The Global Model Test Bed (GMTB) conducted an assessment of the GFS using the Grell-Freitas (GF) cumulus parameterization (GFS-GF) and compared the results against a control run using the GFS FY17, which employs the scale-aware Simplified Arakawa Schubert (SAS) cumulus scheme (GFS-SAS). This test was a follow up to a previous test conducted by GMTB using the GFS FY16, in which GF runs were initialized from the operational GFS analyses (cold starts). In contrast, for the current test, cycled data assimilation was employed to produce initial conditions consistent with the model physics. Additionally, a set of GF cold-started runs was produced using the FY17 GFS (GFS-GFcold) for an assessment of the impact of cycling.

The test, which employed the GMTB Single Column Model (SCM), and the Global Spectral Model (GSM) was planned jointly by the GF scheme developer, GMTB, the NOAA Environmental Modeling Center (EMC), and representatives of the NOAA's Next-Generation Global Prediction System program, with the goal of supporting the development of an advanced physics suite for the GFS. Global forecasts were run at medium resolution (T574, approximately 34 km) without tuning of the physics suite.

The GMTB staff would like to thank the GF developer (Georg Grell of NOAA/ESRL), the NOAA/EMC Global Team led by Vijay Tallapragada, the NGGPS and EMC Strategic Implementation Plan Physics Team co-leads (Jim Doyle of NRL, Shrinivas Moorthi of NOAA/EMC, Chris Bretherton of UW, and Georg Grell), and the NGGPS Program Office led by Fred Toepfer for their engagement in preparing the test plan, defining verification metrics, and supplying code.