The two late 2018 California wildfires both began on 8 November 2018, in Butte County and Los Angeles / Ventura Counties and became known as the Camp Wildfire and the Woolsey / Hill Wildfires.
The Camp Wildfire effectively destroyed the towns of Paradise and Concow, Butte county forcing thousands to flee before it and consuming nearly 19,000 properties, caused at least 86 deaths and is the deadliest and most destructive wildfire in California history to date.
The Woolsey / Hill Wildfire caused damage across a wide area but overall was not as deadly or as destructive with approximately 1,500 properties destroyed.
Both fires were indiscriminate as to the properties which were destroyed with light timber framed buildings, as well as stone and concrete buildings devastated. The damage ranged from complete destruction to damage from large scale exposure to intense heat and smoke.
All the fires ranged for several days, in the case of Camp Wildfire, nearly two weeks, causing large smoke occlusion for satellite based imagery. MIS analysts used data from the United States Geological Survey (USGS), Geospatial Multi-Agency Coordination (GEOMac) and Woodland Fire Support to determine the wildfires perimeters on a daily basis as well as the National Aeronautics and Space Administration (NASA) Visible Infrared Imaging Radiometer Suite (VIIRS) to pinpoint visible hotspots.
By manipulating VIIRS data against extracted building polygons, analysts were able to provide the market with a greater understanding of which buildings may have been in the vicinity of a fire (though not necessarily affected by the fire) long before satellite imagery was available.
The daily use of NASA VIIRS data to pinpoint a wildfires perimeter allows the market to determine where the fire was on a given day, as well as evaluate exposure for risks not yet affected. Though not 100% accurate, by manipulating VIIRS against buildings analysts were able to show which buildings may have been affected before any imagery (drone, handheld or satellite) was available, and then categorise the results with a defined level of confidence.
Created By: Oz Smith © 2019