Post-Helene Change Detection Assessment using Remote Sensing Images in North Carolina

Abstract. The occurrence of hurricanes in the Southern U.S. is increasingly frequent, and assessing the damage caused to forests is essential for implementing protective measures and comprehending recovery dynamics. This work aims to create a data integration framework utilizing LANDSAT 8 and geographic information system (GIS) data for change detection analysis. We propose a composite spectral index method along with univariate difference imaging and ISO data clustering techniques to create a change detection map reflecting disturbances in regions of McDowell County in North Carolina affected by Hurricane Helene. Accuracy assessment is conducted with GIS data and maps from the National Land Cover Database. The proposed technique demonstrated commendable performance, with overall accuracy ranging between 64% and 86% and kappa statistics spanning from 0.2 to 0.64. The results demonstrate a fair correlation with GIS data and can be utilized to evaluate damage severity.