Spatiotemporal Autocorrelation of Acoustic Indices: Implications for Biodiversity Monitoring

Read full text Summarise with VIRAT
Authors: Xinyi Jiang, Yixuan Zhang, Zufei Shu, Daiping Wang
Cited by: 0
References: 54

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Abstract Passive acoustic monitoring (PAM) is increasingly applied in biodiversity research, yet the spatiotemporal dependence inherent in acoustic indices remains poorly understood, limiting their standardized application and interpretation. We conducted a study using a complete kilometer-grid monitoring system covering an entire protected area (100 grids of 1 × 1 km) in a subtropical region of southern China. We used acoustic recordings from 58 valid monitoring sites to characterize the spatial and temporal autocorrelation of six commonly used acoustic indices. We further examined associations between acoustic indices and environmental characteristics. Spatially, autocorrelation ranges of some acoustic indices extended to approximately 4 km (i.e., the Bioacoustic Index (BIO) and Normalized Difference Soundscape Index (NDSI)). Temporally, all indices showed significant autocorrelation over 2-5 days. Environmental responses were also index-specific, with elevation, distance to roads, and vegetation abundance affecting different indices in contrasting ways. These findings demonstrate that acoustic indices capture spatially and temporally distinct dimensions of soundscape variation rather than a uniform representation of biodiversity. Explicitly accounting for index-specific autocorrelation and environmental responses can improve sampling design and statistical inference in long-term PAM studies.

Fields of Study & Subjects

Wildlife Ecology and Conservation Animal Vocal Communication and Behavior Noise Effects and Management

Publication details

Journal / source
bioRxiv (Cold Spring Harbor Laboratory)
Publication date
20 March 2026
Language
EN
Document type
Preprint