US Wildfire Risk: What the NIFC Data Actually Shows

The National Interagency Fire Center (NIFC) maintains the most comprehensive wildfire record in the United States: over 25,000 fire perimeters dating back to 1984, with acreage, location, containment dates, and responsible agency. Combined with Karen Short's FPA-FOD dataset (1.88 million fire records, 1992–2015), this is the dataset that defines what we know about US wildfire frequency, size, and geographic distribution.

The data tells a clear story. Large fires are becoming more frequent, larger, and more destructive — particularly in California and the Pacific Northwest.

California: 3x Increase in Large Fires

Between 1984 and 1999, California averaged approximately 7–10 fires exceeding 10,000 acres per year. Between 2015 and 2024, that average increased to 20–30. The 2020 fire season alone produced the August Complex Fire (1.03 million acres), the SCU Lightning Complex (396,000 acres), and the Creek Fire (379,000 acres) — three of the largest fires in state history, all in a single year.

The driver is documented. Abatzoglou & Williams (2016, PNAS) found that anthropogenic climate change doubled the area of western US forest fires between 1984 and 2015, primarily through increased temperature, reduced humidity, and earlier spring snowmelt. Westerling et al. (2006, Science) documented a 78-day increase in fire season length and a 400% increase in large wildfire frequency since the 1970s.

For property underwriters, this means historical fire frequency from the 1990s systematically underestimates current and future wildfire risk in California. A fire frequency analysis that uses the full 1984–present record will produce a lower hazard score than one that weights the most recent 15 years more heavily.

Pacific Northwest: Emerging Risk Zone

Oregon and Washington historically had fewer large fires than California. The 2020 Labor Day fires changed that assessment. In a single week in September 2020, Oregon burned over 1 million acres — more than the previous 10 years combined. The Beachie Creek, Lionshead, and Holiday Farm fires destroyed over 4,000 structures.

NIFC data shows the Pacific Northwest averaging 8–12 significant fires (≥1,000 acres) per year since 2015, up from 3–5 per year in the 1990s. The region's fire risk profile is converging toward California's, driven by the same climate mechanisms: warming temperatures, drier fuels, and compressed moisture windows.

WUI Exposure: The Loss Multiplier

Fire frequency alone does not determine insured loss. The wildland-urban interface (WUI) — where developed areas meet undeveloped wildland — determines how many structures are exposed. USGS SILVIS Lab data shows approximately 46 million US homes in the WUI as of 2020, a number that has grown 40% since 1990.

California's 2017–2021 wildfire seasons produced over $30 billion in insured losses (California Department of Insurance). Multiple carriers have withdrawn from the California homeowners market entirely. State Farm, the largest home insurer in the state, stopped writing new policies in May 2023. This is not a future risk scenario. It is current market reality.

How CivilSense Uses NIFC Data

CivilSense integrates NIFC data at three levels:

Live monitoring: Active fire perimeters from NIFC WFIGS (Wildland Fire Interagency Geospatial Services) are ingested every 12 hours. These polygons form the basis of wildfire consequence rings and real-time hazard assessment.

Historical frequency: NIFC perimeters from 1984–present are used to compute fire frequency by region. The hazard scorer counts fires within a configurable radius of any US location and weights recent decades more heavily to capture the climate trend.

Stochastic modeling: The wildfire stochastic event set samples synthetic fire events from log-normal size distributions fitted to NIFC historical data. Regional parameters (California: logMean=8.5, PNW: logMean=8.0, Mountain West: logMean=8.2) produce frequency and size distributions consistent with the observed record. A climate trend multiplier adjusts fire frequency for forward-looking scenarios.

What This Means for Underwriters

Any property within 10 km of a NIFC-recorded fire perimeter from the past 20 years should receive a wildfire hazard assessment before binding. Properties in the WUI with historical fire proximity and high-fuel vegetation should be evaluated against climate-adjusted fire frequency, not historical averages.

The NIFC data is public, authoritative, and comprehensive. The question is not whether the data exists. The question is whether your underwriting workflow integrates it. CivilSense computes wildfire hazard scores from NIFC data for any US address in under 10 seconds.