Minutes of the Joint Meeting of the
CCSM Scientific Steering Committee
CCSM Working Group Co-Chairs
Tuesday and Wednesday, 23 and 24 January 2001
NCAR, Boulder, Colorado
Attendees: Maurice Blackmon (Chair), Dave Randall, Byron Boville, Susan Solomon, Bob Dickinson, Peter Gent, Jeff Kiehl, Dick Moritz, Danny McKenna, Ping Chang, Jay Fein, Anjuli Bamzai, Rick Smith, Elizabeth Hunke, Jim Hack, Gordon Bonan, Paul Houser, Bette Otto-Bliesner, Jim Hurrell, Ed Schneider, Scott Doney, Warren Washington, Karl Taylor, Ricky Rood, Cecelia DeLuca, Jim Rosinski, Tony Craig, Rick Anthes, and Lydia Harper
1. Welcome. Maurice Blackmon welcomed the new members of the SSC to the meeting. The new members are Daniel McKenna and Ping Chang.
2. Atmosphere Model Working Group. Jim Hack gave a presentation (copies available upon request) outlining the process used by this group to decide on a recommendation to the SSC for the next version of the atmosphere model for CCSM-2. First the group adopted a standard experimental strategy using 5-year climatological SST forcing to select the top candidates for additional evaluation. Then a 15-year AMIP II simulation was used to select the winning configuration, since a coupled integration to evaluate the coupled model performance was not yet possible.
The group also adopted a standard diagnostic strategy to facilitate a standard analysis of all the candidate models. A standard minimum set of diagnostics was exploited, starting with the WGNE standard climate diagnostics that was enabled with a "turn-key" diagnostic package by Mark Stevens at NCAR. A number of other diagnostic analyses were completed by members of the CCSM community, including an extensive diagnostic analysis from staff at PCMDI. All the candidate model simulations were posted on the AMWG web page and succeeded in entraining the broader community as participants in the analysis and discussions of the configuration options.
The AMWG recommendation consists of a computationally efficient, two-time level Semi-Lagrangian dynamical (SLD) core using a reduced linear grid, 26 vertical levels, prognostic cloud water, generalized cloud overlap, a modern treatment of water vapor absorption in the longwave radiative transfer parameterization, and a version of the Arakawa-Schubert cumulus parameterization developed at Colorado State University. The new cumulus convection parameterization includes the provision for multiple cloud bases and cloud tops, employs an equilibrium cumulus kinetic energy closure, allows for the handling of arbitrary tracer transport, and includes an extendable framework that can be directly coupled to stratiform cloud and boundary layer parameterizations. The simulation exhibits improvements in precipitable water, implied ocean heat transport, polar thermal bias, tropical variability, and the eastern-ocean surface solar energy budget.
There were also some weaknesses noted in the proposed model, including wintertime stationary wave structure; excessive precipitation maxima associated with major precipitation regimes in the eastern and western Pacific; no improvement in the Arctic surface wind field; a cold tropical tropopause, and non-conservative advection of tracers, a deficiency that is similar to CCM3.
The AMWG feels that the proposed atmospheric model provides improvements to the physical and dynamical framework, improvements to the simulation, and optimal computational performance. Perhaps the greatest success of this activity has been the establishment of a "process" for continued model development work that involves the broader national scientific community.
3. Ocean Model Working Group. Rick Smith gave a presentation (copies available upon request) outlining the OMWG recommendation that POP 1.4 become the new ocean component of CCSM-2. POP 1.4 includes the new Jackett and McDougall equation of state, marginal sea balancing, new time manager, netCDF output capability, various minor bug fixes, and a new treatment of the surface layer that exactly conserves tracers, uses natural boundary conditions for freshwater flux, and allows for closed water and salt budgets in the coupled model. The group also recommended the use of the Greenland Pole horizontal grid because it has less distortion, somewhat better resolution in the Kuroshio region, the Hudson's Strait is open, and the 5 years per day target on 32 processors can be achieved. A one-month extension before freezing the model was requested, so that the final tuning of parameterizations can be made.
