Surface and profile soil moisture spatio-temporal analysis during an excessive rainfall period in the Southern Great Plains, USA

In this work we analyze the temporal stability of soil moisture at the field and watershed scales in the Little Washita River Experimental Watershed (LWREW), as part of the remote sensing Cloud and Land Surface Interaction Campaign (CLASIC07) during June 2007 in south-central Oklahoma. Temporal stability of surface and profile soil moisture data were investigated for 20 LWREW soil moisture measurement stations. In addition, daily surface and profile soil moisture measurements were obtained in four 800 m by 800 m fields (remote sensing footprint), including two rangeland sites and two winter wheat fields. The work aimed to analyze the temporal stability of soil moisture at the watershed and field scale and to identify stations within the watershed, as well as locations within each field, that were representative of the mean areal soil moisture content. We also determined the relationship between sites found to be temporally stable for surface soil moisture versus those determined stable for average profile soil moisture content. For the unusually wet experimental period, results at the watershed scale show that LWREW stations 133 and 134 provided stable underestimates, while stations 132 and 154 provided stable overestimates of the watershed mean at all depths. In addition, station 136 had very high non-zero temporal stability at the 25 cm and 45 cm depths indicating that it could be used as representative watershed site provided a constant offset value is used to acquire a watershed mean soil water content value. In general, the deeper depths exhibited higher soil moisture spatial variability, as indicated by the higher standard deviations. At the field scale, measured average profile soil moisture was higher in the winter wheat fields than the rangeland fields with the majority of the winter wheat depth intervals having high non-zero temporal stability. Field scale temporal stability analysis revealed that 4 of the 16 sampling sites in the rangeland fields and 3 of the 16 sampling sites in the winter wheat fields either under or overestimated the field means in the 0-5 and 0-60 cm depth intervals. Field sites considered temporally stable for the surface soil moisture were not stable for the profile soil moisture, except for the LW45 field where two sites were stable at both the surface and profile soil moisture. This finding is significant in terms of soil moisture ground-truth sampling for calibrating and validating airborne remotely sensed soil moisture products under extremely wet conditions. In addition, identification of temporally stable sites at the watershed and field scales in the LWREW provide insight in determining future measurement station locations and field scale ground sampling protocol, as well as providing data sets for hydrologic modeling. Published by Elsevier B. V.