GRSG Conference 2022: Orbit to Outcrop

Title:  Forest Fire Assessment and Correlation Between Vegetation Regrowth and Fire Severity with dNDVI and dNBR

Author:Khalid Abiola Lawal

Abstract:

The effect of forest fires on humans, ecosystems and economies worldwide cannot be downplayed. Spectral index, more importantly, vegetation indices, is one of the crucial remote sensing tools used for detecting, analyzing and quantifying wildfire and its effects. This study addresses the impact of the Dixie wildfire. The study aims to evaluate the severity of fire in the area and the recovery rate of vegetation using the NBR, NDVI, dNDVI and dNBR indices.

A correlation analysis was also performed between the dNBR and dNDVI results to check the relationship between the two indices in evaluating fire severity and recovery rate. Level-2 Sentinel 2 satellite images for 2020 and 2021 were used to perform multitemporal analysis of the study, classifying the indices into five categories: high severity to unburn or regrowth vegetation.

A correlation was established between the dNDVI and dNBR results from the analysis with a correlation coefficient of 64% and a p-value of 0.04. However, the result is not strong enough to believe that the dNDVI index can be a good substitute for dNBR when evaluating burn severity.