Climate Modeling Section Narrative

The mission of the Climate Modeling Section (CMS) is to improve understanding of the global atmosphere and its role in the climate system, through modeling and observational studies, and to represent that understanding in the form of improved numerical models of the atmosphere and broader climate system. The principal modeling tool in CMS is the Community Climate System Model (CCSM) Community Atmospheric Model (CAM). CAM is the global atmospheric model employed in the CCSM, where CMS scientists play a primary role in its ongoing development.

Climate Modeling

Members of CMS (Byron Boville, William Collins, James Hack, Philip Rasch, and David Williamson), in collaboration with colleagues in the university and national laboratory community, contributed broadly to the development of the new Community Atmosphere Model, CAM3. The new model was formally released in June 2004. CAM3 includes a large number of significant improvements and enhancements to the treatment of physical processes, new capabilities such as a Slab Ocean Model (SOM) configuration, and additional improvements to the software engineering implementation of the model. Some of the improvements to the physics include: a substantially revised prognostic cloud water parameterization that includes separate phases for ice and liquid condensate, advection and sedimentation of condensate, and a consistent treatment of condensate in the microphysics and radiative transfer parameterizations. The representation of direct shortwave aerosol forcing is incorporated using an annually repeating aerosol distribution for sulfate, dust, sea salt, and carbonaceous aerosols. Improved representations of shortwave water vapor absorption, longwave absorption, and emission by greenhouse gases are also included. The latent heat of fusion is included in all aspects of the thermodynamics involving the phase transformation of water substances, where the model now conserves energy exactly in all physical parameterizations. The shallow/frontal convection parameterization now interacts more closely with the prognostic cloud parameterization by detraining condensate directly into the stratiform clouds. The new CAM represents a concerted effort by CMS scientists to address several of the more important systematic biases identified in the coupled and uncoupled simulations with CAM2. CAM3 is incorporated as the atmospheric component of CCSM3, which will be used for NCAR's participation in the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report. The new model provides three dynamical cores, including the spectral Eulerian dynamical formulation, which uses a semi-Lagrangian transport scheme, the NASA Data Assimilation Office (DAO) finite-volume dynamical core that was incorporated via a collaboration with Shian-Jiann Lin (NASA Goddard Space Flight Center, GSFC), and a two-time-level, semi-Lagrangian dynamical formulation employing a linear transform grid. For the first time a single-column modeling capability (Single Column Community Atmosphere Model, SCAM) is fully integrated with the evolving CAM physics package and is included in the standard model distribution. This modeling framework also exploits new capabilities in the CAM3 to sample arbitrary columns, or regions of columns, for later exploitation by the SCAM (John Truesdale, CMS). This capability will be essential for the Climate Variability and Predictability program (CLIVAR) Atmospheric Climate Process Team (CPT) work currently underway with the Geophysical Fluid Dynamics Laboratory (GFDL), NASA, and university collaborators. The model continues to be capable of running on a variety of computer architectures popular for climate research applications, and is in the process of being extended to once again exploit re-emerging vector architectures.

Hack, in collaboration with Julie Caron (CMS) and Christine Shields (CCR), has developed climate configurations of the new CAM3 at multiple resolutions. These include T31, T42, T85, and T170 truncations for the spectral dynamical core, where the T85 model will serve as the flagship configuration of the atmospheric model. This will be the highest resolution atmospheric model configuration ever released by NCAR, and leads the way for other global modeling activities. Slab Ocean Model simulations have been completed for the T31, T42, and T85 configurations. Hack continues to work internally and with external collaborators to develop higher resolution configurations of the finite-volume version of the atmospheric model, including 1.0°x1.125° and 0.5°x0.625° configurations of the dynamical core.

