Introduction
It is now a fact that the atmospheric concentrations of several chemically and radiatively important trace constituents (gases and particles) are changing, primarily due to human influences. Several trace constituents like primary aerosols, methane, nitrous oxide, higher nitrogen oxides (NOx), carbon monoxide, nonmethane hydrocarbons, sulfur dioxide, dimethylsulfide and halocarbons have direct anthropogenic and/or natural emissions to the atmosphere. Others, like ozone and secondary aerosols, are produced from chemical reactions of their precursors. In turn, the changing atmospheric concentrations of these trace constituents can affect the radiative balance and climate of the earth. To understand and reliably predict chemical and climate changes in the atmosphere, a thorough understanding of the chemical, physical, biological and climatic processes which affect the distributions of trace constituents in the atmosphere, and of the interactions between these processes is required.

In this context, numerical models are especially useful tools for evaluating the global budgets of trace constituents, for understanding the evolution of trace constituent distributions due to natural and anthropogenic forcings, and for assessing atmospheric chemistry-climate interactions. The Global Integration and Modeling (GIM) Activity is a recently-initiated effort under the IGAC Global Focus which specifically aims to address the development and application of advanced three-dimensional global chemical transport and coupled climate/chemistry models, with an emphasis on tropospheric applications. It is important to note that several mature research efforts focused on global-scale tropospheric chemistry and transport modeling already exist. Thus, the activities to be undertaken as part of GIM will be directed towards facilitating these existing efforts, rather than developing independent modeling capabilities. With this in mind, GIM will pursue its goals by: (i) conducting a series of model intercomparison exercises focusing on key problems in tropospheric chemistry; (ii) conducting workshops specifically targeted to graduate students and postdoctoral researchers and focusing on the latest developments in tropospheric chemistry and climate modeling; (iii) collaborating actively with other on-going model development and evaluation efforts such as those being carried out as part of the World Climate Research Program and the NASA/Global Modeling Initiative; and (iv) establishing close linkages other with on-going IGAC Activities such as the Global Atmospheric Chemistry Survey (GLOCHEM), the Global Emissions Inventory Activity (GEIA), and others.

In the sections below we provide a brief description of tropospheric chemistry and transport models, elaborate on specific GIM efforts that have been initiated or are planned for the near future, and outline the organizational structure of GIM.

Tropospheric Chemistry and Transport Models (CTMs)
Tropospheric CTMs are numerical models which simulate the interactions between processes such as emissions, photochemical production and destruction, convective and synoptic mixing, and dry and wet deposition, which together shape the distribution of trace constituents. These models are time-dependent and, depending on their spatial resolution, can be classified as 1-, 2-, and 3-dimensional models. Depending on the horizontal extent of the area considered, one distinguishes between limited area or mesoscale models and global models. The limited area/mesoscale models cover areas up to a few thousand square kilometers, and can be subdivided into 3 main categories:
  1. The g mesoscale models covering areas of 1-10 km2 (for instance urban core and urban perimeter, and local models) used for studying changes occurring within a few minutes;

  2. The b mesoscale models extending from about 10 to 100 km2 (regional models) used for studying transport and chemical processes occurring on the time-scale of several hours. On this spatial and temporal scale, photochemistry can produce pollution episodes and have a significant impact on relative long lived trace gas concentrations; and

  3. The a mesoscale models covering areas of hundreds to many thousands of km2 (for instance the North American continent) which corresponds to the scale of occurrence of synoptic transport (high and low pressure meteorological systems).
The rapidly increasing computational capabilities of modern-day computers enhances the use of atmospheric chemistry transport models at various spatial resolutions. The same global or mesoscale model can be run with different horizontal and vertical resolutions or can have the option of using a window with higher resolution over a particular area of interest.

Another classification of tropospheric CTMs is based on the meteorology used to drive the model. In decoupled or "off-line" CTMs, chemistry is not allowed to feed back into meteorology. The meteorology used to drive these models is derived either from general circulation models (GCMs) which simulate the ensemble of meteorological events corresponding to a generic time-period or from assimilation-based models which simulate specific time periods. In either case, the meteorology used to drive most off-line CTMs is synoptic in nature, though a few CTMs are climatological in nature with the transport based on monthly-mean winds. The most advanced models used for tropospheric chemistry studies are the fully coupled general circulation/chemistry or "on-line" models in which chemistry, meteorology, and radiative transfer are computed simultaneously and changes in the chemical species concentrations may affect meteorology and climate and vice versa.

