Automated Storm Track and Cyclone-Centered Compositing Analysis for Model Evaluation

Abstract:

The clouds, winds, and precipitation generated by extratropical cyclones have a dominant role in midlatitude climate, and therefore it is important that climate models properly represent these cyclones. This presentation introduces an automated process-oriented diagnostic (POD) for evaluating model generated extratropical cyclone characteristics. This POD will be distributed as part of the NOAA Model Diagnostic Taskforce Diagnostic package.

The POD includes analysis of the Eulerian storm track, which provides a good metric for studying aggregate storm activity and its relationship to time-mean wind and precipitation fields. The POD also contains an analysis of Lagrangian cyclone tracks, i.e., tracking the storm systems, in time and space. The tracking algorithm used in this diagnostic is the MAP Climatology of Midlatitude Storminess. The POD analyzes the statistics of the tracks – location, genesis, intensification. Then it generates cyclone-centered composites. In the composites, variables such as wind, precipitation and clouds are averaged together across a large number of cyclone events in a reference grid in which the cyclone’s low-pressure minimum is at the center of the frame. This allows for a comparison between model output and observations – because the orientation of the composites relative to a cyclone center is same for both model and observations. Also, by averaging over all extratropical cyclone events within 3 or more seasons, factors such as natural variability are minimized. For this POD, model observations are compared with reanalysis fields for wind and satellite retrievals for precipitation and clouds. Results of the POD are shown for output from the NOAA GFDL and NASA GISS climate models.