Real-Time Application of Multi-Satellite Precipitation Analysis For Floods and Landslides
Adler, R.; Hong, Y.; Huffman, G.
NASA Goddard Space Flight Center, Laboratory for Atmospheres

Satellite data acquired and processed in real time now have the potential to provide the space-time information on rainfall needed to monitor flood and landslide events around the world. This can be achieved by integrating the satellite-derived forcing data with hydrological models and landslide algorithms. Progress in using the TRMM Multi-satellite Precipitation Analysis (TMPA) as input to flood and landslide forecasts is outlined, with a focus on understanding limitations of the rainfall data and impacts of those limitations on flood/landslide analyses. Case studies of both successes and failures will be shown, as well as comparison with ground comparison data sets—both in terms of rainfall and in terms of flood/landslide events. In addition to potential uses in real-time, the nearly ten years of TMPA data allow retrospective running of the models to examine variations in extreme events.

The flood determination algorithm consists of four major components: 1) multi-satellite precipitation estimation; 2) characterization of land surface including digital elevation from NASA SRTM (Shuttle Radar Terrain Mission), topography-derived hydrologic parameters such as flow direction, flow accumulation, basin, and river network etc.; 3) a hydrological model to infiltrate rainfall and route overland runoff; and 4) an implementation interface to relay the input data to the models and display the flood inundation results to potential users and decision-makers.

In terms of landslides, the satellite rainfall information is combined with a global landslide susceptibility map, derived from a combination of global surface characteristics (digital elevation topography, slope, soil types, soil texture, and land cover classification etc.) using a weighted linear combination approach. In those areas identified as “susceptible” (based on the surface characteristics), landslides are forecast where and when a rainfall intensity/duration threshold is exceeded. Results are described indicating general agreement with landslide occurrences.


Legal Disclaimer - Glossary - FAQ - Site Map - Contact us Last updated: 09-Feb-2010