National-scale mapping of staple crops in Nepal with multi-year Sentinel-2 L2A imagery and binary transformer models

Accurate mapping of staple crops using satellite imagery is essential for sustainable agricultural management,particularly in data-limited regions. Despite recent advances in Earth observations and deep learning, usingmulti-year time series satellite data and binary deep learning models to produce national-scale multi-class croptype map remains underexplored. In this study, we attempt to map staple crops, other crops, and non-crop at 10m resolution that leverages multi-year, unsmoothed Sentinel-2A/B MSI Level-2A time series imagery, binarytransformer models, and a probability-based thresholding method. Nepal was selected as the case study due to itscomplex topography and heterogeneous landscapes. We developed six experimental setups, and models achievedan average overall accuracy of 91.39%, with class-wise F1-scores ranging from 0.81 to 0.96. The best-performingmodel was then applied to 28 tiles covering Nepal, and a probability-based thresholding approach was used toassemble binary results into a multi-class map. The map overall accuracy was higher than 94%, when validatedwith field data. Sensitivity analyses further confirmed the robustness and spatial consistency of the proposedframework across diverse land-cover and agroecological zones. Finally, the produced national-scale maps indicated 3.91% increase in staple crops areas from 2021 to 2023 in Nepal.