Atmos. Chem. Phys. Discuss., 12, 14089-14114, 2012
www.atmos-chem-phys-discuss.net/12/14089/2012/
doi:10.5194/acpd-12-14089-2012
© Author(s) 2012. This work is distributed
under the Creative Commons Attribution 3.0 License.
Review Status
This discussion paper has been under review for the journal Atmospheric Chemistry and Physics (ACP). Please refer to the corresponding final paper in ACP.
Mapping the uncertainty in global CCN using emulation
L. A. Lee, K. S. Carslaw, K. J. Pringle, and G. W. Mann
Institute for Climate and Atmospheric Science, School of Earth and Environment, University of Leeds, UK

Abstract. In the last two IPCC assessment reports aerosol radiative forcings have been given the largest uncertainty range of all forcing agents assessed. This forcing range is really a diversity of simulated forcings in different models and an essential step towards reducing it is to quantify and attribute sources of model uncertainty at the process level. Here, we use statistical emulation techniques to quantify uncertainty in simulated concentrations of July-mean cloud condensation nuclei (CCN) from a complex global aerosol microphysics model. Specifically, we use Gaussian process emulation to give a full variance-based sensitivity analysis and quantify, for each model grid box, the uncertainty in simulated CCN that results from 8 uncertain model parameters. We produce global maps of absolute and relative CCN sensitivities to the 8 model parameter ranges and derive probability density functions for simulated CCN. The approach also allows us to include the uncertainty from interactions between these parameters, which cannot be quantified in traditional one-at-a-time sensitivity tests. The key findings from our analysis are that model CCN in polluted regions and the Southern Ocean are mostly only sensitive to uncertainties in emissions parameters but in all other regions CCN uncertainty is driven almost exclusively by uncertainties in parameters for model processes. For example, in marine regions between 30° S and 30° N model CCN uncertainty is driven mainly by parameters associated with cloud-processing of Aitken-sized particles whereas in polar regions uncertainties in scavenging parameters dominate. In these two regions a single parameter dominates but in other regions up to 50% of the variance can be due to interaction effects between different parameters. Our analysis provides direct quantification of the reduction in variance that would result if a parameter could be specified precisely. When extended to all process parameters the approach presented here will therefore provide a clear global picture of how improved knowledge of aerosol processes would translate into reduced model uncertainty.

Citation: Lee, L. A., Carslaw, K. S., Pringle, K. J., and Mann, G. W.: Mapping the uncertainty in global CCN using emulation, Atmos. Chem. Phys. Discuss., 12, 14089-14114, doi:10.5194/acpd-12-14089-2012, 2012.
 
Search ACPD
Discussion Paper
    XML
    Citation
    Final Revised Paper
    Share