High-Resolution Bamboo Distribution Map for Mizoram, Northeast India Derived from Integrated LISS-IV and Sentinel-2 Data Using Deep Learning

This dataset provides a high-resolution spatial map of bamboo land cover across Mizoram, Northeast India, generated using a deep learning-based semantic segmentation framework applied to multi-source satellite imagery.The dataset was produced by integrating high-resolution LISS-IV surface reflectance data with Sentinel-2 Level-2A red-edge and narrow near-infrared spectral bands, along with vegetation indices including NDVI and EVI2. These inputs were combined into a nine-channel composite and used to train a U-Net model with a ResNet-34 encoder for binary classification of bamboo and non-bamboo land cover.The resulting bamboo distribution map represents a multi-year composite (2021–2023) derived from post-monsoon imagery to ensure phenological consistency and reduced cloud contamination. The dataset covers the full geographical extent of Mizoram (~21,081 km²), with an estimated bamboo area of approximately 6,262 km², corresponding to about 29.7% of the state’s total area.The map captures both continuous and fragmented bamboo patches across complex mountainous terrain and is suitable for applications in ecological assessment, forest resource management, biodiversity studies, and hazard monitoring related to bamboo flowering events.