Monitoring the Intensity and Extent of Wildfire Using Remote Sensing

نوع: Type: Thesis

مقطع: Segment: masters

عنوان: Title: Monitoring the Intensity and Extent of Wildfire Using Remote Sensing

ارائه دهنده: Provider: Peyman Heidarpour

اساتید راهنما: Supervisors: Dr. Morteza Heidari Mozaffar

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: Dr. Hassan Khotanloo-Dr. Hossein Torabzadeh Khorasani

زمان و تاریخ ارائه: Time and date of presentation: 2025

مکان ارائه: Place of presentation: 43

چکیده: Abstract: Forest fires are one of the greatest environmental challenges, with widespread impacts on natural ecosystems, climate, and human communities. Monitoring the intensity and extent of these fires is essential for crisis management and planning the restoration of affected areas. This study aimed to investigate the intensity and extent of forest fires using remote sensing in three regions (first: Marivan in Iran, second: Boeotia in Greece, and third: Pedrógão Grande in Portugal) with different areas and conditions. In this study, satellite data from Landsat 8 and Sentinel-2 were used to analyze the burned areas. First, the intensity and extent of the fires were classified using the NBR index and in accordance with USGS standards. Then, various indices including the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Moisture Stress Index (MSI), Normalized Difference Moisture Index (NDMI), Soil-Adjusted Vegetation Index (SAVI), Leaf Area Index (LAI), Digital Elevation Model (DEM), and Land Surface Temperature (LST) were calculated using the Google Earth Engine platform. Regression and correlation analyses between these indices and fire intensity (NBR) were evaluated in three states: pre-fire, post-fire, and the difference between them. In the first region, the results of linear regression and the coefficient of determination for fire intensity (dNBR) with dNDVI from Landsat 8 imagery were 0.936 and from Sentinel-2 imagery were 0.844; with dNDWI from Landsat 8 imagery were 0.976 and from Sentinel-2 imagery were 0.960; with dMSI from Landsat 8 imagery were 0.886 and from Sentinel-2 imagery were 0.918; with dNDMI from Landsat 8 imagery were 0.859 and from Sentinel-2 imagery were 0.885; with dSAVI from Landsat 8 imagery were 0.936 and from Sentinel-2 imagery were 0.844; with dLST from Landsat 8 imagery were 0.824. With dLAI derived from NDVI from Landsat 8 imagery, it was 0.936, and from Sentinel-2 imagery, it was 0.844. With the DEM of the region from Landsat 8 imagery, it was 0.144, and from Sentinel-2 imagery, it was 0.159. The results showed that the NDVI, SAVI, NDWI, NDMI, and MSI indices have significant linear correlations with fire intensity and extent. Changes in vegetation cover and moisture are related to fire intensity and extent. The findings indicated that a decrease in vegetation cover (an increase in NBR) has an inverse relationship with an increase in land surface temperature (LST). These changes clearly reflect the effects of fire. The LAI was also calculated in two ways: using the EVI and the NDVI indices. Linear regression of LAI using EVI produced poor results, but using NDVI produced results similar to the linear regression results of NDVI. Furthermore, linear regression was compared with a second-degree polynomial nonlinear regression. The results showed that linear regression performed similarly to second-degree polynomial nonlinear regression.