Beyond the Canopy: Resolving Topographic and Acoustic Complexities with Machine Learning for Karst Avifauna Monitoring
Article Sidebar
Abstract:
Background of study: Tropical karst landscapes harbor exceptional avian biodiversity but pose unique monitoring challenges due to complex topography, cave reverberation, and humidity-driven sound distortion. Conventional ecoacoustic methods fail in these environments, with indices showing weak correlations (r=0.20-0.43) for avian diversity due to insect masking and abiotic interference. Over 83% of karst-endemic birds lack standardized monitoring protocols despite escalating extinction risks.
Aims and scope of paper: This review aims to: (1) quantify limitations of current ecoacoustic methods in karst ecosystems, (2) develop a machine learning-enhanced framework addressing topographic and reverberation effects, and (3) establish conservation-ready protocols for endangered karst avifauna. The study synthesizes evidence from 29 studies across hardware innovation, signal processing, and policy applications.
Methods: We systematically analyzed 29 studies on acoustic monitoring in karst ecosystems, focusing on machine learning innovations, topographic adaptations, and conservation applications.
Result: Topography drives 47% of soundscape variation, surpassing vegetation effects. Machine learning (CNNs/MFCCs) boosts detection accuracy by 22-80% in reverberant caves. Hybrid protocols enable 25-m resolution habitat mapping and precise disturbance monitoring, overcoming tropical "latitude paradox" limitations.
Conclusion: This review establishes the first karst-adapted ecoacoustic framework, integrating machine learning with topographic variables to transform monitoring from biodiversity proxy to precision tool. Critical next steps include developing species-specific call libraries, wind-reverberation filters, and policy integration of acoustic baselines for IUCN assessments. The proposed protocols address urgent conservation needs for Earth's most threatened avian sanctuaries.
Keywords: Conservation acoustics, Ecoacoustic monitoring, Karst avifauna, Machine learning, Topographic calibration
Downloads
Copyright (c) 2025 Anggyta Fitryan, Ahmad Faruq Abdurrahman, Nuryani, Surya Prihanto, Yusril Al Fath, Ayu Aprilia, Junaidi, Arif Surtono
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
References
Allen-Ankins, S., McKnight, D. T., Nordberg, E. J., Hoefer, S., Roe, P., Watson, D. M., McDonald, P. G., Fuller, R. A., & Schwarzkopf, L. (2sni023). Effectiveness of acoustic indices as indicators of vertebrate biodiversity. Ecological Indicators, 147. https://doi.org/10.1016/j.ecolind.2023.109937
Azofeifa-Solano, J. C., Parsons, M. J. G., Brooker, R., McCauley, R., Pygas, D., Feeney, W., Simpson, S., Nedelec, S. L., Croxford, E. M., Meekan, M., & Erbe, C. (2025). Soundscape analysis reveals fine ecological differences among coral reef habitats. Ecological Indicators, 171. https://doi.org/10.1016/j.ecolind.2025.113120
Campos, I. B., Fewster, R., Truskinger, A., Towsey, M., Roe, P., Vasques Filho, D., Lee, W., & Gaskett, A. (2021). Assessing the potential of acoustic indices for protected area monitoring in the Serra do Cipó National Park, Brazil. Ecological Indicators, 120. https://doi.org/10.1016/j.ecolind.2020.106953
Chen, Y. F., Luo, Y., Mammides, C., Cao, K. F., Zhu, S., & Goodale, E. (2021). The relationship between acoustic indices, elevation, and vegetation, in a forest plot network of southern China. Ecological Indicators, 129. https://doi.org/10.1016/j.ecolind.2021.107942
Dröge, S., Fulgence, T. R., Osen, K., Rakotomalala, A. A. N. A., Raveloaritiana, E., Schwab, D., Soazafy, M. R., Wurz, A., Kreft, H., & Martin, D. A. (2024). Understanding acoustic indices as multi-taxa biodiversity and habitat quality indicators. Ecological Indicators, 169. https://doi.org/10.1016/j.ecolind.2024.112909
Francomano, D., Gottesman, B. L., & Pijanowski, B. C. (2020). Biogeographical and analytical implications of temporal variability in geographically diverse soundscapes. Ecological Indicators, 112. https://doi.org/10.1016/j.ecolind.2019.105845
Galappaththi, S., Goodale, E., Sun, J., Jiang, A., & Mammides, C. (2024). The incidence of bird sounds, and other categories of non-focal sounds, confound the relationships between acoustic indices and bird species richness in southern China. Global Ecology and Conservation, 51. https://doi.org/10.1016/j.gecco.2024.e02922
Gibb, K. A., Eldridge, A., Sandom, C. J., & Simpson, I. J. A. (2024). Towards interpretable learned representations for ecoacoustics using variational auto-encoding. Ecological Informatics, 80. https://doi.org/10.1016/j.ecoinf.2023.102449
Gottesman, B. L., Olson, J. C., Yang, S., Acevedo-Charry, O., Francomano, D., Martinez, F. A., Appeldoorn, R. S., Mason, D. M., Weil, E., & Pijanowski, B. C. (2021). What does resilience sound like? Coral reef and dry forest acoustic communities respond differently to Hurricane Maria. Ecological Indicators, 126. https://doi.org/10.1016/j.ecolind.2021.107635
