Contents | Director's Message | Executive Summary | Narratives of Divisional Activities | Publications 
|Educational Statistics | Significant Accomplishments | Community Service| Staff | Visitors | Collaborators |
ASR03 Home

Executive Summary

The Climate and Global Dynamic (CGD) Division’s research emphases include:  development and use of the Community Climate System Model, studies of climate variability and predictability, climate diagnostics and development of climate data sets, basic research and observations on the oceans. CGD also participates strongly in a variety of the new initiatives, including:  Geophysical Statistics Project, Biogeosciences, Data Assimilation, Water Cycle Across Scales, and Weather and Climate Impact Assessment. 

A.     COMMUNITY CLIMATE SYSTEM MODEL (CCSM)

 CCSM2 was released over a year ago.  After examination of various aspects of the 1000 year simulation with that model, it was decided to improve the components of CCSM to reduce some of its systematic errors and to make it a better model for use in the upcoming report of the International Panel on Climate Change.  The resulting model has been named CCSM3.

• Model Development
  

Atmosphere Model

The treatment of solar and infrared radiation in the atmosphere model has been improved by incorporation of a more accurate treatment of the water vapor “continuum.”  A prescribed, time varying aerosol distribution, derived from satellite observations, has been included in the atmosphere mdel. 

The atmosphere component of CCSM3 has been developed in two resolutions, T85 and T42.  The higher resolution model, T85, or approximately 140 km horizontal resolution, will be used for the Intergovernmental Panel on Climate Change (IPCC) experiments (see below), while the T42 version, the same horizontal resolution used in earlier versions of CCSM, will be used in more general climate change experiments.

Ocean Model

A diurnal cycle has been included in the ocean model.  A seasonally and spatially varying, prescribed distribution of absorption of solar radiation, based on chlorophyll observations, has also been included in the ocean model.  This has improved the simulation of the spatially varying sea surface temperature and temperature below the surface of the ocean.

Sea Ice Model

An incremental remapping advection algorithm has been incorporated into the sea ice model and is currently being used for coupled integrations. This transport scheme has improved accuracy, compared to the previously used algorithm, and is also more computationally efficient than previously used transport schemes.


Coupler and Software Engineering


A new coupler has been built for the CCSM. This coupler is more flexible than the previous coupler and it runs on a variety of supercomputing platforms. It allows the exchange of information from “M” processors to “N” processors efficiently. This will also facilitate the implementation of more modules into the coupled CCSM3 and future versions of the model.

Most components of the CCSM3 have been “vectorized”, that is, modified to run efficiently on vector supercomputers, such as the Earth Simulator in Japan, and the new Cray X1 computer being installed at Oak Ridge National Laboratory. Vectorization of the model will be completed by January, 2004. 

 "Comparison of POP ocean model throughput on various platforms as a function of processor count (Aug, 2003). Courtesy of P. Jones, P. Worley, Y. Yoshida, J. White III, J. Levesque." 

• Scientific Applications

Climate Change Experiments Planning for the CCSM contribution to the next IPCC report is nearing completion.  CCSM3 simulations will be run on a variety of computers, including the NCAR IBM, DOE machines (IBM and Cray X1) and the Earth Simulator.  

Biogeosciences and Carbon Cycle

Fully prognostic terrestrial carbon and nitrogen cycles have been added to the Community Land Model (CLM) component of CCSM3.  Model experiments have been conducted to examine the behavior of a carbon-only model and a carbon-nitrogen model. 

A new eco-biogeochemistry model has been developed in the CCSM ocean model. Enhancements include dynamic ecology (phytoplankton, zooplankton, etc.), an active iron cycle, a new particle vertical transport scheme and community structure of key planktonic functional groups (nitrogen fixers, calcifiers, diatoms).

 An earlier coupled carbon cycle model has been put into CSM1.  The first interactive carbon cycle experiments are underway.  Experiments using the newer, more extensive coupled model will begin in early 2004. 

 Paleoclimate 

 A 1000-year run using a paleoclimate version of CCSM has been run.  The forcing inputs for this run include reconstructions of solar variability and volcanic eruptions, plus anthropogenic greenhouse gas emissions for the 19^th and 20^th Centuries.  The global temperature in the model agrees well with proxy data of Earth’s temperature and strongly suggests that the temperature increase over the 20^th Century is due to human influences. 

