HR: 0800h
AN: A41D-0748    [Abstracts]
TI: Parametric Sensitivity Analysis in a New Cloud Resolving Aerosol Model
AU: * Godinez, H C
EM: hgodinez@pdx.edu
AF: Department of Mathematics and Statistics, Portland State University, P.O. Box 751, Portland, OR 97207, United States
AU: Moulton, D J
EM: moulton@lanl.gov
AF: Los Alamos National Laboratory, Mail Stop B284, Los Alamos, NM 87545, United States
AU: Reisner, J M
EM: reisner@lanl.gov
AF: Los Alamos National Laboratory, Mail Stop D401, Los Alamos, NM 87545, United States
AB: Mathematical models of stratus clouds have become more complex due to an increasing demand for realistic quantitative simulations. Most models employ parameterizations to approximate micro-physical processes that take place within a cloud. These parameterizations contain a number of parameters whose values and impact to the model solution is unknown. This presents a problem for the predictive quality of the numerical simulation of the model. Sensitivity analysis helps identify the most relevant parameters providing valuable insight into which of these should be accurately determined. We present the implementation of parametric sensitivity analysis to a newly developed high-resolution cloud resolving aerosol model for stratus clouds. Identifying the relevant parameters significantly contributes towards the improvement of the current cloud aerosol model. The sensitivity of the model with respect to its various parameters, such as activation time scales and fall speed of water droplets, is computed. Additionally, we examine the sensitivity with respect to the artificial diffusion coefficients that are present in the discretization. The model solution as well as the sensitivities are computed using the SUite of Nonlinear and DIfferential/ALgebraic equation Solvers (SUNDIALS) package.
DE: 0545 Modeling (4255)
DE: 0560 Numerical solutions (4255)
DE: 3311 Clouds and aerosols
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting