PM10 Parametresinin Makine Öğrenmesi Algoritmalari ile Mekânsal Analizi, Kayseri İli Örneği

Abstract

Hava kirliliğinin son yıllarda artışı ile alınacak olan erken önlemler dâhilinde hava kirliliği tahmininin yapılması insan ve çevre sağlığına verilebilecek zararın en aza indirilmesinde önemlidir. Bu çalışmada günlük ortalama hava kirliliği miktarının, önemli bir hava kirletici olan PM10 konsantrasyonu üzerinden tahminlenmesi ve hava kirliliğinin çevresel ve mekânsal modellenmesi amaçlanmıştır. Tahminleme modeli, Orta Anadolu Bölgesinde yer alan Kayseri ilinde bulunan 3 istasyondan alınan 2010-2018 yılları arasında ölçülen PM10 konsantrasyonu verileri kullanılarak makine öğrenmesi algoritmaları (kNN DVR, RF, ANN, Lineer Regresyon) ile eğitilmiştir. Kayseri’deki 3 istasyonun 2010-2018 yılları arasındaki PM10 konsantrasyon değerleri girdi olarak verilmiş ve 2019 yılına ait PM10 konsantrasyon değerleri tahmin edilmiştir. En iyi sonuçlar 3 istasyon için de Destek Vektör Regresyonu algoritması ile elde edilmiş olup Trafik bölgesi için R2:0.85, RMSE:17.57, MAE:10.17; Hürriyet bölgesi için R2:0.73, RMSE:34.91, MAE:24.61 ve OSB bölgesi için R2:0.82, RMSE:41.71, MAE:21.62 olarak tespit edilmiştir. Ayrıca elde edilen tahmini konsantrasyon sonuçlarının mekânsal dağılımı (CBS) ve değişimi de analiz edilmiştir.

Keywords

• Makine öğrenmesi algoritmaları
• PM10
• destek vektör algoritması
• CBS
• mekânsal analiz

Spatial Analysis of PM10 Parameter by Machine Learning Algorithms, City of Kayseri

Abstract

With the increase of air pollution in recent years, it is important to make an estimation of air pollution within the scope of early measures to be taken in minimizing the damage to human and environmental health. In this study, it is aimed to estimate the daily average amount of air pollution from the PM10 concentration, which is an important air pollutant, and to model the environmental and spatial air pollution. The prediction model was trained with machine learning algorithms (kNN DVR, RF, ANN, Linear Regression) using PM10 (particulate matter) concentration data measured between 2010-2018 from 3 stations in Kayseri, Central Anatolia. PM10 concentration values of 3 stations in Kayseri between 2010-2018 were given as input and PM10 concentration values for 2019 were estimated. The best results were obtained with Support Vector Regression algorithm for all three stations. For the Traffic region, R2: 0.85, RMSE: 17.57, MAE: 10.17; For Hurriyet region, R2: 0.73, RMSE: 34.91, MAE: 24.61 and for OSB region R2: 0.82, RMSE: 41.71, MAE: 21.62. Also, the spatial distribution and variation of the estimated concentration results were analyzed by the Geographical Information System (GIS).

Keywords

• Machine learning algorithms
• PM10
• Support vector algorithm
• GIS
• spatial analysis

References

1. A. Alkan, Hava Kirliliğinin Ciddi Boyutlara Ulaştığı Kentlere Bir Örnek : An Example of Cıtıes Where Aır Pollutıon Has Reached Serıous Dımensıons : Siirt, (2018) 641–666.
2. Y. DOKUZ, A. BOZDAĞ, B. GÖKÇEK, HavKali̇tesiParametreleri̇ni̇Tahmi̇ni̇ Ve Mekansal Dağilimİçi̇nMaki̇neÖğrenmesiYöntemleri̇ni̇n Kullanilmasi, Ömer Halisdemir Üniversitesi Mühendislik Bilim. Derg. 9 (2020) 37–47. https://doi.org/10.28948/ngumuh.654092.
3. G.Ç. SÜMER, Hava Kirliği Kontrolü: Türkiye’de Hava Kirliliğini Önlemeye Yönelik Yasal Düzenlemele-rin ve Örgütlenmelerin İncelenmesi, Uluslararası İktisadi ve İdari İncelemeler Derg. 13 (2014) 37. https://doi.org/10.18092/ijeas.51643.
4. A. Rahimpour, J. Amanollahi, C.G. Tzanis, Air quality data series estimation based on machine learning approaches for urban environments, Air Qual. Atmos. Heal. 14 (2021) 191–201. https://doi.org/10.1007/s11869-020-00925-4.
5. C. Amuthadevi, D.S. Vijayan, V. Ramachandran, Development of air quality monitoring (AQM) models using different machine learning approaches, J. Ambient Intell. Humaniz. Comput. (2021). https://doi.org/10.1007/s12652-020-02724-2.
6. H. Tian, Y. Zhao, M. Luo, Q. He, Y. Han, Z. Zeng, Estimating PM2.5 from multisource data: A comparison of different machine learning models in the Pearl River Delta of China, Urban Clim. 35 (2021) 100740. https://doi.org/10.1016/j.uclim.2020.100740.
7. Y. Son, Á.R. Osornio-vargas, M.S.O. Neill, P. Hystad, J.L. Texcalac-sangrador, P. Ohman-strickland, Q. Meng, S. Schwander, Land use regression models to assess air pollution exposure in Mexico City using fi ner spatial and temporal input parameters, Sci. Total Environ. 639 (2018) 40–48. https://doi.org/10.1016/j.scitotenv.2018.05.144.
8. Y. Xu, H. Liu, Z. Duan, A novel hybrid model for multi-step daily AQI forecasting driven by air pollution big data, Air Qual. Atmos. Heal. 13 (2020) 197–207. https://doi.org/10.1007/s11869-020-00795-w.

