Climate and Global Dynamics Division
Climate Modeling Section
CGD ASR00 Home | ASR00 Home | CGD Home
The Climate Modeling Section's (CMS) research goal is to increase the understanding of the atmosphere and its role in the climate system through modeling and observational studies and to represent that understanding in atmospheric models. This involves studies of the physical mechanisms governing the global climate system and the numerical modeling techniques required to represent these mechanisms. CMS scientists compare observed atmospheric data with model output data to validate and to improve the models. They have also been principally responsible for developing and testing new versions of the Community Climate Model (CCM), including the Land Surface Model (LSM). These models are the atmosphere and land components of the Community Climate System Model (CCSM). The next generation models, known as CCSM2-Atmosphere (CCSM2-A) and CCSM2-Land (CCSM2-L), are being jointly developed by the CCSM Atmosphere Model Working Group (AMWG) and Land Model Working Group (LMWG), respectively. A CMS scientist serves as co-chair of each group (James Hack, AMGW and Gordon Bonan, LMWG). All of the other CMS scientists are active participants in the AMWG.
CCSM and CCM Development
Within CMS, development work on CCSM2-A is focused on improvements to the deep convection parameterization, alternative dynamical cores, prognostic cloud water and sulfate aerosol schemes, and interactions between clouds and radiation. Improvements to the CCM3 are being carried out in collaboration with other members of the AMWG including scientists from several universities, the Data Assimilation Office (DAO) at NASA Goddard Space Flight Center (GSFC), Oak Ridge National Laboratory (ORNL), Argonne National Laboratory (ANL), and Lawrence Livermore National Laboratory (LLNL).
As co-chair of the AMWG, James Hack has invested a substantial amount of time in coordinating and evaluating the implementation of several modifications to the physical parameterizations in a 30-level prototype version of the CCSM2-A, including longwave and shortwave radiation, cloud overlap, and prognostic cloud water. Part of this effort has included the development of a web-based diagnostic analysis capability by Mark Stevens (CMS). These efforts have helped to ensure that changes to the prototype model will be evaluated by the community in a comprehensive, systematic, and timely way.
William Collins (CMS) and John Truesdale (CMS) collaborated on a new, general treatment of vertical cloud overlap in radiative calculations performed by the CCM3. The new parameterizations compute the shortwave and longwave fluxes and heating rates for random overlap, maximum overlap, or an arbitrary combination of maximum and random overlap. The introduction of the generalized overlap assumptions permits more realistic treatments of cloud-radiative interactions. The parameterizations are based upon representations of the radiative transfer equations that are more accurate than previous approximations. These techniques increase the computational cost of the radiative calculations by approximately 30%.
The methodology has been designed and validated against calculations based upon the independent pixel approximation (IPA). The hourly radiative fluxes and heating rates from the parameterizations and IPA have been compared for a 1-year integration of CCM3. The mean and root mean square (RMS) errors in the hourly longwave top of atmosphere (TOA) fluxes are -0.006 +/- 0.066 W/m2, and the corresponding errors in the shortwave TOA fluxes are -0.20 +/- 1.58 W/m2. Heating rate errors are O(10-3) K/day. In switching from random to maximum/random overlap, the largest changes in TOA shortwave fluxes occur over tropical continental areas, and the largest changes in TOA longwave fluxes occur in tropical convective regions. The effects on global climate are determined largely by the instantaneous changes in the fluxes rather than feedbacks related to cloud overlap.
The new parameterizations are one of several improvements in the physical parameterizations planned for CCSM2-A. The treatment of infrared absorption by water vapor may also be improved by updating the absorptivity-emissivity formulation using line-by-line calculations. Collins, Jeremy Hackney (CMS), and Dave Edwards (Atmospheric Chemistry Division ACD) are using Edwards' General Line-by-Line Atmospheric Transmittance and Radiance Model (GENLN) line-by-line radiative transfer code to develop a new parameterization for water-vapor absorption.
Hack has continued work on understanding the interaction of convection and atmospheric boundary layer processes as simulated in the global model. Single-column model simulations have been central to this activity. Equilibrium configurations of the model have proven to be of great utility for work which addresses the effects of anthropogenic absorbing aerosol on cloud formation. These configurations have also revealed interactions between the convection and atmospheric boundary layer parameterizations that can lead to multiple equilibrium states when subjected to a steady-state forcing. Work continues on a more comprehensive characterization of the strengths and weaknesses of single-column modeling frameworks. Work to date along these lines clearly demonstrates that there are significant limitations for certain experimental configurations due to non-linearities in parameterized physics packages. One goal of this research will be to identify optimal experimental configurations for robustly identifying the sources of system bias in physical parameterization packages.
