zcher-20/wildfire-risk-qgis

Morocco wildfire risk mapping with QGIS, PyQGIS, and Python — fire detection, regional density analysis, and reproducible cartographic workflows # Morocco Wildfire Risk Mapping — QGIS + Python A geospatial wildfire-risk mapping project using QGIS, Python, and open Earth observation data. The project visualizes fire occurrence patterns, prepares spatial datasets, and produces reproducible wildfire-risk maps for Morocco. Complements the RICER wildfire prediction model by adding the geospatial and visual layer to the ML/data-science workflow. --- ## Project structure ``` wildfire-risk-qgis/ ├── data/ │ ├── raw/ # NASA FIRMS CSVs (not committed — see Data section) │ └── processed/ # Cleaned CSV, GeoJSON, GeoPackage ├── qgis/ │ ├── wildfire_mapping_project.qgz # QGIS project file │ └── exported_maps/ # QGIS layout exports ├── notebooks/ │ └── 01_data_cleaning.ipynb # EDA + preprocessing walkthrough ├── scripts/ │ ├── clean_fire_data.py # Filters and cleans raw FIRMS CSV │ ├── export_geojson.py # CSV → GeoJSON / GeoPackage │ └── pyqgis_style_layers.py # PyQGIS styling + map export ├── maps/ # Final exported map images ├── methodology.md # Full methodology writeup └── README.md ``` --- ## Quickstart ### 1. Install dependencies ```bash pip install pandas geopandas matplotlib shapely ``` ### 2. Download fire data Go to NASA FIRMS, select: - **Product:** VIIRS S-NPP 375m - **Region:** draw Morocco bounding box (lon −17.2 to −1.0, lat 21.0 to 36.0) - **Date range:** your target years - **Format:** CSV Save to `data/raw/FIRMS_morocco.csv`. ### 3. Run preprocessing ```bash python scripts/clean_fire_data.py python scripts/export_geojson.py ``` ### 4. Open in QGIS Load `data/processed/morocco_fires.gpkg` into QGIS. Open `qgis/wildfire_mapping_project.qgz` if the project file is present. Run `scripts/pyqgis_style_layers.py` from the QGIS Python console to apply styling. --- ## Data sources | Data | Source | |------|--------| | Active fire detections | NASA FI …