J8.4 (Invited Speaker) Challenges in modeling atmospheric aerosol—climate interactions in Earth System Models

Tuesday, 25 January 2011: 2:15 PM
605/610 (Washington State Convention Center)
Philip J. Rasch, PNNL, Richland, WA; and S. Ghan and X. Liu

Aerosols have been recognized to be a potent forcing agent in the climate system for decades. They interact with the climate components in many ways, playing a direct role in scattering and absorbing radiation and an indirect role as Cloud Condensation Nuclei and Ice Nuclei. They are also sites for important photochemistry, and influence ecosystems by modulating light and providing nutrients. I will outline some of the progress that has been made in modeling aerosol cloud interactions, and identify areas where critical challenges remain. I will try to define some critical criteria that should be met in order to be confident of modeling aerosols/cloud interactions sufficiently accurately to be useful.

It will be impossible to be comprehensive in this brief talk, but I will try to pick topics relevant in answering some questions from the following list: How accurate must one be in characterizing natural versus anthropogenic sources? Which sources are most important? What is the relative importance of characterization of mass versus number of aerosol emissions? Does the answer change with a focus only on direct aerosol forcing, or the indirect effect? How accurate must one be in estimating fire emissions? VOCs? Dust? Sea Salt? How much detail is required to treat organics reasonably? Should one expect to get a reasonable emulation of aerosol cloud interactions in parameterizations that do not include specific treatments of aerosol/cloud/dynamical interactions? Since cloud-aerosol interactions also include cloud effects on aerosol, how sensitive is the forcing to the parameterization of clouds? How can one reconcile the very different response of global and LES models to aerosol perturbations? What characteristics should one require to be consistent in order to be confidence in either type of model?

I will try to highlight our perspective on these questions with example model simulations.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner