Removing Circulation Effects to Assess Land-Atmosphere Interactions in Observations and GLACE-CMIP5

The land surface feedback on surface air temperature (SAT) has been challenging to quantify due to spatially and temporally limited observational records of soil moisture and surface fluxes. SAT, however, is a well-measured quantity, from which a metric reflecting the strength of land surface feedback can be derived. If the influence of atmospheric circulation on SAT is accounted for, the land surface feedback could be described solely in terms of interannual to multidecadal SAT variability. An empirical method to remove the effects of circulation on observed SAT, termed dynamical adjustment, is explored over North America and Europe. A land-atmosphere coupling metric is derived from the dynamically adjusted SAT, which has been shown to relate to indicators of the land surface feedback in the Community Earth System Model (CESM) Large Ensemble. The efficacy of dynamical adjustment in separating SAT into circulation-induced and land surface driven components is evaluated by applying the method to SAT in the prescribed soil moisture Global Land Atmosphere Coupling Experiment (GLACE) CMIP5 experiment. The new index is compared with existing land-atmosphere coupling metrics to see how robustly land-atmosphere coupling can be described simply by dynamically-adjusted SAT. Finally, the evolution of identified hotspot regions in projections of future climate is also assessed.