Climate Processes in CMIP5: A Process-based Analysis of Warm Season Precipitation for Northeast North America

Large scale processes related to warm season precipitation and its variability in the Northeast United States are investigated in observations and in CMIP5 model experiments for the historical period. A framework for evaluating model credibility has been developed which employs process based analysis focussing on the Northeast United States. This framework includes evaluation of models' ability to simulate observed atmospheric circulation features, moisture transport, and its divergence, and SST anomalies involved in generating precipitation and its variability. Analysis of observations indicates that wet summers in the Northeast are associated with negative mid-tropospheric height anomalies centered near the Great Lakes, positive height anomalies in the western Atlantic off of the Mid-Atlantic coast, southwesterly moisture flow and increased moisture convergence along the eastern seaboard, and positive SLP anomalies in the western Atlantic, related to southwest ridging of the North Atlantic subtropical anticyclone. We evaluate CMIP5 models using this process-based analysis, and contrast projections based on the models identified as credible versus those from the full suite of models. We also discuss the implications of process based measures of model credibility.