Wildfire Prediction (Sentinel-2)

Overview

An independent Python proof-of-concept for satellite wildfire analysis. Ingests multispectral Sentinel-2 imagery to compute vegetation and burn indices, classifying burn severity across Yukon Flats, Alaska.

How it works

1. 01

   Multispectral satellite band processing calculating NDVI (Normalized Difference Vegetation Index), NBR (Normalized Burn Ratio), and NDMI (Normalized Difference Moisture Index).

2. 02

   Differenced NBR (dNBR) change detection isolating post-fire perimeter dynamics.

3. 03

   Burn severity classification mapping applied to the Yukon Flats 2022 fire season (~15,000 km² study area).

4. 04

   Geospatial raster array manipulation using rasterio, NumPy array vectorization, and SciPy spatial analysis.