The bias of bat netting

Bats (Chiroptera) are the second largest and by far most diverse order of mammals. Their wide variety of ecological niches is well reflected in their different food habits. In rain forests, bats are of critical importance for pollination and seed dispersal, as well as for predation on insects and small mammals. The Neotropical bat fauna is especially diverse and comprises a particularly large proportion of the mammalian fauna, thus knowledge of their ecology is essential for research on and conservation of tropical forest communities. Most bat studies, whether on ecology, behaviour, natural history, community structure or sensory ecology, entail the capture of the animals. For the past decades mist netting has been the most commonly used and most effective technique to capture bats. Though mist netting is a common method and the presence of a `capture-bias` is known through many anecdotal experiences and often discussed, the biases of mist netting are not well understood. Supposedly there are disparate netting biases for different bat species due to the behaviour, foraging strategy, echolocation skills and spatial memory of the bats as well as environmental factors such as location of the net, light conditions, precipitation and wind. While some species appeare to be fairly well sampled by mist netting, others seem to be underrepresented. To date there are no studies addressing the problem of netting biases directly, and for the analysis of mist netting data, assumptions are often based on anecdotal information. Last field season we monitored a mist net set in the tropical forest of Barro Colorado Island (Panama) for one hour with a thermo camera. We observed 21 bats approaching the net within the first hour after sunset. Only four of these bats were permanently caught in the net. These preliminary results demonstrate the necessity for a study on mist netting biases, since nearly every second bat seemed to detect and avoid the net, and most of bats that did fly into the net were able to free themselves within seconds or minutes. The objective of this project is to study mist netting biases by combining new methods such as filming with thermo- and infrared-cameras, and analyses of acoustic data of bat detectors simultaneously with standardized mist netting. We will study the probabilities of different bat species being captured in mist nets as well as the influence of light conditions, netting site and weather on netting success. The results of this study will provide adequate data on mist net biases, which will make it possible to assess the abundance of bats based upon mist netting data much more accurately. With results of our study taken into account the assessments of relative abundance of bat species within communities may shift considerably in the future. Thus this project will have a significant influence on community and ecology studies of bats.