Collins has served as the co-chair of the CCSM Scientific Steering Committee, and coordinated the final scientific development of CCSM3 and the preparation of the model for its release in June 2004. A description of the new CCSM3 and some of the initial scientific work has been completed. The new atmospheric component CAM3 of the coupled system is described in the NCAR technical note NCAR/TN 464+STR. Collins is now working on a program plan for the near-term scientific exploitation and extension of the CCSM system, including the development of a 1st generation coupled chemistry-climate model.

In support of the upcoming Fourth Assessment Report (AR4) for IPCC, Collins and V. Ramaswamy (GFDL) have organized a radiative transfer model intercomparison project (RTMIP). This project was commissioned by Susan Solomon, one of the co-chairs for the Working Group 1 volume on “Climate Change 2007: The Physical Science Basis.” Estimates of radiative forcing by well-mixed greenhouse gases obtained from coupled atmosphere-ocean general circulation models (AOGCM) are being compared against benchmark forcing calculations from line-by-line (LBL) codes. The goal is to characterize the accuracy of the codes used in all of the model integrations submitted for the Fourth Assessment. NCAR is contributing calculations using the radiative parameterizations in CCSM3 and using the General line by line radiative transfer model (GENLN) LBL code developed in the NCAR Atmospheric Chemistry Division (ACD). Details on the experimental protocol and overall design of the intercomparison are available from ftp://science.arm.gov/outgoing/IPCC-rt. Preliminary results from the intercomparison were presented at the IPCC Workshop on Climate Sensitivity and indicate a wide disparity in estimates of forcing by CH₄ and N₂O at the tropopause and by CO₂ at the surface. Collins will present the results in invited talks at the 85th meeting of the American Meteorological Society, the 2005 Annual Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Program, and the IPCC workshop on multi-model analysis in May 2005. Collins served as a co-chair for the IPCC Climate Sensitivity Workshop and led a breakout group to further refine estimates of radiative forcing for climate simulations submitted for the AR4.

In addition to the research invested in completing the development, documentation, and application of the CAM3, scientists in CMS have been involved in a wide variety of other research activities related to modeling of the climate system.

Collins and David Fillmore (CMS) have continued work on aerosol assimilation and participated in the international Aerosol Model Intercomparison (AEROCOM) program. Their assimilation using new aerosol retrievals from NASA’s MODIS (Moderate Resolution Imaging Spectroradiometer) has been formally adopted for the NASA CERES (Clouds and the Earth's Radiant Energy System) project. The CERES products include vertical profiles of radiation in the atmosphere and at the surface computed using our assimilation results. Collins gave an invited presentation on the effects of aerosols on the coupled climate system at the International Radiation Symposium in August 2004.

Andrew Gettelman, in collaboration with Steve Massie (ACD), Douglass E. Kinnison (ACD), Cora Randall (University of Colorado, CU), and Michael Mills (CU), has been pursuing several avenues to modify and improve the representation of thin clouds in global models. Most of the efforts center on thin cirrus clouds in the upper troposphere. This work includes multi-scale modeling validated with aircraft and satellite observations from detailed box and process models to global models. First steps this year have included detailed satellite validation and preparation for using large data sets. One aspect includes adding supersaturation to global models starting with the Whole Atmosphere Community Climate Model (WACCM) in an investigation of the impact of water vapor changes on climate and chemistry of the upper troposphere and stratosphere. Another avenue of investigation is understanding model representations of polar mesospheric clouds (PMC's) and developing a climatology of them in WACCM.

This figure shows the distribution of clooud ice mass mixing ratio (or ice water content) in Polar Mesospheric Clouds (PMCs) around the south pole at 86km from the WACCM3 GCM. The plot is a instantaneous snapshot from a model simulation. PMCs form at the cold summer mesopause, and due to a combination of mesospheric cooling due to antropogenic greenhouse gases and increases in water vapor at the mesopause (also due to an anthropogenic greenhous gas- methane), these clouds are sensitive indicators of climate change, and possibly also solar cycle influences on climate.

