Seasonal ARIMA Approach for Forecasting Air Pollution of Chennai City in India

IJEP 43(4): 321-329 : Vol. 43 Issue. 4 (April 2023)

Imran Nadeem and P.S. Sheik Uduman*

B.S. Abdur Rahman Crescent Institute of Science and Technology, Department of Mathematics, Chennai, Tamil Nadu – 600 048, India

Abstract

Air pollution is one of the hazards posing a severe threat at the global level in recent years. The forecasting of air pollutants is crucial for the implication of a policy that safeguards the environment from further deterioration in a metropolitan city, like Chennai. The study aims to forecast the rising level of air pollutants from the 15 years of monthly data of each pollutant monitored at the commercial and industrial sites of Chennai city. In addition, the study compared the actual air pollutants concentration with the admissible limit as defined by national ambient air quality standards of India for determining the level of pollution at all five stations. This study aims to employ seasonal autoregressive integrated moving average (SARIMA) approach using Box-Jenkins methodology based on the database monitored at the five sites of Chennai city for forecasting the rising level of RSPM, SO2 and NO2. The evaluation of model statistics obtained in our study shows RMSE values lie in the range of 1.52-23.55, MAPE values lie between 6.43-16.73 and R2 values lie between 84-94% for all the 15 best-fitted models by employing SARIMA approach. The forecasting of the three most prevalent pollutants in the main areas of Chennai city is quite beneficial for forming a policy to handle the pollution level in a better way.

Keywords

Air pollutant, Ljung-box test, Seasonal ARIMA approach, Time series analysis

References

  1. Afroz, R., M.N. Hassan and N.A. Ibrahim. 2003. Review of air pollution and health impacts in Malaysia. Env. Res., 92(2): 71-77.
  2. Kumar, A. and P. Goyal. 2011. Forecasting of daily air quality index in Delhi. Sci. Total Env., 409(24): 5517-5523.
  3. Nurulilyana, S., et al. 2011. Statistical analysis of PM10concentrations at different locations in Malaysia. Env. Monit. Assess., 180(1): 573-588.
  4. Adebiyi, A.A., A.O. Adewumi and C.K. Ayo. 2014. Comparison of ARIMA and artificial neural networks models for stock price prediction. J. Appl. Mathematics. 2(1):1–7.
  5. Chatfield, C. 1996. The analysis of time series- An introduction (5th edn). Chapman and Hall CRC, London.
  6. Pankratz, A. 2009. Forecasting with univariate Box-Jenkins models: Concepts and cases. John Wiley and Sons.
  7. Jolliffe, I.T. 2002. Principal component analysis (2nd edn). Springer, New York.
  8. Kim, J. and C.W. Mueller. 1986. Factor analysis: Statistical methods and practical issue. Sage Publication, Beverly Hills.
  9. Kaplunovsky, A.S. 2005. Factor analysis in environmental studies. HAIT J. Sci. Eng. B. 2(1-2):54-94.
  10. Huang, J., M. Ho and P. Du. 2011. Assessment of temporal and spatial variation of coastal water quality and source identification along Macau peninsula. Stochastic Env. Res. Risk Assess., 25(3): 353-361.
  11. Box, G.E.P. and G.M. Jenkins. 1976. Time series analysis: Forecasting and control (revised edn). Holden Day, San Francisco.
  12. Chatfield, C. 2000. Time-series forecasting. Chapman and Hall CRC, Boca Raton.
  13. McBerthouex, P. and L.C. Brown. 2002. Statistics for environmental engineers. Lewis Publishers, Boca Raton.
  14. Jian, L., et al. 2012. An application of ARIMA model to predict submicron particle concentrations from meteorological factors at a busy roadside in Hangzhou, China. Sci. Total Env., 426:336-345.
  15. Liu, P.W.G. 2009. Simulation of the daily average PM10concentrations at Ta-Liao with Box-Jenkins time series models and multivariate analysis. Atmos. Env., 43:2104-2113.
  16. Duenas, C., et al. 2011. Stochastic model to forecast ground-level ozone concentration at urban and rural areas. Chemosphere. 61(10):1379–1389.
  17. Sharma, P., A. Chandra and S.C. Kaushik. 2009. Forecasts using Box-Jenkins models for the ambient air quality data of Delhi city. Env. Monit. Assess., 157(1–4):105–112.
  18. Kumar, U. and V.K. Jain. 2010. ARIMA forecasting of ambient air pollutants (O3, NO, NO2and CO). Stochastic Env. Res. Risk Assess., 24:751-760.
  19. Kadilar, G. and C. Kadilar. 2017. Assessing air quality in Aksaray with time series analysis. AIP conference. Proceeding, 1833(1):20-112.
  20. Cujia, A., et al. 2019. Forecast of PM10time-series data: A study case in Caribbean cities. Atmos. Poll. Res., 10(6): 2053-2062.
  21. Giridhar, V.V. 2001. Coastal ocean pollution monitoring programme. Proceedings of UGC course on marine sciences and environment conducted by the Department of Applied Geology, University of Madras, Chennai, India.
  22. Brockwell, J.B. and R.A. Davis. 2002. Introduction to time series and forecasting. Springer, New York.
  23. Schwarz, G. 1978. Estimating the dimension of a model. Annal. Stat., 6:461-464.
  24. Burnham, K.P. and D.R. Anderson. 2004. Multimodel inference: Understanding AIC and BIC in model selection. Sociol. Methods Res.,33:261–304.
  25. Ljung, G.M. and G.E. Box. 1978. On a measure of lack of fit in time series models. Biometrika. 65(2):297–303.
  26. Robeson, S.M. and D.G. Steyn. 1990. Evaluation and comparison of statistical forecast models for daily maximum ozone concentrations. Atoms. Env., 24B:303–312.
  27. Cai, M., Y. Yin and M. Xie. 2009. Prediction of hourly air pollutant concentrations near urban arterials using artificial neural network approach. Transportation Res. Part D: Transport Env., 14:32–41.
  28. Guttikunda, S.K., R. Goel and P. Pant. 2014. Nature of air pollution, emission sources and management in the Indian cities. Atmos. Env., 95:501-510.