MODELLING THE PM2.5 CONCENTRATION WITH ARTIFICIAL INTELLIGENCE-BASED ENSEMBLE APPROACH

Abstract

Fine particulate matter (PM2.5) has been linked to a number of adverse health effects, hence its prediction for epidemiological studies has become very crucial. In this study, a novel ensemble technique was proposed for the prediction of PM2.5 concentration in cities with high traffic noise using traffic noise as an input parameter. Air pollutants concentration (P), meteorological parameters (M) and traffic data (T) simultaneously collected from seven sampling points in North Cyprus were used for conducting the study. The modelling was done in 2 scenarios. In scenario I, PM2.5 was modelled using 4 different input combination without traffic noise as input parameter while in scenario II, traffic noise was added as an input variable for 4 input combinations. The models were evaluated using 4 performance criteria including Nash-Sutcliffe Efficiency (NSE), Root Mean Square Error (RMSE), Correlation Coefficient (CC) and Bias (BIAS). Modelling PM2.5 with combined relevant input parameters of P, M and T could improve the performance of the model developed with only one set of the parameters by up to 12, 17 and 29% for models containing only P, M and T respectively. All the models in scenario II have demonstrated high prediction accuracy than the corresponding model in scenario I by up to 12% in the verification stage. The Support Vector Regression-based Ensemble model (SVR-E) could improve the performance accuracy of single models by up to 17% in the verification stage.

Keywords

• Traffic noise
• Sensitivity analysis
• Ensemble modelling
• Air quality

Öz

İnce partikül madde (PM2.5) bir dizi olumsuz sağlık etkisi ile ilişkilendirilmiştir, bu nedenle epidemiyolojik çalışmalar için öngörüsü çok önemli hale gelmiştir. Bu çalışmada, giriş parametresi olarak trafik gürültüsü kullanılarak trafik gürültüsü yüksek şehirlerde PM2.5 konsantrasyonunun tahmini için yeni bir topluluk tekniği önerilmiştir. Çalışmanın yürütülmesi için Kuzey Kıbrıs'taki yedi örnekleme noktasından eş zamanlı olarak toplanan hava kirletici konsantrasyonu (P), meteorolojik parametreler (M) ve trafik verileri (T) kullanılmıştır. Modelleme 2 senaryoda yapılmıştır. Senaryo I'de PM2.5, trafik gürültüsü olmadan 4 farklı giriş kombinasyonu kullanılarak giriş parametresi olarak modellenirken, senaryo II'de trafik gürültüsü 4 giriş kombinasyonu için giriş değişkeni olarak eklenmiştir. Modeller, Nash-Sutcliffe Verimliliği (NSE), Ortalama Kare Hatası (RMSE), Korelasyon Katsayısı (CC) ve Bias (BIAS) olmak üzere 4 performans kriteri kullanılarak değerlendirildi. PM2.5'in ilgili P, M ve T girdi parametreleriyle modellenmesi, yalnızca bir parametre seti ile geliştirilen modelin performansını yalnızca P, M ve T içeren modeller için sırasıyla %12, 17 ve %29'a kadar iyileştirebilir. Senaryo II'deki tüm modeller, doğrulama aşamasında senaryo I'deki karşılık gelen modelden %12'ye kadar yüksek tahmin doğruluğu göstermiştir. Support Vector Regresyon tabanlı Ensemble modeli (SVR-E), doğrulama aşamasında tekli modellerin performans doğruluğunu %17'ye kadar artırabilir.

References

1. 1. Arhami, M., Kamali, N. & Rajabi, M.M. 2013. Predicting hourly air pollutant levels using artificial neural networks coupled with uncertainty analysis by Monte Carlo simulations. Environmental Science & Pollution Research, 207: 4777-4789.
2. 2. Cai, M., Yin, Y. & Xie, M. 2009. Prediction of hourly air pollutant concentrations near urban arterials using artificial neural network approach. Transportation Research Part D: Transport & Environment, 141: 32-41.
3. 3. Danciulescu, V. & Bucur, E. 2015. Correlations between noise level & pollutants concentration in order to assess the level of air pollution induced by heavy traffic pollutants concentration in order to assess the level of air pollution induced by heavy. Journal of Environmental Protection & Ecology, 163: 815-823.
4. 4. Doǧan, E. & Akgüngör, A. P. 2013. Forecasting highway casualties under the effect of railway development policy in Turkey using artificial neural networks. Neural Computing & Applications, 225: 869-877.
5. 5. Dong, G.-H., Zhang, P., Sun, B., Zhang, L., Chen, X., Ma, N., Yu, F., Guo, H., Huang, H., Lee, Y.L., Tang, N. & Chen, J. 2012. Long-Term Exposure to Ambient Air Pollution & Respiratory Disease Mortality in Shenyang, China: A 12-Year Population-Based Retrospective. Respiration, 84: 360-368.
6. 6. Dunea, D., Pohoata, A. & Iordache, S. 2015. Using wavelet–feedforward neural networks to improve air pollution forecasting in urban environments. Environmental Monitoring & Assessment: 1877.
7. 7. Elkiran, G., Nourani, V., Abba, S.I. & Abdullahi, J. 2018. Artificial intelligence-based approaches for multi-station modelling of dissolve oxygen in river. Global Journal of Environmental Science & Management, 44: 439-450.
8. 8. European Environment Agency. 2012. Road Traffic’s Contribution to Air Quality in European Cities, ETC/ACM Technical Paper 2012/14.

