Meteosat Active Fire Monitoring (FIR) product update delayed



Changes to the Meteosat Active Fire Monitoring (FIR) product are now due to take place in August (previously end of July).

Last Updated

12 April 2022

Published on

18 July 2019

The active fire monitoring product is a fire detection product indicating the presence of fire within a pixel. The underlying concept of the algorithm takes advantage of the fact that SEVIRI channel IR3.9 is very sensitive to hot spots which are caused by fires.

The two main improvements to the algorithm will be:

Meteosat Active Fire Monitoring example
Figure 1: Example of the fire product
  1. Fire detection closer to the coastlines will be improved by introducing a new enhanced higher resolution land-sea mask.
  2. The threshold tests will adopt a Bayesian approach. Each individual test will have a minimum and maximum threshold. Values below the minimum threshold indicate no fire (confidence=0), while values above the maximum threshold indicate fire (confidence=1). Values between the thresholds are linearly interpolated between 0 and 1. Combination of all the tests together will yield a 'fire probability' between 0 and 100%.

Due to this new Bayesian approach, the new FIR GRIB product will have an additional 'fire probability' parameter encoded with parameter number of 192 of GRIB table, currently reserved for local use. There will be no changes for the FIR CAP product. The use of this local parameter, as well as its migration to using a WMO template, will be documented soon.

The new FIR product will be rolled-out as part of the MSG MPEF (Meteosat Product Extraction Facility) Release 2.8, planned for 6-8 August. More details about this release will be communicated in due course.

Sample data of the new FIR GRIB product for all Meteosat services (0 degree, RSS and IODC) are available on our ftp site:

Note: These sample data are provided for users to familiarise with the new product format. The new FIR product is still being validated and the algorithm test thresholds may require some fine tuning during the validation campaign.

For more information, contact our User Service Helpdesk.