Quick Links


VIIRS AF Products

VIIRS AF replacement product

This product is generated using the latest MODIS Collection 6 algorithm which has been adapted for VIIRS. The replacement VIIRS AF product offers an active fire mask that flags fires and other relevant pixels (e.g. cloud), fire radiative power (FRP), as well as QA and metadata details related to calibration and granule specific information. The figures below are examples from the Rim fire, California, between August 17th - September 8th, 2013. The top figure shows the FRP product generated from the replacement code while the bottom chart demonstrates the difference in fire detection counts between the replacment product and the standard VIIRS AF product (which is based on the MODIS collection 4 algorithm).


While the Visible Infrared Imaging Radiometer Suite (VIIRS) sensor, aboard the Suomi-NPP satellite, and Aqua have similar orbital characteristics and compatible sampling of the diurnal cycle of fire activity, sensor differences, including pixel sizes, along-scan aggregation, and swath width result in inherent differences in the expected fire observations.

The standard product is comprised solely of a list of fire‐pixel locations and the current the VIIRS detection algorithm is a “stripped down” version of the MODIS Collection 4 production code (see below).

The MODIS (MOderate Resolution Imaging Spectroradiometer) sensors is on-board NASA’s EOS Terra and Aqua satellites. More information is available on the NASA MODIS website.

The MODIS active fire detections are based on the contextual algorithm developed by Giglio et al. (2003) and currently is up to Collection 6 in the reprocessing chain. Reprocessing of the entire MODIS data archive is periodically performed to incorporate better calibration, algorithm refinements, and improved upstream data into all MODIS products. The updated MODIS data archive resulting from each reprocessing is referred to as a collection. Later collections supersede all earlier collections.

Each active fire location represents the center of a 1 km pixel that is flagged by the algorithm as containing a fire within the pixel.

The collection 6 improvements include:

  • Reduce false alarms in Amazon caused by small forest clearings.
  • Processing to extend to oceans and other large water bodies.
  • Detect off-shore gas flaring.
  • Dynamically adjust potential fire thresholds.
  • Detect smaller and/or cooler fires.
  • Improved cloud mask.
  • Improved detection confidence estimate.
  • Updated FRP retrieval

VIIRS vs MODIS example from Idaho on September 2012:

Nearly ten fire complexes were burning in Idaho on September 9th and both Suomi-NPP VIIRS and Aqua-MODIS detected numerous hot spots. Given the close timing in overpass (Aqua was only 20 minutes behind S-NPP) and similar viewing angle, the agreement in detections is quite good, as evident in the grid plot. The shading reflects the VIIRS scan angle. The darker the shade of blue, the greater the scan angle. The white numbers are binned in 2° x 2° cells to summarize the coincident detections between VIIRS and MODIS-Aqua. The first number (N/x/x) is the VIIRS detections; the second (x/N/x) is the number of VIIRS pixels with a coincident MODIS fire detection; and the third number (x/x/N) is the MODIS-Aqua detections. The image below the grid plot is an RGB (M-bands 5-4-3) from VIIRS while the 3rd image is from Aqua-MODIS.