9. 9. European Environment Agency. 2014. EEA Report No 10/2014 - Noise in Europe 2014.
10. 10. Gan, W.Q., McLean, K., Brauer, M., Chiarello, S.A. & Davies, H.W. 2012. Modeling population exposure to community noise & air pollution in a large metropolitan area. Environmental Research, 116: 11-16.
11. 11. Ghaffari, A., Abdollahi, H., Khoshay, M., Bozchalooi, I., Dadgar, A. & Rafiee-Tehrani, M. 2006. Performance comparison of neural network training algorithms in modeling of bimodal drµg delivery. International Journal of Pharmaceutics, 327: 126-138.
12. 12. Jahani, B. & Mohammadi, B. 2019. A comparison between the application of empirical & ANN methods for estimation of daily global solar radiation in Iran. Theoretical & Applied Climatology, 137: 1257-1269.
13. 13. Khan, J., Kakosimos, K., Solvang, S., Hertel, O. & Sørensen, M. 2020. Science of the Total Environment The spatial relationship between traffic-related air pollution & noise in two Danish cities: Implications for health-related studies. Science of the Total Environment, 726: 138577.
14. 14. Khan, J., Ketzel, M., Kakosimos, K., Sørensen, M. & Jensen, S.S. 2018. Road traffic air & noise pollution exposure assessment – A review of tools & techniques. Science of the Total Environment, 634: 661-676.
15. 15. Klingberg, J., Broberg, M., Strandberg, B., Thorsson, P. & Pleijel, H. 2017. Influence of urban vegetation on air pollution & noise exposure – A case study in Gothenburg, Sweden. Science of the Total Environment, 599: 1728-1739.
16. 16. Kumar, P., Nigam, S.P. & Kumar, N. 2014. Vehicular traffic noise modeling using artificial neural network approach. Transportation Research Part C: Emerging Technologies, 40: 111-122.
17. 17. Lin, M., Guo, Y., Chen, Y., Chen, W., Young, L., Lee, K., Wu, Z. & Tsai, P. 2018. An instantaneous spatiotemporal model for predicting traffic-related ultra-fine particle concentration through mobile noise measurements. Science of the Total Environment, 636: 1139-1148.
18. 18. Ljungman, P. L. & Mittleman, M. A. 2014. Ambient air pollution & stroke. Stroke, 4512: 3734-3741.
19. 19. Lou, C., Liu, H., Li, Y., Peng, Y., Wang, J. & Dai, L. 2017. Relationships of relative humidity with PM2.5 & PM10 in the Yangtze River Delta, China. Environmental Monitoring & Assessment, 18911.
20. 20. Maschke, C., & Widmann, U. 2013. The Effects of Sound on Humans, pp. 69-86. In Müller, G. & Möser, M. (eds.), Handbook of Engineering Acoustics. Springer-Verlag Berlin Heidelberg, x + 576 pp.
21. 21. Moriasi, D.N., Arnold, J.G., Liew, M.W. Van, Bingner, R.L., Harmel, R.D. & Veith, T.L. 2007. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 503: 885-900.
22. 22. Newby, D.E., Mannucci, P.M., Tell, G.S., Baccarelli, A.A., Brook, R.D., Donaldson, K., Forastiere, F., Franchini, M., Franco, O.H., Graham, I., Hoek, G., Hoffmann, B., Mills, N., Pekkanen, J., Peters, A., Piepoli, M.F., Rajagopalan, S. & Storey, R.F. 2015. Expert position paper on air pollution & cardiovascular disease. European Heart Journal, 36: 83-93.
23. 23. Nourani, V. & Fard Sayyah, M. 2012. Sensitivity Analysis of the artificial neural network outputs in simulation of the evaporation process at different climatologic regimes. Advances in Engineering Software, 47: 127-129.
24. 24. Nourani, V. & Sharghi, Z.A.E. 2020. Sensitivity analysis & ensemble artificial intelligence ‑ based model for short ‑ term prediction of ­ NO2 concentration. International Journal of Environmental Science & Technology, 18: 2703-2722.
25. 25. Nourani, V., Gökçekuş, H. & Umar, I.K. 2020a. Artificial intelligence-based ensemble model for prediction of vehicular traffic noise. Environmental Research, 180: 108852.
26. 26. Nourani, V., Gökçekus, H., Umar, I.K. & Najafi, H. 2020b. An emotional artificial neural network for prediction of vehicular traffic noise. Science of the Total Environment, 707: 136134.
27. 27. Ilgurel, N., Akdag, N.Y. & Akdag, A. 2016. Evaluation of noise exposure before & after noise barriers, a simulation study in Istanbul. Journal of Environmental Engineering and Landscape Management, 244: 293-302.
