Research Catalog: CGD's Climate Modeling

Climate Modeling

Scientific modeling activities

Scientific activities in the Climate Modeling Section have generally been focused on the improvement, analysis and documentation of the CCSM Community Atmosphere Model (CAM), fundamental studies aimed at improving the understanding of key processes in the climate system, and contributing directly to national and international activities focused on advancing climate science by coordinating and conducting broad community initiatives. In addition to their contributions to CAM development, CMS staff play an integral role in the development and improvement of the Whole Atmosphere Community Model (WACCM), a comprehensive model of the atmosphere from the Earth’s surface to about 150 km, which includes interactive chemistry and physical processes throughout the model column.

The standard CAM3 and CCSM3 simulations and contributions to the Fourth Assessment Report (AR4) of the Intergovernmental Panel on Climate Change (IPCC) have been extensively analyzed and documented in the peer reviewed literature by members of the Climate Modeling Staff. In addition to this broad collection of contributions to the detailed analysis of the energy, water, and dynamical aspects of the CCSM’s mean climate and its variability, Collins has served as lead author and expert reviewer of the IPCC AR4 and as a contributing author for the U.S. Climate Change Science Program (CCSP).

CAM simulations

CMS scientists continue to explore innovative ways to evaluate the quality of CAM simulations. One of the more unique approaches recognizes that Numerical Weather Prediction frameworks can provide an excellent method of examining parameterization methods as it allows direct comparison of the parameterized variables (e.g. clouds, precipitation) with observations from field campaigns such as ARM early in the forecast while the forecast state is still near that of the atmosphere. In collaboration with staff members of PCMDI Williamson and Olson have developed the capability to apply the Community Atmosphere Model (CAM) in forecast mode without developing a complete NWP forecast/analysis system. The forecasts are initialized from reanalyses or operational NWP analyses for the atmosphere with the land spun up to be consistent with the atmosphere. They have completed in depth studies of the CAM for several ARM IOPs, concentrating on moisture and cloud aspects of the forecasts. These studies have examined the balance of terms in the moisture and temperature prediction equations during the forecasts at the ARM CART site for different synoptic situations. Although these analyses can not attribute a unique mapping of forecast error to parameterization component deficiencies, they do identify which model components should be further examined to determine the cause of their anomalous behaviors.

CAM development with very simplified surface conditions

In addition to short-term deterministic diagnostic techniques, Williamson and Olson have developed versions of the CAM with very simplified surface conditions, allowing examinations of physical parameterization behavior with both the surface and the large-scale dynamical core. These "aqua-planet" experiments in which the surface is specified to be all ocean with a simple, often zonal, specified sea surface temperature distribution is a useful configuration since the atmosphere retains its full complexity, but eliminated the complexities associated with sea-ice, land, orography, and land-ocean contrasts. This work has led to an internationally coordinated activity known as the Aqua-Planet Experiment (APE) which is being conducted under the auspices of WGNE with collaborators Brian Hoskins and Mike Blackburn (University of Reading), Peter Gleckler (Program for Climate Model Diagnosis and Intercomparison, LLNL) and Richard Neale (NCAR). 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 parameterized physics and dynamics. In addition to this simplified configuration of CAM, Williamson and Olson have been exploring methods for evaluating the quality of dynamical approximations in global models. While most model development groups devise and apply tests during their model development and model documentation, the tests themselves are often not specified in enough detail that another group can completely reproduce the setup or analysis for comparison. To date there is no commonly adopted set of tests with specified metrics for baroclinic cores such as the set commonly used with the shallow water equations. Jablonoski proposed one easy to apply test involving the growth of a specified perturbation in a baroclinicly unstable flow. Jablonoski and Williamson (2006) studied the convergence-with-resolution characteristics four very different dynamical cores, three of which are options in the CAM, and evaluated the uncertainty of the high resolution reference solutions produced by the cores. They continue to develop and evaluate baroclinic core test cases which have potential to become standard approaches for testing new dynamical approximations.

