Enhancing Convection Parameterization for Next Generation E3SM

Motivations: Convective parameterization is one of the major factors responsible for biases in simulations of present-day climate and uncertainty in future climate projection using global climate models (GCM). As GCM resolution increases, the stochastic behavior of convection becomes important and many important assumptions in conventional parameterization will break down. Therefore, a stochastic convective parameterization along with scale-awareness is highly desirable for high resolution GCMs such as E3SM.  Cloud microphysical processes are at the heart of aerosol, cloud and climate interaction. The representation of microphysical processes in convection is another critical component missing or poorly treated in convective parameterization.

Objectives: The goal of this project is to improve and understand the climate simulation and reduce model biases in precipitation in the E3SM by improving the representation of convective processes. Towards this goal, we have the following specific scientific tasks to accomplish:

  1. We will first transplant a stochastic convective parameterization that we have implemented in the NCAR CAM5 to E3SM to improve the simulation of precipitation and its variability. Since the bias of “too much drizzle and too little heavy precipitation” in the E3SM is inherited from CAM5, and this bias has been remarkably alleviated in the CAM5 with stochastic convective parameterization, we are hopeful that the implementation of the stochastic convective scheme will reduce the precipitation biases in the E3SM. In addition, we will investigate sensitivity of the simulated precipitation and its variability to tunable parameters and resolution dependence of the effect of stochastic convective parameterization.
  2. We will implement the two-moment, four-species (cloud drops, cloud ice, rain and snow) microphysics parameterization for convective updrafts developed by the PIs into the E3SM model and then further improve it. We will develop a two-moment microphysics parameterization for downdrafts to better represent the evaporation of rain and melting of snow and will parameterize the impact of microphysics on cumulus dynamics, which is currently missing in most convection schemes used in GCMs including E3SM. We will represent the impact of drag forces of settling hydrometeors on buoyancy, the impact of ice-phase latent heating on the strength and depth of convective updrafts, and the impact of evaporative/melting cooling on the strength of convective downdrafts. The parametric uncertainty of convective microphysics scheme in E3SM will be systematically evaluated. This work will help us understand the impact of microphysics scheme on the simulation of precipitation and climate in E3SM. Inclusion of convective microphysics in convection schemes makes it possible to study the interaction of aerosol and convection.
  3. We will improve the scale-awareness of major components of existing convection scheme, including convection trigger, closure and horizontal transport by convection, based on the scale-awareness analyses of cloud-resolving model (CRM) simulation averaged over various sizes equivalent of different GCM resolutions. The improved scheme will be evaluated in E3SM to understand what effects the scale-awareness of convective scheme has on ITCZ precipitation and cloud simulation, including spatial distribution, frequency and intensity of convection, and partitioning between convective and grid-scale precipitation.

Potential impacts: The proposed work will improve the simulation of precipitation distribution and statistics and improve the scale-awareness of convective parameterization for use at higher resolutions in E3SM. It will also improve the representation of microphysical processes in convection and facilitate the investigation of aerosol-convection interaction.

This project is in progress.

Project Term: 
2018 to 2021
Project Type: 
University Funded Research

Publications:

None Available

Research Highlights:

None Available