Understanding the Relative Contributions of Aerosol Properties to Cloud Droplet Number: Adjoint Sensitivity Intercomparison of Activation Parameterizations in a Global Circulation Model
Abstract
The complex interaction between atmospheric aerosols and clouds, and their impact on the planetary radiation budget is a major contributor to the uncertainty in climate predictions and to estimates of climate change. Key to the representation of aerosol-cloud interactions in Atmospheric Global Circulation Models (A-GCMs) is the activation of cloud droplets from an underlying aerosol population. Physically based cloud droplet number activation parameterizations are currently used to represent this process in GCMs. However, the internal complexity of GCMs in itself as well as the many different parameterizations employed by different models to perform aerosol indirect effect estimates further obscure the apportionment of uncertainty. Adjoint-sensitivity analysis has emerged as a new tool that can expedite our efforts to identify the major sources of uncertainty in the aerosol-cloud interaction problem, and to determine which set of parameters carry the largest sensitivity in cloud droplet number estimates (and therefore in aerosol indirect effect estimates). In this work, we make use of the newly developed adjoints of a number of physically based droplet activation parameterizations [Abdul-Razzak and Ghan (2000), Fountoukis and Nenes (2005), Kumar et al (2009)]. Aerosol input fields from the Global Modeling Initiative (GMI) chemical transport model were used to drive off-line comparisons between the sensitivities of the droplet number predicted by these parameterizations against predictions with detailed numerical simulations of the activation process to identify process-level discrepancies between the activation schemes in the absence of feedbacks. Important differences were observed particularly in their response to number concentration and hygroscopicity of fine and coarse mode aerosols. Overall, the parameterizations were capable of capturing the response of droplet concentrations to perturbations in aerosol number concentration, and updraft velocity better than to chemical composition variations. The adjoint sensitivities predicted by these activation parameterizations were then implemented in the framework of the Community Atmospheric Model version 5 GCM, which includes interactive aerosols and detailed cloud microphysics. These simulations exhibit clear patterns of high-sensitivity areas to aerosol number concentration coinciding with relatively clean environments, as well as continental and relatively polluted areas showing saturation with respect to aerosol load changes.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.A21L..08M
- Keywords:
-
- 0305 ATMOSPHERIC COMPOSITION AND STRUCTURE / Aerosols and particles;
- 0321 ATMOSPHERIC COMPOSITION AND STRUCTURE / Cloud/radiation interaction;
- 1626 GLOBAL CHANGE / Global climate models;
- 3311 ATMOSPHERIC PROCESSES / Clouds and aerosols