Comparison of yellow pan, Malaise, and ethanolic traps for sampling parasitoid Vespoidea and other Hymenoptera in a semideciduous forest fragment of the Brazilian Atlantic Rainforest

Abstract

Parasitoid wasps within the Vespoidea superfamily have been chronically under surveyed, thanks to the bigger attention given to the social species within that group or to the other parasitoids within the megadiverse Parasitica infraorder. To address that, we test a new sampling technique for the capture of parasitoid Vespoidea and other Hymenoptera in comparison to other two well stablished designs. Between the consecutive dry and wet seasons of 2014-2015 we placed sets of Malaise (MT), yellow pan (YPT) and ethanolic (ET) traps in a fragment of semideciduous Brazilian Atlantic Rainforest. We hypothesised that MT is the most efficient method for sampling Hymenoptera families and lower taxa of Vespoidea, while YPT and ET would be more efficient in attracting specific taxa, given their characteristic luring mechanisms. We calculated taxa accumulation curves to evaluate expected richness. Average Taxonomic Distinctiveness (Delta(+)) was used as the continuous dependent variable in two-way ANOVAs. Faunal similarity was inspected through nonmetric multidimensional scaling (NMDS). All tests were performed taking season and trap design as explanatory variables. MT was the most efficient in sampling total richness and Delta(+) of Hymenoptera and Vespoidea lower taxa. It captured all 39 families recorded in the present study and 60 out of the 73 Vespoidea lower taxa. YPT tended to lure Diapriidae, Dryinidae, Ponerinae and Myrmicinae ants and a few taxa of Pompilidae and Mutillidae, proving its efficiency in sampling wingless and short-flying hymenopterans as well as the predators and parasites of other non-grass feeding insects. ET, although successful in estimating Hymenoptera Delta(+), did not have a sufficient capture rate to give a precise estimate of total richness. It attracted, however, the social wasps Polybia jurinei Saussure and Synoeca surinama (L.), species traditionally captured by active search. Synthesis and Applications: MT is confirmed as the most efficient capture method for Hymenoptera sampling and is recommended when a survey of the whole community is intended. ET was not successful in capturing any parasitoid Vespoidea as hypothesised but, as well as the YPT, it could be used as an alternative to active searches when the taxa listed above are the focus of the survey. These results provide a better understanding of passive capture methods for Hymenoptera sampling and can be considered in future surveys aiming to investigate their diversity, distribution and improve their conservation.