Hack has been involved in the detailed analysis of prototype configurations of the CCM containing higher vertical resolution, modified convection schemes, alternate longwave radiative transfer schemes, and generalizations to cloud overlap. He has also implemented a version of the Relaxed Arakawa-Schubert parameterization scheme in both the CCM3 and prototype CCSM2-A models. Simulations exhibit similar systematic errors including a highly-zonal South Pacific Convergence Zone (SPCZ) during December, January, and February (DJF); the tendency to move rainfall away from the equator by producing double Intertropical Convergence Zone (ITCZ)-like rainfall distribution patterns; the anomalous rainfall rates over tropical land regions; and the anomalous and excessive precipitation over southern and southeast Asia during the summer monsoon.
A revised closure on the Zhang-McFarlane cumulus parameterization scheme has also been implemented and tested in the CCM. This closure was developed using the CCM Single Column Model (SCM) in collaboration with Minghua Zhang (State University of New York (SUNY), Stony Brook) and Shaocheng Xie (LLNL). Although the revised formulation yields reductions in certain systematic errors exhibited by the CCM, additional investigation suggests that the current implementation has some undesirable properties with regard to the detailed transient behavior of parameterized convection. We have also determined that the closure changes have had almost no effect on low-frequency behavior of the model. Furthermore, the undesirable characteristics of the basic scheme appear to be exacerbated with higher vertical resolution.
Hack has also begun exploring another Arakawa-Schubert treatment of convection. The main emphasis for this implementation is twofold: to give shallow convection a larger role in the reduction of boundary-layer based Convective Available Potential Energy (CAPE), which will be dealt with using a linear optimization approach, and to come up with a computational approach that allows the closure to be implemented as a function of the change in CAPE as opposed to a relaxation approach toward climatological values (a common implementation of the quasi-equilibrium closure).
Philip Rasch (CMS) was also involved in the integration of convection and cloud water parameterizations in the CCSM model framework. He worked with S.J. Lin and colleagues of the NASA-DAO group in the integration and tuning of the CCSM with NASA-DAO, dynamical module. He also modified the Zhang-McFarlane convective parameterizations to incorporate a "trigger function that suppresses convection when parcels entering cloud base are negatively buoyant." Simulations were performed in versions of the CCSM with and without an interactive ocean. Rasch and Charlotte Demott (Colorado State University) have analyzed the impact of the modifications in the CCSM and demonstrated that the modification has a significant and beneficial impact on timescales of transient features ranging from the diurnal to the seasonal timescales in both coupled and uncoupled versions of the CCSM.
Bonan continued his study of the ecological and hydrological processes by which natural and human-mediated changes in vegetated landscapes affect climate. Previous modeling studies by Bonan showed that temperate deforestation cools the climate of the U.S. These studies were followed by analyses of observed temperatures for the U.S. in relation to land-use. The analyses confirmed the results of the model simulations and showed cropland areas are cooler than forested areas. Bonan continued to develop and apply a land surface process model for use with the CCSM. This model accounts for the ecological effects of different vegetation types and the thermal and hydrologic effects of different soil types. The model was expanded to include river routing, dust emissions, and volatile organic carbon emissions. Three major foci for further model development are:
(a) Arctic regions, using data collected during field campaigns in boreal forest and tundra ecosystems to identify the key ecological and hydrological processes important for climate modeling and how to best parameterize these processes (in collaboration with F.S. Chapin III, University of Alaska, and Amanda Lynch, University of Colorado); (b) inclusion of ecosystem dynamics in land models; and (c) development of the Community Land Model that combines sub-areas of biogeophysics, biogeochemistry, river routing, and ecosystem dynamics into a unified treatment of terrestrial processes for use with CCSM (in collaboration with the LMWG and the CCSM Biogeochemistry Working Group).
