2001), Interpolating sparse surface measurements for calibration and validation of satellite-derived snow water equivalent
Abstract Geostatistical methods are used to interpolate and average point snow water equivalent (SWE) measurements to a spatial resolution appropriate for comparison with estimates from Special Sensor Microwave Imager (SSM/I) images. Block kriging, a form of optimal interpolation, is applied to stat...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.1062.8634 http://hydrologie.org/redbooks/a267/iahs_267_0093.pdf |
| Summary: | Abstract Geostatistical methods are used to interpolate and average point snow water equivalent (SWE) measurements to a spatial resolution appropriate for comparison with estimates from Special Sensor Microwave Imager (SSM/I) images. Block kriging, a form of optimal interpolation, is applied to station snow course measurements in and around the Ob' River basin in Russian Siberia, giving spatially averaged SWE estimates at the resolution of the sensors, 25 x 25 km cells. For the two dates studied, interpolated station data and SSM/I estimates agree well in flat, low-lying regions. In the Ural Mountains, the SSM/I estimates significantly underestimate SWE, with respect to the station data. In the Altai Mountains, the SSM/I algorithm indicates higher SWE than elsewhere, but comparison is difficult, due to the sparsity of stations there. The results will be used to adjust the SSM/I SWE retrieval algorithm for use in complex terrain, where remotely sensed snow data are particularly needed. |
|---|