Abstract: Global biodiversity is undergoing rapid change and sustained loss, creating an urgent need for ecological monitoring technologies capable of continuous operation over large spatial scales and long time series. Passive acoustic monitoring (PAM) records environmental sounds in an unattended manner, enabling the simultaneous capture of biotic, geophysical, and anthropogenic sounds. It thus provides a vital tool for identifying changes in biological community activity, soundscape structure, and ecosystem dynamics under anthropogenic disturbance. In recent years, PAM has been widely applied to biodiversity monitoring in the context of disturbances such as traffic, construction, and recreational noise, land-use change, and habitat fragmentation. However, changes in the soundscape cannot be directly equated with changes in biodiversity. Human-generated noise may affect recording results through acoustic masking and reduced detectability, while animals may compensate behaviorally by adjusting call frequency, singing duration, or activity rhythms; habitat disturbance, in turn, may trigger community restructuring and changes in the spatiotemporal patterns of soundscapes by altering vegetation structure, resource availability, landscape connectivity, and sound propagation conditions. Based on this, this paper proposes an analytical framework of disturbance type—mechanism of action—acoustic response—ecological implications—validation pathway. Focusing on four pathways—direct noise disturbance, changes in land use and management, alterations in landscape pattern, and the superposition of multiple disturbances—it systematizes typical responses of the bioacoustic landscape and their ecological interpretations. The paper further discusses the key challenges facing PAM in the interpretation of acoustic indices, the development of monitoring standards, and its application in ecological early warning systems. It proposes recommendations to enhance the reliability and operability of soundscape monitoring in biodiversity assessment, including strengthening multi-source data validation, refining data collection and analysis standards, establishing long-term baselines, and improving model interpretability.

Key words: passive acoustic monitoring, ecoacoustics, anthropogenic disturbance, biological soundscape, biodiversity evaluation