Prediction of size-fractionated airborne particle-bound metals using MLR, BP-ANN and SVM analyses.

Size-fractionated heavy metal concentrations were observed in airborne particulate matter (PM) samples collected from 2014 to 2015 (spanning all four seasons) from suburban (Xianlin) and industrial (Pukou) areas in Nanjing, a megacity of southeast China. Rapid prediction models of size-fractionated metals were established based on multiple linear regression (MLR), back propagation artificial neural network (BP-ANN) and support vector machine (SVM) by using meteorological factors and PM concentrations as input parameters. About 38% and 77% of PM2.5 concentrations in Xianlin and Pukou, respectively, were beyond the Chinese National Ambient Air Quality Standard limit of 75 μg/m3. Nearly all elements had higher concentrations in industrial areas, and in winter among the four seasons. Anthropogenic elements such as Pb, Zn, Cd and Cu showed larger percentages in the fine fraction (ø≤2.5 μm), whereas the crustal elements including Al, Ba, Fe, Ni, Sr and Ti showed larger percentages in the coarse fraction (ø > 2.5 μm). SVM showed a higher training correlation coefficient (R), and lower mean absolute error (MAE) as well as lower root mean square error (RMSE), than MLR and BP-ANN for most metals. All the three methods showed better prediction results for Ni, Al, V, Cd and As, whereas relatively poor for Cr and Fe. The daily airborne metal concentrations in 2015 were then predicted by the fully trained SVM models and the results showed the heaviest pollution of airborne heavy metals occurred in December and January, whereas the lightest pollution occurred in June and July.

[1]  Szymon Hoffman,et al.  Short-Time Forecasting of Atmospheric NOx Concentration by Neural Networks , 2006 .

[2]  Hui-ming Li,et al.  Heavy metals in atmospheric particulate matter: a comprehensive understanding is needed for monitoring and risk mitigation. , 2013, Environmental science & technology.

[3]  E. Fireman,et al.  Respiratory hospitalizations of children living near a hazardous industrial site adjusted for prevalent dust: a case-control study. , 2015, International journal of hygiene and environmental health.

[4]  Trace elements in size-segregated urban aerosol in relation to the anthropogenic emission sources and the resuspension , 2014, Environmental Science and Pollution Research.

[5]  J. Chow,et al.  A hybrid ARIMA and artificial neural networks model to forecast particulate matter in urban areas: The case of Temuco, Chile , 2008 .

[6]  Costas Papaloukas,et al.  Assessment and prediction of exposure to benzene of filling station employees , 2007 .

[7]  Yu Qin,et al.  Association of cardiopulmonary health effects with source-appointed ambient fine particulate in Beijing, China: a combined analysis from the Healthy Volunteer Natural Relocation (HVNR) study. , 2014, Environmental science & technology.

[8]  Ali S. Kamal,et al.  Insights Into the Mechanisms and Mediators of the Effects of Air Pollution Exposure on Blood Pressure and Vascular Function in Healthy Humans , 2009, Hypertension.

[9]  P. Khadikar,et al.  Prediction of intrinsic solubility of generic drugs using MLR, ANN and SVM analyses. , 2010, European journal of medicinal chemistry.

[10]  Jing-chun Duan,et al.  Characteristics of size-fractionated atmospheric metals and water-soluble metals in two typical episodes in Beijing , 2015 .

[11]  Ari Karppinen,et al.  Evaluation of a multiple regression model for the forecasting of the concentrations of NOx and PM10 in Athens and Helsinki. , 2011, The Science of the total environment.

[12]  Yufang Wang,et al.  A novel hybrid decomposition-and-ensemble model based on CEEMD and GWO for short-term PM2.5 concentration forecasting , 2016 .

[13]  Gokmen Tayfur,et al.  Forecasting Ambient Air SO2 Concentrations Using Artificial Neural Networks , 2006 .

[14]  Jichun Wu,et al.  Size Distribution and Source Apportionment of Airborne Metallic Elements in Nanjing, China , 2013 .

[15]  D. Sarigiannis,et al.  Total exposure to airborne particulate matter in cities: the effect of biomass combustion. , 2014, The Science of the total environment.

[16]  Yu Qin,et al.  Chemical constituents of fine particulate air pollution and pulmonary function in healthy adults: the Healthy Volunteer Natural Relocation study. , 2013, Journal of hazardous materials.

[17]  D. Dockery,et al.  Increased Particulate Air Pollution and the Triggering of Myocardial Infarction , 2001, Circulation.

[18]  Huiming Li,et al.  Chemical characterization and source apportionment of PM2.5 aerosols in a megacity of Southeast China , 2016 .

