High-resolution estimation of ambient sulfate concentration over Taiwan Island using a novel ensemble machine-learning model

Abstract

Heavy loadings of sulfate aerosol trigger haze formation and pose great damage to human health in Taiwan Island. Nevertheless, high-resolution spatiotemporal variation of ambient sulfate across Taiwan Island still remained unknown because of the scarce monitoring sites. Thus, we developed a novel ensemble model named extreme gradient boosting coupled with geographically and temporally weighted regression (XGBoost-GTWR) to predict the high-resolution sulfate concentration (0.05°) based on satellite data, assimilated meteorology, and the output of chemical transport models (CTMs). The result suggested that XGBoost-GTWR model outperformed other five models in predicting the sulfate concentration with the highest R2 value (R2 = 0.58) and the lowest relative mean square error (RMSE = 1.96 μg/m3). Besides, the transferability of the XGBoost-GTWR model was also validated based on the ground-level sulfate data in 2019. The result suggested that the R2 value of the extrapolation equation (0.53) did not show notable decrease compared with the 10-fold cross-validation result (0.58), indicating that the model was robust to predict the sulfate concentration. The ambient sulfate concentration in Taiwan Island displayed featured spatial variation with the highest one in Southwest Taiwan and the lowest one in Northeast Taiwan, respectively. It was assumed that the higher anthropogenic emission combined with the adverse meteorological condition led to the higher sulfate level in the southwestern coastal region. The ambient sulfate concentration exhibited significantly seasonal variation with the highest value in spring (5.65 ± 0.84 μg/m3), followed by those in winter (5.45 ± 1.25 μg/m3) and autumn (4.60 ± 0.80 μg/m3), and the lowest one in summer (3.80 ± 0.65 μg/m3). The higher sulfate concentration in spring was mainly contributed by the dense biomass burning and scarce rainfall amount. The present study develops a novel model to capture the high-resolution sulfate map and provides basic data for effective regulations of air pollution and epidemiological studies.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Data availability

Not applicable.

References

  1. Baker KR, Foley KM (2011) A nonlinear regression model estimating single source concentrations of primary and secondarily formed PM2.5. Atmos Environ 45:3758–3767

    CAS  Article  Google Scholar 

  2. Chen T, Guestrin C (2016) Xgboost: a scalable tree boosting system. Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. ACM, pp 785–794

  3. Chen JP, Chen IJ, Tsai IC (2016) Dynamic feedback of aerosol effects on the East Asian summer monsoon. J Clim 29:6137–6149

    Article  Google Scholar 

  4. Chen G, Li S, Knibbs LD, Hamm N, Cao W, Li T, Guo J, Ren H, Abramson MJ, Guo Y (2018a) A machine learning method to estimate PM2.5 concentrations across China with remote sensing, meteorological and land use information. Sci. Total Environ 636:52–60

    CAS  Article  Google Scholar 

  5. Chen ZY, Zhang TH, Zhang R, Zhu ZM, Ou CQ, Guo Y (2018b) Estimating PM2.5 concentrations based on non-linear exposure-lag-response associations with aerosol optical depth and meteorological measures. Atmos Environ 173:30–37

    CAS  Article  Google Scholar 

  6. Chen ZY, Zhang TH, Zhang R, Zhu ZM, Yang J, Chen PY, Ou CQ, Guo Y (2019) Extreme gradient boosting model to estimate PM2.5 concentrations with missing-filled satellite data in China. Atmos Environ 202:180–189

    CAS  Article  Google Scholar 

  7. Cheng Y, Zheng G, Wei C, Mu Q, Zheng B, Wang Z, Gao M, Zhang Q, He K, Carmichael G (2016) Reactive nitrogen chemistry in aerosol water as a source of sulfate during haze events in China. Sci Adv 2:e1601530

    Article  CAS  Google Scholar 

  8. Chuang HL (2016) Spatiotemporal variation, chemical fingerprint, and source identification of atmospheric fine particles long-range transported toward the intersectional region of Taiwan Strait and South China Sea

    Google Scholar 

  9. Chuang MT, Chou CCK, Lin NH, Takami A, Hsiao TC, Lin TH, Fu JS, Pani SK, Lu YR, Yang TY (2017) A simulation study on PM2.5 sources and meteorological characteristics at the northern tip of Taiwan in the early stage of the Asian haze period. Aerosol Air Qual. Res 17:3166–3178