9. S. Van Roode, J.J. Ruiz-Aguilar, J. González-Enrique, I.J. Turias, An artificial neural network ensemble approach to generate air pollution maps, Environ. Monit. Assess. 191 (2019). https://doi.org/10.1007/s10661-019-7901-6.
10. J.Y. Yang, W.F. Ip, C.M. Vong, P.K. Wong, Effect of choice of kernel in support vector machines on ambient air pollution forecasting, Proc. 2011 Int. Conf. Syst. Sci. Eng. ICSSE 2011. (2011) 552–557. https://doi.org/10.1109/ICSSE.2011.5961964.
11. W. Lu, W. Wang, A.Y.T. Leung, S.M. Lo, R.K.K. Yuen, Z. Xu, H. Fan, Air pollutant parameter forecasting using support vector machines, Proc. Int. Jt. Conf. Neural Networks. 1 (2002) 630–635. https://doi.org/10.1109/ijcnn.2002.1005545.
12. A. Sotomayor-Olmedo, M.A. Aceves-Fernández, E. Gorrostieta-Hurtado, C. Pedraza-Ortega, J.M. Ramos-Arreguín, J.E. Vargas-Soto, Forecast Urban Air Pollution in Mexico City by Using Support Vector Machines: A Kernel Performance Approach, Int. J. Intell. Sci. 03 (2013) 126–135. https://doi.org/10.4236/ijis.2013.33014.
13. H. Sun, D. Gui, B. Yan, Y. Liu, W. Liao, Y. Zhu, C. Lu, Assessing the potential of random forest method for estimating solar radiation using air pollution index, Energy Convers. Manag. 119 (2016) 121–129. https://doi.org/10.1016/j.enconman.2016.04.051.
14. D. Kumar, ScienceDirect ScienceDirect Evolving Differential evolution method with random forest for Evolving Differential evolution method with random forest for prediction of Air Pollution prediction of Air Pollution, Procedia Comput. Sci. 132 (2018) 824–833. https://doi.org/10.1016/j.procs.2018.05.094.
15. C. Gariazzo, G. Carlino, C. Silibello, M. Renzi, S. Finardi, N. Pepe, P. Radice, F. Forastiere, P. Michelozzi, G. Viegi, M. Stafoggia, A multi-city air pollution population exposure study: Combined use of chemical-transport and random-Forest models with dynamic population data, Sci. Total Environ. 724 (2020) 138102. https://doi.org/10.1016/j.scitotenv.2020.138102.
16. M.D. Mallet, Meteorological normalisation of PM10 using machine learning reveals distinct increases of nearby source emissions in the Australian mining town of Moranbah, Atmos. Pollut. Res. (2020) 0–1. https://doi.org/10.1016/j.apr.2020.08.001.
17. V. Chaudhary, A. Deshbhratar, V. Kumar, D. Paul, Time Series Based LSTM Model to Predict Air Pollutant ’ s Concentration for Prominent Cities in India, Udm’18. (2018).
18. C. Wen, S. Liu, X. Yao, L. Peng, X. Li, Y. Hu, T. Chi, A novel spatiotemporal convolutional long short-term neural network for air pollution prediction, Sci. Total Environ. 654 (2019) 1091–1099. https://doi.org/10.1016/j.scitotenv.2018.11.086.
19. Z. Qin, C. Cen, X. Guo, Prediction of Air Quality Based on KNN-LSTM, J. Phys. Conf. Ser. 1237 (2019). https://doi.org/10.1088/1742-6596/1237/4/042030.
20. A. Marjovi, A. Arfire, A. Martinoli, High resolution air pollution maps in urban environments using mobile sensor networks, Proc. - IEEE Int. Conf. Distrib. Comput. Sens. Syst. DCOSS 2015. (2015) 11–20. https://doi.org/10.1109/DCOSS.2015.32.