Simulation quality as a function of horizontal and vertical resolution

Another major evaluation focus in CMS is on simulation quality as a function of horizontal and vertical resolution. Hack, Caron, and Truesdale have been examining the quality of the mean climate and its variability characteristics using the spectral dynamical core for horizontal resolutions ranging from T31 through T341. This work has demonstrated clear improvements in the simulated mean dynamical circulation when more a more traditional climate resolution of T42 is doubled to T85 (Hack et al., 2006). Variability metrics, however, are generally unchanged at the higher resolution. Furthermore, continued increases in horizontal resolution exhibit weaker responses in terms of simulation improvement, particularly with regard to the most serious systematic simulation errors. This appears to point to deficiencies in the treatment of parameterized physics as the principal source of these systematic errors. The resolution studies have demonstrated significant differences in the interaction of the dynamical core and the parameterized physics package as a function of horizontal resolution. To better explore these issues, Truesdale has incorporated modifications to the CAM so that a single-column model version of the CAM (known as SCAM) can be more seamlessly exploited. Studies using the CAM and SCAM at multiple resolutions are underway for important climate regimes.

Analysis of the Madden-Julian Oscillation (MJO)

In addition to exploring alternative configurations of the CAM for diagnostic purposes, CMS staff continue to extend standard turn-key and other diagnostic capabilities. Caron has been developing quantitative methods for analyzing the structure of the Madden-Julian Oscillation (MJO) using the eastward propagating MJO-period OLR as an index (as in Wheeler & Kiladis 1999). Caron has also been developing diagnostic techniques for examining the diurnal cycle of warm-season convection in T31, T42, T85, and T170 CAM for the Southern Great Plains as well as other regions around the globe, such as the Tropical Western Pacific Ocean, and Amazonia where data exist for comparison from various field experiments. An important component of this diagnostic work has been a comparison of the individual terms of the warm-season water vapor budget for a point in the Southern Great Plains in CAM simulations with IOP data (from Zhang et al 2001). Caron has also been working in collaboration with Junhong Wang, and Dave Parsons to look at the structure of the diurnal cycle of the Low-Level Jet over the Southern Great Plains compared to results from IHOP.

The role of Polar Mesospheric Clouds (PMCs)

Gettelman has been exploring the role of Polar Mesospheric Clouds (PMCs) that form at the summer mesopause (~80km altitude) when temperatures reach below 150K. These clouds are modulated by temperature and by water vapor concentrations, and may be affected by particles acting as nuclei. The variability of these clouds provides a critical test of the coupling of dynamics and chemistry of the mesopause, and sheds important light on variability due to solar cycles, or due to long term climate changes in the upper atmosphere. Gettelman modified the cloud condensation routines in CAM & WACCM to parameterize Polar Mesospheric Clouds (PMC’s) which can serve as a sensitive diagnostic of model processes in the mesopause region strongly affecting trace gas chemistry. PMC’s are being used as a model diagnostic, as well as for historical climate simulations to look at the sensitivity of PMC’s to climate changes over the last 150 years.

Climate feedbacks

Climate feedbacks in the upper troposphere are some of the largest remaining uncertainties in quantifying the climatic response to anthropogenic forcing. With collaborators at the University of Washington and University of Reading, Gettelman has continued to explore the radiation balance of the Tropical Tropopause Layer (TTL) and the implications for understanding upper tropospheric water vapor feedbacks in observations and in climate models. Current work last year has investigated differences in humidity between observations and models and is continuing to analyze co-variations of clouds, humidity and temperature to try to understand climate sensitivity. This work suggests that humidity in the upper troposphere increases as the surface temperature increases. Significantly, CAM simulations show a similar picture. Much of this work exploits newly available satellite sensors, in particular from NASA EOS satellites (Aqua and Aura). Extensive work with AIRS and data validation has continued this year. Validation of EOS Aura data (MLS and eventually HIRDLS) is progressing and has led to numerous collaborations on the validation side with in situ data: University of Idaho, NASA-JPL, NOAA, Kyoto University, CNRS (France), IAP-Beijing, as well as with several other groups working with the data, including the University of Washington, and Johns Hopkins University.

The study and improvement of atmospheric transport

Another major area of scientific research in the Climate Modeling Section involves the study and improvement of atmospheric transport exploiting observational opportunities associated with tracer transport. Rasch, in collaboration with P. Hess and F. Vitt, has developed an offline transport version of CAM for the purpose of exploring the broad range of transport pathways as well as transport associated with resolved-scale motions. The latter topic is integrally coupled to studies of dynamical cores conducted by other CGD and NCAR investigators (Williamson, Chen, Lauritzen, Nair).

