Analysis of the Spatial and Temporal Correlation Statistics of Soil Moisture Observed in the Russian River Basin

Statistical analyses have been used to study the temporal and spatial variability of soil moisture in the Russian River watershed. The variability found in soil moisture measurements has been compared to soil moisture variability simulated using the National Weather Service (NWS) Hydro Lab Research Distributed Hydrological Model (HL-RDHM). Soil moisture observation data for 2011 were obtained from the National Oceanic and Atmospheric Administration (NOAA) Hydrometeorology Testbed (HMT). These data were collected at seven soil moisture observing stations in the Russian River Basin. The model was run using a-priori parameter estimates and atmospheric forcing grids supplied by the NOAA Office of Hydrological Development (OHD) and the California Nevada River Forecast Center (CNRFC). These analyses were necessary during the development of successive-correction data assimilation techniques.

The temporal variability of soil moisture in the basin was analyzed using the temporal autocorrelation functions calculated for each of the seven HMT observing stations during the study period. The spatial soil moisture variability was examined using correlation coefficients calculated for each possible station data pair in the network. Results from these two procedures indicate that coherent spatial and temporal features of the soil moisture field in the Russian River basin can be inferred from HMT soil moisture observations. The third part of the analysis involved the construction of a covariance matrix from the spatial and temporal differences between the simulated and observed data. A positive definite test on this matrix revealed that it is possible to minimize the difference between the HL-RDHM simulations and the corresponding soil moisture observations from the HMT network. This poster will demonstrate how these analyses have been used to help determine whether or not soil moisture observations can be assimilated into the HL-RDHM, and thereby improve the quality of its soil moisture and streamflow simulations.