This work was carried out within the framework of contract no. O3.02.03-24-2282 in partnership with CSDIngénieurs and focuses on the development of a bat activity reference framework at the scale of Wallonia, by species, particularly those sensitive to wind energy, based on existing acoustic data.

The objective of this study is to clearly identify, using a robust methodological approach, classes of bat activity in order to facilitate the assessment of potential relevance for the following species: Greater mouse-eared bat, Daubenton’s bat, Natterer’s bat, Common noctule, Leisler’s noctule, Common pipistrelle, Nathusius’ pipistrelle, and Common serotine. In addition, we have developed a reference framework for each of the following six acoustically similar species groups: pipistrelles, serotine–noctule bats, Myotis bats, long-eared bats, horseshoe bats, and a general group including all species.

The first part of the study consists of an international review of scientific literature dealing with bat activity reference frameworks based on acoustic monitoring and the biases encountered. It identifies the key elements considered when developing such a framework and highlights existing differences between countries.

The second part describes the biases related to equipment that must be taken into account. These factors are related to technical configuration (battery, microphone type, deployment setup, SD card), device settings (sampling rate, trigger window, minimum and maximum frequency, etc.), as well as data format and post-processing (deployment period, call duration, classifiers, etc.).

The third part explains the methodology used to establish activity classes: data collection and cleaning, filtering and categorisation using logistic regressions, filtering and temporal segmentation (swarming vs. gestation periods), consideration of equipment-related filters (Wildlife Acoustics range), spatial segmentation by habitat (open, edge, forest), and grouping into species groups—all of which were taken into account in the creation of activity classes.

The fourth part presents the results and compares the resulting framework with existing studies. In addition, an application was developed to facilitate easier visualization of the results.