Marviii1/FloodSense-Lokoja-A-Python-Based-Flood-Risk-Intelligence-and-Decision-Support

Development of a Python-Based Flood Risk Intelligence and Decision Support System Using Hydrological GIS, Sentinel-1 SAR Validation, Exposure Modelling and WebGIS # FloodSense Lokoja **A Python-based flood risk intelligence and decision support system for Lokoja, Kogi State, Nigeria.** FloodSense is not just a flood map. It is a reproducible workflow that identifies flood-prone areas, validates modelled flood zones with Sentinel-1 SAR evidence, estimates exposed people and infrastructure, ranks intervention priorities, and presents results in an interactive WebGIS dashboard. --- ## Study area - City: Lokoja, Kogi State, Nigeria - Validation event: October 2022 Niger–Benue flood - Projected CRS: EPSG:32632 (UTM Zone 32N) --- ## Project concept | Stage | What it does | |---|---| | DEM analysis | Studies the shape of the land | | Rainfall analysis | Estimates how much water enters the system | | Flow accumulation | Identifies where water gathers | | Flood susceptibility | Predicts likely flood-prone areas | | Sentinel-1 validation | Checks predictions against satellite flood evidence | | Exposure analysis | Counts who and what may be affected | | Priority index | Decides where intervention should happen first | | WebGIS dashboard | Presents results to decision-makers | --- ## Folder structure ``` config/ config.yaml Master configuration — paths, weights, CRS, scenarios data/ raw/ Original downloaded datasets (boundary, dem, sentinel1, ...) interim/ Clipped, reprojected, and cleaned datasets processed/ Analysis-ready rasters, vectors, tables, dashboard layers notebooks/ Portfolio Jupyter notebooks — one per analysis stage scripts/ Executable workflow scripts (00–07 + run_pipeline.py) src/floodsense/ Reusable Python package (imported by scripts and notebooks) dashboard/ Streamlit + Folium WebGIS app outputs/ Maps, tables, validation results, reports docs/ Methodology, validation report, portfolio summary tests/ Unit tests for c …