A192-04
Exploring Lateral Boundary Forcings on High Resolution Regional Sub-Seasonal Prediction Performance.

Tuesday, 15 December 2020: 05:42
Virtual
Carlos Cruz1, Craig Pelissier1, Zhifeng Yang2 and Milton Halem3, (1)NASA Goddard Space Flight Center, Computational and Information Science & Technology Office, Greenbelt, MD, United States, (2)University of Maryland Baltimore County, Physics Department, Baltimore, MD, United States, (3)University of Maryland Baltimore County, Computer Science, Baltimore, MD, United States
Abstract:
We will present results of a cloud resolving regional subseasonal (32 days) model forecast of precipitation and temperature anomalies over the US by employing the NASA Unified Weather Research and Forecasting (NU-WRF) system and coupling the lateral boundary conditions (LBC) with both MERRA and GEOS-5 provided global boundary forcings. The model is run with one domain with a 4km spatial resolution and 64 vertical layers This combination of LBC prescribes an upper bound on forecast skill and a lower bound depending on the deterioration of the global subseasonal model as the boundary prediction deteriorates.

The NU-WRF framework provides several physics parameterization packages and we have selected the following; a two bin bulk cloud microphysics scheme to parameterize condensation, sublimation, evaporation and sedimentation of liquid and ice (Bacmeister et al., 2006), the GOCART aerosol model (JJ Shi et.al., 2015). For these initial simulations, we are excluding both chemistry and photochemistry parameterizations as well as coupling to the land information system framework because of their intensive demand on the computational resources.

Employing a cloud resolving resolution is important to the hydrologic cycle for resolving the sub-grid scale cloud droplet formation of precipitation parameterization that represents the ice-water phase and mixed-phase processes resulting from the interaction of ice and water particles. It is recommend to use using these mixed-phase process effects for grid scales below 10 km. We therefore expect to see an impact of the cloud resolving microphysics parameterization resulting in potential improvements to the precipitation forecast and in turn produce a larger public benefit of subseasonal forecasts.

The forecasts are being implemented on 1000 processor cores of the NCCS HPC Discover system and require ~2hrs/sim/day. Thus, this exploration of a cloud resolving subseasonal forecast with a GEOS-5 model provided LBC offers a reasonable approach to evaluate the comparison of a Hi-Res forecast with an ensemble of Low-Res spatial forecast and the influence of microphysics on precipitation. In addition, we will archive all the prognostic variables and additionally all the diagnostic variables in the aerosol and microphysics parameterizations.

<< Previous Abstract | Next Abstract >>