ABSTRACT
We present an application of the space-beats-time (SBT) statistical framework to analyze the impact of Hurricane Rita (2005) on vegetation in coastal Louisiana parishes. SBT identifies areas of abrupt change and recovery after a disruptive event based on the interplay between spatial and temporal components of data. In this case study, we used the normalized difference vegetation index (NDVI) product from the MODIS sensor to generate separate temporal-only and spatial-only model predictions. We compare the performance of each model’s prediction before and after Hurricane Rita in terms of generated map patterns of prediction error in order to identify areas of abrupt change and recovery. Results suggest that SBT can be useful for identifying any potential areas of interest for targeting post-hurricane relief and recovery monitoring. Our case study shows conspicuous SBT effects in the aftermath of Hurricane Rita, including areas where flooding was identified by the Federal Emergency Management Agency (FEMA).