Both off-line and on-line models are useful. On-line models are advantageous for studying chemistry-climate interactions, while off-line models offer a computationally efficient tool for studying the global distributions and budgets of a number of important tropospheric species which are influenced by climate only indirectly.

Tropospheric CTMs are useful in unraveling the complex physical and chemical interactions that shape tropospheric trace constituent distributions. A key component of any model development and evaluation exercise is evaluation against relevant field measurements. Such measurements include both in situ and remote data and, depending on the nature of the CTM, may be climatological or episodic in nature. The CTMs in turn can be used for designing field measurement campaigns, and for interpreting and analyzing the data gathered from these campaigns.

Specific GIM Efforts
The development and application of comprehensive CTMs for tropospheric ozone and aerosols is of considerable scientific importance. Tropospheric ozone plays a central role in determining the oxidizing power of the atmosphere and is also an important greenhouse gas. Tropospheric aerosols are believed to significantly affect the Earth's radiative balance, and can affect gas phase chemical composition via heterogeneous reactions. Furthermore, in some polluted regions both ozone and fine-particles are believed to pose risks to human health as well as agricultural crops and forest resources. Over the last decade or so, there has been considerable research in the development and application of CTMs for tropospheric ozone. Similarly, a number of models have been developed for evaluating the mass distribution of sulfate aerosols. Less advanced is our capability to simulate particle size distributions in CTMs, though again there is considerable ongoing research in this area. Keeping in mind GIM's objectives of facilitating rather than developing tropospheric CTMs, three distinct activities have been developed in the general areas of tropospheric ozone and aerosol modeling. These are: (i) a tropospheric ozone global CTM intercomparison exercise; (ii) an aerosol dynamics model intercomparison exercise; and (iii) a tropospheric global aerosol modeling workshop and intercomparison exercise. Of these, the tropospheric ozone intercomparison is well underway, and is discussed in detail below. The aerosol modeling activities are in the planning stages, and are also briefly discussed below.

Tropospheric Ozone Global CTM Intercomparison Exercise

The objective here is to evaluate systematically the capabilities of the current generation of global tropospheric ozone CTMs, and to identify key areas of uncertainty in our understanding of the tropospheric ozone budget. To accomplish this goal, a tropospheric ozone modeling intercomparison exercise was organized during the late summer/early fall of 1996 involving research groups from a number of institutions. The strategy was to investigate the convergence of the models and the extent to which the models reproduce observed characteristics of tropospheric ozone. The twelve global 3-dimensional CTMs involved, together with the contributing scientists, are listed in Table 1.

Model Name Contact
IMAGES J.-F. Muller, G.P. Brasseur, C. Granier
GFDL H. Levy II
HARVARD D. Jacob
ECHAM G.-J. Roelofs
TM3 F. Dentener, S. Houweling
IMAU3 M. Krol
CTMK W.M.F. Wauben
MATCH M.G. Lawrence, P.J. Crutzen
MOGUNTIA N. Poisson, M. Kanakidou
MOZART D.A. Hauglustaine, G.P. Brasseur
UKMETO R.G. Derwent, C.E. Johnson, W.J. Collins, D.S. Stevenson
UIO T.K. Berntsen, I. Isaksen

Table 1.
3-dimensional CTMs that participated in the tropospheric ozone global modeling GIM intercomparison exercise.

A meeting to discuss the preliminary results from this exercise was held at Gif-sur-Yvette in France in November, 1996. Figure 1 shows examples of model comparisons for surface ozone and carbon monoxide, respectively, at the Barrow, Alaska, and Cape Grim, Tasmania baseline stations. Observed mixing ratios and standard deviations are plotted for comparison purposes.


Figure 1.
Results of model simulations of the annual cycles of ozone and carbon monoxide at the baseline observatories at Barrow, Alaska (USA) and Cape Grim, Australia. Observed mixing ratios (heavy solid lines) and their standard deviations (heavy dashed lines) are plotted for comparison purposes.

