A simple retrieval method for land surface temperature and fraction of water surface determination from satellite microwave brightness temperatures in subarctic areas. Remote Sensing of Environment
Abstract A strong linear relationship is found between Special Sensor Microwave/Imager (SSM/I) microwave (19 and 37 GHz) surface emissivities at horizontal and vertical polarizations over snow-and ice-free land surfaces. This allows retrieving the land surface emissivity and temperature from satelli...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Published: |
2003
|
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1061.5804 http://aramis.obspm.fr/%7Eprigent/2003_RSE_wetland_fily.pdf |
| Summary: | Abstract A strong linear relationship is found between Special Sensor Microwave/Imager (SSM/I) microwave (19 and 37 GHz) surface emissivities at horizontal and vertical polarizations over snow-and ice-free land surfaces. This allows retrieving the land surface emissivity and temperature from satellite microwave brightness temperatures after atmospheric corrections. Over the Canadian sub-arctic continental area, we show that the main factor modifying the emissivity is the fraction of water surface (FWS) within a pixel. Accordingly, a map of the fraction of water surface across the Canadian landmass is derived, given a correspondence within 6% as compared to the 1 km 2 Canadian National Topographic Database of water-covered areas. The microwave-derived surface temperatures are compared to synchronous in situ air and ground surface temperatures and also with independent satellite IR measurements over areas without snow or ice. Root mean square differences range between 2j and 3.5j, with mean bias error of the order of 1 -3j. Better results are always obtained with the 37 GHz channel rather than with the 19 GHz channel. Over dense vegetation, the microwave-derived surface temperature is closer to the air temperature (at surface level) than to the ground temperature. The proposed simple retrieval algorithm, not sensitive to cloud cover, appears very useful for monitoring summer interannual or seasonal trends of the fraction of surface water, as well as the daily land surface temperature variation, which are very important parameters in environmental change analysis. D |
|---|