traits represented within samples. On average, species recorded by pitfall traps were larger, more night-active, more carnivorous, and less colourful and had lower flight abilities. Strong method-specific biases in the representation of species traits can pose a serious problem for studies focused on functional traits. Applied alone, the pitfall trapping technique seems to be not optimal for biological surveys that aim to produce comprehensive species lists, although it is currently widely adopted. To collect more species and cover a more complete spectrum of species traits, we recommend to combine complementary sampling techniques, e.g., pitfall trapping and hand searching, in a single study.