Explicit Prediction of Continental Convection in a Skillful Variable‐Resolution Global Model

Abstract

We present a new global‐to‐regional model, cfvGFS, able to explicitly (without parameterization) represent convection over part of the Earth. This model couples the Geophysical Fluid Dynamics Laboratory Finite‐Volume Cubed‐Sphere Dynamical Core (FV3) to the Global Forecast System physics and initial conditions, augmented with a six‐category microphysics and a modified planetary boundary layer scheme. We examine the characteristics of cfvGFS on a 3‐km continental U. S. domain nested within a 13‐km global model. The nested cfvGFS still has good hemispheric skill comparable to or better than the operational Global Forecast System, while supercell thunderstorms, squall lines, and derechos are explicitly represented over the refined region. In particular, cfvGFS has excellent representations of fine‐scale updraft helicity fields, an important proxy for severe weather forecasting. Precipitation biases are found to be smaller than in uniform‐resolution global models and competitive with operational regional models; the 3‐km domain also improves upon the global models in 2‐m temperature and humidity skill. We discuss further development of cfvGFS and the prospects for a unified global‐to‐regional prediction system.

GFDL SHiELD: A Unified System for Weather-to-Seasonal Prediction

“…The fundamental characteristics of SHiELD compared to previous-generation and existing operational models are documented in this and other publications. For some applications, we have previously demonstrated capabilities similar to that of existing modeling systems, such as severe-storm prediction in C-SHiELD (Harris et al, 2019) and tropical cyclone intensity prediction in T-SHiELD (Hazelton, Bender, et al, 2018;. We have shown significant improvements over existing models, especially over existing global models, for large-scale and hurricane prediction skill in 13-km SHiELD (Chen, Lin, Magnusson, et al, 2019;Zhou et al, 2019), and the diurnal cycle and MJO prediction in S-SHiELD.…”

“…The UH threshold giving the best score for C-SHiELD is higher than in many other convective-scale models due to the significantly higher updraft helicities in FV3-based models (Potvin et al, 2019). This in turn is likely due to the emphasis on vorticity in the horizontal discretization as described in Harris et al (2019).…”

“…This "cold starting" from the hydrostatic, spectral GFS could potentially leave the convective-scale configurations (T-SHiELD and C-SHiELD) at a comparative disadvantage to models with native, specialized convective-scale data assimilation. This issue is minimized here due to the ability of FV3-based models to "spin up" their convective scales within a few hours of initialization and experience little degradation thereafter (Harris et al, 2019;Hazelton, Bender, et al, 2018;Marchok et al, 2018;Zhang et al, 2019).…”

“…Global atmosphere models lack the lateral boundary errors that contaminate the solutions of regional models after a few days of simulation. They thus allow us to extend mesoscale and storm-scale predictions into the medium range and beyond (Harris & Lin, 2013;Harris et al, , 2019Zhou et al, 2019). Global modeling also brings many new challenges-one cannot "throw your garbage in the neighbor's yard" in global modeling, so to speak.…”

GFDL SHiELD: A Unified System for Weather‐to‐Seasonal Prediction

“…The fundamental characteristics of SHiELD compared to previous‐generation and existing operational models are documented in this and other publications. For some applications, we have previously demonstrated capabilities similar to that of existing modeling systems, such as severe‐storm prediction in C‐SHiELD (Harris et al, 2019) and tropical cyclone intensity prediction in T‐SHiELD (Hazelton, Bender, et al, 2018; Hazelton, Harris, & Lin, 2018). We have shown significant improvements over existing models, especially over existing global models, for large‐scale and hurricane prediction skill in 13‐km SHiELD (Chen, Lin, Magnusson, et al, 2019; Zhou et al, 2019), and the diurnal cycle and MJO prediction in S‐SHiELD.…”