Peter Gent also reported that the horizontal viscosity still needs further work; the Gent-McWilliams and KPP implementations have been made consistent and G-M now allows different coefficients in advection and diffusion terms; and the river runoff and marginal sea balancing code needs to be tested and incorporated into POP1.4. He said that the new coupler interface was done and needs to be tested. The final x1 version of the ocean model should be ready by the end of February 2001. Copies of his viewgraphs are available upon request.
Future plans are to complete runs with the Visbeck et al. scheme that has been implemented and that has the diffusivity coefficients as functions of space; complete runs with the Beckmann and Doscher scheme that has been implemented to improve the bottom boundary layer; complete runs with the improved partial bottom cells scheme; and do more ocean-alone runs with new boundary conditions under sea-ice before running the x3 ocean model coupled to sea ice to equilibrium.
4. Land Model Working Group. Paul Houser gave a presentation (copies available upon request) on the work that has been done on the new Common Land Model (CLM) by this working group. The CLM will have common data sets and processes; biogeophysics; biogeochemistry; ecosystem dynamics that include leaf phenology, carbon allocation, biomass, composition, and structure (height); and river flow, lakes, wetlands, and irrigation. The CLM is a one-dimensional model that includes one vegetation layer with Bonan's conductance model; 10 uneven soil layers to 3.5 meters depth that include a thin surface layer (1.75 cm) for realistic diurnal cycle and remote sensing comparison and water, ice, and temperatures states in each layer; five snow layers with water flow, refreezing, compacting, and aging allowed; two stream canopy radiative transfer; turbulence considered above, within, and below the canopy; Bonan's lake model; topographic enhanced streamflow; subgrid tile approach; and ALMA land modeling data standards.
Gordon Bonan gave a presentation (copies available upon request) on the merged version of the CLM and the NCAR Land Surface Model (LSM) that has been recommended by this working group as the land component of CCSM-2. The data structures issue was solved by adopting the CLM data structures and processing, and this task has been completed. The variable names and units issue was solved by using a standard set of variable names and units, and this task has been completed in CLM and is in progress in LSM. The constants issue of different names, different values, and different accessing was solved by using the LSM approach, and the CLM work is in progress. The vegetation and soil types issue was solved by using the LSM approach, and the LSM is complete and the CLM is in progress. The albedo issue was solved by using the LSM albedo and radiative transfer approach.
5. Polar Climate Working Group. Elizabeth Hunke gave a presentation (copies available upon request) on development and recent improvements to the sea ice model. Several members of the Polar Climate Working Group (PCWG) have worked closely to update and validate the model for release with CCSM-2. Improvements include merging thermodynamics and ice thickness distribution models from C. Bitz into CICE, the Los Alamos National Laboratory (LANL) sea ice model chosen as the base sea ice model for CCSM-2, and Briegleb's development of an active-ice-only (AIO) framework for testing the model. Upon agreement of the PCWG, Hunke converted the code to F90 and cleaned it up, including implementing a standard convention for ice state variables; debugging restarts, input/output, coupling, flux calculations, timing routines, etc.; and implementing a linear remapping ice thickness distribution scheme contributed by Lipscomb. Elizabeth also significantly improved code performance in response to load-balancing concerns in the coupled configuration. The code is being tested and validated using CCM3 and NCEP forcing, and documentation material is being updated to be consistent with the new ice model.
Dick Moritz reported on the formulation of the sea ice model, which includes enhanced sea ice thermodynamics that resolve vertical temperature profiles, energy conserving vertical boundary conditions, and explicit brine pocket parameterization; multi-category sea ice thickness distribution; plastic ice rheology with an elliptical yield curve that represents shear and normal stresses; uses a generalized orthogonal curvilinear coordinate grid; and runs efficiently on distributed shared-memory platforms. His viewgraphs are available upon request.