Studies of and improvements to the parameterized treatment of physical processes

Studies of and improvements to the parameterized treatment of physical processes is a broad area of CMS research. Collins and Conley have developed a new mathematical method for approximating transmission in radiative transfer calculations, perhaps the most fundamental quantity for describing the propagation of light and heat through the atmosphere. It has proven quite difficult to represent transmission in numerical models because of the enormous range in the optical opacity depending on the color, or wavelength, of light and the chemical composition of the atmosphere. This work has yielded a new approximation for transmission based upon a novel mathematical analysis, and is designed to overcome the severe mathematical and physical limitations of previous ad hoc techniques. It also yields new, mathematical insights into the issue of “anomalous”, or enhanced, shortwave absorption in clouds and into the origin of relatively transmissive parts of the solar and terrestrial spectra known as atmospheric “windows”. Collins and Conley are now applying this new method to the design of new radiative parameterizations for CCSM. In addition to his IPCC work on radiative forcing of the climate system, Collins collaborated with Dr.Ramaswamy to lead a Radiative Transfer Model Intercomparison Project (RTMIP) (Collins et al, 2006). The objective in RTMIP was to compare the forcings computed by the radiative parameterizations of AOGCMs and against benchmark line-by-line (LBL) codes. Findings of this work show that in order to improve the interpretation of climate change simulations from multimodel ensembles, it will be necessary to collect much more complete information on the forcings applied to each of the AOGCMs than has been customary.

Radiative effects of aerosols

Collins has also been examining the radiative effects of aerosols, primarily focused on the effects of dust. He collaborated with members of CGD’s Terrestrial Sciences Section (TSS) on studies examining how dust has affected the Sahelian drought (Yoshioka, 2005) and how it has affected and might affect the planetary energy budget during the last-glacial maximum, the pre-industrial period, and future warmer climate regimes (Mahowald, 2005). Rasch has also been focused on exploring improvements to Aerosol formulations, including an improved understanding of the atmospheric transformation of anthropogenic aerosol (in collaboration with Tami Bond and Gazala Habib), exploring formulations for the representations of natural and anthropogenic aerosol (in collaboration with Natalie Mahowald and Jean-Francois Lamarque), the role of black carbon emissions from forest fires and convection in the upper troposphere and lower stratosphere (with M. Fromm NRL), and improving the understanding of the climate forcing and response to black carbon by examining aerosol deposition on snow and ice (in collaboration with Charlie Zender, Mark Flanner, and Jim Randerson, UC-Irvine).

Improving the parameterization of moist convection in the CAM

J. Richter has begun working on improving the parameterization of moist convection in the CAM. This work has been focused on the implementation of a parameterization of convective momentum transport. This parameterization improves several of the long-standing model biases, including features in the surface wind field such as (a) the easterly bias in the tropics, b) the westerly bias in the north-eastern pacific, c) and the westerly bias in the 60S jet. Additionally, as a result of added convective momentum transport, the representation of tropical precipitation in CAM is improved in the Indian Ocean and the equatorial Western Pacific. In connection with this work, Richter has begun a collaboration with Chris Bretherton (University of Washington) to evaluate parameterizations of convective momentum transport in the single column CAM (SCAM) against cloud resolving model (CRM) simulations. Current comparisons are being carried out with the Colorado State University’s CRM and will transition to the WRF in the next fiscal year.

Microphysics parameterization for CAM

Gettelman has been leading a major collaborative development effort for a new microphysics parameterization for CAM. The goal of this effort is to develop an advanced microphysics package which can represent the size of cloud drops, and how cloud drops are affected by the distribution of aerosols. The ultimate goal is to quantify aerosol indirect effects in CAM and CCSM. This work dovetails with other studies in ACD, MMM and CGD of cloud microphysics, both in observations and in models. The current project is a collaboration between MMM, CGD, ACD and ASP. An important component of the broader activity is the development of better satellite data sets of cloud microphysical properties for comparing the CAM simulations to observations.

Gravity wave research

Richter, in collaboration with Pennsylvania State University graduate student Alex Hassiotis configured the Weather Research and Forecasting model (WRF) to simulate gravity wave generation near northern Australia during the Darwin Area Wave Experiment (DAWEX). This simulation is quite novel to gravity wave research, as it resolves gravity waves with horizontal scales on the order of tens to thousands of kilometers in one simulation. This work elucidates the importance of surface terrain and the interaction of various wave sources to the gravity wave spectrum propagating into the stratosphere. The modeled gravity wave field is being examined using Fourier and wavelet analysis and is being compared to an existing convective source spectrum parameterization. During the course of this work, it was found that the top boundary condition in WRF is not ideally suited for gravity wave studies as wave reflections occur from the top boundary even in the presence of a damping layer. Richter plans to work with Hassiotis, Joe Klemp (MMM), and Hanli Liu (HAO) to implement a radiative top boundary condition in WRF.

