Break-out Group on deep and shallow/turbulence convectionFrom Chin-Hoh MoengDate February 11, 2007 Hi, The research part of CMMAP has been re-organized into four new themes with four Break-out Groups. Our Break-out Group, called 'Focus on Deep and Shallow Convection, and Turbulence,' is charged with the task of better understanding and representing the effects of shallow convection and turbulence on deep convection. In the MMF framework, deep convection systems are relatively well resolved, but small convection and turbulence (inside deep convection systems and in the PBL) remain subgrid scale. Understanding the interactions and feedbacks between deep and shallow convection/turbulence is crucial for the development of these SGS representations in a cloud-resolving GCM. We propose to perform a large-eddy simulation that explicitly resolves nearly all convective eddies (from deep convection scale down to large turbulent eddy scale) and use this simulation as a benchmark dataset for the above study. To reach this goal, we ask you to join us and to bring your ideas to the Break-out Group during the Kauai meeting. You are also welcome to email us your ideas before the meeting. Thank you. Break-out Group co-leaders, Steve Krueger and Chin-Hoh Moeng From Chris Bretherton Date February 11, 2007 Dear Chin-hoh, I know of two published papers, both by members of CMMAP, that have addressed the shallow to deep convection problem at LES resolution, and which could provide the benchmark datasets you alluded to in your email: Kuang, Z., and C. S. Bretherton, 2006: A mass flux scheme view of a high-resolution simulation of transition from shallow to deep cumulus convection. J. Atmos. Sci., 63, 1895-1909. Khairoutdinov, M. F., and D. A. Randall, 2006: High-resolution simulation of shallow-to-deep convection transition over land. J. Atmos. Sci., 63, 3421-3436. Chris From Wojciech Grabowski Date February 11, 2007 Chin-Hoh and the Team, Perhaps Chris made this point in his e-mail (I cannot see his html attachment from home): Simulations that you propose to do (i.e., covering the wide range os scales) already exist. They concern, however, only the transition from shallow to deep convection: Chris was involved in such a study ; I performed a costom-designed simulations with gradually increasing horizontal domain to simulate daytime convective development over land based on LBA data; Marat simulated the same case with LES-type resolution and a horizontal domain large enough for deep convection. There is also a large body of work in the NWP arena (e.g., using WRF) that suggests sensible approaches to represent convective development over land in a convection-permitting model. I think we need to have a set of clearly defined scientific problems and design such simulations accordingly. Daytime convective deveopment over land is one issue. Interactions between BL processes, shallow convection, and deep convection in the Tropics is another one. Recent 1-week long japanese simulation using cloud-resolving GCM and realistic land and SSTs (resulting in unrealistic development of tropical cyclones) highlights this very issue. And for the cloud microphysics, even LES simulations have way taoo coarse grid, an aspect that needs to be recognized (and worked on) as well. I look forward to the doscussions at the meetinng... Wojtek. From Steven Ghan Date February 12, 2007 Regarding microphysics, we've been awarded NASA IDS funding to a team that includes Vince Larson. Our plans are to apply Vince's latin hypercube framework to subgrid microphysics in the MMF. We won't be able to get started until we get the funding, probably this summer or maybe next fall, but we hope to contribute to the MMF effort eventually. -Steve Ghan From Chin-Hoh Moeng Date February 12, 2007 Hi, Wojtek: Yes, as you pointed out here, all those previous LESs of convection concern only the transition from shallow to deep convection. For CMMAP, I think how to represent interactions between BL processes, shallow convection and deep convection is a more relevant issue. Japanese cloud-resolving GCM, which reveals strong sensitivity of resolved deep convection to SGS small convection (i.e., shallow convection and PBL turbulence), is a good example that shows the importance of this issue. I think this very basic problem (on how to represent small convection in a model where large convection is resolved) can be better studied from an LES that is in quasi-equilibrium with large-scale forcing. But of course this is open for discussion. Chin-Hoh From Steve Krueger Date February 12, 2007 I will show some LES results from a pilot study of the type that Chin-Hoh is proposing at the meeting. The case is a precipitating trade Cu case based on RICO (but is not the GCSS intercomparison case, which is very weakly precipitating and showed no mesoscale organization in large domain simulations), in a 40 km x 40 km domain with 100 m grid size, run for 24 hours. The last 15 h are in quasi-equilibrium. We also ran the same case using grid sizes of 500 m, 1000 m, 2000 m, and 4000 m. Steve |