Burn Severity Analyst
Version: v0.1.5b
App: Burn Severity Analyst
Tutorial/manual: Fire severity analyst - SeverusPT
Burn Severity Analyst is a Google Earth Engine app developed to explore fire severity and recovery through satellite image time series. Instead of producing a single burn severity map, the app focuses on temporal profiles: users can inspect how spectral indices change before and after a fire and compare the post-fire trajectory of vegetation or burned surfaces across different areas.
The app supports analyses based on the Normalized Burn Ratio (NBR) and the Normalized Difference Vegetation Index (NDVI), two indices commonly used to detect fire effects and vegetation condition. The time-series workflow is useful for identifying abrupt breaks associated with fire events, tracking early post-fire response, and comparing recovery patterns over weeks, months, or years.
Main functionality
- Generate NBR or NDVI time series for a selected area of interest.
- Work with user-drawn geometries, coordinate-based circular buffers, or existing burned area polygons.
- Use satellite image collections from MODIS, Landsat 5, Landsat 7, Landsat 8, and Sentinel-2.
- Add annual burned area layers for mainland Portugal from EFFIS and ICNF sources.
- Select a burned area polygon directly on the map and use it as the analysis geometry.
- Compute pre- and post-fire deltas based on user-defined temporal windows.
- Export generated charts from the Earth Engine interface.
Typical use cases
The app is designed for exploratory post-fire assessment. A user can add a burned area layer, select a fire scar, choose a satellite sensor and spectral index, and define the analysis dates. After the time series is plotted, the secondary delta chart helps summarize how much conditions changed relative to the selected fire date and how that difference evolves over time.
This makes the app especially useful for screening burned areas before more detailed mapping, comparing recovery between locations, or communicating fire effects with time-series evidence rather than static maps alone.