28. 28. Raaschou-nielsen, O., Andersen, Z.J., Beelen, R., Samoli, E., Stafoggia, M., Weinmayr, G., Hoff, B. & Fischer, P. 2013. Air pollution and lung cancer incidence in 17 European cohorts: prospective analyses from the European Study of Cohorts for Air Pollution Effects ESCAPE. The Lancet Oncology, 149: 813-822.
29. 29. Rumelhart, D.E., Hinton, G.E. & Williams, R. 1986. Learning representations by backpropagating errors. Nature, 323: 533-536.
30. 30. Schlittmeier, S., Feil, A., Liebl, A. & Hellbrück, J. 2015. The impact of road traffic noise on cognitive performance in attention-based tasks depends on noise level even within moderate-level ranges. Noise & Health, 1776: 148.
31. 31. Sørensen, M., Andersen, Z.J., Nordsborg, R.B., Becker, T., Tjønneland, A., Overvad, K. & Raaschou-Nielsen, O. 2013. Long-term exposure to road traffic noise & incident diabetes: a cohort study. Environmental Health Perspectives, 1212: 217-22.
32. 32. Suleiman, A., Tight, M.R. & Quinn, A.D. 2016. Hybrid Neural Networks & Boosted Regression Tree Models for Predicting Roadside Particulate Matter. Environmental Modeling & Assessment, 216: 731-750.
33. 33. Suleiman, A., Tight, M.R. & Quinn, A.D. 2019. Applying machine learning methods in managing urban concentrations of traffic-related particulate matter PM10 & PM2.5. Atmospheric Pollution Research, 101: 134-144.
34. 34. Sun, W. & Li, Z. 2020. Hourly PM2.5 concentration forecasting based on feature extraction & stacking-driven ensemble model for the winter of the Beijing-Tianjin-Hebei area. Atmospheric Pollution Research, 116: 110-121.
35. 35. Taylor, K. E. 2001. Summarizing multiple aspects of model performance in a single diagram. Journal of Geophysical Research, 106: 7183-7192.
36. 36. Tenailleau, Q.M., Bernard, N., Pujol, S. & Parmentier, A. 2016. Do outdoor environmental noise and atmospheric NO2 levels spatially overlap in urban areas? Environmental Pollution, 214: 767-775.
37. 37. Umar, I. K., Nourani, V. & Gokcekus, H. 2021. A novel multi-model data-driven ensemble approach for the prediction of particulate matter concentration. Environmental Science & Pollution Research, 28: 49663-49677.
38. 38. Uzoigwe, J.C., Prum, T., Bresnahan, E. & Garelnabi, M. 2013. The emerging role of outdoor & indoor air pollution in cardiovascular disease. North American Journal of Medical Sciences, 58: 445-453.
39. 39. Van D.A., Martin R.V. & Park R.J. 2006. Estimating ground-level PM2.5 using aerosol optical depth determined from satellite remote sensing. Journal of Geophysical Research, 111:1-10.
40. 40. Volk, H.E., Lurmann, F., Penfold, B., Hertz-Picciotto, I. & McConnell, R. 2013. Traffic-related air pollution, particulate matter, & autism. JAMA Psychiatry, 701: 71-77.
41. 41. Wang, J. & Ogawa, S. 2015. Effects of meteorological conditions on PM2.5 concentrations in Nagasaki, Japan. International Journal of Environmental Research & Public Health, 128: 9089-9101.
42. 42. Wang, W., Xu, D., Chau, K.W. & Chen, S. 2015. Improved annual rainfall-runoff forecasting using PSO–SVM model based on EEMD. Journal of Hydroinformatics, 154: 1377-1390.
43. 43. Wang, X., Zhang, R. & Yu, W. 2019. The Effects of PM2.5 Concentrations & Relative Humidity on Atmospheric Visibility in Beijing. Journal of Geophysical Research: Atmospheres, 1244: 2235-2259.
44. 44. Whalley J., Zandi S. 2016. Particulate Matter Sampling Techniques and Data Particulate Matter Sampling Techniques and Data Modelling Methods. pp. 29-54. In Sallis, P. (ed.) Air Quality - Measurement and Modeling. INTECH open science, Croatia, xii + 185 pp.
45. 45. Winkler, R.L. & Makridakis, S. 1983. The Combination of Forecasts. Journal of the Royal Statistical Society, 1462: 150-157.
46. 46. Yangyang, X., Bin, Z., Lin, Z. & Rong, L. 2015. Spatiotemporal variations of PM2.5 & PM10 concentrations between 31 Chinese cities and their relationships with SO2, NO2, CO & O3. Particuology, Chinese Society of Particuology, 20: 141-149.
47. 47. Yaseen, Z.M., Deo, R.C., Hilal, A., Abd, A.M., Bueno, L.C., Salcedo-Sanz, S. & Nehdi, M.L. 2018. Predicting compressive strength of lightweight foamed concrete using extreme learning machine model. Advances in Engineering Software, 115: 112-125.