Hack and Truesdale have completed development of the second generation of the single column model version of the NCAR Community Atmosphere Model, SCAM. The entire software package, including the redesigned Graphical User Interface (GUI), has been fully incorporated in the CAM development library. As such, the SCAM is now routinely distributed as a part of the CAM3, where maintenance responsibility has been transferred to the NCAR CCSM Software Engineering Group (CSEG) and the CCSM Atmospheric Model Working Group (AMWG) liaison.

With help from NCAR initiative funding, Hack and Caron have accelerated the development of a more comprehensive turn-key diagnostic framework to support process studies using the SCAM. The initial focus has been on the examination and improvement of the diurnal representation of warm season convection over the southern central great plains. This work exploited the infrastructure investment that allows for routine high-temporal regional sampling of state and process behavior in the global CAM. This data can also be used to force the single-column version of the respective CAM physics package in order to allow more detailed investigations of process behavior. Initial work documenting the processes involved in regulating moist convection focused on the CAM3 physics package near the ARM Southern Great Plains (SGP) observational facility. CAM3 convection is much more vigorous, is much less persistent throughout the day, and the maximum in convective precipitation rate occurs later in the day. The cloud field is also quite different, with a nighttime maximum in total cloud cover and considerably more lower tropospheric liquid water cloud throughout the day. We also note that the surface forcing is quite different in the new model, where the latent heat flux plays a much greater role in the total surface turbulent heat flux. Experiments to isolate and understand the role of cloud and land surface improvements to the model in altering the behavior of the diurnal cycle have been conducted.

High-resolution configurations of the spectral dynamical core version of CAM3 are extremely expensive to run in terms of computational resources. The principal reason for this is related to the inability to sub-cycle the dynamics calculation, i.e., the physics and dynamics computations are tightly linked to each other, and the dynamics is strongly constrained by Courant-Friedrichs-Lewy (CFL) linear stability requirements. Hack, Williamson, and James Rosinski (CMS) explored ways to decouple the physics and dynamics time steps, as is done in the two-time-level, finite-volume version of the atmosphere. The goal is to identify a suitable strategy for implementing such a decoupling without exciting the computational mode associated with the three-time-level scheme employed by the spectral dynamical core . Hack, Williamson, and Rosinski have also revived the reduced-grid form of the spectral model and are in the process of testing the configuration at multiple resolutions. This configuration of the model will save approximately 25% in computer time. The hope is that this configuration will allow efficient exploration of resolution parameter space up to the equivalent of a T341 spectral truncation if computational resources can be identified.

Work over the last year continued to confirm the results of previous studies, which demonstrated a strong simulation sensitivity to horizontal resolution. Strong sensitivities to horizontal resolution of time-mean features of the simulation are observed, like extratropical cloud forcing, as well as transient behavior and climate sensitivities. Similar sensitivities are also seen to the choice of dynamical cores. Hack, Caron, Truesdale, and Williamson are working with Jeffrey Kiehl (CCR) and Cecile Hannay (CCR) to better understand the reasons for this behavior.

Rasch has been working on the diagnosis and understanding of relatively short time scale phenomena associated with variations in cloud processes. This work, in collaboration with Aiguo Dai (CAS), Lucrezia Ricciardulli (Remote Sensing Systems), Rob Wood (University of Washington), and Patrick Minnis (NASA Langley), has focused on understanding diurnal variability and partitioning between convection and stratiform cloud processes radiation and precipitation properties. Rasch and Ben Johnson (NCAR Advanced Study Program ASP) have also collaborated on the development of a new formulation for cloud fraction based on probability distribution functions for subgrid-scale variability of clouds.

Rasch is also collaborating with Kerry Emanuel (Massachusetts Institute of Technology, MIT) on the exploration of alternate formulations of parameterized convection for climate modeling, and the impact of these alternate formulations on the distribution of tracers. Two schemes in particular, the Emanuel convection scheme and the Kain/Fritsch convection scheme, are now working in the context of both SCAM and CAM.

