Energy Balance Partitioning and Net Radiation Controls on Soil Moisture–Precipitation Feedbacks

Abstract

A series of model runs using the University of Oklahoma?s Advanced Regional Prediction System (ARPS) were conducted to investigate the relative impacts of energy balance partitioning and net radiation on soil moisture?precipitation feedbacks in the U.S. central plains and to examine how the dominant physical processes are affected by changes in mean soil moisture and spatial resolution. Soil temperature and Bowen ratio are influenced nonlinearly by soil moisture, and by varying the mean soil moisture in the model it was possible to examine the relationship between soil moisture and the scaling characteristics of these fields using the statistical moments. Information theory metrics were used to provide an indication of the uncertainty associated with varying model resolutions. It was determined that energy balance partitioning plays a dominant role in the occurrence of soil moisture?precipitation feedback, while net radiation was not impacted by mean soil moisture. A strong relationship was seen between soil moisture and the scaling properties of Bowen ratio, while soil moisture did not appear to influence the scaling characteristics of soil temperature. Spatial resolution had a large effect on the representation of boundary layer turbulence, with coarser resolutions unable to capture turbulent motions, which are necessary for convective processes. The ability of the model to capture boundary layer turbulence will alter the dynamics of soil moisture?precipitation feedback as the horizontal transport of moisture by turbulent motions will affect the spatial and temporal scales over which feedback occurs. Higher-resolution runs are generally associated with a higher information content. This may provide a methodology for monitoring land?atmosphere feedbacks via remotely sensed soil moisture and vegetation fields through statistical knowledge of the dependency of the resulting precipitation signal on soil moisture and vegetation fields at the resolution they were observed.