Using MODIS imagery to map coastal floods from Hurricane Irene
Abstract
Hurricanes are one of the most severe natural disasters to impact the US coastal areas. Intense rainfall and storm surge brought by hurricanes often results in extensive floods, which pose a significant threat to infrastructure and human lives. Hurricane Irene was listed as a strong category 1 storm in the 2011 Atlantic hurricane season. The impacts of Hurricane Irene include extensive flooding and sizable wind damage, landslides that collapsed roads and wiped away stream gauges, and week-long loss of economic activity due to power outages and limited transport mobility.
Remote sensing is an invaluable tool in hazard monitoring, mitigation design, disaster response, and damage assessment. Due to the large spatial coverage and frequent revisit, no data cost, and low risk to gain access to the observed area, satellite remote sensing is especially helpful for relief and rescue efforts. To improve the quality of post-hurricane flood mapping and assessment, we developed an inundation-mapping demonstration project to demonstrate how MODIS imagery can aid in incident detection and post-hurricane flood analysis by applying the VIIRS flood algorithms, including the cloud shadow and terrain shade removal techniques and downscaling model. In this study, MODIS flood maps were generated at the original 500 m resolution and subsequently downscaled to 10 m spatial resolution along Hurricane Irene's impact zones across the the mid-Atlantic Bight of the US East Coast. Overall, flood maps at both 500-m and 10-m resolution were reasonably favorable upon evaluative comparison with hydrodynamic inundation model outputs and aerial photographs. The results demonstrate that MODIS flood map at the original 500 m resolution can provide a big picture over large area along the entire US east coastal area, which can provide guidance for FEMA to conduct high water mask survey, and request SAR data, which is usually with limited spatial and temporal coverage but can watch flood through clouds. While the downscaled flood map at 10 m resolution can significantly enhance the capability of moderate resolution sensors, like MODIS and VIIRS, and detect small scale features of flooding waters over local locations. This work can be particularly helpful for researchers and scientists working on flood modeling for validation and evaluation.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.H41M2275S
- Keywords:
-
- 3354 Precipitation;
- ATMOSPHERIC PROCESSESDE: 1821 Floods;
- HYDROLOGYDE: 1855 Remote sensing;
- HYDROLOGYDE: 4335 Disaster management;
- NATURAL HAZARDS