6. Software Engineering Working Group. Cecelia DeLuca reported (copies of viewgraphs available upon request) on the progress made on the issues in the Software Engineering Plan, including management issues, coding practices, code infrastructure, and user interface and support. A Software Engineering Manager position will be established. The SEWG is creating a CCSM Code Developer's Guide that will serve as a repository for software conventions and information on configuration management, testing, and validation. Each SEWG meeting is focusing on a specific topic, such as configuration management, software documents, and testing and validation. She also reported that a person will be hired in SCD to work on developing utilities for robust fault tolerance, time management, timing, etc. SCD and CGD in collaboration with the University of Michigan and MIT have submitted a GUI proposal that includes CCSM as a testbed.
Tony Craig discussed the status of plans to integrate all the new component models and the updated coupler. The NCAR Programmer's Group has established testing and validation procedures, version control procedures, and data file management procedures. They have ported the model to the IBM; improved the CPL performance; improved portability, flexibility, and usability of the scripts; tested performance; identified bottlenecks; verified "suitable" performance for production; and updated the coupling interface. Tony reported on the components of the current version of CCSM (1.5.2) and outlined the target components of CCSM-2 (2.0.x). The target is to be able to run the new coupled model 5 years per day on several processors. Once CCSM-2 components are provided, this group hopes to have a working system up and running within a few weeks. A working system means fully coupled, exact restarts, and demonstrating production runs. He reported the new model uses more computational resources, and that some performance testing and validation will need to be done once the final components are settled. His viewgraphs are available upon request.
Jim Rosinski reported on the performance of the atmospheric component of CCSM since it is the most costly. At the time of the Semi-Lagrangian dynamical (SLD) core selection for CCSM, the Lin-Rood option only ran on shared-memory machines, such as NCAR's SGI Origin 2000. Because this machine is slow compared to the IBM SP and Compaq ES40 cluster, the SLD dynamical core was strongly favored from the standpoint of speed with which an integration could proceed. The cost variation between dynamical core candidates was substantially greater than the cost variation between physics options on the table at the time. The cost of input/output (i.e., writing history tapes) is minimal in the default configuration (monthly averages), so that a two-dimensional decomposition has some promise for giving reasonable speedup once it is available. His viewgraphs are available upon request.
7. Paleoclimate Working Group. Bette Otto-Bliesner reported on the progress of this group (copies available upon request) in their use of the CSM, which helped them exponentially to be able to do experiments. They have completed 1500+ years of coupled simulations and have many university collaborators (University of California-Santa Cruz, University of Wisconsin, University of Massachusetts, Yale, University of Toronto, University of Texas, University of Chicago, and University of Southern California). Another collaborative group working with the Paleoclimate Working Group is PMESH (Project for Modeling Earth System History). Bette also reported that they will move to the new version of the model as soon as they can.
8. Climate Change and Assessment Working Group. Warren Washington reported (viewgraphs available upon request) that this group is running experiments to look at greenhouse gases, sulfate aerosols, stratospheric ozone, land surface changes, volcanic forcing, biomass burning, and various energy/emissions use strategies.
He also reported on the merging of the Climate System Model (CSM) and the Parallel Climate Model (PCM), which is DOE funded. The collaborators in this effort are Los Alamos National Laboratory, NCAR, Naval Postgraduate School, Oak Ridge National Laboratory, University of Texas at Austin, Scripps Institution of Oceanography, DOE Program on Climate Model Diagnostics and Intercomparison (PCMDI), U.S. Army Cold Regions Research and Engineering Laboratory, National Energy Research Supercomputer Center, and Lawrence Berkeley National Laboratory.
9. Climate Variability Working Group. Ed Schneider gave a presentation (copies available upon request) about the history of this working group and reported that the working group co-chairs helped in the evaluation of the new atmospheric model for CCSM-2. He also updated the group on the COLA AGCM coupled to MOM3.