Grinfeder, E., Lorenzi, C., Teytaut, Y., Haupert, S., & Sueur, J. (2025). Introducing evascape, a model-based soundscape assembler: impact of background sounds on biodiversity monitoring with ecoacoustic indices. Ecological Indicators, 178. https://doi.org/10.1016/j.ecolind.2025.113882
He, X., Deng, Y., Dong, A., & Lin, L. (2022). The relationship between acoustic indices, vegetation, and topographic characteristics is spatially dependent in a tropical forest in southwestern China. Ecological Indicators, 142. https://doi.org/10.1016/j.ecolind.2022.109229
Holgate, B., Maggini, R., & Fuller, S. (2021). Mapping ecoacoustic hot spots and moments of biodiversity to inform conservation and urban planning. Ecological Indicators, 126. https://doi.org/10.1016/j.ecolind.2021.107627
Huancapaza Hilasaca, L. M., Gaspar, L. P., Ribeiro, M. C., & Minghim, R. (2021). Visualization and categorization of ecological acoustic events based on discriminant features. Ecological Indicators, 126. https://doi.org/10.1016/j.ecolind.2020.107316
Lakdari, M. W., Ahmad, A. H., Sethi, S., Bohn, G. A., & Clink, D. J. (2024). Mel-frequency cepstral coefficients outperform embeddings from pre-trained convolutional neural networks under noisy conditions for discrimination tasks of individual gibbons. Ecological Informatics, 80. https://doi.org/10.1016/j.ecoinf.2023.102457
Metcalf, O. C., Baccaro, F., Barlow, J., Berenguer, E., Bradfer-Lawrence, T., Chesini Rossi, L., do Vale, É. M., & Lees, A. C. (2024). Listening to tropical forest soils. Ecological Indicators, 158. https://doi.org/10.1016/j.ecolind.2024.111566
Okamoto, R., & Oguma, H. (2025). A simulation based approach for enhancing automatic detection of cicada songs in challenging chorus conditions. Ecological Informatics, 90. https://doi.org/10.1016/j.ecoinf.2025.103299
Oliveira, E. G., Ribeiro, M. C., Roe, P., & Sousa-Lima, R. S. (2021). The Caatinga Orchestra: Acoustic indices track temporal changes in a seasonally dry tropical forest. Ecological Indicators, 129. https://doi.org/10.1016/j.ecolind.2021.107897
Pan, W., Goodale, E., Jiang, A., & Mammides, C. (2024). The effect of latitude on the efficacy of acoustic indices to predict biodiversity: A meta-analysis. Ecological Indicators, 159. https://doi.org/10.1016/j.ecolind.2024.111747
Quinn, C. A., Burns, P., Gill, G., Baligar, S., Snyder, R. L., Salas, L., Goetz, S. J., & Clark, M. L. (2022). Soundscape classification with convolutional neural networks reveals temporal and geographic patterns in ecoacoustic data. Ecological Indicators, 138. https://doi.org/10.1016/j.ecolind.2022.108831
Sánchez-Giraldo, C., Correa Ayram, C., & Daza, J. M. (2021). Environmental sound as a mirror of landscape ecological integrity in monitoring programs. Perspectives in Ecology and Conservation, 19(3), 319–328. https://doi.org/10.1016/j.pecon.2021.04.003
Sarker, M. A. R., McKnight, D. T., Ryder, D., Walcott, A., Ocock, J. F., Spencer, J. A., Preston, D., Brodie, S., & Bower, D. S. (2022). The effect of inundation on frog communities and chorusing behaviour. Ecological Indicators, 145. https://doi.org/10.1016/j.ecolind.2022.109640
Scarpelli, M. D. A., Ribeiro, M. C., & Teixeira, C. P. (2021). What does Atlantic Forest soundscapes can tell us about landscape? Ecological Indicators, 121. https://doi.org/10.1016/j.ecolind.2020.107050
Scarpelli, M. D. A., Roe, P., Tucker, D., & Fuller, S. (2023). Soundscape phenology: The effect of environmental and climatic factors on birds and insects in a subtropical woodland. Science of the Total Environment, 878. https://doi.org/10.1016/j.scitotenv.2023.163080
Siddagangaiah, S., Chen, C. F., Hu, W. C., Danovaro, R., & Pieretti, N. (2021). Silent winters and rock-and-roll summers: The long-term effects of changing oceans on marine fish vocalization. Ecological Indicators, 125. https://doi.org/10.1016/j.ecolind.2021.107456
Vega-Hidalgo, Á., Flatt, E., Whitworth, A., & Symes, L. (2021). Acoustic assessment of experimental reforestation in a Costa Rican rainforest. Ecological Indicators, 133. https://doi.org/10.1016/j.ecolind.2021.108413
Wang, J., Li, C., Lin, Y., Weng, C., & Jiao, Y. (2023). Smart soundscape sensing: A low-cost and integrated sensing system for urban soundscape ecology research. Environmental Technology and Innovation, 29. https://doi.org/10.1016/j.eti.2022.102965
Wang, J., Li, C., Yao, Z., & Cui, S. (2024). Soundscape for urban ecological security evaluation. Basic and Applied Ecology, 76, 50–57. https://doi.org/10.1016/j.baae.2024.02.007
Yoh, N., Mbamy, W., Gottesman, B. L., Froese, G. Z. L., Satchivi, T., Obiang Ebanega, M., Carlson, L., Koto, S. E., Özdoğan, M., Seaman, D. J. I., Maicher, V., Malinowski, H., Poulsen, J., Ebang Mbélé, A., & Buřivalová, Z. (2024). Impacts of logging, hunting, and conservation on vocalizing biodiversity in Gabon. Biological Conservation, 296. https://doi.org/10.1016/j.biocon.2024.110726
Zhang, C., Zhang, Y., Zheng, X., Gao, X., & Hao, Z. (2024). Influence of recording devices and environmental noise on acoustic index scores: Implications for bird sound-based assessments. Ecological Indicators, 159. https://doi.org/10.1016/j.ecolind.2024.111759
