Toward reducing systematic errors in NWP – cross-evaluation of common physics from 6h-regional to 6d-global to 6mon-coupled applications

Abstract:

An integrated evaluation system against gridded data and observations is being applied against global models (FIM, GFS) and regional models (WRF-ARW applications for RAP/HRRR). An overview will be presented on wind, relative humidity, and temperature model errors as measured against rawinsonde and aircraft observations in common at 12h forecast duration for global and regional models. Systematic errors common to both applications will be presented. A common problem with deficient cloud cover has been evident in both 6h (3km HRRR-WRF-ARW) regional forecasts and 6-month coupled-global (FIM-HYCOM) forecasts, allowing improvements in a common deep/shallow convection scheme (Grell-Freitas) with subgrid-scale clouds to be evaluated across time scales.