Jim Hurrell gave a presentation that talked about the updated CCM3.6 runs where observed sea surface temperatures (SSTs) and sea ice have been prescribed as lower boundary conditions from 1950-1999. There are now 12 member ensembles for both global SST and tropic-only SST forcing. These data have been distributed, at least informally, to members of the community. Marty Hoerling (NOAA/CDC) and Jim have used these runs, for instance, to offer a new paradigm on the origin of the secular change in the North Atlantic Oscillation in recent decades.
Jim reported that he participated in both AMWG meetings involving the selection of the new atmospheric model. His analysis primarily focused on the simulation of precipitation and planetary waves. The new model (Semi-Lagrangian dynamical core plus CSU convection scheme) looks better than CCM3.6 in some respects, but worse in others. Concerning the latter, of particular concern are excessive precipitation rates during boreal summer, a degradation of the planetary wave structure throughout the year, especially during boreal summer over the Atlantic, and a poorer simulation of the SPCZ and SACZ during boreal winter.
10. Biogeochemistry Working Group. Scott Doney gave a presentation (copies available upon request) on this group's Flying Leap series experiments. The Leap 0 experiment is a joint IGBP-WCRP project including radiative CO2 in the atmosphere and co-evolution of the climate and CO2. The leaders of the Leap 0 experiment are Inez Fung and Scott Doney.
The Leap 1 experiment includes the Leap 0 experiment plus dust and enhanced ecosystem dynamics. It will have links among climate and hydrology, land cover, terrestrial sources, climate, marine productivity, and CO2. The leaders for the Leap 1 experiment are still to be determined.
The Leap 2 experiment includes the Leap 1 experiment plus sulfur, nitrogen, and CH4. It will have links among the biosphere, atmospheric chemistry, clouds, climate, etc.
This group overlaps with the CCSM Land Working Group in the areas of terrestrial photosynthesis, allocation, and respiration; coupled carbon, water, and nitrogen cycles; prognostic leaf phenology; and dynamic vegetation. A search for an NCAR terrestrial biogeochemical scientist is underway to more formally continue this work between the CCSM Land and Biogeochemical Working Groups.
This group will meet at the University of California-Berkeley on 29 and 30 March 2001 to develop a plan for a community biogeochemical model.
11. Organization of the Workshop. Some suggestions from the working group co-chairs on this issue were: hold the workshop every 18 months instead of 12 months, alternate the venue between Breckenridge and other Colorado destinations, more science talks rather than routine working group reports, science talks spread out throughout the workshop, invited speakers for the plenary session(s), look at Cheyenne Mountain Resort in Colorado Springs as a venue, have poster sessions instead of working group reports, post working group reports on the web rather than have working group reports at the last plenary session, and each working group take 10 minutes each to present their future plans at the end of the workshop. The action items are (1) Harper will disseminate information about Cheyenne Mountain Resort; (2) there will be a plenary session every morning with invited speakers; and (3) working group co-chairs should tell Harper what working groups are working together, so the working group meetings can be better scheduled so as not to conflict with each other.
12. Working Group Co-Chair Responsibilities: It was decided that a formal list of responsibilities was not needed. Most agreed that the e-mail sent by Blackmon on 11 January 2001 was adequate. Any suggestions to be added to the e-mail should be sent to Blackmon via e-mail. Here is the text of the Blackmon e-mail:
"The co-chairs should:
- be leaders for the various important activities that make up CCSM.
- try to develop a community of scientists interested in participating in the activity of their particular working group.
- invite people to participate by analyzing model data, developing model components, or suggesting projects for the working group to carry out.
- interact with other co-chairs if there are mutual interests between two or more working groups.
- expect to contribute to the writing of the CSL (computer time) proposal once per year or so.
- be thinking of the scientific program of their particular working group and the entire project as well: what interesting science can be done?