21. F. Taşpınar, Improving artificial neural network model predictions of daily average PM10 concentrations by applying principle component analysis and implementing seasonal models, J. Air Waste Manag. Assoc. 65 (2015) 800–809. https://doi.org/10.1080/10962247.2015.1019652.
22. B. Choubin, M. Abdolshahnejad, E. Moradi, X. Querol, A. Mosavi, S. Shamshirband, P. Ghamisi, Spatial hazard assessment of the PM10 using machine learning models in Barcelona, Spain, Sci. Total Environ. 701 (2020) 134474. https://doi.org/10.1016/j.scitotenv.2019.134474.
23. E. Sharma, R.C. Deo, R. Prasad, A. V. Parisi, A hybrid air quality early-warning framework: An hourly forecasting model with online sequential extreme learning machines and empirical mode decomposition algorithms, Sci. Total Environ. 709 (2020) 135934. https://doi.org/10.1016/j.scitotenv.2019.135934.
24. M. Stafoggia, T. Bellander, S. Bucci, M. Davoli, K. De Hoogh, F. De Donato, C. Gariazzo, A. Lyapustin, P. Michelozzi, M. Renzi, M. Scortichini, A. Shtein, G. Viegi, I. Kloog, J. Schwartz, Estimation of daily PM 10 and PM 2 . 5 concentrations in Italy , 2013 – 2015 , using a spatiotemporal land-use random-forest model, Environ. Int. 124 (2019) 170–179. https://doi.org/10.1016/j.envint.2019.01.016.
25. A. Ertürk, A. Ekdal, M. Gurel, K. Yuceil, A. Tanik, Use of mathematical models to estimate the effect of nutrient loadings on small streams, Fresen. Environ. Bull. 13 (2004) 1361–1370.
26. M. ATAOL, Burdur Gölü’nde Seviye Değişimleri, Co. 8 (2010) 077–092. https://doi.org/10.1501/cogbil_0000000105.
27. T. Yomralıoğlu, Coğrafi Bilgi Sistemleri: Temel Kavramlar ve Uygulamalar, Seçil Ofset, İstanbul, 2000.
28. S. İncekara, M. Karakuyu, A. Karaburun, ORTAÖĞRETİM CoğrafyaDersleri̇nde Yaparak ÖğrenmeyeBi̇r Örnek:CoğrafiBi̇lgSi̇stemleri̇ni̇n ProjeTemelli̇Öğreni̇mde Kullanilmasi, Elektron. Sos. Bilim. Derg. 8 (2009) 305–322. https://doi.org/10.17755/esosder.76302.
29. E. Kapluhan, COĞRAFİ BİLGİ SİSTEMLERİ’NİN(CBS) COĞRAFYA ÖĞRETİMİNDE KULLANIMININ ÖNEMİ VE GEREKLİLİĞİ, MARMARA COĞRAFYA DERGİSİ. 29 (2014) 34–59.
30. C. Güney, Mekansal zekanın getirdiği paradigma değişimi, J. Geod. Geoinf. 6 (2019) 128–142. https://doi.org/10.9733/jgg.2019r0006.t.
31. W.J. Requia, B.A. Coull, P. Koutrakis, Evaluation of predictive capabilities of ordinary geostatistical interpolation , hybrid interpolation , and machine learning methods for estimating PM 2 . 5 constituents over space, Environ. Res. 175 (2019) 421–433. https://doi.org/10.1016/j.envres.2019.05.025.
32. C.C. Lim, H. Kim, M.J.R. Vilcassim, G.D. Thurston, T. Gordon, L. Chen, K. Lee, M. Heimbinder, S. Kim, Mapping urban air quality using mobile sampling with low-cost sensors and machine learning in Seoul , South Korea, Environ. Int. 131 (2019) 105022. https://doi.org/10.1016/j.envint.2019.105022.
33. ÇED ve Çevre İzinleri Şube Müdürlüğü, Çevre ve Şehircilik İl Müdürlüğü Kayseri İli 2018 Yılı Çevre Durum Raporu, (2019).
34. G.H. Marks Hall, WEKA: Practical Machine Learning Tools and Techniques with JAva Implementations, (1994). https://researchcommons.waikato.ac.nz/bitstream/handle/10289/1040/uow-cs-wp-1999-11.pdf?sequence=1&isAllowed=y.