Rasch, in collaboration with David Mitchell (Desert Research institute) and Paul Lawson (SPEC Incorporated, Boulder), worked on new formulations for ice particle formation, sedimentation, and cloud optical properties. New formulations have been proposed for each of these processes, based on measurements in a number of field campaigns, including bimodal size distributions representative of both tropical and midlatitude cirrus regimes.

Rasch, in collaboration with Steve Ghan (Pacific Northwest Laboratory, PNL), worked on the development of a prognostic cloud drop number density distribution for CAM that extends the formulations currently in use in CAM. These formulations allow a much improved representation for the first and second indirect effect of aerosols on cloud radiative and precipitation properties.

Rasch, in collaboration with Natalie Mahowald (TSS) has been working on improving representations of dust and sea salt. Others involved in this scientific work include Chao Luo (University of California, Santa Barbara) and Tami Bond (University of Illinois), who are focused on carbonaceous aerosol emissions, and their space and time evolution particularly from Asia. A major emphasis of this work is understanding the sensitivity of aerosols to scavenging and the interaction of scavenging and cloud precipitation processes.

Rasch, Boville, and James McCaa (CMS) have been exploring the Finite Volume (FV) version of CAM in coupled and uncoupled simulations with the goal of moving it closer towards becoming the next generation CCSM. The focus has been on exploring the differences in heat and moisture distributions in the FV version of the model, and its impact on the distribution of surface fluxes of heat and moisture. These differences subsequently effect the sea ice distributions in the coupled simulations. Formulations of form and wave drag, along with their impacts on the momentum budget and systematic biases in the simulated stationary wave structure, are also being explored.

Williamson and Jerry Olson (CMS) continue to work on developing the capability to apply CAM3 in forecast mode. The objective is to gain insight into errors in parameterizations by directly comparing parameterized variables, such as clouds and precipitation, early in the forecasts with observations from field campaigns. They are working in collaboration with scientists from the Program for Climate Model Diagnosis and Intercomparison at Lawrence Livermore National Laboratory (PCMDI/LLNL), and the project has been labeled the Climate Change Prediction Program (CCPP)-ARM Parameterization Testbed (CAPT). Their approach is to run CAM in forecast mode using analyses or reanalyses from National Weather Prediction (NWP) centers for recent years. By using analyses from several centers, information becomes available on the sensitivity of the parameterized variables to the initial conditions. Parameterization and the Community Land Model (CLM) initial conditions are obtained by spinning up the land and parameterized atmospheric variables to be consistent with the evolution of the analysed atmospheric state variables. They established that, for the large scales contained in climate models, the forecasts are actually comparable to operational NWP forecasts as demonstrated using skill scores. This is important to demonstrating that the parameterizations under investigation are not being driven by grossly incorrect atmospheric states. Several other in-depth studies of the CAM for several ARM Intensive Observing Periods (IOPs) have been completed and can be found at http://www-pcmdi.llnl.gov/projects/capt/publications.php.

This figure shows the anomaly correlations for forecasts with CAM and with reanalyses models. It illustrates that the CAM forecasts of the large scales are comparable to those from the operational NWP models used to produce the reanalysis. The green and yellow curves are from CAM2 starting from ERA40 and NCEP R2 reanalyses respectively. The blue and red curves are form the ECMWF and NCEP models used to produce the ERA40 and NCEP R2 reanalyses respectively.

