Forest Fire Mapping using Random Forest based on Sentinel-2 and Sentinel-1 data

Forest fire mapping is a critical tool for environmental monitoring and disaster management, enabling the timely detection and assessment of burned areas. This service is build upon techniques described in the research paper by Zhou, Bao et al., which introduces a machine learning–based approach using Sentinel-2 imagery. Their method combines spectral, topographic, and textural features to improve classification accuracy, particularly emphasising GLCM texture features extracted from Sentinel-2’s short-wave infrared band.Thus, the UDP performs forest fire mapping using a pre-trained Random Forest model in openEO. It combines Sentinel-1 and Sentinel-2 features, applies the model, and outputs the predicted fire mapping results.