On the portability of the RST-FIRES technique to higher resolution EUMETSAT systems for early fire detection

 

Wildfires are a worldwide phenomenon with local and global effects. They may pose a risk for life and infrastructures, degrading air quality and perturbing large areas over a wide variety of biomes. The fire severity, frequency of occurrence, and duration of fire seasons have increased in recent decades. Climate change has undoubtedly played a role in this growth, as rising temperatures, changes in precipitation patterns and winds, and more extended drought periods have all contributed to increased fire danger. Many satellite-based methods for fire detection and monitoring have been developed to provide systematic and accurate information about fire locations and space-time evolutions. In order to detect and monitoring short-living events or fires characterized by very rapid evolution times, geostationary satellites have to be used, offering a very high observation frequency, i.e., a temporal resolutions of 30 up to 5 minutes.  Among the number of fire detection techniques based on this technology, the RST-FIRES, a change detection multi-temporal approach, has already demonstrated a significant improvement in terms of small/starting fire detection using EUMETSAT Meteosat Second Generation (MSG) SEVIRI data with a 15 minutes of temporal resolution. In this work, the RST-FIRES porting on the MSG Rapid Scan Service (RSS) data, offering 5 minutes of revisit time, is experimented and its possible impact in early fire detection is assessed and quantified. To do that, a first study case has been selected, analysing results achieved over the Calabria Region (Southern Italy) during July 2022 and comparing them with the outcomes of the standard RST-FIRES algorithm. Preliminary results suggest that RSS data would allow for a quite systematic earlier detection and a better sensitivity (doubled) than MSG 0deg data because of the improved temporal (and spatial) resolutions.