    CAS  Google Scholar 

  10. Chuang MT, Lee CT, Hsu HC (2018) Quantifying PM2.5 from long-range transport and local pollution in Taiwan during winter monsoon: an efficient estimation method. J Environ Manag 227:10–22

    CAS  Article  Google Scholar 

  11. Dai QL, Bi XH, Song WB, Li TK, Liu BS, Ding J, Xu J, Song CB, Yang NW, Schulze BC, Zhang YF, Feng YC, Hopke PK (2019) Residential coal combustion as a source of primary sulfate in Xi'an. China Atmos Environ 196:66–76

    CAS  Article  Google Scholar 

  12. Díaz-de-Mera Y, Aranda A, Martínez E, Rodríguez AA, Rodríguez D, Rodríguez A (2017) Formation of secondary aerosols from the ozonolysis of styrene: effect of SO2 and H2O. Atmos Environ 171:25–31

    Article  CAS  Google Scholar 

  13. Dutkiewicz VA, Das M, Husain L (2000) The relationship between regional SO2 emissions and downwind aerosol sulfate concentrations in the northeastern US. Atmos Environ 34:1821–1832

    CAS  Article  Google Scholar 

  14. Di Q, Koutrakis P, Schwartz J (2016) A hybrid prediction model for PM2.5 mass and components using a chemical transport model and land use regression. Atmos Environ 131:390–399

    CAS  Article  Google Scholar 

  15. Ding A, Huang X, Fu C (2017) Air pollution and weather interaction in East Asia. Oxford Research Encyclopedia of Environmental Science

    Google Scholar 

  16. Duan L, Yu Q, Zhang Q, Wang Z, Pan Y, Larssen T, Tang J, Mulder J (2016) Acid deposition in Asia: Emissions, deposition, and ecosystem effects. Atmos Environ 146:55–69

    CAS  Article  Google Scholar 

  17. Fu H, Chen J (2017) Formation, features and controlling strategies of severe haze-fog pollutions in China. Sci Total Environ 578:121–138

    CAS  Article  Google Scholar 

  18. Fu C, Ding A, Wu J (2017) Review on studies of air pollution and climate change interactions in Monsoon Asia, The Global monsoon system: Research and forecast. World Scientific, pp 315–326

  19. Gao M, Carmichael GR, Wang Y, Ji D, Liu Z, Wang Z (2016) Improving simulations of sulfate aerosols during winter haze over Northern China: the impacts of heterogeneous oxidation by NO2. Front Env Sci Eng 10:16

    Article  CAS  Google Scholar 

  20. Hsu CY, Chiang HC, Lin SL, Chen MJ, Lin TY, Chen YC (2016) Elemental characterization and source apportionment of PM10 and PM2.5 in the western coastal area of central Taiwan. Sci. Total Environ 541:1139–1150

    CAS  Article  Google Scholar 

  21. Huang G, Huang GB, Song S, You K (2015) Trends in extreme learning machines: a review. Neural Netw 61:32–48

    Article  Google Scholar 

  22. Huang Y, Yang Y, Hu H, Xu M, Liu H, Li X, Wang X, Yao H (2019) A deep insight into arsenic adsorption over γ-Al2O3 in the presence of SO2/NO. P Combust Inst 37:2951–2957

    CAS  Article  Google Scholar 

  23. Itahashi S, Uno I, Irie H, Kurokawa JI, Ohara T (2018) Impacts of biomass burning emissions on tropospheric NO2 vertical column density over continental Southeast Asia, Land-Atmospheric Research Applications in South and Southeast Asia. Springer, pp 67–81

  24. Jeon W, Choi Y, Mun J, Lee SH, Choi HJ, Yoo JW, Lee H, Lee HW (2018) Behavior of sulfate on the sea surface during its transport from Eastern China to South Korea. Atmos Environ 186:102–112

    CAS  Article  Google Scholar 

  25. Jung CR, Hwang BF, Chen WT (2018) Incorporating long-term satellite-based aerosol optical depth, localized land use data, and meteorological variables to estimate ground-level PM2.5 concentrations in Taiwan from 2005 to 2015. Environ Pollut 237:1000–1010

