Mean soil water content estimation using measurements from time stable locations of adjacent or distant areas

• Published in: Journal of hydrology (Amsterdam) Vol. 497; pp. 234 - 243
• Main Authors: Hu, Wei, Shao, Ming An, Hou, Mei Ting, She, Dong Li, Si, Bing Cheng
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier B.V 08.08.2013; Elsevier
• Subjects: Earth sciences; Earth, ocean, space; ecological value; Exact sciences and technology; Hydrology. Hydrogeology; loess; neutron probes; plateaus; prediction; Remote sensing; Scaling; soil depth; soil heterogeneity; Soil water content; soil water potential; Temporal variability; Time stability; vegetation; water analysis; water content; watersheds; Time stability; Scaling; Temporal variability; Soil water content; Remote sensing; loess; remote sensing; vegetation; variability; drainage basins; depth; plateaus; water content; soils; errors; stability; neutron probe
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2013.05.046

•Mean soil water content estimation with a time stable point of other areas is feasible.•This method can ignore the impact of variability in soil properties.•More accurate estimation can be available for larger areas and deeper soils.•This method was validated by predicting mean soil water content of larger watershed. Quick and accurate estimates of spatial mean volumetric soil water content (θ) are essential for validating remotely-sensed soil water content and water budget analyses. The objective of this study was to test and validate a methodology that utilizes measured θ from the Most Time Stable Locations (MTSLs) to estimate mean θ in an adjacent or distant area while negating the impact of variability in soil, vegetation and topographic properties. Soil water content measured by a neutron probe at depths of 0.1, 0.2, 0.4, 0.6 and 0.8m in Laoyemanqu watershed on the Chinese Loess Plateau was used to test our methodology. This method predicts mean θ of one area with measured θ at the MTSL from another area. Estimation errors depend on size of the study areas, number of measurement times in the target area and soil depth. A more accurate estimation of mean θ was found when using larger areas and deeper soils. Our method was also validated by predicting mean θ of a larger watershed (Liudaogou watershed) using the θ measurement at the MTSL at Laoyemanqu watershed. The proposed method has great potential for soil water upscaling with socio-economic, environmental and geo-political values.