It is encouraging that the model simulations reproduce the general variations of the measured monthly-mean concentrations. However, major differences in the model outputs are obvious and remain to be analyzed in terms of the ozone budget in the free troposphere and in particular close to the tropopause. These differences are also reflected in the OH distribution calculated by the various models. This is illustrated in Table 2 which shows the methane lifetimes calculated by the models.

  Range Median
Global 6.4-10
(15*) 		7.5
NH 5.8-9.1
(14*) 		6.9
SH 7.1-10.3
(19*) 		8.6

Table 2.
CH4 lifetime (in years) as computed by the CTMs listed in Table 1.

* only one model result, not taken into account for the median calculation.

A more detailed diagnostic analysis of the ozone intercomparison results is now underway. A follow-up meeting of the GIM Ozone Action Committee is planned for the fall of 1997 to summarize the results of this exercise and finalize the preparation of a manuscript detailing these results of the analysis. Further details can be obtained from Dr. Maria Kanakidou, mariak@lmce.saclay.cea.fr.
Aerosol Dynamics Model IntercomparisonExercise.

At the November 1996 workshop in France, there was a general consensus that the organization of an aerosol modeling intercomparison exercise would be of considerable value to the atmospheric sciences community. A proposal defining a limited intercomparison (focusing on the treatment of sulfuric acid-water aerosol formation and growth in box models) was prepared and circulated by P. Kasibhatla. In June 1997 a zero-dimensional intercomparison exercise to test the numerical fidelity of various techniques used to parameterize the aerosol size distribution in large-scale CTMs has been defined as a collaborative effort between the European Aerosol Assembly (EAA) and GIM. Further details of this exercise can be obtained from the World Wide Web at http://www.tropos.de in the Cooperation Section.

Tropospheric Global Aerosol Modeling Workshop and Intercomparison Exercise.

The goal of this exercise is to perform a detailed evaluation of the capabilities of the current generation of global aerosol models that are used to characterize the distribution of sulfates, soil dust, and black carbon. This exercise will be conducted jointly with the WCRP's Working Group on Numerical Experimentation (WGNE), and is a follow-up to the WCRP-sponsored 1995 workshop on Transport and Scavenging of Trace Constituents by Clouds in Global Atmospheric Models. As part of this exercise, it is currently planned to hold a workshop in September 1998 in Nova Scotia. Part of the workshop will be devoted to invited talks on the parameterization of various relevant physical and chemical processes in global aerosol models, and part of the workshop will be focused on model evaluation and intercomparison activities. Further information on this can be obtained from Dr. Len Barrie, len.barrie@ec.gc.ca.

In addition to the above-mentioned activities, Drs. Martin Heimann, martin.heimann@dkrz.de and Dana Hartley, hartley@voir.eas.gatech.edu have taken the lead in organizing a Workshop on Inverse Methods in Global Biogeochemical Cycles as a joint effort between the IGBP Global Analysis, Interpretation, and Modeling (GAIM) Framework Activity and GIM. The main objective of this workshop will be to teach students entering the field about the mathematics of inverse problems and the issues of a priori constraints. The long-term benefit will be more experienced researchers who will be able to address some of the outstanding problems in global biogeochemical cycles with the proper scientific tools. The workshop is currently planned for March 16-20, 1998, in Crete, Greece. The workshop will follow a format of invited expert lectures, as well as practical problem-solving exercises. It is anticipated that the expert lectures will be printed in a book together with the practical examples and their solutions. Potentially, both the problems and the software will be made available in electronic form (e.g., as a CD-ROM) together with the book.

It is anticipated that other topics such as heterogeneous reactions in the atmosphere and biosphere/atmosphere interactions will be the focus of future GIM activities. Such activities will involve a close coordination between GIM and other related activities of IGAC and IGBP.
GIM Leadership
The current Co-Convenors of GIM are Dr. Maria Kanakidou and Dr. Prasad Kasibhatla, psk9@duke.edu. The other inaugural members of the GIM Coordinating Committee are: Carmen Benkovitz, USA , Frank Dentener, The Netherlands, Laura Gallardo-Klenner, Chile, Claire Granier, France & USA, Ivar Isaksen, Norway, Jack Kaye, USA, Kathy Law, UK, Jennifer Logan, USA, and Jack McConnell, Canada. This Committee will be expended in near future to include scientists from Asia, Australia, and other locales.

Back to previous page.