    CAS  Article  Google Scholar 

  26. Kheirbek I, Johnson S, Ross Z, Pezeshki G, Ito K, Eisl H, Matte T (2012) Spatial variability in levels of benzene, formaldehyde, and total benzene, toluene, ethylbenzene and xylenes in New York City: a land-use regression study. Environ Health 11:51

    CAS  Article  Google Scholar 

  27. Lee HH, Bar-Or RZ, Wang C (2017) Biomass burning aerosols and the low-visibility events in Southeast Asia. Atmos Chem Phy 17:965–980

  28. Li TC, Yuan CS, Lo KC, Hung CH, Wu SP, Tong C (2015) Seasonal variation and chemical characteristics of atmospheric particles at three islands in the Taiwan Strait. Aerosol Air Qual Res 15:2277–2290

    CAS  Article  Google Scholar 

  29. Li H, Duan F, He K, Ma Y, Kimoto T, Huang T (2016a) Size-dependent characterization of atmospheric particles during winter in Beijing. Atmosphere 7:36

    Article  CAS  Google Scholar 

  30. Li TC, Yuan CS, Huang HC, Lee CL, Wu SP, Tong C (2016b) Inter-comparison of seasonal variation, chemical characteristics, and source identification of atmospheric fine particles on both sides of the Taiwan Strait. Sci Rep 6:22956

    CAS  Article  Google Scholar 

  31. Li K, Chen L, White SJ, Han K, Lv B, Bao K, Wu X, Gao X, Azzi M, Cen K (2017a) Effect of nitrogen oxides (NO and NO2) and toluene on SO2 photooxidation, nucleation and growth: a smog chamber study. Atmos Res 192:38–47

    CAS  Article  Google Scholar 

  32. Li R, Cui L, Li J, Zhao A, Fu H, Wu Y, Zhang L, Kong L, Chen J (2017b) Spatial and temporal variation of particulate matter and gaseous pollutants in China during 2014-2016. Atmos Environ 161:235–246

    CAS  Article  Google Scholar 

  33. Li TC, Yuan CS, Huang HC, Lee CL, Wu SP, Tong C (2017c) Clustered long-range transport routes and potential sources of PM2.5 and their chemical characteristics around the Taiwan Strait. Atmos Environ 148:152–166

    CAS  Article  Google Scholar 

  34. Li J, Li R, Cui L, Meng Y, Fu H (2019a) Spatial and temporal variation of inorganic ions in rainwater in Sichuan province from 2011 to 2016. Environ Pollut 254:112941

    CAS  Article  Google Scholar 

  35. Li R, Cui L, Hongbo F, Li J, Zhao Y, Chen J (2019b) Satellite-based estimation of full-coverage ozone (O3) concentration and health effect assessment across Hainan Island. J Clean Prod 118773

  36. Li R, Cui L, Meng Y, Zhao Y, Fu H (2019c) Satellite-based prediction of daily SO2 exposure across China using a high-quality random forest-spatiotemporal Kriging (RF-STK) model for health risk assessment. Atmos Environ 208:10–19

    CAS  Article  Google Scholar 

  37. Li R, Cui L, Zhao Y, Zhang Z, Sun T, Li J, Zhou W, Meng Y, Huang K, Fu H (2019d) Wet deposition of inorganic ions in 320 cities across China: spatio-temporal variation, source apportionment, and dominant factors. Atmos Chem Phys 19:11043–11070

    CAS  Article  Google Scholar 

  38. Li R, Wang Z, Cui L, Fu H, Zhang L, Kong L, Chen W, Chen J (2019e) Air pollution characteristics in China during 2015-2016: spatiotemporal variations and key meteorological factors. Sci Total Environ 648:902–915

    CAS  Article  Google Scholar 

  39. Li X, Huang L, Li J, Shi Z, Wang Y, Zhang H, Ying Q, Yu X, Liao H, Hu J (2019f) Source contributions to poor atmospheric visibility in China. Resour Conserv Recycl 143:167–177

    Article  Google Scholar 

  40. Lin H, Tao J, Du Y, Liu T, Qian Z, Tian L, Di Q, Zeng W, Xiao J, Guo L (2016) Differentiating the effects of characteristics of PM pollution on mortality from ischemic and hemorrhagic strokes. Int J Hyg Environ Health 219:204–211