Williamson continued to organize a coordinated Aqua-Planet Experiment (APE) under the auspices of the Working Group on Numerical Experimentation (WGNE) with Richard Neale (NOAA Climate Diagnostics Center), Brian Hoskins and Mike Blackburn (University of Reading), and Peter Gleckler (PCMDI/LLNL). The project is intended to provide a benchmark of current model behaviors, and more importantly, to stimulate research to understand the cause of differences arising from different models, different subgrid-scale parameterization suites, different dynamical cores, and different methods of coupling the two. Groups that have announced their intention to participate to date include the UK Meteorological Office, the Australian Commonwealth Scientific and Industrial Research Organization (CSIRO), the Department of Numerical Mathematics, Russian Academy of Sciences (DNM), GFDL, NASA GSFC, the Institute of Atmospheric Physics at the Chinese Academy of Sciences (LASG), Frontier Research System for Global Change, the European Center for Medium Range Weather Forecasts (ECMWF), NCAR, the Centre for Global Atmospheric Modelling, University of Reading (CGAM), the Main Geophysical Observatory, St. Petersburg, Russian Federation (MGO), and the Institut fur Physik der Atmosphere, Universitat Mainz, although not all have run the experiments and submitted data. Details of APE can be found at the APE Homepage. Williamson and Olson have carried out the standard APE experiments with CAM3 and submitted the data to PCMDI. They have also carried out a variety of experiments in addition to the standard APE experiment to determine the affect of resolution, parameterization tuning, dynamical core, and time step on the aqua planet simulations. They are just beginning to examine these experiments, but have already discovered a very intriguing low-frequency wavenumber five mid-latitude phenomenon in the aqua-planet simulations.

Climate Research

Using observations from aircraft and global chemical and climate models, Gettelman, in collaboration with Kinnison, Timothy Dunkerton (Northwest Research Associates, NWRA) and Guy Brasseur (Max Planck Institute for Chemistry, Hamburg), has examined in detail the transport characteristics of the Asian monsoon complex on the upper troposphere and lower stratosphere. This work is continuing by incorporating observations from satellites. He has also proposed specific satellite validation activities surrounding the Asian monsoon in conjunction with colleagues at University of Colorado and in Japan.

Hack and Caron, in collaboration with Kiehl, have applied the Microwave Sounding Unit (MSU) channel 2 temperature retrieval technique to three different global climate models: CSM1, Parallel Climate Model (PCM), and CCSM3, to quantify the extent to which stratospheric temperature trends contaminate tropospheric temperature retrievals. The modeling results confirm earlier hypotheses that the channel 2 MSU retrieval of tropospheric temperature trends is likely to be significantly contaminated by the stratospheric cooling trends, a signature of global warming in response to increasing levels of greenhouse gases. A recent retrieval algorithm proposed by Qiang Fu (University of Washington) does a very good job of eliminating this contamination and accurately capturing the true middle-tropospheric temperature trend. The results suggest that the technique is quite robust, and produces similar ratios of tropospheric warming to surface warming when compared with what has been derived from the observational record by Fu and colleagues.

Climate and Chemistry

Rasch, in collaboration with Mahowald, Williamson, and Lin, has developed a suite of tests in CCSM for understanding the mechanisms that affect trace constituent distributions in the atmosphere. These constituents provide a framework in which changes to CCSM and CAM can be evaluated with respect to relevant problems in chemistry, biogeochemistry, and climate. An ozone-like constituent is used for studying differences in the stratosphere-troposphere exchange. Emissions similar to those of CO₂ from the neutral biosphere provide an important test of the rectification associated with correlations between transport and sources and sinks with strong diurnal and seasonal variation.

Rasch has been working on mechanisms for convective transport of chemical species in collaboration with Mark Lawrence (Max Planck Institute for Chemistry in Mainz). They have shown that two formulations (bulk and ensemble formulations) used to represent convective transport can have a significant effect on subsequent tracer evolution in atmospheric models. These formulations have historically been viewed as equivalent in the meteorological community. This work has shown that the assumptions of equivalence of formulation is not true for many important constituent distributions in the atmosphere, and that it is important to acknowledge and attempt to understand these differences in the picture of atmospheric convection.

Rasch has been collaborating with Peter Hess and Francis Witt of ACD on the development of an offline transport modeling capability for CAM. The intention is to produce a model that can replace the Model of Atmospheric Transport and Chemistry (MATCH) and the Model for OZone And Related chemical Tracers (MOZART) offline transport models, and allow the CCSM to be used for a variety of new chemistry/climate transport problems.

