Detecting wildfire damage from space is an important and impactful use case for remote sensing and satellite data.
So, when I saw that Day 15 of the #30DayMapChallenge is themed Fire, I immediately decided to build a small geospatial pipeline to detect wildfire burn scars and assess property damage using geospatial data.
Using ESA Sentinel satellite data, NASA FIRMS fire observations, and OSM building data via osmnx, I created a reference map one month before the fire and a burned area map during the wildfire. Then, I computed Normalized Burn Ratio (NBR) differences to detect damage, filtered noisy patches, and mapped these areas interactively against building footprints.
As a target area, I focused on the devastating wildfires in the Los Angeles area during the summer of 2025. This methodology shows an example of how wildfire can be quickly and globally detected - and as you will see, the results, even the efficient steps to contain the wildfire, are visible. Hopefully, with more geospatial intelligence, in the future, we can prevent even more damage caused by natural disasters.
Code: https://open.substack.com/pub/milanjanosov/p/detecting-wildfire-damage-with-python?r=3o5qdz&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
#30DayMapChallenge #WildfireDetection #Geospatial #RemoteSensing #PythonGIS #SentinelHub #NBR #FIRMS #OpenStreetMap #DisasterMapping #SpatialDataScience #EarthObservation #WildfireDamage