    CAS  Article  Google Scholar 

  41. Liu Y, Schichtel BA, Koutrakis P (2009) Estimating particle sulfate concentrations using MISR retrieved aerosol properties. IEEE J Selected Topics Appl Earth Observ Remote Sens 2:176–184

    Article  Google Scholar 

  42. Liu X, Sun K, Qu Y, Hu M, Sun Y, Zhang F, Zhang Y (2015) Secondary formation of sulfate and nitrate during a haze episode in megacity Beijing. China Aerosol Air Qual Res 15:2246–2257

    CAS  Article  Google Scholar 

  43. Liu CY, Kuo SC, Lim A, Hsu SC, Tseng KH, Yeh NC, Yang YC (2016) Optimal use of space-borne advanced infrared and microwave soundings for regional numerical weather prediction. Remote Sens 8:816

    Article  Google Scholar 

  44. Lu HY, Lin SL, Mwangi JK, Wang LC, Lin HY (2016) Characteristics and source apportionment of atmospheric PM2.5 at a coastal city in southern Taiwan. Aerosol Air Qual. Res 16:1022–1034

    CAS  Google Scholar 

  45. Meng X, Hand JL, Schichtel BA, Liu Y (2018) Space-time trends of PM2.5 constituents in the conterminous United States estimated by a machine learning approach, 2005-2015. Environ Int 121:1137–1147

    CAS  Article  Google Scholar 

  46. Meng Y, Zhao Y, Li R, Li J, Cui L, Kong L, Fu H (2019) Characterization of inorganic ions in rainwater in the megacity of Shanghai: spatiotemporal variations and source apportionment. Atmos Res 222:12–24

    CAS  Article  Google Scholar 

  47. Polezer G, Tadano YS, Siqueira HV, Godoi AF, Yamamoto CI, de André PA, Pauliquevis T, de Fatima Andrade M, Oliveira A, Saldiva PH (2018) Assessing the impact of PM2.5 on respiratory disease using artificial neural networks. Environ Pollut 235:394–403

    CAS  Article  Google Scholar 

  48. Qu Z, Henze DK, Li C, Theys N, Wang Y, Wang J, Wang W, Han J, Shim C, Dickerson RR (2019) SO2 emission estimates using OMI SO2 retrievals for 2005-2017. J Geophys Res Atmos 124:8336–8359

    CAS  Article  Google Scholar 

  49. Sarwar G, Fahey K, Kwok R, Gilliam RC, Roselle SJ, Mathur R, Xue J, Yu JZ, Carter WPL (2013) Potential impacts of two SO2 oxidation pathways on regional sulfate concentrations: aqueous-phase oxidation by NO2 and gas-phase oxidation by Stabilized Criegee Intermediates. Atmos Environ 68:186–197

    CAS  Article  Google Scholar 

  50. Shi X, Zhao C, Jiang JH, Wang C, Yang X, Yung YL (2018) Spatial representativeness of PM2. 5 concentrations obtained using observations from network stations. J Geophys Res Atmos 123:3145–3158

    CAS  Article  Google Scholar 

  51. Si YD, Li SS, Chen LF, Yu C, Wang HM, Wang YP (2019) Impact of precursor gases and meteorological variables on satellite-estimated near-surface sulfate and nitrate concentrations over the North China Plain. Atmos Environ 199:345–356

    CAS  Article  Google Scholar 

  52. Sun W, Shao M, Granier C, Liu Y, Ye C, Zheng J (2018) Long-term trends of anthropogenic SO2, NOx, CO, and NMVOCs emissions in China. Earth’s Future 6:1112–1133

    CAS  Article  Google Scholar 

  53. Tseng C-H, Lüthgens C, Tsukamoto S, Reimann T, Frechen M, Böse M (2016) Late Pleistocene to Holocene alluvial tableland formation in an intra-mountainous basin in a tectonically active mountain belt-a case study in the Puli Basin, central Taiwan. Quat Sci Rev 132:26–39

    Article  Google Scholar 

  54. Unger N, Shindell DT, Koch DM, Streets DG (2006) Cross influences of ozone and sulfate precursor emissions changes on air quality and climate. P Natl Acad Sci 103:4377–4380

    CAS  Article  Google Scholar 

  55. Wang Y-C, Lin Y-K (2016) Mortality and emergency room visits associated with ambient particulate matter constituents in metropolitan Taipei. Sci Total Environ 569:1427–1434

