Recent regional numerical modeling work has explored the role of feedbacks with soil moisture in warm season continental precipitation. The focus has largely been on how domain-average initial soil moisture amount conditions future precipitation on monthly to seasonal timescales (with corresponding implications for precipitation predictability). While these investigations have greatly increased our understanding of the relevant feedback pathways, the role of initial soil moisture spatial distribution, rather than simply total amount, has been given less attention.

In the work presented here, we aim to quantify the relative influences of initial soil moisture amount and initial soil moisture distribution in the GCIP domain. A series of numerical simulations are made for a domain covering the U.S. great plains and southwest using the Regional Atmospheric Modeling System (RAMS) for the GCIP warm season Enhanced Seasonal Observing Periods (ESOPs) of 1995, 1996, and 1997. Control simulations are initialized with soil moisture and temperature from the NCEP reanalysis, as well as with soil texture and soil hydraulic properties from the LDAS database, and are validated against various GCIP datasets including precipitation observations from the Arkansas-Red Basin River Forecast Center (ABRFC). We show the results of various numerical experiments, which are based on these control simulations and are designed to investigate the relative sensitivity of simulated precipitation (higher-order statistics as well as space and time means) to both the initial domain-average soil moisture amount and the spatial distribution of initial soil moisture. We also investigate the persistence of these effects. We find that simulated precipitation is highly sensitive to initial spatial distribution as well as total amount, because the spatial variability enhances the three-dimensional (large-scale dynamic) feedbacks that amplify the one-dimensional (vertical thermodynamic) convective feedbacks. In addition, we discuss the sensitivity of the results to model convective parameterization and simulation configuration.