David Williamson (CMS) and James Rosinski (CMS) completed the implementation and examination of the reduced grid in the Eulerian and semi-Lagrangian spectral transform versions of the CCM. They defined an n-digit reduced grid for spectral transform models in which the relative error made by omitting certain associated Legendre functions from the Gaussian quadrature poleward of some latitude is less than 10-n. They demonstrated that an adiabatic Eulerian spectral transform model run on an n-digit grid is accurate to n digits for short integrations. The error introduced by the reduced grid grows as expected for a turbulent atmosphere, and the growth is not accelerated by the reduced grid. The errors from an adiabatic semi-Lagrangian spectral transform model can be significantly larger than those from an Eulerian model because the interpolation component of the semi-Lagrangian method will not maintain more than a few digits of accuracy. The adiabatic model errors were put in perspective by comparing with errors introduced by arbitrary aspects of model specification, such as the longitude of the first grid point. This also provided a first indication of the size of aliasing errors in baroclinic spectral models. The errors in a cosine bell advection test were also studied. All reduced grids tested have the same error for Eulerian spectral transform advection. The semi-Lagrangian advection shows some increase in error with greater reduction in the grid, but the variation is comparable to the difference in error from choosing different interpolants or different time steps. Finally, multiple year climate simulations with full and reduced grids were performed. Overall the mean climates produced by the models run on the reduced grids are all very similar to those produced with the full grid. There is no indication that the reduced grids introduce pathological errors that contaminate the simulations. Their results indicate that a 1-digit grid is suitable for climate modeling with both Eulerian and semi-Lagrangian spectral transform models providing a potential savings in computer resources of 30%. The reduced grid codes have been fully implemented in the development version of the CCM, which is being used by the community for the development of CCSM2-A.
Jerry Olson (CMS) and Williamson continued development of two-time-level semi-Lagrangian approximations for CCM. They performed a set of 26 and 49 level, T42 Eulerian and semi-Lagrangian simulations with the idealized Held-Suarez and Boer-Denis forcings. Their original semi-Lagrangian formulation produced very high variance in the tropical vertical motion that was not seen in the Eulerian simulations. Investigations showed that the noise was associated with the trajectory calculation that used tri-cubic interpolation. When tri-linear was used the noise was not present and the characteristics of the simulation were very close to those of the Eulerian simulation. This change was introduced into the semi-Lagrangian version of the CCM.
Williamson re-examined the method of coupling subgrid-scale parameterizations to the dynamical core of the CCM. Some aspects of the coupling in CCM3 can be traced back 30 years to the original grid point models first developed at NCAR. The long evolution of the parameterizations lead to a hesitancy to change the coupling since it was not well understood, but worked, and everyone was familiar with it. He proposed a new design for the CCM in which the parameterizations can be coupled to the dynamics in either a time-split or process-split manner. Although the new design, which involves algorithmic changes, is for the convenience of computer science aspects, it is important to remember that it must be justified by scientific arguments. Truesdale and Williamson carried out simulations to study the effect of these approaches on model climate and climate balance. These simulations provide the scientific justification for the new model design. They also provide some indication of the size of time truncation errors with the modest time step of the Eulerian CCM3. Their relatively short (only 5 years) simulations indicated that the regional differences in the modeled climates introduced by the different time approximations are smaller than even crude measures of natural variability except in a few isolated regions. They are extending the simulations to determine the significance of regional differences. However, even with the short runs, overall the differences are significantly smaller than differences introduced by changes in parameterizations, such as from CCM2 to CCM3, differences introduced by changes within a parameterization (e.g., convection), and differences introduced by the tuning of parameterizations during model development to achieve, say, reasonable TOA energy balances. In addition, the differences are neither consistently good nor bad in terms of simulation quality. Thus, at least for the near term, the errors introduced by the different time approximations allowed in the new design will not be a dominant error in the simulations. Based on the analyses to date, the coupling between the dynamics and parameterizations has been redesigned in the CCM. This allows a structure with the two components cleanly separated and allows the comparison of different dynamical cores in the same model framework with the same parameterizations. This new structure has made possible the Department of Energy (DOE)/NASA-DAO/NCAR collaboration on atmospheric model development.
Byron Boville and Williamson collaborated with Ian Foster (ANL), John Drake (ORNL), and other computer scientists at ANL, ORNL, Los Alamos National Laboratory (LANL), and LLNL on the design and implementation of a performance portable version of the full CCSM, and CCSM2-A in particular. Requirements and design documents have been written for the CCSM2-A software. The design is object oriented and cleanly separates the dynamics from the physical parameterizations, allowing multiple dynamical formulations to be easily tested and maintained. When complete, in early 2001, the dynamics and physics may be separately decomposed for optimal perfomance on distributed memory computers. A simple software framework to isolate the required data transpositions is being developed, primarily at ORNL.
