Sensitivity of summer ensembles of fledgling superparameterized U.S. mesoscale convective systems to cloud resolving model microphysics and grid configuration

Abstract

The sensitivities of simulated mesoscale convective systems (MCSs) in the central U.S. to microphysics and grid configuration are evaluated here in a global climate model (GCM) that also permits global‐scale feedbacks and variability. Since conventional GCMs do not simulate MCSs, studying their sensitivities in a global framework useful for climate change simulations has not previously been possible. To date, MCS sensitivity experiments have relied on controlled cloud resolving model (CRM) studies with limited domains, which avoid internal variability and neglect feedbacks between local convection and larger‐scale dynamics. However, recent work with superparameterized (SP) GCMs has shown that eastward propagating MCS‐like events are captured when embedded CRMs replace convective parameterizations. This study uses a SP version of the Community Atmosphere Model version 5 (SP‐CAM5) to evaluate MCS sensitivities, applying an objective empirical orthogonal function algorithm to identify MCS‐like events, and harmonizing composite storms to account for seasonal and spatial heterogeneity. A five‐summer control simulation is used to assess the magnitude of internal and interannual variability relative to 10 sensitivity experiments with varied CRM parameters, including ice fall speed, one‐moment and two‐moment microphysics, and grid spacing. MCS sensitivities were found to be subtle with respect to internal variability, and indicate that ensembles of over 100 storms may be necessary to detect robust differences in SP‐GCMs. These results emphasize that the properties of MCSs can vary widely across individual events, and improving their representation in global simulations with significant internal variability may require comparison to long (multidecadal) time series of observed events rather than single season field campaigns.

Improvement in the Modeled Representation of North American Monsoon Precipitation Using a Modified Kain–Fritsch Convective Parameterization Scheme

“…For example, the convective environment is much more favorable for MCSs to reach the urban areas of Tucson and Phoenix and westward into the low southwest deserts and tropics, as compared to a more traditional Arakawa-Schubert scheme [37,38]. More pertinent to this work, super-parameterization has demonstrated value added in improving the representation of MCS-driven convective precipitation in the central U.S. [39,40], but there are still uncertainties in the sensitivities of the microphysics formulation and the MCS-like signal they simulate is a 'fledging one. ' The super-parameterization approach is also two to three orders of magnitude more expensive than current climate models [41].…”

Global Microphysical Sensitivity of Superparameterized Precipitation Extremes

“…Two previous studies have investigated the sensitivity of microphysics schemes within super-parameterized models, though with a regional focus on CONUS. While Elliott et al [2016] found no discernible signal in summertime MCSs, Charn et al [2020] found significant differences in extreme precipitation distributions.…”

“…To the best of our knowledge, these two studies are also the only ones to have examined the effects of different parameterizations of microphysics within the SPCAM framework. Elliott et al [2016] investigated summertime mesoscale convective systems (MCSs) within CONUS and found that sensitivities in MCS event counts and in precipitation rates were overshadowed by interannual variability. Charn et al [2020] looked at extreme precipitation within CONUS and found significant differences, mostly when comparing onemoment and two-moment microphysics schemes, as a result of feedbacks onto the largescale circulation.…”

The Ongoing Need for High-Resolution Regional Climate Models: Process Understanding and Stakeholder Information

“…Parameterizations of processes at the smallest scales, such as cloud microphysics, also strongly impact the structure and lifetime of MCSs (Feng et al 2018), but these sensitivities can be small relative to natural variability (Elliott et al 2016), necessitating multiyear simulations. Investigation across all relevant scales, in the context of natural variability, thus presents a computational challenge that regional modeling can approach through technical advances (GPU computing and machine learning), creative solutions for complexity tradeoffs (embedded cloud-resolving models and variable resolution), and explicit large-eddy simulation models.…”