Smith, A. J., 2007: Using Large-Eddy Simulations to analyze microphysical behavior in midlevel, mixed-phase clouds. Master's thesis, 68 pp., Univ. of Wisc.---Milwaukee.
A PDF version of my thesis can be downloaded here.
My defense presentation (in PowerPoint 2002 format) can be downloaded here.
Larson, V. E., A. J. Smith, M. J. Falk, K. E. Kotenberg and J.-C. Golaz, 2006: What determines altocumulus lifetime? J. Geophys. Res., 111, D19207, doi:10.1029/2005JD007002.
A PDF version of this paper can be downloaded here.
Smith, A. J., B. M. Griffin, J.-C. Golaz and V. E. Larson, 2006: "Comparison of large-eddy simulations with a single-column model: Implications for mid-level cloud parameterization." Preprints, Twelfth Conference on Cloud Physics, Madison, Wisconsin, American Meteorological Society.
A PDF version of the conference poster can be downloaded here.
The extended manuscript for the poster can be downloaded in PDF format here.
Smith, A. J., B. M. Griffin, J.-C. Golaz, V. E. Larson, 2007: Analyzing a mid-level cloud layer: Comparison of 1D parameterized and large eddy simulations. Presented as a poster at the Center for Geosciences /Atmospheric Research Annual Review, 17-19 April 2007, Fort Collins, CO.
I am currently working with the Naval Research Laboratory's (NRL) Coupled Ocean Atmosphere Mesoscale Prediction System (COAMPS) Large Eddy Simulation (COAMPS-LES) model. As an extension of my undergraduate research, I am now examining the microphysics scheme that is written into the COAMPS-LES code. This scheme is based on Rutledge and Hobbs (1983), and provides a detailed analysis of liquid and ice processes within clouds. We have used this scheme to analyze the phase behavior associated with three mixed-level clouds. After simulating these clouds using COAMPS-LES, we have analyzed the results and developed a series of methods to determine microphysical behaviors. These methods include a series of detailed budgets and a group of analytic equations that predict snow properties using simple parameters.
I also presented two posters, one at the 2006 AMS Cloud Physics Conference, July 10-14, 2006 in Madison, WI, and a second at the 2007 Center for Geosciences / Atmospheric Research Annual Review, 17-19 April 2007 in Fort Collins, CO. The posters detail a comparison between a large-eddy simulation (LES) and a single-column model (SCM). Results for the SCM were qualitatively very similar to the LES, both in a control simulation and in a related sensitivity study. These results demonstrate that our SCM accurately simulates at least one mid-level cloud, in addition to other clouds already examined in Golaz et al. (2002a & 2002b).
From April 2004 to January 2006, I participated in research to understand the behavior of altocumulus by using Large-Eddy Simulations (LES). In a previous study, specific forcings were identified as the major causes of dissipation for a specific altocumulus. However, we wanted to understand these forcings in more detail, especially which forcings are truly the most influential in causing clouds to die.
Using the COAMPS-LES model, we generated a control simulation of a cloud sampled during the fifth of the Complex Layered Cloud Experiments (CLEX-5). From these simulation results, we created a liquid water budget from conserved variables, and used it to determine which forcings have the strongest influence on cloud life. We also developed sensitivity simulations, which involve imposing one perturbation (forcing change) from the control Nov.11 case. From the sensitivity studies, we interpolated how much cloud death would be affected by a specific change in forcings. For example, if we change subsidence (sinking air) velocity by 1 cm/s, cloud death time would decrease by XX minutes. Finally, we developed a feedback matrix, which compares forcing magnitudes between sensitivity studies. From this matrix, we can determine whether some terms resist changes brought on by our imposed perturbations.
Our results have been presented in the paper "What determines altocumulus lifetime?", which is referenced above.