Peaks Over Threshold Method Application on Airborne Particulate Matter (PM10) and Sulphur Dioxide (SO2) Pollution Detection in Specified Regions of İstanbul

Abstract

In this study, we investigate the application of peak over threshold (POT)  method on extreme events which usually appears with low frequently but high effects. Daily averages of PM₁₀ and SO₂ pollutants are measured at 5 permanent monitoring stations in İstanbul (Beşiktaş, Yenibosna, Alibeyköy, Esenler, Aksaray). The SO₂ and PM₁₀ concentration dataare obtained from İstanbul Municipality through a period from January 2009 to December 2015.Daily averages of theconcentrations are analyzed by using peaks over threshold methods of extreme value theory and then predicted for the largest concentrations for the following 12 months. We find thatPOT methods can provide useful information about the occurrence of limit exceedances of air pollution in Istanbul and these models can easily be used to make short term predictions about limit exceedances.As a consequence, we can say that predicting the air pollutant levels of SO₂ and PM₁₀ will be beneficial for the decision makers which help them to develop advanced policies to control and prevent the air pollution.

Keywords

• Extreme value theory

Peaks Over Threshold Method Application on Airborne Particulate Matter (PM10) and Sulphur Dioxide (SO2) Pollution Detection in Specified Regions of İstanbul

Abstract

In this study, we investigate the application of peak over threshold (POT)  method on extreme events which usually appears with low frequently but high effects. Daily averages of PM₁₀ and SO₂ pollutants are measured at 5 permanent monitoring stations in İstanbul (Beşiktaş, Yenibosna, Alibeyköy, Esenler, Aksaray). The SO₂ and PM₁₀ concentration dataare obtained from İstanbul Municipality through a period from January 2009 to December 2015. Daily averages of theconcentrations are analyzed by using peaks over threshold methods of extreme value theory and then predicted for the largest concentrations for the following 12 months. We find thatPOT methods can provide useful information about the occurrence of limit exceedances of air pollution in Istanbul and these models can easily be used to make short term predictions about limit exceedances.As a consequence, we can say that predicting the air pollutant levels of SO₂ and PM₁₀ will be beneficial for the decision makers which help them to develop advanced policies to control and prevent the air pollution.

Keywords

• POT,extreme Value Theory,İstanbul air quality,PM10

References

1. Arnold CB (2015) Pareto Distribution. , Wiley Stats Ref: Statistics Reference Online. doi 10.1002/9781118445112.stat01100.pub2.
2. Bader B,Yan J,Zhang X (2016) Automated Threshold Selection for Extreme Value Analysis via Goodness-of-Fit Tests with Application to Batched Return Level Mapping,Cornell University Library Publications.https://arxiv.org/abs/1604.02024.Accessed 08.12.2016.
3. Balkema A, De H (1974) "Residual life time at great age", Annals of Probability, 2, 792–804.
4. Beguería S (2005) Uncertainties in partial duration series modelling of extremes related to the choice of the threshold value, J. Hydrol., 303, 215–230.
5. Bommier E (2014) Peaks-Over-Threshold Modelling of Environmental Data.Https://uu.diva-portal.org/smash/get/diva2:760802/FULLTEXT01.pdf. Accessed 01.12.2016.
6. Capraz O,Efe B,Deniz A (2006) Study on the association between air pollution and mortality in İstanbul, 2007–2012, Atmos. Poll. Res. 7(1). 147-154.doi.http://dx.doi.org/10.1016/j.apr.2015.08.006.
7. Chock DP, Sluchak PS (1986) Estimating extreme values of air quality data using different fitted distributions. Atmos. Environ., 20, pp. 989–993.
8. Coles S (2001) An Introduction to Statistical Modelling of Extreme Values, 208 pp., Springer, London.

