Soil moisture persistence and its implications for predictability and skill in the GFDL forecast system

A key issue in the study of continental climates is the extent to which land surface processes impact predictability across timescales ranging from synoptic to interannual. In this study, we analyze a suite of ensemble simulations from the Geophysical Fluid Dynamics Laboratory (GFDL) forecast system to quantify the persistence of simulated soil moisture and assess its physical controls. Using a simple e-folding time measure, we investigate several aspects of soil moisture persistence that may be relevant to predictability, including the (i) seasonal evolution of soil moisture persistence; (ii) relationships between anomaly persistence and local climatology and remote large-scale sea surface temperature forcing; and (iii) asymmetries in the growth and decay of positive and negative soil anomalies.