Return to the Research Projects at NSIDC page

Snow Cover Validation Case Study: Tibet Plateau

NASAThis project is funded by NASA


To remeasure snow cover over the Tibetan Plateau, as optical sensors may in fact undermeasure the snow cover due to uncertainties resulting from cloud cover


Richard Armstrong

Project Summary

Previous studies have shown that the Tibetan Plateau is one the areas of largest disagreement between optical and microwave satellite snow products. At this point it is not entirely clear which remote sensing data set is correct, due to the fact that only a very limited amount of surface station data have been available to support objective validation. For the purposes of this study, the optical (MODIS) data is treated as “truth,” or at least closer to “truth” than results coming from the investigator's current microwave algorithm which appears to overestimate snow cover. However, the optical sensors may in fact undermeasure the snow cover over the Plateau due to uncertainties resulting from cloud cover. Therefore, the investigators are currently acquiring new surface snow observations over this region through a collaborative study with the Cold and Arid Regions Environmental and Engineering Research Institute (CAREERI) in Lanzhou, China. In addition, they are receiving feedback on prototype daily SSM/I Tibetan Plateau snow maps from CAREERI and the Remote Sensing Center of the Tibetan Meteorological Bureau in Lhasa, Tibet.

The apparent overestimation of snow cover by the passive microwave data may be partially explained by the effect on the microwave algorithms of an enhanced spectral gradient resulting from decreased atmospheric absorption in this region of extremely high elevation. Most current microwave snow algorithms are based on the assumption that the atmosphere is essentially transparent at frequencies around 18 and 36 GHz. However, recent work by Wang and Manning (2003) demonstrates that the atmosphere should be considered when validating microwave satellite retrieval algorithms based on surface and low-elevation aircraft measurements. In the case of the Tibet Plateau, the reduced atmosphere between the orbiting sensor and the ground surface at the mean elevation of 4,000 meters would significantly affect the spectral gradient at 18 and 36 GHz. The investigators applied the corrected temperature gradient, provided by Dr. James Wang, to pixels on the Plateau resulting in a microwave-retrieved snow extent that agrees more closely with that of the optical sensor.

Related Resources


Contact NSIDC User Services for more information.