    Article  CAS  Google Scholar 

  56. Wang G, Zhang R, Gomez ME, Yang L, Zamora ML, Hu M, Lin Y, Peng J, Guo S, Meng J (2016) Persistent sulfate formation from London Fog to Chinese haze. P Natl Acad Sci 113:13630–13635

    CAS  Article  Google Scholar 

  57. Xiao Q, Chang HH, Geng G, Liu Y (2018) An ensemble machine-learning model to predict historical PM2.5 concentrations in China from satellite data. Environ Sci Technol 52:13260–13269

    CAS  Article  Google Scholar 

  58. Yang Y, Wang H, Smith SJ, Zhang R, Lou S, Qian Y, Ma P-L, Rasch PJ (2018) Recent intensification of winter haze in China linked to foreign emissions and meteorology. Sci Rep 8:2107

    Article  CAS  Google Scholar 

  59. Yang LJ, Xu HQ, Jin ZF (2019) Estimating ground-level PM2.5 over a coastal region of China using satellite AOD and a combined model. J. Cleaner Prod 227:472–482

    CAS  Article  Google Scholar 

  60. Zhan Y, Luo Y, Deng X, Zhang K, Zhang M, Grieneisen ML, Di B (2018) Satellite-based estimates of daily NO2 exposure in China using hybrid random forest and spatiotemporal Kriging model. Environ Sci Technol 52:4180–4189

    CAS  Article  Google Scholar 

  61. Zhang J, Cheng M, Ji D, Liu Z, Hu B, Sun Y, Wang Y (2016) Characterization of submicron particles during biomass burning and coal combustion periods in Beijing. China Sci Total Environ 562:812–821

    CAS  Article  Google Scholar 

  62. Zhang Z, Wang W, Cheng M, Liu S, Xu J, He Y, Meng F (2017) The contribution of residential coal combustion to PM2.5 pollution over China's Beijing-Tianjin-Hebei region in winter. Atmos Environ 159:147–161

    CAS  Article  Google Scholar 

  63. Zhang SP, Xing J, Sarwar G, Ge YL, He H, Duan FK, Zhao Y, He KB, Zhu LD, Chu BW (2019) Parameterization of heterogeneous reaction of SO2 to sulfate on dust with coexistence of NH3 and NO2 under different humidity conditions. Atmos Environ 208:133–140

    CAS  Article  Google Scholar 

  64. Zheng G, Duan F, Su H, Ma Y, Cheng Y, Zheng B, Zhang Q, Huang T, Kimoto T, Chang D (2015) Exploring the severe winter haze in Beijing: the impact of synoptic weather, regional transport and heterogeneous reactions. Atmos Chem Phys 15:2969–2983

    CAS  Article  Google Scholar 

Download references

Acknowledgments

This work was supported by National Natural Science Foundation of China (Nos. 91744205, 21577022, 21177026, 21777025) and National Key R&D Program of China (2016YFC0202700). The land use data were downloaded from National Earth System Science Data Center, National Science & Technology Infrastructure of China (http://www.geodata.cn).

Funding

National Natural Science Foundation of China (Nos. 91744205, 21577022, 21177026, 21777025) and National Key R&D Program of China (2016YFC0202700).

Author information

Affiliations

Authors

Contributions

Lulu Cui: writing-original draft, Qingwei Ma: writing-original draft, Rui Li: data analysis, Hongbo Fu: conceptualization, Ziyu Zhang: formal analysis, Liwu Zhang: formal analysis, Ying Chen: conceptualization.

Corresponding authors

Correspondence to Hongbo Fu or Ying Chen.

Ethics declarations

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent to publish

Not applicable.

Competing interest

The authors declare that they have no competing interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Responsible Editor: Gerhard Lammel

Supplementary Information

ESM 1

(DOCX 1132 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cui, L., Ma, Q., Li, R. et al. High-resolution estimation of ambient sulfate concentration over Taiwan Island using a novel ensemble machine-learning model. Environ Sci Pollut Res (2021). https://doi.org/10.1007/s11356-021-12418-7

Download citation

Keywords

  • Sulfate
  • Extreme gradient boosting
  • Geographically and temporally weighted regression
  • Taiwan Island