Repository logo
 

Development of a third-order closure turbulence model with subgrid-scale condensation

Date

2009

Authors

Firl, Grant J. (Grant Jacob), author
Randall, David A. (David Allan), 1948-, advisor
Estep, Donald J., 1959-, committee member
Denning, A. Scott, committee member

Journal Title

Journal ISSN

Volume Title

Abstract

Boundary layer clouds play an important role in the Earth's climate system due to their local effect on the radiation budget and their expansive geographic extent. Their poor representation within many general circulation models provides motivation for the development of a turbulence parameterization capable of better simulating these clouds and their effect on the turbulent structure of the boundary layer. With this goal in mind, this study presents the development and testing of such a model. Since most boundary layer clouds result from convective processes, it is important that the new turbulence model be able to accurately predict the growth of the boundary layer in a convective regime. Previous studies have shown that third-order closure models provide sufficient detail for realistic boundary layer growth in such a regime. For this reason, the development of the new turbulence model is based on this level of detail. The current model predicts the evolution of 10 second-order moments and diagnoses the values of 28 third-order moments. Further, a subgrid-scale condensation scheme is utilized to diagnose the cloud fraction and liquid water content. This scheme also allows the diagnosed cloud cover to interact with the turbulence variables by modifying the buoyancy production terms in their predictive equations. Finally, the diagnosed cloud cover participates in warm rain processes and a novel procedure is used to account for rain falling through partial cloudiness on the subgrid-scale. The new turbulence model is tested as both a single column model and as a turbulence parameterization within a host three-dimensional mesoscale model. For the single column model, five cases are simulated in order to test the model's ability in different boundary layer regimes: a clear convective case, a stratocumulus-like smoke cloud case, a nocturnal drizzling stratocumulus case, a non-precipitating trade-wind cumulus case with low cloud fraction, and a precipitating trade-wind cumulus case. It is demonstrated that the new model simulates all regimes satisfactorily as the mean and turbulent states of the simulated boundary layer are compared with results from large-eddy simulation intercomparison studies. Finally, the new turbulence model is used as a parameterization within a three-dimensional model and two of the previous cases are run. The results are compared to those from both large-eddy simulation intercomparison studies and the host model run with its standard parameterizations. It is shown that the modified version of the three-dimensional model improves upon the results from the same model with its standard parameterization suite.

Description

Department Head: Richard Harlan Johnson.
2009 Fall.
Includes bibliographical references (pages 218-224).

Rights Access

Subject

Citation

Associated Publications