Mean soil moisture estimation using temporal stability analysis

Abstract

In the present work, we estimated the mean soil moisture content in an area of 1285 km 2 (Soil Moisture Measurement Stations Network of the University of Salamanca, REMEDHUS) and in an experimental catchment of 0.62 km 2 (Rinconada Experimental Catchment). Both field sites are located in the River Duero basin (Spain). The work aimed to analyse the temporal stability of soil moisture and to identify the representative mean soil moisture measuring stations (RMSM) from each network of stations. Another aim of the work was to compare the results obtained using this methodology at two very different scales. In the study, we used long observation series that include several complete seasonal cycles. For the calibration (identification of the RMSM) and validation (estimation of mean soil moisture contents by RMSM) phases, we used the data from 23 soil moisture-measuring stations (REMEDHUS network) over a period of 52 months (from June 1999 to September 2003) and data from a network of 18 stations over a period of 41 months (May 2000 to September 2003) at the Rinconada E.C. At each station, soil moisture was measured fortnightly by TDR at four depths in the profile (5, 25, 50 and 100 cm). In the analysis, we used the data on depth-averaged soil moisture content. In the REMEDHUS network, the stations are independent of one another from the point of view of hydrology and are spatially distributed on the basis of physiographic and pedological criteria. At Rinconada, the stations are positioned on three transects perpendicular to the valley of the catchment and are located on the upper, middle and lower parts of the slope. Analysis of temporal stability clearly showed which stations were representative of wet and dry conditions and the persistence of representative stations even in highly contrasted situations. Station J14 was representative of the mean conditions at REMEDHUS, and station RM4 of those at Rinconada. Both stations are good estimators of the mean soil moisture in the two networks analysed. Finally, one of the most outstanding results of the study was the observation that 1 year (a complete seasonal cycle) of measurements is sufficient to determine the RMSM.