AAAR 34th Annual Conference
October 12 - October 16, 2015
Hyatt Regency
Minneapolis, Minnesota, USA
Abstract View
The Effects of Model Spatial Resolution on Cloud Condensation Nuclei and Ultrafine Number Concentrations Simulated in a Global Model
MARGUERITE COLASURDO MARKS, Peter Adams, Carnegie Mellon University
Abstract Number: 525 Working Group: Aerosols, Clouds, and Climate
Abstract The aerosol indirect effect is considered the largest source of forcing uncertainty in current climate models. Assessing levels of the atmospheric particles responsible for this effect (cloud condensation nuclei, or CCN) requires knowledge of aerosol levels and their global distribution, size distributions, and composition.
A key tool for advancing our understanding of CCN is the use of global aerosol microphysical models. These models seek to simulate the physical processes that control aerosol size distributions: nucleation, condensation/evaporation, and coagulation. Previous studies have found important differences in CO (Chen, D. et al., 2009) and ozone (Jang, J., 1995) modeled at different spatial resolutions, and it is reasonable to believe that short-lived, highly localized aerosol species will be similarly – or more – susceptible to model resolution effects. Therefore, detailed evaluation of these models against observations will depend on achieving sufficiently high spatial resolution in the model to allow meaningful comparisons.
The goal of this modeling study is to determine how the predicted number concentrations of CCN and ultrafine particles are affected by model spatial resolution. We examine both the enhanced spatial and temporal variability captured with higher model resolutions and how this affects comparisons against observations. Simulations were performed using the global chemical transport model GEOS-Chem (v9-02). The years of 2008 and 2009 were simulated at 4° x 5° and 2° x 2.5° globally, as well as at 0.5° x 0.667° over both Europe and North America. Results were evaluated against surface-based particle size distribution measurements from the European Supersites for Atmospheric Aerosol Research project.
Results suggest that the coarse model simulations predict systematically lower CCN levels than do the fine-resolution simulations. The fine-resolution model simulates more spatial and temporal variability in CCN and ultrafine concentrations, better resolving topographic features and demonstrating time variability much more consistent with observations. As a result, significant differences are evident with respect to model-measurement comparisons.