Land-atmosphere coupling and seasonal forecast skill over the Great Plains and the Southeast United States

Droughts represent a significant source of social and economic damage in many parts of the United States. Having sufficient warning of these extreme events enables managers to prepare for and potentially mitigate the severity of their impacts. A seasonal hydrologic forecast system can provide such warning, but current forecast skill is low during the convective season when precipitation is affected by regionally-varying land surface heat flux contributions. Recent work developed a new classification methodology of coupling suitable at daily time scales based on the joint probability space of surface soil moisture, convective triggering potential and the low-level humidity index. The methodology was demonstrated over the Southeast United States using satellite remote sensing, reanalysis, and hydrological model data which demonstrated that a coupling-based drought index shows good agreement with the temporal and spatial variability of drought. Recent work shows that the forecast precipitation bias is consistent with derived coupling classification. In this work we explore the characteristics of coupling in the forecast model and compare it with that observed from the reanalysis model, with a particular emphasis on the role that land atmosphere coupling plays in the persisting hydrologic extremes. To do this, we consider the NCEP Climate Forecast System (CFS) reanalysis (CFSR) and the reforecasts (from CFS version 2) from 1982-2008 over two study domains, the Southeast and the Central Great Plains in the United States. .132 on 7-30-2012-->