Boville worked with Cecelia De Luca (Scientific Computing Division, SCD), Max Suarez (NASA/GSFC), Arlindo DaSilva (NASA/GSFC), Jeffrey Anderson (NOAA/Geophysical Fluid Dynamics Laboratory, GFDL), Mark Iredell (NOAA/National Centers for Environmental Prediction, NCEP), Chris Hill (Massachusetts Institute of Technology, MIT), Phil Jones (LANL), and Jay Larson (ANL) on the preliminary design of a more complete software framework, the Earth System Modeling Framework (ESMF), to be developed jointly over the next several years. The ESMF will replace the simple framework being used in CCSM2-A, in addition to replacing much of the CCSM2 coupler. The ESMF will provide high level tools to facilitate the construction and coupling of atmosphere and ocean models and data assimilation systems. Use of the ESMF will also greatly simplify the task of moving major components between different modeling systems.
Fabrizio Sassi (CMS) and Boville collaborated with Rolando Garcia (ACD) and Douglas Kinnison (ACD) and Raymond Roble (High Altitude Observatory, HAO) on the development of a Whole Atmosphere CCM (WACCM). WACCM will eventually represent the atmosphere from the surface to ~500 km, including interactive chemistry. In the first year of the project, the vertical domain of the CCM has been extended upward to 140 km from the previous top of the Middle Atmosphere CCM3 at ~80 km. For this extension, the CCM longwave radiation parameterization has been merged in the mesosphere with Fomichev's (1998) nonlocal thermodynamic equilibrium parameterization from Roble's Thermosphere Ionosphere Mesosphere Electrodynamics General Circulation Model (TIME GCM). Solar heating for wavelengths shorter than 200 nm and chemical heating, primarily due to the delayed recombination of atomic oxygen, has also been adapted from the TIME GCM. Molecular diffusion, which becomes a dominant process above ~120 km, has been added. The parameterization of gravity waves has been extended to the model top, including gravity waves dissipation by molecular diffusion. A simplified parameterization of ion drag, important above ~120 km, has been included. The initial version of WACCM, a dynamical model with specified constituent distributions, has been integrated for 20 years and is working quite well. The circulation from the WACCM simulation is now being used by Kinnison in an offline transport model to test the chemical model that will be included over the next year.
Climate and Chemistry Research
Members of the CMS have participated in a variety of projects to understand the interactions between components of the climate system. These projects used a variety of models to understand various aspects of the components of the earth system and their impact on climate.
Collins and Rasch developed a variant of the Model for Atmospheric Transport and Chemistry (MATCH) that included a satellite assimilation procedure to help understand aerosol distributions. The procedure was initially utilized for the Indian Ocean Experiment (INDOEX) and subsequently on a larger class of problems. These MATCH calculations include a representation of sea-salt, sulfate, carbonaceous, and soil-dust aerosols.
The Indian Ocean region includes some of the most densely populated countries on Earth. The pollution from fossil-fuel and biomass burning in these countries is already having a significant influence on the regional atmospheric chemistry. The anthropogenic aerosols released from this region are projected to become the dominant component of anthropogenic aerosols worldwide in the next 40 years. INDOEX was designed to measure the characteristics, distribution, and radiative effects of these aerosols. The aerosol distributions were obtained from a global aerosol simulation including assimilation of satellite retrievals of aerosol optical thickness (AOT). The time-dependent, three-dimensional aerosol distributions were derived with MATCH driven with meteorological analyses for a particular time period. The surface albedos were obtained from the latest version of the NCAR LSM and forced with an identical meteorological analysis and satellite-derived rainfall and insolation. The calculations are consistent with in situ observations of the surface insolation over the central Indian Ocean and with satellite measurements of the reflected shortwave radiation. Rasch compared the assimilation estimates to in situ measurements and derived budgets for each aerosol component for the INDOEX region. They identified major contributors to aerosol loading and the influence of the meteorological fields in modulating the aerosol distribution in this region. The study suggested that the Madden Julian Oscillation was strongly correlated with the aerosol distribution during INDOEX. The aerosol assimilation procedure has now been extended to operate over a global domain, rather than just the INDOEX region, and a variety of projects are ongoing with this model.
