Methodology & Data Sources

How we collect, filter, and present wildfire data

Current Wildfires is built entirely on public, authoritative data. This page explains exactly where our information comes from, how we turn raw satellite detections into a usable map and readable fire pages, and — just as importantly — what our data can and cannot tell you.

Live fire detection (the map)

The live map is powered by NASA FIRMS (Fire Information for Resource Management System). We use active-fire detections from the VIIRS instruments aboard the Suomi-NPP and NOAA-20 satellites, which scan the entire planet several times a day and flag thermal anomalies — points on the ground much hotter than their surroundings. Each point on our map is one such detection (a "hotspot"). The underlying data refreshes continuously throughout the day.

How we filter to real wildfires

Satellites detect heat, not "wildfires" as such — the same sensors pick up gas flares, industrial sites, and agricultural burning. To keep the map focused on genuine wildfire activity, we apply a filter based on two signals that distinguish a wildfire from routine industrial heat: whether the detection occurs at night, and how much radiative power (FRP) it emits. Detections that look like persistent industrial heat sources are screened out. This filter is deliberately conservative: we would rather occasionally miss a small fire than fill the map with gas flares and factories.

Named fires — the Wildfire Newsroom

For significant fires we go beyond raw hotspots and build a dedicated page using official incident data:

• United States — NIFC / WFIGS (National Interagency Fire Center).
• Canada — CWFIS / NRCan (Canadian Wildland Fire Information System).
• Europe — EFFIS (European Forest Fire Information System, part of Copernicus).

We include fires above a size threshold for each source, so the Newsroom stays focused on events that matter.

Satellite imagery

Where available, each fire page includes before/after satellite imagery from Sentinel-2 (European Space Agency / Copernicus). We show two views: natural color, and a short-wave infrared (SWIR) view that cuts through smoke and highlights actively burning areas and burn scars. Imagery is generated for the fire's exact coordinates, so you see that specific fire — not a generic regional view.

Plain-language summaries

Each fire page includes a written summary in English and Spanish. These are generated with Google Gemini from the structured data we hold about the fire — its location, size, containment, how long it has been burning, its recent growth, nearby satellite heat activity, and (in Europe) the type of land that burned. The summaries are original text written for this site; we do not copy descriptions from other services. Like any automated writing, they can contain errors — which is why every page also links back to the official source and timestamps its data.

Updates and timestamps

The live map updates throughout the day as new satellite passes come in. Newsroom fire pages are refreshed on a regular schedule and show when their data was last updated. Because satellites pass over any given location only a few times per day, there is always some delay between an event on the ground and its appearance here.

Limitations and accuracy

Sources

• NASA FIRMS — active-fire satellite detection (VIIRS/MODIS).
• NIFC / WFIGS — United States wildfire incidents.
• CWFIS (NRCan) — Canada wildfires.
• EFFIS (Copernicus) — Europe wildfires and burned areas.
• Copernicus Sentinel-2 — before/after satellite imagery.