Improving the capabilities and simulation fidelity of WACCM

A substantial component of CMS research is related to improving the capabilities and simulation fidelity of WACCM. J. Richter has been analyzing in detail the momentum budget and the resolved wave spectra of the stratosphere and mesosphere in WACCM3. Such analyses are crucial to validating the processes that are driving the dynamical behavior in WACCM. The extra-tropical stratosphere in WACCM is dominated by momentum deposition from stationary planetary waves and parameterized gravity waves. The tropical stratosphere is forced by a greater variety of wave modes. Recognizing the importance of gravity wave forcing mechanisms, J. Richter developed, and implemented in WACCM, a physically based source spectrum parameterization for convectively generated gravity waves. This parameterization significantly improves the representation of the stratospheric Semi-Annual Oscillation (SAO) and the mesospheric semi-annual oscillation (MSAO) in WACCM. The true impact, however, is probably underestimated since convectively generated waves are underrepresented in WACCM due to the low variability of tropospheric convection. WACCM3 is successful in reproducing the dominant tidal features in the mesosphere/lower thermosphere such as the diurnal and semidiurnal tide. WACCM’s tides are being compared to a simpler model (GSWM) and observations in collaboration with Loren Chang (graduate student at the University of Colorado), Scott Palo (University of Colorado), and Maura Hagan (HAO). WACCM3 does not represent well the two-day wave. This is due to the fact that the stratospheric/mesospheric shear zones are two weak, not allowing for wave growth in baroclinically unstable regions.

WACCM response to changes in the gravity wave parameterization

F. Sassi conducted a set of simulations that describe the WACCM response to changes in the gravity wave parameterization during FY06. These simulations show that the gravity wave processes can have first order effects on the climate of the middle atmosphere, influencing the occurrence of stratospheric warmings, the magnitude of some model biases (cold pole problem), and ultimately affecting the stratosphere-troposphere coupling mechanisms. It is also shown that the inclusion of processes-consistent sources of gravity waves in the troposphere helps for example the simulation of the tropical oscillations. Other major collaborative activities included studies of the effect of solar variability in the middle atmosphere including historical simulations of the past 50 years (in collaboration with R. Garcia and D. Marsh in the Atmospheric Chemistry Division); investigations of dynamical variability in the middle atmosphere (in collaboration with L. Polvani and A. Charlton at Columbia University); continued work comparing WACCM simulation data and lidar observations to better evaluate the gravity wave parameterization in the mesosphere-lower thermosphere region (in collaboration with C-Y She at Colorado State University), and completion of a model inter-comparison that investigates the effects of climate change on the Brewer-Dobson circulation (in collaboration with N. Butchart at the Met Office – UK). Also during FY06, the greenhouse component of the WACCM model was completely redesigned to allow integration of the model without interactive chemistry and with specified heating rates above 60 km. Not only this component is about twice as fast the older version, but it is possible now to run WACCM with boundary fields generated by the fully interactive model to extend studies of the effect of interactive chemistry to the mesosphere and lower thermosphere region.

Model intercomparison activities

CMS Scientists were also involved in a number of model intercomparison activities. These included several GEWEX intercomparisons, including Case 5 of the Deep Convection Working Group that concentrates on the TOGA CORE period, and the Cross Pacific Intercomparison case, which spans the range from deep convection to stratocumulus (Williamson, Olson, Hannay, Kiehl, and Hack). Gettelman was the co-coordinator (w/ V. Eyring, DLR, Germany) of an international project to evaluate coupled chemistry climate models, a project conducted under the auspices of the Stratospheric Processes and their Role in Climate (SPARC) project of the World Climate Research Program (WCRP). These activities included organizing several major international workshops and contributing to ongoing scientific assessment efforts, including several publications feeding into the next WMO ozone assessment. The WACCM group co-authored a major paper on this subject, and I served as co-author of a chapter of the WMO ozone assessment and a reviewer of 2 chapters. CMS Scientists also continue to play leadership roles as members or chairs of many national and international committees including IGAC (Rasch, Chair), CLIVAR PSMIP (Hack, co-chair), AMS Committee on Atmospheric Radiation (Collins, Chair), DOE ASCAC (Hack), and WGNE (Hack).

Additional Information

Print version (.pdf)

CGD Sectional Research

CAS | CCR | CDP | CMS | CSEG | OS | TSS