This figure (20K) shows a
reduction in sunlight reaching the surface averaged over January-March, 1999, the period
of INDOEX. The forcing is calculated by combining results from the NCAR Land Surface
Model, aerosol assimilation model, and Column Radiation Model.
Collins, Jeffrey Kiehl (CMS), and Rasch subsequently used the aerosol assimilation to examine the aerosol radiative forcing of the Indian Ocean region. In this study, Collins and collaborators use a model that is consistent with the field observations to derive the direct aerosol radiative forcing over the INDOEX region. The calculations show that the surface insolation is reduced by as much as 40 W/m2 over the Indian subcontinent by anthropogenic aerosols. This reduction in insolation is accompanied by an increase in shortwave flux absorbed in the atmosphere by 25 W/m2.
The large magnitude of the forcing by aerosols has several implications. First, the aerosol forcing is as large or larger than other heat sources and sinks, for example, shortwave cloud forcing, which is currently included in the LSM and meteorological analysis system. Second, aerosol forcings of the magnitude observed during INDOEX could have a significant impact on the regional atmospheric and oceanic dynamics. Studies of the coupled ocean-atmosphere response are needed to understand the effects on the regional climate system. Finally, it is clear that anthropogenic aerosols have significantly altered the radiative transfer over India and China. Modeling studies of the aerosol forcing over the next century that are consistent with INDOEX observations are urgently needed for climate change assessment.
Kiehl, in collaboration with Hack and V. Ramanathan (University of California, San Diego, Scripps Institution of Oceanography, SIO), has studied the role of absorbing aerosols in affecting the atmospheric thermodynamic state over the Indian Ocean region. Measurements during INDOEX indicate the pervasive presence of absorbing aerosols over a large region. Measured single scattering albedos were typically 0.85-0.95, much lower than that resulting from pure sulfate aerosols. The study employs the CCM SCM in conjunction with NCEP analyses for the INDOEX time period of 1999. Simulations indicate that the presence of these aerosols leads to a more stable lower troposphere during noontime, which significantly reduces shallow convection. This reduction in convection leads to a reduction in convective cloud cover and a build up of boundary layer humidity. At nighttime this increase in humidity leads to an increase in stable low cloud amount. These features were all observed during INDOEX.
Kiehl, Hack, and Ramanathan carried out simulations with the CCM3 coupled to a slab ocean model to study the effects of absorbing aerosols on the general circulation in the Indian Ocean region. These studies employ a prescribed distribution of absorption that matches the magnitude observed during the INDOEX time period. The simulations show a coherent and statistically significant response of precipitation, free atmosphere and surface temperature, and the large-scale low-level circulation to the presence of observed absorbing aerosol. The presence of the aerosol has a large impact on the position of the deep convection occurring in the ITCZ. Global simulations confirm the local phenomenological processes that lead to the larger scale circulation changes, first identified in SCM simulations, and observed during the INDOEX field program.
This figure (319K) shows the change in the surface temperature (K) due to the presence of absorbing aerosols for a 10-year ensemble January-February-March average. Results are from the CCM3 coupled to a slab ocean model. Aerosol properties are based on observations from the Indian Ocean Experiment (INDOEX).
Kiehl, in collaboration with Lisa Sloan (University of California, Santa Cruz), studied the effect of polar stratospheric clouds (PSCs) on radiative forcing during the Eocene. Kiehl carried out equilibrium chemical calculations to predict the optical depth of these clouds for thermal conditions predicted from the CCM3 with Eocene surface boundary conditions. The preliminary results indicate low optical depths.
The MATCH offline transport model underwent substantial revision during the last year by Rasch. The modifications were made to increase the number of physical problems for which it is appropriate, to improve its computational performance, and to increase the number of platforms on which it runs. In addition to the aerosol studies mentioned above, Rasch worked with Mark Kritz (SUNY, Albany) to compare modeled Radon distributions with measurements over the continental U.S., and with Natalie Mahowald (University of California, Santa Cruz) and Raymond Plumb (MIT) on transport in an isentropic vertical coordinate framework. The model is also being used in the CCSM biogeochemistry modeling effort for carbon cycle and by an increasingly large community of users around the world for a variety of transport problems.
Top of Page | CGD ASR00 Home | ASR00 Home | CGD Home