A multi-pronged investigation of the isotopic composition of water in the atmosphere has been extended in the past year. Gettelman, in collaboration with Christopher Webster (NASA-Jet Propulsion Laboratory), David Noone (CU), Andrew Dessler and Sun Wong (University of Maryland), and Mahowald, has extended an analytic model of the tropical tropopause layer to include isotopes of water and compared against observations of isotopes from in-situ aircraft. The agreement is quite good, indicating that the observations can be explained using existing isotopic physics. This group has been working as part of a team with university collaborators to put isotopes in the CCSM. The atmospheric portion of this work is nearly complete. Isotopes will also be integrated into WACCM. As part of this work, Gettelman and Mahowald coordinated a workshop on “Isotopes in the Earth System” at NCAR in January 2004, which brought in 25 international experts on modeling isotopes.

This figure shows the monthly mean Oxygen-18 isotopic composition of precipitation from CAM3 with isotopes (top panel) and from the GNIP observations database for January. The model reproduces the basic features, with more depleted air at higher latitudes, and over continents. This correspondence makes the model useful for paleoclimate studies and comparisons to ice cores.

WACCM Development

Boville and Fabrizio Sassi (CMS) have worked closely with ACD and the High Altitude Observatory (HAO) colleagues toward the implementation of interactive chemistry and of a solar module in WACCM. Implementation of the interactive chemistry includes the in-line calculation of heating rates to 120 nm, which has been led by Kinnison in ACD. This extension to WACCM is now completed, and a study investigating the effect of interactive ozone chemistry in stratosphere has been presented at the 3rd Stratospheric Processes And their Role in Climate (SPARC) General Assembly by Sassi and will soon to be submitted for publication. In collaboration with Raymond Roble (HAO), the solar cycle module has now been tested and will be ready shortly.

A large number of scientific studies using WACCM are now well underway or completed. They include impacts of El Niño/Southern Oscillation ( ENSO) on the middle atmosphere (Sassi, Kinnison, Boville, Rolando Garcia (ACD), and Roble), seasonal variability of trace species in the Upper Tropospheric-Lower Stratospheric region (Mijeong Park (Seoul National University), Kinnison, Boville, Garcia, and W. Choi (Seoul National University)), stratospheric influence on seasonal cycles of tropospheric tracers (Cynthia Nevison (ACD), Kinnison, R. Weiss (Scripps Institution of Oceanography)), monsoon impacts on stratosphere/troposphere exchange (Gettelman, Kinnison, Dunkerton, and Brasseur (MPI-Hamburg)), parameterization of gravity wave fluxes due to convective excitation (Jadwiga Beres (ASP), Boville, Garcia, Sassi), stratospheric trends in temperature and water vapor (Daniel Marsh (ACD), Garcia, Kinnison, and Sassi), chemical transition across the extratropical tropopause (Liwan Pan (ACD), J Wei (Goddard SFC), Kinnison), structure and variability of the SE Asia monsoon system (Park, William Randel (ACD), Kinnison, and Garcia), and the effect of interactive ozone chemistry in simulations of the middle atmosphere (Sassi, Boville, Garcia, and Kinnison).

This figure illustrates the temperature difference between two 20-year simulations of WACCM3. Ozone is fully interactive in the first simulation. A zonal mean and monthly mean ozone climatology is constructed from the interactive simulation. This ozone climatology is used for the radiative heating rate calculations in the second simulation. The temperature difference is for the fully interactive minus the not-interactive ozone simulation. Contour interval is 1 K. Shading indicates the regions where differences are statistically significant at least at the 90% level, based on a t-test with 20 degrees of freedom.

With the help of software engineers Stacy Walter (ACD) and Jeff Lee (CDP) and other members of the WACCM development team, plans to release to the community WACCM version 3 around the beginning of 2005. Work to study the effects of the solar cycle on the middle atmosphere, and to explore improvements to tropical wave variability (in collaboration with Ricciardulli), is also underway with the new version of WACCM.