Make your own free website on


1 Brief description (key words)

1 Policy issue

urban air quality

2 Application type

air quality assessment

regulatory purposes and compliance

policy support

public information

scientific research

3 Model output


4 Air pollution source

array of line source


5 Release type


6 Spatial scale


7 Simulation character


8 Pollutants modelled




9 Processes considered

vertical structure


chemical transformation

10 Computer platform






2 Long description

1. Basic information

Model name

CAR-FMI (Contaminants in the Air from a Road – Finnish Meteorological Institute)

Model versions and status

Last update: April 1998


Finnish Meteorological Institute (FMI)

Contact persons (providing all necessary technical support)

Jari Härkönen

Ari Karppinen

Contact address

Finnish Meteorological Institute

Air Quality Research

Sahaajankatu 20 E

FIN-00810 Helsinki


Phone number

+358 9 1929 11

Fax number

+358 9 1929 5403

E-mail address


Technical support

Provided by contact person

2. Intended field of application

The model is a local scale Gaussian model, which evaluates atmospheric dispersion and chemical transformation from a network of line sources. The model requires information on emissions and meteorological parameters on an hourly basis. It computes time series of concentrations and related statistics at user-specified receptor points.

3. Model type and dimension

Multiple line-source Gaussian plume model, combined with a meteorological pre-processing model based on atmospheric boundary layer scaling.

4. Model description summary

The modelling system includes an emission model, a dispersion model, statistical analysis of the computed time series of concentrations and a graphical Windows-based user interface. The meteorological data for the model is evaluated by the FMI meteorological pre-processor. The dispersion model allows for a network of finite line sources. It includes a treatment of chemical transformation (for NO, NO2 and O3).

The Gaussian dispersion parameters are functions of boundary layer variables, and their dependence on source height is also taken into account. The model includes the influence of a finite mixing height on the plume dispersion. The system computes statistical concentration parameters from the hourly time series, which can directly be compared to air quality guidelines and limit values.

5. Model limitations

The concentration distributions are assumed to be Gaussian in both the horizontal and vertical directions. The model does not allow for the influence on dispersion of individual buildings and obstacles, or inhomogeneous terrain. The current model version does not include dry or wet deposition, or dispersion of particles.

6. Resolution

Temporal resolution

1 hour

Horizontal resolution

Domain dimension: up to 10 km, adjustable calculation grid

Vertical resolution

Domain dimension: mixing height, grid size not explicitly limited

7. Schemes


Gaussian plume


Atmospheric boundary layer scaling theory.


The version currently distributed does not allow for deposition.


The chemical transformation of nitrogen oxides is modelled by using the so-called receptor-oriented discrete parcel method. This model includes the basic reactions of nitrogen oxides, oxygen and ozone, but the influence of hydrocarbons and other compounds is neglected.

8. Solution technique

Not applicable

9. Input requirements


Required parameters include: the pollutant species, the hourly emission time-series, the effective source height and the geographical coordinates of the line sources.


Meteorological measurements are processed by a pre-processor (MPP-FMI); the input includes three-hourly data from synoptic stations and twice-daily vertical profiles (of temperature, wind and humidity) from a radiosonde station.


Specified as terrain heights at receptor locations.

Initial conditions

Not applicable.

Boundary conditions

Not applicable.

Other input requirements

Receptor locations.

10. Output quantities

The model output consists of hourly concentrations at each 3D-grid point.

11. User interface availability

The CAR-FMI program has an easy-to-use Windows interface. In order to obtain the program, please contact the model developers.

12. User community

The model is technically very easy-to-use, but the proper interpretation of results requires a sufficient knowledge in air pollution science.

13. Previous applications

The modelling system has been applied widely nationally in air quality assessments. It has also been applied world-wide in commissioned work, using meteorological data from the global observational network of the World Meteorological Organisation.

Examples of model applications have been discussed in the references. For instance, Karppinen et al. (1997a,b and 1998b) numerically evaluated the spatial distribution of concentrations and the statistical concentration parameters of NO2- and CO in the Helsinki metropolitan area.

14. Documentation status

Level 3. The model is documented in publicly available reports and publication series (for instance, Härkönen et al, 1996).

15. Validation and evaluation

Level 4. The predictions of the model have been validated against the results of two field measurement campaigns, conducted near major roads (Walden, et al., 1995, Härkönen et al.,1997). The modelling system has also been tested extensively against results from urban measurement networks (Karppinen et al., 1997a,b and 1998b).

16. Frequently Asked Questions

17. Portability and computer requirements


The PC-version of the model can be run in any PC, which is equipped with the Windows 95 or Windows NT –operating system.

The mainframe version of the model has been written in Fortran 77. The model has been written in Fortran 77. A parallel processor version of the program has been written for the Cray C-97 supercomputer. The program can also be executed (without the parallel processor routines) in any mainframe or workstation computer.

CPU time

The computation time depends on the number of sources, the number of receptor points and the extent of the emission and meteorological time series. The computational times vary from seconds to a couple of tens of hours (for very extensive applications) CPU – time on Cray C-97.

18. Availability

In order to obtain the CAR-FMI program, please contact the model developers at the Finnish Meteorological Institute.

19. References

Härkönen, J., Valkonen, E., Kukkonen, J., Rantakrans, E., Lahtinen, K., Karppinen, A. and Jalkanen, L., 1996. A model for the dispersion of pollution from a road network. Finnish Meteorological Institute, Publications on Air Quality 23. Helsinki, 34 p.

Härkönen J., Walden J. and Kukkonen, J., 1997. Comparison of model predictions and measurements near a major road in an urban area. International Journal of Environment and Pollution, Vol. 8, Nos. 3-6, p. 761-768.

Karppinen, A., Kukkonen, J., Konttinen, M., Härkönen, J., Valkonen, E. , Rantakrans, E., Koskentalo, T., and Elolähde, T., 1997a. The emissions, dispersion and chemical transformation of traffic-originated nitrogen oxides at the Helsinki metropolitan area. In: Joumard, R., Proceedings of the 4th international scientific symposium "Transport and Air Pollution" in Avignon, France, 9-13 June 1997. Avignon, pp.121-127.

Karppinen, A., Kukkonen, J., Konttinen, M., Rantakrans, E., Valkonen, E., Härkönen, J., Koskentalo, T. and Elolähde, T., 1997b. Comparison of dispersion model predictions and the results from an urban air quality measurement network. In: Power, H., Tirabassi, T., and Brebbia, C. A. (eds.). Air Pollution V. CMP, Southampton, pp. 405-411.

Karppinen, A., Kukkonen, J., Konttinen, M., Härkönen, J., Valkonen, E., Koskentalo, T., Elolähde, T., 1998. Development and verification of a modelling system for predicting urban NO2 concentrations. In: Gryning, S.-E. and Chaumerliac, N. (eds.), Air pollution modelling and its application XXII, NATO, Challenges of Modern Society, Volume 22. New York and London, pp. 567-574.

Kukkonen, J., Härkönen, J., Valkonen, E., Karppinen, A. and Rantakrans, E., 1997. Regulatory dispersion modelling in Finland. International Journal of Environment and Pollution, Vol. 8., Nos. 3-6, p. 782-788.

Walden, J., Härkönen, J., Pohjola, V., Kukkonen, J. and Kartastenpää, R., 1995. Vertical concentration profiles in urban conditions - comparison of measurements and model predictions. In: Anttila, P. et al. (ed.), Proceedings of the 10th World Clean Air Congress, Espoo, Finland, May 28 - June 2, 1995. Vol. 2. The Finnish Air Pollution Prevention Society, Helsinki, p. 268 (4 pages).