Complementarity between mist-netting and low-cost acoustic recorders to sample bats in Amazonian rainforests and savannahs

Effective survey methods are paramount to measure changes in species distribution, populations dynamics and to guide conservation. Mist-netting and passive acoustic monitoring are two of the most used techniques to sample bats assemblages. Yet, despite the great potential of low-cost autonomous ultrasound recorders in surveying bat assemblages, we lack thorough assessments of their performance in relation to more established survey methods. Taking advantage of the rich bat fauna of the northeastern Brazilian Amazon, we set out to i) investigate the complementarity of mist-netting and acoustic surveys in sampling bats in forest and savannah habitats in the Savannahs of Amapá and ii) undertake a cost-effectiveness evaluation of using one, two or three recorders per sampling site to simultaneously survey bat assemblages. The two methods show complementary, and overall species diversity recorded with mist nets was higher than with acoustic recorders. However, species diversity was higher with acoustic recorders than with mist nets when considering a reduced (n < 3) number of transects. In addition, we found a gain in species diversity when using more than one acoustic recorder in forest habitats, despite the low cost-effectiveness. However, there were no differences between the diversity using one, two or three acoustic recorders in savannah. Due to possible device malfunction, we recommend the use of at least two acoustic recorders in both habitats to reduce the likelihood of data loss. The use of low-cost bioacoustic recorders in bat surveys can help to address critical knowledge gaps for poorly known aerial-hawking insectivores and support evidence-based conservation strategies.