Antecedent wetness conditions based on ERS scatterometer data

Abstract

Summary Soil moisture is widely recognized as a key parameter in environmental processes mainly for the role of rainfall partitioning into runoff and infiltration. Therefore, for storm rainfall–runoff modeling the estimation of the antecedent wetness conditions ( AWC ) is one of the most important aspect. In this context, this study investigates the potential of scatterometer on board of the ERS satellites for the assessment of wetness conditions in three Tiber sub-catchments (Central Italy), of which one includes an experimental area for soil moisture monitoring. The satellite soil moisture data are taken from the ERS/METOP soil moisture archive. First, the scatterometer-derived soil wetness index ( SWI ) data are compared with two on-site soil moisture data sets acquired by different methodologies on areas of different extension ranging from 0.01 km 2 to ∼60 km 2 . Moreover, the reliability of SWI to estimate the AWC at a catchment scale is investigated considering the relationship between SWI and the soil potential maximum retention parameter, S , of the Soil Conservation Service-Curve Number (SCS-CN) method for abstraction. Several flood events occurred from 1992 to 2005 are selected for this purpose. Specifically, the performance of the SWI for S estimation is compared with two antecedent precipitation indices ( API ) and one base flow index ( BFI ). The S values obtained through the observed direct runoff volume and rainfall depth are used as benchmark. Results show the great reliability of the SWI for the estimation of wetness conditions both at the plot and catchment scale despite the complex orography of the investigated areas. As far as the comparison with on site soil moisture data set is concerned, the SWI is found quite reliable in representing the soil moisture at layer depth of 15 cm, with a mean correlation coefficient equal to 0.81. The characteristic time length parameter variations, as expected, is depended on soil type, with values in accordance with previous studies. In terms of AWC assessment at catchment scale, based on selected flood events, the SWI is found highly correlated with the observed maximum potential retention of the SCS-CN method with a correlation coefficient R equal to −0.90. Besides, SWI in representing the AWC of the three investigated catchments, outperformed both API indices, poorly representative of AWC , and BFI . Finally, the classical SCS-CN method applied for direct runoff depth estimation, where S is assessed by SWI , provided good performance with a percentage error not exceeding ∼25% for 80% of investigated rainfall–runoff events.