Rapid mapping of landslides in the Western Ghats (India) triggered by 2018 extreme monsoon rainfall using a deep learning approach

A deep learning approach significantly improves the accuracy of landslide mapping in the Western Ghats, India. By integrating PlanetScope imagery with convolutional neural networks (CNNs) and topographic slope data, researchers achieved a 78% mean accuracy in identifying rainfall-induced landslides triggered by the 2018 extreme monsoon season—an increase of 12.5% compared to spectral-only methods. This methodology effectively reduces false positives, distinguishing landslide areas from barren lands and riverbeds without compromising true positive detection. The study demonstrates a scalable solution for rapid hazard assessment in landslide-prone regions, offering potential applications for disaster mitigation strategies worldwide.

Meena, S. R., Ghorbanzadeh, O., van Westen, C. J., Nachappa, T. G., Blaschke, T., Singh, R. P., & Sarkar, R. (2021). Rapid mapping of landslides in the Western Ghats (India) triggered by 2018 extreme monsoon rainfall using a deep learning approach. Landslides, 18(5), 1937–1950. https://doi.org/10.1007/s10346-020-01602-4