Passive acoustic monitoring reveals seasonal patterns in European green toad calling activity but fails to accurately reflect population abundance

Amphibians are among the most endangered groups of vertebrates worldwide; therefore, effective population monitoring is crucial for understanding the factors driving these unfavourable trends. In this study, we evaluated the effectiveness of fully automated passive acoustic monitoring in estimating the distribution and population size of the European green toad (Bufotes viridis)-an anuran species in which males vocalize intensively during the breeding season. We conducted visual counts with human observers, recorded the soundscape using autonomous sound recorders, applied automatic call detection, and calculated soundscape complexity using acoustic indices adjusted to the frequency range of European green toad vocalizations. Our algorithm generated a calling intensity index, which detected 36% of calls with 96% precision. The calling intensity index showed a significant correlation with the number of calls detected manually (r = 0.727). Additionally, we found moderate but significant correlations between acoustic indices (Acoustic Complexity Index and Bioacoustic Index) and both the manually and automatically detected call counts. The calling intensity index and acoustic indices did not significantly correlate with the number of males counted by human observer, but they reflected the phenological changes in species' reproductive activity. Our study demonstrates that passive acoustic monitoring enables the assessment of presence and distribution of the European green toad, as well as daily and seasonal patterns in calling activity. However, due to the specific biology and ecology of the species, calling intensity alone is insufficient for accurately estimating male abundance.