Species characteristics predict the effectiveness of avian survey methods in an Afromontane sky-island

Biodiversity conservation requires effective monitoring of ecological communities in remote locations, where limited accessibility often restricts survey efforts. Passive acoustic monitoring (PAM) is becoming an established method for measuring biodiversity, facilitated by the increased accessibility of autonomous recording units. Comparing the performance of PAM and traditional survey methods can provide insights into how species characteristics such as foraging strata, vocal behaviour and taxonomy influence the detection ability of each method. Here, avian species occurrences were collected using PAM and point count surveys (PCS) across three forest fragments in the Taita Hills, an Afromontane sky-island in southeastern Kenya. These montane forests contain high species-richness and endemism, but they have been severely fragmented and degraded as a result of deforestation, making comprehensive monitoring critical for conservation. We grouped detected species into clusters based on their foraging strata, using partitioning around medoid cluster analysis. From 25 survey sites, PAM and PCS detected 60 and 57 species, respectively, indicating that PAM is as effective as PCS for surveying montane tropical birds. However, species that primarily foraged on the ground and secondarily used the understorey, and species that primarily used the understorey and secondarily used the canopy, were more likely to be recorded by PAM than by PCS. Species that only used the understorey were more likely to be recorded by PCS. Investigation of broad taxonomic groupings showed that PAM was 20% more likely to record non-passerines than PCS, while passerines were equally likely to be recorded by either method. This study highlights how species characteristics, such as foraging strata and taxonomic group, influence the performance of surveying methods. By identifying which species are best monitored by each method, this study can inform more targeted monitoring strategies, improving accuracy and supporting biodiversity management efforts to mitigate ongoing species loss.