9. Cox WM, Chu SH (1993) Meteorologically adjusted ozone trends in urban areas - a probabilistic approach. Atmos. Environ. Part B-Urban Atmosphere, 27, pp. 425–434.
10. Embrechts P, Kluppelberg C, Mikosch (1997) Modelling Extremal Events for Insurance and Finance, Springer–Verlag, Berlin , p. 650.
11. Ferreiara A and De Haan L (2015) On the block maxima method in extreme value theory: pwm estimators. The Ann. of Stat..43 (1), 276-298. doi: 10.1214/14-AOS1280.
12. Fisher RA, Tippett LHC (1928) Limiting forms of the frequency distribution of the largest or smallest member of a sample Proceedings of the Cambridge Philosophical Society, 24, pp. 180–190.
13. Gencay R, Selcuk F (2004) Extreme value theory and Value-at-Risk: Relative performance in emerging markets. Int J of Fore. 20.2, pp. 287-303.
14. Gnedenko R (1943) Sur la distribution limite du terme maximum d'une serie aleatoire Annals of Mathematics, 44, pp. 423–453.
15. Gilleland E, Nychka D (2005) Statistical models for monitoring and regulating ground-level ozone Environmetrics, 16 ,pp. 535–546.
16. Grimshaw S (1993) Computing Maximum Likelihood Estimates for the Generalized Pareto Distribution. Technometrics , 35(2), 185-191. doi:10.2307/126966.3.
17. Güler Ç,Cobanoglu Z (1994) Outdoor Air Pollution.Ministry of Health of Turkey Publications.
18. Gumbel EM (1954) Statistical Theory of Exreme Values and some Practical Observations. National Bureau of Standards, Appl. Math. Series 33, U.S. Government Printing Office.
19. Gumbel EJ (1958) Statistics of Extremes, Columbia University Press, Columbia, p. 395.20 Horowitz J (1980) Extreme values from a non–stationary stochastic process: an application to air quality analysis Technometrics, 22, pp. 469–478.21 Hosking JRM, Wallis JR, Wood EF (1985) Estimation of the generalized extreme-value distribution by the method of probability-weighted moments Technometrics, 27, pp. 251–261.
20. ISSE (2009) Institute For the Study of Society and Environment.https://www.isse.ucar.edu/. Accessed 01.11.2016.
21. Joe H (1994) Multivariate extreme-value distributions with applications to environmental data. The Canad. J of Statistics. 22 (1). doi.https:10.2307/3315822.
22. Kottegoda NT, Rosso R (1997) Statistics, Probability and Reliability for Civil and Environmental Engineers. McGraw-Hill.
23. Kysely J, Picek J and Beranova R (2010) Estimating extremes in climate change simulations using the peaks-over-threshold method with a non-stationary threshold, Glo. and Planet. Chan. 72, (1-2), 55-68.doi:10.1016/j.gloplacha.2010.03.006.
24. Jakeman AJ, Bai J, Miles GH (1991) Prediction of seasonal extremes of one-hour average urban CO concentrations Atmos. Environ. Part B-Urban Atmosphere, 25, pp. 219–229.
25. Luceño A., Menéndez M, Méndez FJ (2006), The effect of temporal dependence on the estimation of the frequency of extreme ocean climate events, Proc. R. Soc., 462, 1683–1697.
26. Lu HC, Fang GC (2003) Predicting the exceedances of a critical PM10 concentrationan - a case study in Taiwan Atmos. Environ. 37, pp. 3491–3499
27. Mehrannia H, Pakgohar A (2014) Using easy Fit Software For Goodness-of-Fit Test and Data Generation.Int. J of Math. Arc.-5(1), 2014,pp. 118-124.
28. Mc Neil AJ. Saladin T (1997). The peaks over thresholds method for estimating high quantiles of loss distributions. Proceedings of 28th International ASTIN Colloquium.
29. Myung IJ (2001). Tutorial on maximum likelihood estimation". J of Math Psychology (47), 90-100.
30. Pickands J (1975) Statistical inference using extreme order statistics, Annals of Statistics, 3, 119–131.
31. Rieder HE (2014) Extreme Value Theory: A primer.http://www.ldeo.columbia.edu/~amfiore/eescG9910_f14_ppts/Rieder_EVTPrimer.pdf. Accessed 04.12.2016.
32. Roberts EM. (1979a) Review, of statistics of extreme values with applications to air quality data. Part I: review. J of the Air Pol Cont. Assoc. 29, 632–637.
33. Roberts, EM. (1979b) Review, of statistics of extreme values with applications to air quality data. Part II: application. J of the Air Pol Cont. Assoc. 29, 733–740. 36 Schittkowski K (2003)EASY-FIT: a software system for data fitting in dynamical systems. Struct Multidisc Optim.23: 153. doi:10.1007/s00158-002-0174-6.
34. Sfetsos A, Zoras S, Bartzis JG, Triantafyllou AG (2006) Extreme value modeling of daily PM10concentrations in an industrial area. Fresenius Environmental Bulletin, 15, pp. 841–845.
35. Sharma P, Khare M, Chakrabarti SP (1999) Application of extreme value theory for predicting violations of air quality standards for an urban road intersection Transportation Research Part D. Transport and Environment, 4, pp. 201–216.
36. Shively TS (1990) An analysis of the long-term trend in ozone data from 2 Houston, Texas monitoring sites. Atmos. Environ. Part B-Urban Atmosphere, 24, pp. 293–301.
37. Smith RL, Huang L (1993) Modelling High Threshold Exceedances of Urban Ozone, Technical Report 6,National Institute of Statistical Sciences, Research Triangle Park.
38. Smith RL, Shively TS (1995) Point process approach to modeling trends in tropospheric ozone based on exceedances of a high-threshold Atmos Environ. 29, pp. 3489–3499.
39. Smith RL (1989) Extreme value analysis of environmental time series: an application to trend detection in ground level ozone. Stats Sci, 4, pp. 367–377.
40. Surman PG, Bodero J, Simpson RW (1987) The prediction of the numbers of violations of standards and the frequency of air-pollution episodes using extreme value theory. Atmos Environ, 21, pp. 1843–1848.
41. Wang QJ (1991) The POT model described by the generalized Pareto distribution with Poisson arrival rate. J of Hydro 129.1: 263-280.
42. Yolsal H (2016) Estimatıon of the Air Quality Trends in İstanbul. Marmara Univ J. 38, (1).doi: 10.14780/iibd.98771.