Using NOAA AVHRR data to assess flood damage in China

Area of Science:

• Remote Sensing
• Environmental Monitoring
• Hydrology

Background:

• Satellite remote sensing offers a valuable tool for monitoring large-scale natural disasters like floods.
• Optical sensors like AVHRR face limitations due to cloud cover during flood events.
• The Normalized Difference Vegetation Index (NDVI) can be adapted for flood detection due to water's distinct spectral properties.

Purpose of the Study:

• To assess flood damage in the middle and lower reaches of China's Changjiang River (Yangtze River) in 1998 using satellite data.
• To evaluate the effectiveness of the NDVI derived from AVHRR data for flood inundation mapping.
• To quantify the extent of inundated cultivated fields.

Main Methods:

• Utilized NOAA-14 AVHRR datasets from before (May 26) and after (August 22) the 1998 flood.
• Applied geometric, atmospheric, and angular corrections to the AVHRR data.
• Calculated NDVI differences between pre- and post-flood images to identify inundated areas, considering land-use data.

Main Results:

• Identified 358,867 hectares of inundated land, including 207,556 hectares of cultivated fields.
• Results showed a higher inundation extent compared to previously reported figures.
• Validation with JERS-1 synthetic aperture radar (SAR) data indicated a 67.26% agreement in identified inundated areas.

Conclusions:

• The NDVI method using AVHRR data is a viable approach for flood damage assessment, even with cloud limitations.
• The study highlights the significant impact of the 1998 Yangtze River flood on agricultural land.
• Remote sensing techniques provide crucial data for disaster management and impact assessment.