Near Real Time Data

Visualize Near Real Time Data

Two platforms provide mapping capabilities for NRT FEDS fire tracking data.

NASA FIRMS US/Canada - FEDS output for North America (continental US, Alaska, Canada, and Northern Mexico) are viewable on the Experimental tab within NASA’s Fire Information for Resource Management System (FIRMS) US/Canada map application under the name “VIIRS Modeled Fire Perimeters.” Using FIRMS, you can view FEDS perimeter information from the current calendar year.

Disclaimer: The VIIRS Modeled Fire Perimeters product is intended to provide situational awareness for ongoing fire events in the US and Canada, not a precise estimate of the fire perimeter for emergency response. This research product estimates individual fire event perimeters and properties every 12 hours; the spatial and temporal resolution of each modeled fire perimeter reflects the characteristics and availability of VIIRS 375 m active fire detections from FIRMS.

Fire Events Explorer - Visualize FEDS fire tracking from the beginning of this year to the present, including fire perimeters, the active portion of the fire perimeter, and VIIRS active fire detections from the Suomi-NPP and NOAA-20 satellites.

Access Near Real Time Data

NRT data are available via OGC API. An example notebook that demonstrates how to subset, analyze, and visualize the data outside of FIRMS is available on NASA’s VEDA platform.

The API status Binder notebook also provides a quick way to visualize recent NRT outputs from the OGC API in your browser, without the need for any environment setup or downloads.

Data FAQ

Note

FEDS fire tracking data is labeled as “VIIRS Modeled Fire Perimeter” on the NASA FIRMS application.

Data FAQs

1. What is the source of the FEDS fire tracking data?

A: The FEDS algorithm was developed by Yang Chen and colleagues at the University of California-Irvine, NASA’s Goddard Space Flight Center (GSFC), Cardiff University (UK), and Universidad del Rosario (Colombia), as described in a scientific paper published in 2022 (Chen et al., 2022).

3. What is the temporal frequency of the FEDS fire tracking data?

A: FEDS fire tracking data are updated twice per day using the VIIRS 375 m active fire information from the morning (01:30) and afternoon (13:30) Suomi-NPP and NOAA-20 satellite overpasses over each fire event. The vector output from the FEDS algorithm is produced within approximately 4 hours following the availability of VIIRS 375 m active fire data in FIRMS.

4. What is the spatial resolution of the FEDS fire tracking data?

A: Although the vector data in the FEDS fire tracking data has no defined spatial resolution, the product is derived from the VIIRS 375 m active fire detections, and therefore reflects the spatial resolution and geolocation accuracy of those products from the Suomi-NPP and NOAA-20 satellites. The FEDS algorithm uses an alpha hull approach to cluster new active fire detections and model the updated fire perimeter every 12 hours. The alpha hull approach groups individual fire detection points from the VIIRS sensors on two satellite platforms to model their combined spatial extent and distribution at each 12-hour time step. Compared to a convex hull, the alpha hull approach is better able to capture irregular shapes of active fire events and changes in shape over time. See Chen et al., (2022) for additional details.

5. Why are FEDS fire tracking data only available for the Continental United States (CONUS) and Canada?

A: The FEDS algorithm was originally developed and tested for large fires in California (Chen et al., (2022)). Our team has scaled the production of the fire event tracking approach to cover CONUS and Canada, with data available via OGC API. Data for additional regions, including Alaska and Hawaii, will be released through FIRMS when available.

6. What is the temporal extent of the FEDS fire tracking data?

A: FEDS fire tracking data are currently available from 2018-2021 for the Western US and from Jan 1, 2025 to present for North America (continental US, Alaska, Canada, and Northern Mexico) via OGC API. This example notebook demonstrates how to subset, analyze, and visualize these data.

FEDS fire tracking perimeters are viewable through FIRMS beginning January 1 of the current calendar year.

FEDS fire tracking perimeters, active fireline, and VIIRS active fire detections are viewable through the Fire Events Explorer beginning January 1, 2025.

