ASSESSING TROPICAL CYCLONE DAMAGE USING MODERATE SPATIAL RESOLUTION SATELLITE IMAGERY: CYCLONE

assessment ABSTRACT: Tropical cyclones are a common natural disaster, and are predicted to increase in intensity and frequency under future climate change scenarios in many coastal areas across the world. Satellite remote sensing can provide a capability for large area (10,000’s km2) coverage and derivation of essential map products at high to moderate spatial resolution (5 – 30 m pixels) on a regular basis weekly to monthly. These data are ideal for assessing damage produced by the cyclone and can be used to derive appropriate information for planning and directing relief efforts in a short time frames. This study developed and tested approaches for assessing multiple damages caused in 2007 by tropical Cyclone Sidr in Sarankhola upazila (151.24 km2) in Bangladesh from moderate spatial resolution satellite imagery. Object based image classification techniques were applied to map cover types in pre-and post-cyclone SPOT 5 satellite imagery. Post classification change detection techniques identified types of land cover changes. Our results indicate that around 60 % of the study area was significantly damaged by Sidr. About 31 % of croplands were flooded by storm surges and with vegetation, settlements and infrastructure of the area all being fully or partially damaged. The methods developed may be used in future to assess the damages caused by tropical cyclones in Bangladesh and other countries