Optimising Survey Strategies for the Endangered Romer's Tree Frog (Liuixalus romeri): Effects of Diel, Seasonal and Weather Variation on Acoustic Detection.

Accurate estimation of species distribution and abundance depends on detection probability, which can vary with environmental conditions. For amphibians, calling activity and detectability are strongly influenced by diel, seasonal and weather conditions. However, these relationships remain poorly quantified for many species, limiting the optimisation of survey planning. We used passive acoustic monitoring (PAM) to investigate environmental factors of call detection in the endangered Romer's tree frog (Liuixalus romeri) in Hong Kong. Acoustic data were collected continuously across multiple sites to quantify the effects of time of day, season and weather variables on detection probability. We further compared PAM detections with active call surveys to assess concordance between methods. Detection probability exhibited strong diel and seasonal patterns, peaking during the wet season (June-August) and at night, with highest detection probability during early evening (20:00-22:00) and a secondary peak before dawn (03:00-04:00). Rainfall effects varied across the diel cycle, with positive effects during peak calling periods but negative or neutral effects at other times. Detection probability showed a non-linear relationship with temperature, peaking at approximately 27.5°C, and declined with increasing wind speed. Agreement between PAM and active call surveys was moderate (Cohen's Kappa = 0.47), with concordant call agreement for 73% of survey nights. This indicates that the two methods are complementary rather than interchangeable. Based on these findings, we recommend conducting manual surveys during peak activity periods (June-August, 20:00-22:00), under conditions of low wind and following recent rainfall. PAM provides a valuable tool for continuous monitoring and detecting temporal changes in call detection, while manual surveys remain important for estimating abundance. Integrating both approaches can improve monitoring protocols for L. romeri and other amphibians.