Advanced Testing and Evaluation by the Developmental Testbed Center towards Physics Unification in the UFS

Under the UFS-R2O physics subproject, the Developmental Testbed Center (DTC) works closely with the physics developers to provide testing and evaluation support to the UFS physics development. The targeted physics include pre-implementation enhancements and innovations/updates towards a long-term goal of the UFS becoming a unified, convection-allowing Earth System Modeling system. The work is focused on two major physics suites, P8 [GFS_v17_p8 in Common Community Physics Package (CCPP) v6.0; an experimental suite for GFS v17/GEFS v13] and RRFS v1beta, along with several physics enhancements such as prognostic and scale-adaptive cumulus convection closure (dubbed progsigma) and the updated Thompson microphysics and MYNN PBL schemes. Real cases over both the land and ocean were simulated using the limited-area model (LAM) provided by the UFS Short-Range Weather (SRW) Application and the CCPP Single-column model (SCM) at both 13- and 3-km grid spacings to facilitate investigations of physics unification and scale adaptiveness. These cases capture processes of moist physics, atmospheric boundary layer, surface conditions, and their interplays, and represent phenomena under various cloud/weather/climate regimes. They include the ARM LASSO non-precipitating shallow cumulus cases, Atlantic tropical cyclones (TCs; Hurricanes Florence, Ian and Laura), the DYNAMO case with a transition from shallow to deep convection during the Madden-Julian Oscillation (MJO) initiation, and the Marine ARM GPCI Investigation of Clouds (MAGIC) case featuring a transition from stratocumulus off the west coast to cumulus clouds in the trade wind regime.

We aim to conduct evaluations against relatively reliable benchmarks that can help constrain parameterized processes and inform further physics improvement. Selected findings are described as follows. In the tropical convection cases (e.g. Hurricane Florence and DYNAMO), the 3-km LAM runs using the P8 suite generate a weaker TC with right-of-track errors and excessive large-wavenumber signals during the MJO initiation compared to the observations. Further analysis shows that more grid-scale precipitation and low-level clouds, and increased hydrometeor contents (incl. cloud liquid, rain, snow, and ice mixing ratios) are produced in the 3-km runs relative to their 13-km counterparts. Using progsigma results in less scale sensitivity, showing improved TC rainfall, track and intensity. It is possible that this cumulus convection closure tends to prevent microphysics from dominating rainfall production as grid spacing increases. This stresses the importance of interplays between moist physics components when considering scale awareness. The challenging MAGIC case persistently exhibits a lack of boundary layer decoupling and marine stratocumulus-to-cumulus transition, and an overestimation of rainfall in all of the CCPP SCM simulations with various physics options using the P8 suite. Adopting the updated Thompson microphysics and progsigma do not alleviate the problems. Sensitivity testing confirms that entrainment rate and cloud condensation nuclei in the shallow cumulus scheme of the P8 suite are not the culprits either. On the other hand, simulations with the RRFS v1beta suite are associated with improved representation of clouds, PBL development, and precipitation across the CONUS shallow cumulus, MAGIC, and DYNAMO cases. Additional in-depth diagnostics demonstrate physics sensitivity to vertical coordinate configuration, another important aspect of physics scale adaptiveness.