Post-hurricane forest damage assessment using satellite remote sensing
This study developed a rapid assessment algorithm for post-hurricane forest damage estimation using moderate resolution imaging spectroradiometer (MODIS) measurements. The performance of five commonly used vegetation indices as post-hurricane forest damage indicators was investigated through statistical analysis. The Normalized Difference Infrared Index (NDII) was identified as the optimal damage indicator among these vegetation indices. An approach for detecting forest damage at a regional scale, without relying on ground inventory or sampling, was designed and validated. The validation showed that the relative change of pre- and post-hurricane NDII was linearly related to the damage severity estimated by the ground inventory with the coefficient of determination 0.79. This approach was applied to evaluate forest damage severity and the impacted region caused by Hurricane Katrina.