Dataset for: A cloud-based Geo-AI framework for automated high-resolution flood mapping with explainable machine learning

This repository contains the spatial dataset, statistics files, and ML modelling data supporting the manuscript A cloud-based Geo-AI framework for automated high-resolution flood mapping with explainable machine learning. The study presents a hybrid cloud-local geospatial artificial intelligence (Geo-AI) workflow for rapid, high-resolution (10 m) flood mapping. The framework integrates Sentinel-1 synthetic aperture radar (SAR), Sentinel-2 optical imagery, and Copernicus digital elevation model (DEM) topography within Google Earth Engine (GEE). It automatically derives high-confidence training samples using a conservative Z-score threshold and benchmarks ~17 machine-learning algorithms (including LightGBM, Random Forest, Gradient Boosting, and Extra Trees). Model predictions and multi-sensor feature contributions are interpreted using SHapley Additive exPlanations (SHAP).More content to be updated soon in next version!