Variation of ANN Model's Predictive Performance Concerning Short-term (<24 hrs) $SO_2$ Concentrations with Prediction Lagging Time

  • Park, Ok-Hyun (Department of Environmental Engineering, Pusan National University) ;
  • Sin, Ji-Young (Department of Environmental Engineering, Pusan National University) ;
  • Seok, Min-Gwang (Department of Environmental Engineering, Pusan National University)
  • Published : 2008.12.31


In this study, neural network models (NNMs) were examined as alternatives to dispersion models in predicting the short-term $SO_2$ concentrations in a coastal area because the performances of dispersion models in coastal areas have been found to be unsatisfactory. The NNMs were constructed for various combinations of averaging time and prediction time in advance by using the historical data of meteorological parameters and $SO_2$ concentrations in 2002 in the coastal area of Boryeung, Korea. The NNMs were able to make much more accurate predictions of 1 hr $SO_2$ concentrations at ground level in the morning in coastal area than the atmospheric dispersion models such as fumigation models, ADMS3 and ISCST3 for identical conditions of atmospheric stability, area, and weather. Even when predictions of 24-h $SO_2$ concentrations were made 24 hours in advance, the predictions and measurements were in good accordance(correlation coefficient=0.65 for n=216). This accordance level could be improved by appropriate expansion of training parameters. Thus it may be concluded that the NNMs can be successfully used to predict short-term ground level concentrations averaged over time less than 24 hours even in complex terrain. The prediction performance of ANN models tends to improve as the prediction lagging time approaches the concentration averaging time, but to become worse as the lagging time departs from the averaging time.



  1. Bishop, C.M. (1995) Neural Networks for Pattern Recognition, Clarendon Press, Oxford, New York
  2. Boznar, M., M. Lesjak, and P. Mlakar (1993) A neural network-based method for short-term predictions of ambient $SO_2$ concentrations in highly polluted industrial areas of complex terrain, Atmospheric Environment, 27(2), 221-230
  3. Chang, J.C. and S.R. Hanna (2004) Air quality model performance evaluation, Meteorology and Atmospheric Physics, 87, 167-196
  4. Chelani, A.B., C.V. Chalapati Rao, K.M. Phadke, and M.Z. Hansan (2002) Prediction of sulfur dioxide concentration using artificial neural networks, Environmental Modeling & Software, 17, 161-168
  5. Choi, H.J., H.W. Lee, and K.H. Sung (2007) Air quality modeling of ozone concentration according to the roughness length on the complex terrain, J. KOSAE, 23(4), 430-439
  6. Christine, A. and M. John (2001) The importance of meteorological data for modeling air pollution using ADMS-Urban, Meteorological Application, 8, 345-350
  7. Collet, R.S. and K. Oluyemi (1997) Air quality modeling: a technical review of mathematical approaches, Meteorological Applications, 4, 235-246
  8. Comrie, A.C. (1997) Comparing neural networks and regression models for ozone forecasting, Journal of Air and Waste Management, 47, 653-663
  9. Gardner, M.W. and S.R. Dorling (1998) Artificial neural networks (the multilayer perception)-A review of applications in the atmospheric science, Atmospheric Environment, 32, 2627-2636
  10. Hanna, S.R., B.A. Egan, C.J. Vaudo, and A.J. Curreri (1984) A complex terrain dispersion model for regulatory applications at the WESTVACO LUKE mill, Atmospheric Environment, 18(4), 685-699
  11. Hanna, S.R., B.A. Egan, J. Purdum, and J. Wagler (2001) Evaluation of the ADMS, AERMOD, and ISC3 dispersion models with the OPTEX, Duke Forest, Kincaid, Indianapolis and Lovett field datasets, International Journal of Environment and Pollution, 16(1-6), 301-314
  12. Hormik, K., M. Stinchcombe, and H. White (1989) Multilayer feedforward networks are universal approximators, Neural Networks, 2(5), 395-366
  13. James, O.P., G.P. Steven, and J.B. Donna (1992) CTDMPLUS: A dispersion model for sources near complex topography. Part II: performance characteristics, Journal of Applied Meteorology, 31, 646-660<0646:CADMFS>2.0.CO;2
  14. Kim, Y.G. and C.B. Lee (1994) Development of neural network model for prediction of daily maximum ozone concentration in Summer, J. KAPRA, 10(4), 224-232
  15. Lyon, W.A. and H.S. Cole (1973) Fumigation and plume trapping on the shores of Lake Mishigan during stable onshore flow, Journal of Applied Meteorology, 12, 494-510<0494:FAPTOT>2.0.CO;2
  16. Mok, K.M. and S.C. Tam (1998) Short-term prediction of $SO_2$ concentration in Macau with artificial neural networks, Energy and Building, 28, 279-286
  17. Park, O.H. and C.H. Kim (1984) Prediction of urban air pollution using a simple urban dispersion model combined with statistical method, Journal of Korean Society of Environmental Engineers, 6(2), 1-12
  18. Park, O.H. and M.G. Seok (2007) Selection of an appropriate model to predict plume dispersion in coastal area, Atmospheric Environment, 41, 6095-6101
  19. Rege, A.R. and R.W. Tock (1996) A simple neural network for estimating emission rates of hydrogen sulfide and ammonia from single point sources, Journal of Air and Waste Management Association, 46, 953-962
  20. Sin, J.Y. (2006) Thesis for Masters in Eng., Pusan National University
  21. Tamada, T., S. Bunker, and M. Moss (1992) Numerical simulations of atmospheric transport and diffusion over coastal complex terrain, Journal of Applied Meteorology, 31, 565-578<0565:NSOATA>2.0.CO;2
  22. US EPA (2000) Requirements for preparation, adoption, and submittal of state implementation plans (Guideline on Air Quality Models); proposed rule, Federal Register 65(28), 21505-21546
  23. van Dop, H., R. Steenkist, and F.T. Nieuwstadt (1979) Revised estimates for continuous shoreline fumigation, Journal of Applied Meteorology, 18, 133-137<0133:REFCSF>2.0.CO;2
  24. Yi, J. and V.R. Prybutok (1996) A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area, Environmental Pollution, 92, 349-357
  25. Zannettii, P., P. Melli, and E. Runca (1976) Meteorological factor affecting $SO_2$ pollution levels in Venice, Atmospheric Environment, 11, 605-616
  26. Zawar-Rezam, P., S. Kimgham, and J. Pearce (2005) Evaluation of a year-long dispersion modeling of $PM_{10}$ using the mesoscale model TAPM for Christchurch, New Zealand, Science of Total Environment, 349, 249-259
  27. Ziomas, I.C., P. Tzoumaka, and D. Balis (1998) Ozone episodes in Athens, Greece. A modeling approach using data from the medcaphot-trace, Atmospheric Environment, 32, 2313-2321