Evaluating and Improving Convective Parameterization for GCMs using ARM Observations

Principal Investigator(s):
Guang Zhang, The Regents of the University of California-SIO

Xiaoliang Song, The Regents of the University of California-SIO

Our ASR-supported research resulted in a comprehensive convective microphysics parameterization scheme that has been implemented in the National Center for Atmospheric Research Community Atmosphere Model version 5 (NCAR CAM5), and a better understanding of what controls convection in the context of convective parameterization. In this project, we propose to further evaluate and improve the convection scheme used in the model in conjunction with other widely used convection triggers and closures in major climate modeling centers in the world using Atmospheric Radiation Measurement (ARM) observations, cloud-resolving model, the NCAR CAM5 and DOE’s Accelerated Climate Model for Energy (ACME). This work will help to improve the simulation of the present climate and the projection of future climate change. It contributes directly to the objectives of the ASR program to use ARM observations to improve DOE-supported global climate models by providing a better, observationally-evaluated convection scheme that can be used across scales. Specifically, we propose to:

  • Improve convection trigger functions with the ARM data for CAM5 and ACME

We developed a methodology to evaluate convective trigger functions using ARM observations. In this work we will aim to improve them. We will use ARM observations from field experiments and long-term forcing data for single column models. For each of the convection regime, we will separate organized, large-scale forced convection from local, thermally driven convection, and examine the performance of the convection trigger functions. The analysis aims to understand the triggering of convection under differing environmental conditions. This knowledge will then be used to improve the existing triggers or devise new triggers, which will be tested with NCAR CAM5 and DOE ACME.

  • Determine and improve the resolution-independence of convective parameterization using observations and cloud-resolving model simulations

High-resolution climate modeling is becoming a trend. As the global climate model (GCM) resolution increases, it is not clear whether current convective parameterization schemes will work adequately, and if not, how to make them suitable for high-resolution GCMs. We will use ARM observations and cloud-resolving model (CRM) simulations to determine the resolution-independence of critical elements of convection schemes for a wide range of GCM resolutions. The NCAR CAM5 at different resolutions (2° and 1°) and DOE ACME at 0.25° will be used to test and evaluate the improved triggers and closures in a global model setting.

  • Improve the representation of vertical velocity in convective updrafts

Convective parameterization for high resolution GCM requires knowledge of vertical motion inside convective updrafts. Ice nucleation in convective updrafts also strongly depends on updraft vertical velocity. In the past, pressure gradient force for calculating updraft speed is parameterized through a “virtual mass” coefficient. We will use a spectral updraft model and incorporate a more accurate formulation of the effect of non-hydrostatic pressure gradient force in the vertical momentum equation. The vertical velocity then will be compared with wind profiler and radar observations from ARM field campaigns.

  • Improve and evaluate convective microphysics parameterization and its interaction with aerosols

In this work, we will further improve the convective microphysics parameterization we developed. We will extend it to multiple updrafts using vertical velocities from the above task. This will directly impact the representation of aerosol-convection interaction in the updraft ensemble. We will examine aerosol-microphysics-convection interaction and test its sensitivity to uncertain parameters to define the range of uncertainties in aerosol effects on convection, and by extension, on climate.