This figure shows global annual average surface temperature anomalies from 1850-2000 AD simulated by PaleoCSM compared instrumental record (black) and Jones et al. reconstruction (yellow).   PaleoCSM1 simulations are small solar (red), large solar (blue), and large solar without anthropogenic forcing after 1870 (green).

Given the good match over the last 1000 years including the instrumental period, Ammann then extended the 'best-guess' simulations in collaboration with Warren Washington (CCR)and Gerald Meehl (CCR) to the year 2100 using the IPCC scenario A2 for future anthropogenic forcing. Natural forcings were held constant at 2000 AD conditions.

• CCSM Community Support and Planning

The CCSM Workshop continues to grow, with over 300 participants in the most recent workshop.  The newly established focused Working Group meetings have been successful.  One workshop focused on the development of a Climate/Chemistry component for CCSM and the establishment of a Climate and Chemistry Working Group.

A Strategic Plan and a Business Plan have been developed for CCSM.  These plans describe the science to be accomplished and what resources are needed to accomplish these goals, respectively.  These plans are available in print form or on CGD web page.

    B.   OTHER SCIENCE IN CGD

Climate Variability and Predictability

R. Saravanan (CGD) and P. Chang (Texas A&M) have studied aspects of ocean-atmosphere interaction in the tropical Atlantic. Coupling between the atmosphere and the ocean involves the exchange of both momentum and heat. They have studied the role of thermodynamic air-sea coupling using an atmospheric general circulation model (CCM3) coupled to a slab ocean model and they have found that thermodynamic coupling leads to amplification and increased persistence of surface wind variability in the deep tropical Atlantic region. This effect is anisotropic, being stronger in the meridional component than in the zonal component of the surface wind, which suggests a role for the wind-evaporation sea-surface-temperature (SST) feedback in this region. 

Climate Diagnostics and Data

K. Trenberth (CGD) continues his work on the reanalysis data set being produced by the European Centre for Medium-Range Forecasts (ECMRF).  This massive data set, when completed, will provide the most accurate estimate of the state of the atmosphere over the past three decades.  A copy of this data will be delivered to NCAR when completed. 

Ocean Basic Research and Observations

F. Bryan (CGD) and collaborators have developed a 0.1 degree resolution global ocean model, based on the Parallel Ocean Program (POP), which is the same code used in the CCSM, but with coarser resolution.  They have run several centuries of model simulations to examine the interaction of ocean eddies and the ocean mean state.  A variant of this model is likely to be used in a future generation of CCSM. 

Whole Atmosphere Community Climate Model (WACCM)

Version 1b of the Whole Atmosphere Community Climate Model (WACCM1b) has been completed.  This is a non-interactive version of WACCM, whose dynamical outputs can be used to drive the MOZART-3 offline chemical model.  WACCM2, a fully-interactive model that incorporates the Model for Ozone and Related Chemical Tracers (MOZART)-3 chemistry mechanism, NLTE longwave parameterization, shortwave heating, radiation with wavelengths less than 200 nm, full accounting of chemical potential heating and airglow losses, and an auroral parameterization (for heating and NO_X production), is currently being tested.   WACCM2 will be used to investigate problems where coupling between chemistry and dynamics is important, e.g., the response of the winter stratosphere to ozone depletion, the effect of increasing greenhouse gases, and the response to solar variability over the 11-year solar cycle.  Eventually, WACCM will be used as an atmosphere component in future versions of CCSM. 

Geophysical Statistics Project

C. Tebaldi (CGD/ESIG), in collaboration with Linda Mearns (ESIG) and Richard Smith (University of North Carolina), have developed a Bayesian, random effects model to combine the results of different climate model simulations and quantify the uncertainty. This research is in the context of synthesizing the numerical experiments from different climate modeling centers in a way that produces credible probabilistic forecasts of future climate change.  The basic concept is to consider the individual climate model results as independent samples from a super population of models that are unbiased with respect to the true climate. 

C.   STRATEGIC INITIATIVES

• Biogeosciences 

Much of CGD’s activities in support of the Biogeosciences initiative are included in the Narrative Section, CCSM and the Terrestrial Sciences Section.

