Elsevier, Journal of Hydrology, (422-423), p. 63-75, 2012
DOI: 10.1016/j.jhydrol.2011.12.039
Full text: Download
The characterization of the spatial-temporal variability of soil moisture is of paramount importance in many scientific fields and operational applications. However, due to the high variability of soil moisture, its monitoring over large areas and for extended periods through in situ point measurements is not straightforward. Usually, in the scientific literature, soil moisture variability has been investigated over short periods and in large areas or over long periods but in small areas. In this study, an effort to understanding soil moisture variability at catchment scale (>100km 2), which is the size needed for some hydrological applications and for remote sensing validation analysis, is done. Specifically, measurements were carried out in two adjacent areas located in central Italy with extension of 178 and 242km 2 and over a period of 1year (35 sampling days) with almost weekly frequency except for the summer period because of soil hardness. For each area, 46 sites were monitored and, for each site, 3 measurements were performed to obtain reliable soil moisture estimates. Soil moisture was measured with a portable Time Domain Reflectometer for a layer depth of 0-15cm. A statistical and temporal stability analysis is employed to assess the space-time variability of soil moisture at local and catchment scale. Moreover, by comparing the results with those obtained in previous studies conducted in the same study area, a synthesis of soil moisture variability for a range of spatial scales, from few square meters to several square kilometers, is attempted. For the investigated area, the two main findings inferred are: (1) the spatial variability of soil moisture increases with the area up to ~10km 2 and then remains quite constant with an average coefficient of variation equal to ~0.20; (2) regardless of the areal extension, the soil moisture exhibits temporal stability features and, hence, few measurements can be used to infer areal mean values with a good accuracy (determination coefficient higher than 0.88). These insights based on in situ soil moisture observations corroborate the opportunity to use point information for the validation of coarse resolution satellite images. Moreover, the feasibility to use coarse resolution data for hydrological applications in small to medium sized catchments is confirmed. © 2012 Elsevier B.V.