Studies that address the dynamics of fires in the state of Alagoas and their relationship with meteorological conditions remain scarce, especially ones that use Machine Learning in their methodology. Thus, to analyze the spatiotemporal dynamics of fires in the state of Alagoas and their relationship with meteorological variables, daily data from the VIIRS active fire detection product (on board the S-NPP and NOAA-20 satellites) were used to analyze fires from 2012 to 2023 in the state of Alagoas. The dataset was filtered, including only Fire Radiative Power (FRP) values above zero and considering only detections that were greater than 24 hours. Subsequently, the clustering methodology of Nascimento et al. was used, using the DBSCAN algorithm to identify and group fires. To analyze the behavior of meteorological variables, a range of ERA5 reanalysis variables were used. The results showed significant variability in the climatology of the FRP and Fire Radiative Energy (FRE), where the highest FRP averages occurred in the months of October to March, with the highest records occurring mainly in January (~10 MW/h), February (~9 MW/h), March (~9 MW/h), and December (~9 MW/h). During these months, the climatology of temperature at 2 meters presented the highest records, mainly in the sertão (~37°C). In addition, the climatology of relative humidity varied between 0 and 50% throughout these months in the sertão, coastal, and forest zone regions. The climatology of the FRE presented the highest averages in the months of November (~9000 MJ) and December (~8000 MJ), where the highest records of wind speed were obtained, mainly on the coast and in the forest zone (~6 m/s). The highest concentrations of FRP and FRE and the longest stretch of days of fires occurred on the coast and in the forest zone.
Previous Article in event
Previous Article in session
Next Article in event
Next Article in session
AN ANALYSIS OF FIRE DYNAMICS IN THE STATE OF ALAGOAS AND THEIR RELATIONSHIP WITH METEOROLOGICAL VARIABLES
Published:
30 May 2025
by MDPI
in The 7th International Electronic Conference on Atmospheric Sciences
session Meteorology
Abstract:
Keywords: Fire; Machine Learning; Remote Sensing; VIIRS;
Comments on this paper