D. Schimel (CGD) has continued development of data assimilation techniques for biogeochemistry and carbon cycle studies.  A local scale model, developed with Braswell (University of New Hampshire), assimilates CO₂ flux observations and has been used to analyze seasonal and interannual controls at the Harvard Forest in Massachusetts and at flux sites in Brazil as part of the LBA.  The Colorado State University assimilation system,  (Regional Atmospheric Modeling System and Data Assimilation System - RAMDAS) has been ported to NCAR and has been used to design optimal approaches for sampling CO₂ concentrations in order to retrieve surface fluxes in complex terrain.  The RAMDAS system will support design and execution of surface and airborne field studies for an NSF Biocomplexity-sponsored research program on carbon fluxes in mountain landscapes, in collaboration with the University of Colorado, Colorado State University and University of Miami, Florida.

D. Baker (CGD) and collaborators have developed a global Carbon Data Assimilation Model, which is now operational and couples a simple atmospheric transport model and an optimization scheme to estimate surface fluxes globally from observations of concentrations. Studies using the 4-dimensional variational assimilation approach have begun to unravel the interactions between the atmospheric sampling scheme, assumptions about spatial and temporal coherence of concentrations in the atmosphere, and the resolution of the estimated flux field.  Preliminary results strongly suggest that: 1) relatively high sampling density is needed to resolve regional fluxes, independent of model skill, 2) continuous observations are far more influential on the model solution than episodic (e.g. flask) measurements, and 3) assumptions about time and space correlations have a large impact and must be based on careful analyses. 

• Data Assimilation

The NCAR Data Assimilation Initiative (DAI) has begun building a data assimilation research testbed (DART), an initial version of which was completed during Fiscal Year 03 (FY03). DART provided the exercises for the FY03 Advanced Study Program Summer Colloquium that was co-sponsored by DAI. DART has also been used to build ensemble filter assimilation systems for a variety of atmosphere models, both regional and global.

DAI has also begun efforts to evaluate the characteristics of new observational sets including GPS radio occultation observations by developing forward observation operators and an assimilation testing capability. 

• Water Cycle Across Scales 

A. Dai (CGD) and Trenberth have produced updated and improved estimates of fresh water discharge into the oceans, and they have also evaluated the evaporation minus precipitation (E-P) fields derived from  European Centre for Medium-range Weather Forecasting (ECMWF) and NCEP/NCAR reanalyses. The river flow and discharge data are available from the Climate Analysis Section data catalog (http://www.cgd.ucar.edu/cas/catalog/). The new discharge estimates have already been used by Geophysical Fluid Dynamics Laboratory and other groups to evaluate climate models.

Dai and Trenberth have used the new discharge data and the E-P fields derived from the ECMWF and the National Centers for Environmental Prediction and NCAR reanalyses to derive new estimates of meridional transport of fresh water by the oceans. Preliminary results show improved agreement with in situ data based on direct estimates. Compared with earlier indirect estimates, the new estimates show increased southward transport in the Atlantic Ocean and increased northward transport in the South Pacific.

• Weather and Climate Impacts

G. Meehl (CGD), Tebaldi, Mearns, D. Nychka (CGD), and J. Arblaster (CGD)  have analyzed changes in climate variability and extremes in ensembles of future climate projections. This work used statistical analyses of extremes, as well as threshold methods, to study changes in weather and climate extremes in the Parallel Climate Model (PCM). An analysis of frost days in the PCM showed that anomalous sea level pressure patterns in the future climate, indicative of regional atmospheric circulation changes, were prime contributors to the pattern of reductions in frost days in future climate, with soil moisture and clouds playing secondary roles.

A simple coupled gas-cycle/energy-balance model Model for the Assessment of Greenhouse-gas Induced Climate Change (MAGICC) has been developed by T. Wigley.  It has been embedded in a user-friendly shell and is now available for downloading from the NCAR/CGD web page. MAGICC takes emissions scenarios as input (for CO₂, CH₄, N₂O, CO, NO_x, VOCs, SO₂, and a number of halocarbons) and produces atmospheric composition, radiative forcing, global-mean temperature, and sea level changes as output, together with uncertainty bounds for these estimates. MAGICC can be used to reproduce results published in the IPCC Third Assessment Report, and to extend these results to other emissions scenarios. MAGICC also drives a scaling algorithm (SCENGEN – SCENario GENerator) to produce spatially detailed scenarios of future temperature and precipitation change on a 5° by 5° latitude/longitude grid. SCENGEN can be used interactively to validate AOGCMs (from its library of data from 17 AOGCMs) against a library of observed temperature and precipitation data; produce scenarios of changes in variability; quantify uncertainties in terms of temporal and inter-model signal-to-noise ratios; and produce probabilistic projections of climate change.