An Operational Tool for Global Monitoring of Inundation Using NPP ATMS Data

Abstract:

The goal of this study is to introduce an operational microwave-based tool for the detection and monitoring of inundation across the globe using passive microwave observations from the Advanced Technology Microwave Sounder (ATMS) sensor onboard SUOMI NPP. ATMS surface sensitive channels, namely, the 23 GHz and the 89 GHz are used in this study. The inundation detection approach is based on the analysis of the standardized anomalies of a soil wetness index that is determined from the gradient between 89 and 23 GHz brightness temperatures. The dimensionless index is sensitive to extreme wetness conditions. Appropriate threshold-based techniques were implemented in the developed tool to detect and eliminate rainy pixels as well as snow and ice covered pixels. An automated tool was developed to process, analyze the data, develop the inundation product, and disseminate the detected inundated area through a web-based interface.

 The outputs of the developed algorithm were verified against records from the Darthmouth Flood Observatory data archive. The agreement was acceptable with POD reaching 80 % globally for flood with durations longer than 5 days. The analysis of the flood records showed that the most frequent flood events have a duration of 3 days. The flood detection and mapping system was able to reports more short duration events that lasted 1 day or less than what is in the flood observatory records. The inundation global mapping tool was deployed operationally using real time readouts from NOAA-CREST satellite receiving station in New York, USA.