Spatial analysis and mapping of potential wildfires from Landsat satellite data

Elibek Atrauovich Asangaliev, Baitak Apshikur, Sergey Lutay, Zhanna Aleksandrovna Assylkhanova

2024

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.35595/2414-9179-2024-1-30-476-490
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

In 2023 there was a severe forest fire on the territory of the Semipalatinsk Forest. More than 60 thousand ha were burned. This resulted in high environmental costs, destroying trees, real estate, recreational and even human lives. One of the most pressing issues was the determination of the origin and spread of forest fires. In such cases, remote sensing data is a spatial and temporal measure to obtain fast and accurate data to prevent the further spread of the fire and to neutralize this natural disaster. Using such geospatial information, it is possible to prioritize preventive measures to reduce the risk of forest fires and identify mitigating factors to increase the likelihood of immediate fire suppression in threatened areas. This work proposes to assess the fire potential and to determine the hazard potential. This is done by analyzing and mapping the area of the fire that has occurred. The mapping of the fire potential was approached from a remote sensing point of view by estimating and mapping the total vegetation cover using Landsat-8–9 OLI/TIRS DATA and the geographic information systems QGIS, SAGAGIS. A morphometric, spatial analysis of conditions was also conducted, taking into account many factors affecting fire potential—land exposure, aspect, wind direction, fire statistics, population density, climatic characteristics, etc. As a result, an attempt was made to create indices RBR, RdNBR, dNDVI, dGNDI, GEMI and BAI, which could be the basis for the determination of fire potential. These indices are based on the type of tree, the vegetation and the topographic features. These characteristics make it possible, after the classification of the Landsat images, to evaluate the reliability of the information obtained by determining the area of the fire through the Object Based Image Analysis segmentation method and by assessing the accuracy of the detected data. The index values identified were consistent with reliable information for identifying fire locations and monitoring estimated fire risk. They could be used to map fire potential from regional to local scales.