Innovative Application of Radar Camera Technology in Monitoring Amphibian and Reptile Diversity: A Case Study of Lishui City

Habitat fragmentation, anthropogenic disturbance, and climate warming continue to drive population declines in amphibians and reptiles. Traditional line-transect surveys and trapping methods are labor-intensive and difficult to sustain for long-term, continuous monitoring. Infrared cameras, which rely on temperature differences to trigger detection, are particularly prone to missing ectothermic animals such as frogs and snakes, creating a major technical bottleneck for automated monitoring. In this study, six representative monitoring sites in Lishui, Zhejiang Province, China, were selected as the study area. A total of 28 self-developed 905-nm LiDAR cameras were deployed in a grid-based configuration, and field monitoring was conducted from June to September 2023. An amphibian and reptile index (ARI) was developed, and radar-based monitoring results were compared with historical baseline data on regional species occurrences to systematically evaluate the monitoring performance of the technology. A total of 2 099 valid images were obtained from the study area. Anura was overwhelmingly dominant among the recorded taxa, while Lacertilia and Serpentes were detected only at Fengyang Mountain. The ARI was significantly higher in native forest habitats in hilly areas than in urban parks, paddy fields, and high-altitude wetlands, indicating significant differences in biodiversity among habitat types. Compared with infrared cameras and traditional line-transect surveys, LiDAR cameras offer advantages including laser-echo-based triggering, independence from body temperature, non-invasive operation, all-weather monitoring, adaptability to complex mountainous habitats, and low-cost long-term deployment. However, several limitations remain, including false triggers caused by vegetation and flowing water, insufficient detection sensitivity for Caudata and Testudines, and the current limitation of AI-based recognition to the taxonomic order level. The results demonstrate that LiDAR cameras can objectively characterize the spatial distribution and relative abundance of amphibians and reptiles. Future improvements should focus on species-level AI recognition, multisensor fusion, and the establishment of standardized field monitoring protocols to facilitate large-scale and operational applications. This study provides a technological framework for intelligent, long-term monitoring of amphibians and reptiles in China.