NASA FIRMS is a useful starting point for wildfire research, but a hotspot needs a timestamp, sensor and context before it becomes evidence. This guide shows how to read the map and avoid common mistakes.
How to use NASA FIRMS for wildfire research
NASA FIRMS (Fire Information for Resource Management System) brings satellite observations of active fires and thermal anomalies into maps, alerts and downloadable datasets. Its MODIS and VIIRS detections answer a specific question: where did a sensor observe a sufficiently strong thermal signal at its overpass time? They are useful for screening activity and building a time-stamped evidence trail. A detection alone does not identify the ignition point, confirm the cause, draw the final burned perimeter or issue an evacuation instruction.
Open the official NASA FIRMS Fire Map ↗. For an emergency decision, consult the responsible fire authority and local alerts instead of treating a satellite map as a live safety instruction.
Read one hotspot in five checks
- Time: record the acquisition time and the time you retrieved the data. “Near real time” describes delivery after a satellite observation, not continuous observation of the ground.
- Sensor: record whether the point came from MODIS or VIIRS and the product used. Their detection footprints and algorithms differ, so apparent point counts are not interchangeable measures of fire size.
- Location: treat the plotted point as a satellite detection, not a surveyed fire edge or exact origin. Examine the pixel footprint and surrounding detections before making spatial claims.
- Missing detections: no point does not prove no fire. A fire may occur between overpasses or be obscured by clouds, heavy smoke or canopy; a signal may also be below a sensor’s detection threshold.
- Ground truth: compare with dated incident reports, official perimeters and other observations. Keep conflicting or later revised evidence attached to its original timestamp.
What FIRMS can and cannot establish
| Question | Careful interpretation |
|---|---|
| Was thermal activity detected? | Yes, for a named sensor, acquisition time and product, subject to its algorithm and quality fields. |
| Where is the entire fire perimeter? | The hotspot layer cannot establish it. Use a dated, sourced perimeter product and check its mapping method. |
| How did this fire start? | FIRMS cannot determine ignition cause. Use the responsible investigation and preserve unresolved states. |
| Is it safe to travel or stay? | No safety inference follows from hotspot presence or absence. Follow official warnings, road information and orders. |
Build a defensible wildfire timeline from FIRMS records
Practical method added 19 October 2026. The worked example below is fictional and is not an incident reconstruction.
Before comparing dots on a map, separate three clocks: when the satellite acquired the observation, when a source published its information, and when you retrieved it. A download time is not an observation time. NASA’s VIIRS attribute documentation defines acquisition time in UTC. Preserve the original date and time together; record any local-time conversion separately, including its UTC offset.
Worked example: which observation came first?
Assume a fictional download contains a detection acquired at 01:20 UTC on 10 July. An incident bulletin was published at 18:45 on 9 July, explicitly UTC−07:00. Converting that bulletin gives 01:45 UTC on 10 July. The satellite acquisition precedes the bulletin’s publication by 25 minutes. If you downloaded both at 03:00 UTC, that retrieval time does not change their sequence.
Supported conclusion: the acquisition timestamp is earlier than the bulletin publication timestamp. Unsupported conclusion: the fire ignited at 01:20, or the satellite detected it before anyone on the ground. Publication time is not necessarily discovery time, and the first detection in your downloaded selection is not necessarily the first observation of the incident.
Record enough information to reproduce the selection
| Field | What to retain |
|---|---|
| Identity | Source URL, product, satellite and version as supplied |
| Time | Original acquisition date/time; timezone; retrieval date/time |
| Selection | Geographic extent, date range and filters used |
| Detection | Coordinates and available quality fields, interpreted using that product’s documentation |
| Comparison | Incident identifier, perimeter source, effective time and mapping method when available |
| Interpretation | What the item supports, unresolved questions and later revisions |
Compare layers without inventing a fire edge
A detection acquired after a perimeter’s effective time is not a simultaneous measurement of that boundary. First align the times and check the perimeter metadata. NWCG’s mapping workflow calls for source, acreage and effective-date information. If an essential timestamp is missing, mark the comparison unresolved rather than silently substituting the webpage’s update date.
NASA also documents active-fire caveats, including displaced detections associated with superheated plumes. A dot outside a mapped polygon therefore needs investigation; it is not, by itself, proof of a new ignition or an escaped fire front.
Three checks before publishing a map or case study
- Keep the original export. Save the unmodified file alongside your filtered working copy. Record the selection and retrieval time so another reader can understand what was included.
- Separate observations from interpretation. Use wording such as “a detection in this dataset at this time,” and identify any incident association as sourced or provisional. Do not turn the earliest selected point into an ignition coordinate.
- Keep missing observations visible as uncertainty. NASA explains that heavy cloud or smoke can hide activity. A gap in the detection layer does not establish a pause in burning.
For the next interpretation step, read fire perimeter versus burned area and MODIS versus VIIRS. This workflow supports research and source checking; it is not a live evacuation or travel assessment.
Common questions
Why does FIRMS show a point where there is no reported wildfire? An active-fire product detects thermal anomalies; not every observation is a wildfire. Check the product attributes and context before assigning an incident identity.
Why is there no point over a reported fire? Satellite timing, cloud, smoke, canopy, fire size or temperature can limit detection. The absence of a detection is not a negative incident report.
Is FIRMS the same as NASA EONET? No. FIRMS provides satellite fire and thermal-anomaly observations; EONET is a curated catalogue of natural events with source references. A catalogue event and a sensor detection are different evidence objects.
Explore related Tecnobosque explanations: satellite hotspot, VIIRS, MODIS and MODIS vs VIIRS. Keep the NASA FIRMS glossary entry as a short definition and route detailed source questions here.
Primary references: NASA FIRMS · NASA fire alert notes and detection limits · NASA Earthdata FIRMS overview. Reviewed 26 September 2026.
For the concise source assessment, see the Tecnobosque NASA FIRMS source profile. For a comparison with incident reports, see CAL FIRE vs InciWeb vs NASA FIRMS.