OBAS

. The S2 data consists of 13 multispectral bands with spatial resolutions of 10 m, 20 m, and 60 m with two spacecrafts (Sentinel-2A and 2B) flying in tandem. Water was masked out from the S2 surface reflectance (SR) data (Chastain et al., 2019) using a global surface water mask (Pekel et al., 2016). Furthermore, clouds and shadows were masked out using the S2 Scene Classification Layer (SCL) shadow (SCL class3) and cloud (SCL class 8-9) masks (ESA, 2025). All surface reflectance bands were processed in Google Earth Engine (Gorelick et al., 2017) with 16 bits unsigned. A raster data cube with S2 multispectral data was used to feed the CNN model and predict the burned area.