35. M. Yilmaz, R. Kanit, M. Erdal, S. Yildiz, A. Bakiş, K.H. Okulu, İ.M. Bölümü, Bina Bakım Onarım Ödeneklerinin Etkin Kullanımı Maksadıyla İhale Bedelini Etkileyen Faktörlerin Yapay Sinir Ağları ve Lineer Regresyon Yöntemleri ile Belirlenmesi Determination of The Factors Effecting The Tender Price by way of Artificial Neural Networks, 19 (2016) 461–470..
36. H. Drucker, C.J.C. Burges, L. Kaufman, A. Smola, V. Vapoik, W. Long, B. Nj, Support Vector Regression Machines w ) tw , 1 (n.d.).
37. L.E.O. Breiman, Random Forests, (2001) 5–32.
38. B. Lantz, Machine Learning with R, Packt Publishing Ltd, UK, 2013.
39. A. Jain, J. Mao, Artifical Neural Networks: A Tutorial., Comput. (Long. Beach. Calif). 29 (1996) 31–44.
40. J.Y. Son, M.L. Bell, J.T. Lee, Individual exposure to air pollution and lung function in Korea: Spatial analysis using multiple exposure approaches, Environ. Res. 110 (2010) 739–749. https://doi.org/10.1016/j.envres.2010.08.003.
41. M. Wu, J. Huang, N. Liu, R. Ma, Y. Wang, L. Zhang, A Hybrid Air Pollution Reconstruction by Adaptive Interpolation Method, in: Proc. 16th ACM Conf. Embed. Networked Sens. Syst., 2018: pp. 408–409.
42. W.J. Requia, M.D. Adams, A. Arain, S. Papatheodorou, P. Koutrakis, M. Mahmoud, Global Association of Air Pollution and Cardiorespiratory Diseases: A Systematic Review, Meta-Analysis, and Investigation of Modifier Variables, Am. J. Public Health. 108 (2018) S123–S130. https://doi.org/10.2105/AJPH.2017.303839.
43. K. Shukla, P. Kumar, G.S. Mann, M. Khare, Mapping spatial distribution of particulate matter using Kriging and Inverse Distance Weighting at supersites of megacity Delhi, Sustain. Cities Soc. 54 (2020) 101997. https://doi.org/10.1016/j.scs.2019.101997.
44. A. Bozdağ, Y. Dokuz, Ö.B. Gökçek, Spatial prediction of PM10 concentration using machine learning algorithms in Ankara, Turkey, Environ. Pollut. 263 (2020). https://doi.org/10.1016/j.envpol.2020.114635.
45. Url, Kayseri İl Kültür ve Turizm Müdürlüğü, (2020). https://kayseri.ktb.gov.tr/TR-54966/cografi-yapi.html.
46. M. Ashayeri, N. Abbasabadi, M. Heidarinejad, B. Stephens, Predicting intraurban PM2.5 concentrations using enhanced machine learning approaches and incorporating human activity patterns, Environ. Res. (2020) 110423. https://doi.org/10.1016/j.envres.2020.110423.
47. M. Lovrić, K. Pavlović, M. Vuković, S.K. Grange, M. Haberl, R. Kern, Understanding the true effects of the COVID-19 lockdown on air pollution by means of machine learning, Environ. Pollut. (2020). https://doi.org/10.1016/j.envpol.2020.115900.
48. S. Kumar, S. Mishra, S.K. Singh, A machine learning-based model to estimate PM2.5 concentration levels in Delhi’s atmosphere, Heliyon. 6 (2020) e05618. https://doi.org/10.1016/j.heliyon.2020.e05618.
49. A.O. Karababa, B. ATLI, Ç. Çağlayan, G. Varol, G. Ersoy, F. Gacal, N. Etiler, P. Özfırat, S. Ayta, Kara Rapor 2020: Hava Kirliliği ve Sağlık Etkileri, Temiz Hava Hakkı Platformu. (2020). https://www.temizhavahakki.com/kara-rapor/.
50. F.J. Chang, L.C. Chang, C.C. Kang, Y.S. Wang, A. Huang, Explore spatio-temporal PM2.5 features in northern Taiwan using machine learning techniques, Sci. Total Environ. 736 (2020) 139656. https://doi.org/10.1016/j.scitotenv.2020.139656.