Large are variability in climate-based soil moisture estimates and implications for remote sensing |
| |
| Authors: | Manfred Owe Dr Bhaskar J Choudhury Dr James P Ormsby Dr |
| |
| Institution: | (1) Hydrological Sciences Branch, Laboratory for Terrestrial Physics, NASA/Goddard Space Flight Center, 20771 Greenbelt, MD, USA |
| |
| Abstract: | Although the basic physical relationships between various remotely sensed signatures and soil are sound, the representativeness of pixel-averaged wetness used to interpret the remotely sensed data is often suspect and at times even doubtful. To study the affect of local variation in soil moisture on the large-area average, mean areal precipitation, temperature and modelled soil moisture in 1/4° grid cells were analyzed for a 700 km by 400 km region in the US Southern Great Plains. Variability in five years of mean daily grid cell data was investigated for 150 km2 study sites, approximately corresponding in size to a satellite microwave pixel. The number of wetted grid cells and the magnitude of precipitation in grid cells within a site were highly variable. The variability of each of these factors is critical, and will not only influence mean site wetness, but also how the area will be seen from space. Analysis indicated that similar mean site precipitation values may result from numerous widely differing combinations of precipitation magnitude and watted gric cell population. Consequently, mean site moisture is also highly variable in the distribution and magnitude of the individual cell moisture values. Extreme values in several grid cells may severely bias the quadrant mean, while influencing a remote measurement much less. Although large-scale surface wetness may eventually be reliably inferred from satellite measurements, a degree of uncertainty will always exist, especially in semi-arid areas. One may be required to associate mean soil moisture with an expected distribution of wetness within a pixel. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
