Using ARM Observations to Evaluate Land-Atmosphere Coupling in CAM5 Model Hindcasts on the U.S. Southern Great Plains

Statistically significant interactions between summertime soil moisture and a number of atmospheric surface and boundary-layer variables have been observed at the U.S. Southern Great Plains Central Facility (SGP CF) site that is maintained by the Department of Energy's Atmospheric Radiation Measurement (ARM) program in northern Oklahoma (Phillips and Klein, 2014 JGR). The land-atmosphere coupling (LAC) strength was assessed by means of correlation coefficients R and “sensitivity indices” I (a measure of the comparative change in an atmospheric variable for a one-standard-deviation change in soil moisture).

In the current study, we evaluate similar features of LAC in global hindcasts generated by version 5.1 of the Community Atmosphere Model (CAM5.1), when coupled to the CLM4 land model and downscaled to the ARM SGP site. Each day's hindcast was generated after initializing the CAM5 atmosphere with ERA Interim reanalysis state variables, while other needed variables were obtained from a nudging simulation. In addition, the CLM4 daily initial conditions were determined by running the land model offline using observed surface net radiation, precipitation, and wind as forcings.

Different aspects of LAC in the CAM5 will be compared with those found in the ARM observations during the summers of 2003-2011, when 3 independent measurements of soil moisture are available to provide an estimate of the inherent uncertainties in the LAC strengths determined from the ARM observations. This evaluation may uncover some unrealistic aspects of LAC in the CAM5 model that point toward potential deficiencies in its land or atmospheric model parameterizations.

Acknowledgments This work was funded by the U.S. Department of Energy Office of Science and was performed at the Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.