Exploring the potential of occupancy modelling using passive acoustics in Coua gigas and Coua coquereli

In highly threatened habitats such as the dry deciduous forests of western Madagascar, it is essential to develop new approaches to detect population changes and evaluate conservation measures. Passive acoustic monitoring (PAM) is such a promising approach. This method has many advantages over conventional methods, such as time efficiency, money savings, and reduced wildlife disturbance. It is especially suitable for studying occupancy and activity patterns of vocalizing species such as birds. Our study analyzed data recorded with autonomous sound recorders in 2018 in Kirindy Forest for the territorial calls of Coua gigas and Coua coquereli. We modeled occupancy and detection probability for both species in the study area. We also examined activity patterns and found that the peak of vocal activity for Coua coquereli is at 700h and for Coua gigas at 1100h. To also test the value of PAM in relation to ecological factors we modeled occupancy and included logging status as a site covariate. We detected a positive influence of logging in occupancy of Coua gigas. Our study provides guidelines for future occupancy studies using PAM in the two coua species. We conclude that PAM will improve the ecological monitoring of soniferous animals in Madagascar.