7. What validation of the the FEDS fire tracking data has been conducted?

A: Chen et al., (2022) provide a rigorous validation of the FEDS algorithm for large fire events in California between 2012-2020. The study compared the final FEDS perimeter data to official year-end fire perimeter data from the Fire and Resource Assessment Program (FRAP), established by the California Department of Forestry and Fire Protection. Table 6 in Chen et al. (2022) summarizes the comparison between FEDS and FRAP data for large fires in California in 2018 using standard comparison metrics (e.g., Accuracy, Precision, Recall, and Intersection Over Union). Overall, the FEDS data compare favorably to the FRAP year-end data. Validation for other regions and other fire types is an area of ongoing research by the Wildfire Tracking team, and updates from these ongoing studies will be added, as available.

8. How does the FEDS fire tracking data (i.e., the VIIRS Modeled Fire Perimeter Data product in FIRMS) differ from the USA Fire Perimeter layers in FIRMS?

A: The VIIRS Modeled Fire Perimeter Data product provides situational awareness for large fire events in CONUS and Canada every 12 hours, based on the FEDS algorithm and available 375 m VIIRS active fire detection data from Suomi-NPP and NOAA-20. The modeled perimeter is an estimate of the fire-affected area, active portion of the fire perimeter, and metrics of fire behavior. The VIIRS Modeled Fire Perimeter data also provides a history of modeled large fire growth every 12 hours for all fire events detected by VIIRS in CONUS and Canada. By contrast, the USA Fire Perimeter layer is the most recent official incident perimeter data. Official incident data provide a more precise estimate of the perimeter of large fire events in the US than the VIIRS Modeled Fire Perimeter data. The USA Fire Perimeter layer is updated periodically with new official incident perimeter data.

9. What are the attributes of the VIIRS Modeled Fire Perimeter data based on the FEDS algorithm?

Column Description Type/Unit
fireid Fire ID. Unique for each fire. Matches fireid. Numeric ID
pixden Pixel density, the number of pixels divided by area of the estimated perimeter. pixels/km2
duration Number of days since first observation of fire. Fires with a single observation have a duration of zero. Days
flinelen Length of active fire line, based on new pixels. If no new pixels are detected, flinelen is set to zero. km
fperim Length of estimated fire perimeter. km
farea Area within estimated fire perimeter. km2
n_newpixels Number of pixels newly detected since last overpass. pixels
n_pixels Number of pixel detections in history of fire. pixels
isactive Have new fire pixels been detected in the last 5 days? Boolean
ogc_fid The ID used by the OGC API to sort perimeters. Numeric ID
geometry The shape of the estimated perimeter. Geometry

10. What is the rationale for the 5-day threshold on new fire pixels?

A: The FEDS algorithm uses spatial and temporal search criteria to cluster VIIRS 375 m active fire detections into existing or new fire objects. The use of the 5-day threshold to associate new fire detections with an existing fire object reflects the nature of large fire behavior and periodic gaps in active fire detection data. Large fire events often exhibit episodic growth in response to changes in fire weather, suppression activity, and fuels. Periods with less intense fire activity may not generate active fire detections from VIIRS, leading to temporal gaps (12-hour, 24-hour, etc.) in fire detection information. The 5-day threshold allows the FEDS algorithm to connect future fire detections to the same fire object, or event, rather than splitting a large fire into multiple objects. In addition to periods of lower fire intensity, gaps in fire detection can occur due to satellite outages or observing conditions such as clouds that impact the VIIRS fire detection algorithm. The 5-day threshold reduces the influence of these data gaps on the fragmentation of large fire events in the FEDS algorithm output.

11. How can I find more information about the FEDS algorithm, data product versions, and related analyses?

A: Additional information about the FEDS algorithm can be found in Chen et al., (2022). The source code used to generate the VIIRS Modeled Fire Perimeter data product is available on GitHub, and an example notebook that demonstrates how to subset, analyze, and visualize the FEDS output outside of FIRMS is available on NASA’s VEDA platform. Finally, the data products from the FEDS algorithm are also available via OGC API.