This page's content is no longer actively maintained, but the material has been kept on-line for historical purposes.
The page may contain broken links or outdated information, and parts may not function in current web browsers.

GACP Projects

GACP 2nd Year Progress Report

NAME: Andrew Ackerman, Co-PI Brian Toon

TITLE: Theoretical studies of microphysical processes in marine boundary-layer clouds

AFFILIATION: NASA Ames Research Center, University of Colorado

ADDRESS:

Andrew Ackerman
NASA Ames Research Center, MS 245-4
Moffett Field, CA 94035-1000

Brian Toon
University of Colorado, Campus Box 392
Boulder, CO 80309-0392

2ND YEAR PROGRESS REPORT:

Our GACP activities for the second year focused on two tasks: (1) our landmark study of the effects of aerosol-induced solar heating on the coverage of trade cumulus clouds, and (2) development of our parallel large-eddy simulation model.

For the first task, we wrapped up our initial analysis of the effects of absorbing aerosols on trade-cumulus cloud coverage and published the results in _Science_. Considering only the effect of soot in the hazes measured over the Indian Ocean during the winter monsoon of 1998 (much less pollution than in 1999), our simulations indicate that the total radiative (solar plus infrared) diurnal average forcing at the top of the atmosphere is enhanced by nearly a factor of 10 when clouds are present (to 3.4 W/m^2) compared to the clear sky forcing of soot (0.4 W/m^2). The magnitude of the forcing due to cloud-burning by soot is comparable and opposed to the conventional indirect forcing due to doubling the cloud droplet concentration.

The publication of this result in _Science_ generated a good bit of media attention, with Ackerman giving interviews aired on National Public Radio and the Canadian Broadcast Corporation, as well as coverage in the Washington Post, CNN.com, and a number of electronic environmental news outlets. A reprint of the article can be downloaded from http://sky.arc.nasa.gov/~ack/cloudburning.

For the second task, we continued development of the parallel version of our large-eddy simulation (LES) model. Previously we had incorporated our explicit aerosol and cloud microphysics model and our detailed radiative transfer model into a serial version of the LES, and had run a large number of simulations for our investigation of the breakup and collapse of marine stratocumulus at low aerosol concentrations (our 1st Year GACP Significant Highlight). We also have written documentation describing the fully coupled model. However, we postponed completion of that manuscript (describing the model and presenting those simulations) in deference to the urgency of the first task (described above) and have opted to finish the development of our parallel model (no differences in the underlying numerics) before completing that manuscript.

The second task involved two sub-tasks:
(a) design, purchase, and implementation of a "Beowulf" cluster (of commodity personal computers running the Linux operating system and operating in parallel), and
(b) software development of dharma (Distributed Hydrodynamic Aerosol and Radiative Modeling Application), our parallel LES.

For (a), we first developed a prototype system using three existing dual-processor PCs linked through a dedicated hub, on which we installed MPI, the Message Passing Interface that has become the standard for massively-parallel scientific computing. We then ported dharma (the parallel LES written by Dave Stevens of Lawrence Livermore National Laboratory) to that prototype system. After extensive testing, we then created an 8-node cluster of dedicated single-processor PCs. The dedicated system includes a PBS (Portable Batch System) queueing front end, as found in typical production environments.

We have also made progress on software development of dharma. To test that the new parallel model reproduces the results of the serial model (within the noise of turbulence) we incorporated the surface and radiative forcings used for our previous trade cumulus simulations, and found general agreement. We have also merged our detailed radiative transfer model into dharma. Test runs show that the simulation timing speeds up linearly with the number of processors, which is the theoretical maximum efficiency. We have also added an arbitrarily long vector of advected species to dharma as a first step towards merging in our explicit microphysics model.

3RD YEAR STATEMENT OF WORK:

Our proposed tasks for the 3rd year fall into two categories: (1) model development, and (2) model simulations and analysis.

For (1) we will finish the merger of our explicit aerosol and cloud microphysics model with dharma, the massively-parallel version of our LES model.

For (2), the first application of the completed model will be a brief return to the collapse and breakup of marine stratocumulus at low aerosol concentrations, which will provide a framework for documenting the model in the scientific literature and describing that interesting result.

We also plan to use the fully coupled model to investigate the reverse problem: formation of deep cloud-lines (Type 1 ship tracks in the taxonomy of Ackerman et al., JGR, 1995) due to the injection of aerosols, as observed during the Monterey Area Ship Tracks experiment and documented by Taylor and Ackerman (QJRMS, 2000).

We will also use dharma to further investigate the dependence of trade- cumulus cloud coverage on aerosol-induced solar heating. For this follow up, instead of using parameterized precipitation as in our original study, we will use explicit aerosol and cloud microphysics. We will investigate the sensitivity of the cloud-burning by soot to different environmental regimes of trade-cumulus, such as a further downstream case (in the tropical circulation) with lower cloud fractions (e.g., BOMEX), a more upstream case with higher cloud fractions (e.g., ASTEX), and a representative case for the Indian Ocean during the winter monsoon.

We also plan to use the fully coupled dharma to investigate parameterizations of cloud-fraction used in larger-scale models (such as general circulation models), as well as evaluating precipitation parameterizations used in cloud-resolving models.

In collaboration with graduate student Kari Klein we have also recently begun coupling with dharma a chemical model of aerosol nucleation in the marine atmosphere, and we will continue that project during the 3rd year of our GACP work.

BIBLIOGRAPHY:

A. Publications

Ackerman, A. S., O. B. Toon, D. E. Stevens, A. J. Heymsfield, V. Ramanathan, E. J. Welton, Reduction of tropical cloudiness by soot, Science, 288, 1042-1047, 2000.

Ackerman, A. S., O. B. Toon, J. P. Taylor, D. W. Johnson, P. V. Hobbs, and R. J. Ferek, 2000: Effects of aerosols on cloud albedo: Evaluation of Twomey's parameterization of cloud susceptibility using measurements of ship tracks, J. Atmos. Sci., 57, 2684-2695.

Ferek, R. J., T. Garrett, P. V. Hobbs, S. Strader, K. Nielsen, D. Johnson, J. P. Taylor, A. S. Ackerman, Y. Kogan, Q. Liu, B. A. Albrecht, and D. Babb, 2000: Drizzle suppression in ship tracks, J. Atmos. Sci., 57, 2707-2728.

Taylor, J. P., and A. S. Ackerman: A case study of pronounced perturbations to cloud properties and boundary layer dynamics due to aerosol emissions, Q. J. R. Meteorol. Soc., 125, 2643-2662, 1999

B. Presentations at professional society meetings and conferences

Ackerman, A. S., O. B. Toon, and D. E. Stevens, Reduction of trade-cumulus cloud cover due to solar heating by dark haze, AGU Fall Meeting (San Francisco, California), 1999.

Ackerman, A. S., O. B. Toon, D. E. Stevens, A. J. Heymsfield, V. Ramanathan, E. J. Welton, Reduction of tropical cloudiness by soot, 13th International Conference on Clouds and Precipitation (Reno, Nevada), 2000.

Back to Individual Projects page