[19]  Fei Song,et al.  Size distributions of trace elements associated with ambient particular matter in the affinity of a major highway in the New Jersey–New York metropolitan area , 2011 .

[20]  Kazuhiko Sakamoto,et al.  Estimation of the contribution of ultrafine particles to lung deposition of particle-bound mutagens in the atmosphere. , 2011, The Science of the total environment.

[21]  Richard Jensen,et al.  QSRR modeling for diverse drugs using different feature selection methods coupled with linear and nonlinear regressions. , 2012, Journal of chromatography. B, Analytical technologies in the biomedical and life sciences.

[22]  J. Ondov,et al.  Accumulation of metals, trace elements and semi-volatile organic compounds on exterior window surfaces in Baltimore. , 2003, Environmental pollution.

[23]  J M Gorriz,et al.  Prediction of CO maximum ground level concentrations in the Bay of Algeciras, Spain using artificial neural networks. , 2008, Chemosphere.

[24]  T. Nawrot,et al.  Acute changes in pulse pressure in relation to constituents of particulate air pollution in elderly persons. , 2012, Environmental research.

[25]  X. Querol,et al.  Size distribution and chemical composition of metalliferous stack emissions in the San Roque petroleum refinery complex, southern Spain. , 2011, Journal of hazardous materials.

[26]  Marco Casazza,et al.  Size resolved metal distribution in the PM matter of the city of Turin (Italy). , 2016, Chemosphere.

[27]  Dezhi Sun,et al.  Ozone concentration forecast method based on genetic algorithm optimized back propagation neural networks and support vector machine data classification , 2011 .

[28]  Spyros P. Karakitsios,et al.  Bayesian Algorithm Implementation in a Real Time Exposure Assessment Model on Benzene with Calculation of Associated Cancer Risks , 2009, Sensors.

[29]  G. S. Kapur,et al.  Establishing Quantitative Structure−Property Relationships (QSPR) of Diesel Samples by Proton-NMR & Multiple Linear Regression (MLR) Analysis , 2001 .

[30]  Hui-ming Li,et al.  Chemical speciation and human health risk of trace metals in urban street dusts from a metropolitan city, Nanjing, SE China. , 2013, The Science of the total environment.

[31]  Roy M Harrison,et al.  Estimation of the contributions of brake dust, tire wear, and resuspension to nonexhaust traffic particles derived from atmospheric measurements. , 2012, Environmental science & technology.

[32]  X. Y. Zhang,et al.  Application of support vector machine (SVM) for prediction toxic activity of different data sets. , 2006, Toxicology.

[33]  Ian G. McKendry,et al.  Evaluation of Artificial Neural Networks for Fine Particulate Pollution (PM10 and PM2.5) Forecasting , 2002, Journal of the Air & Waste Management Association.

[34]  C. Yang,et al.  Chemical speciation of fine particle bound trace metals , 2009 .

[35]  Sancho Salcedo-Sanz,et al.  Prediction of hourly O3 concentrations using support vector regression algorithms , 2010 .

[36]  Bijan Yeganeh,et al.  Prediction of CO concentrations based on a hybrid Partial Least Square and Support Vector Machine model , 2012 .

[37]  B. Xing,et al.  Size distribution and anthropogenic sources apportionment of airborne trace metals in Kanazawa, Japan. , 2006, Chemosphere.

[38]  Xiangping Zhang,et al.  A quantitative prediction of the viscosity of ionic liquids using S(σ-profile) molecular descriptors. , 2015, Physical chemistry chemical physics : PCCP.

[39]  Wei-Zhen Lu,et al.  Potential assessment of the "support vector machine" method in forecasting ambient air pollutant trends. , 2005, Chemosphere.

[40]  Jens Sadowski,et al.  Comparison of Support Vector Machine and Artificial Neural Network Systems for Drug/Nondrug Classification , 2003, J. Chem. Inf. Comput. Sci..

[41]  P. J. García Nieto,et al.  Application of an SVM-based regression model to the air quality study at local scale in the Avilés urban area (Spain) , 2011, Math. Comput. Model..

[42]  A. Peters,et al.  Short-term Impact of Ambient Air Pollution and Air Temperature on Blood Pressure Among Pregnant Women , 2011, Epidemiology.

[43]  Gavin C. Cawley,et al.  Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations, compared with a deterministic modelling system and measurements in central Helsinki , 2003 .

[44]  Luc Int Panis,et al.  Blood Pressure and Same-Day Exposure to Air Pollution at School: Associations with Nano-Sized to Coarse PM in Children , 2015, Environmental health perspectives.