二氧化氮是一種高活性的空氣污染氣體，且是與臭氧和懸浮微粒的形成有關的初級污染物，它的主要來源為交通和工業中的化石燃料燃燒。人們若是接觸到高濃度的二氧化氮會對健康造成不良的影響，其中包括呼吸系統和心血管問題。空氣污染物分布推估在流行病學研究分析污染物對人類健康影響時扮演著重要角色。然而，過去的空氣污染推估模型只著重在日平均的解析度，這會使得結果無法考慮污染物單日之內的濃度變化。二氧化氮生命週期短，且其在一天中的每個小時都可能隨著氣象、人類活動等其他因素而變化。基於此，本研究發展出集成混合空間模型(EMSM)，並將其用在長時間二氧化氮白天、夜晚、全天的濃度時空變異推估，模型產出之推估圖之網格解析度為50m。模型建置的過程中係以過去26年的二氧化氮觀測值做為依變數；地理空間土地利用圖層、氣象、社會、季節等因子作為模型中排放源代表之解釋變數。在建模方法學上，本研究提出的集成混合空間模型之優勢是夠綜合克里金空間內插、土地利用迴歸、機器學習和集成學習等空間推估方法學的優點，以建置白天、夜晚、全天三個時間段的空污濃度模型。在本研究的結果中，我們產出了可靠又穩定的推估模型，模型解釋力在白天、夜晚及全天分別達到了94%、92%、93%的準確程度。本研究成果不但可以運用於流行病學研究中分析二氧化氮對國人健康的影響，在環境污染的政策規劃上也能當作重要的科學參考依據。

Nitrogen dioxide (NO2), a harmful pollutant, is also associated to the formation of fine particulate matter and ozone. Its primary sources are fossil fuel combustion from traffic and industries. Human exposure to high levels of NO2 may it be short and long term can cause damage to respiratory and cardiovascular systems. Pollutant estimates are essential in epidemiological studies to examine these said effects. However, most of the air pollution models focus only on daily resolutions, without accounting the changes within the day where exposure hazards may be underestimated. NO2 is short-lived and varies at every hour within the day along with other factors like meteorology and human activities. In this study, ensemble mixed spatial model (EMSM) for three time windows of exposure (daytime, nighttime, daily) were developed to examine the spatial and temporal variations of long-term NO2 at 50m resolution. These models utilized a 26-year worth of observed NO2 measurements and meteorological parameters, together with the geospatial layers and social and season-dependent variables as representative of emission sources. This study was able to establish the advantage of combining several spatial modelling methods including spatial interpolation technique, land use regression, machine learning, and ensemble technique in estimating the spatio-temporal variations of long-term NO2 concentrations for the three time periods. This study produced reliable and stable models with an adjusted R2 of 0.94 (daily), 0.92 (daytime), and 0.93 (nighttime). The results of this study can be used not only in epidemiological studies, but also in policy making and environmental planning.

Atkinson, R. W., Butland, B. K., Anderson, H. R., & Maynard, R. L. (2018). Long-term concentrations of Nitrogen Dioxide and mortality. Epidemiology, 29(4), 460–472. https://doi.org/10.1097/ede.0000000000000847

Awang, N. R., Ramli, N. A., Yahaya, A. S., & Elbayoumi, M. (2015). High nighttime ground-level ozone concentrations in Kemaman: No and NO2 concentrations attributions. Aerosol and Air Quality Research, 15(4), 1357–1366. https://doi.org/10.4209/aaqr.2015.01.0031

Brimblecombe, P., & Lai, Y. (2021). Diurnal and weekly patterns of primary pollutants in Beijing under COVID-19 restrictions. Faraday Discussions, 226, 138–148. https://doi.org/10.1039/d0fd00082e

Chen, T. H., Hsu, Y. C., Zeng, Y. T., Candice Lung, S. C., Su, H. J., Chao, H. J., & Wu, C. da. (2020). A hybrid kriging/land-use regression model with Asian culture-specific sources to assess NO2 spatial-temporal variations. Environmental Pollution, 259. https://doi.org/10.1016/j.envpol.2019.113875

Cooper, M. J., Martin, R. v., Hammer, M. S., Levelt, P. F., Veefkind, P., Lamsal, L. N., Krotkov, N. A., Brook, J. R., & McLinden, C. A. (2022). Global fine-scale changes in ambient NO2 during COVID-19 lockdowns. Nature, 601(7893), 380–387. https://doi.org/10.1038/s41586-021-04229-0

Cordioli, M., Pironi, C., de Munari, E., Marmiroli, N., Lauriola, P., & Ranzi, A. (2017). Combining land use regression models and fixed site monitoring to reconstruct spatiotemporal variability of NO2 concentrations over a wide geographical area. Science of the Total Environment, 574, 1075–1084. https://doi.org/10.1016/j.scitotenv.2016.09.089

Diener, A., & Mudu, P. (2021). How can vegetation protect us from air pollution? A critical review on Green Spaces' mitigation abilities for air-borne particles from a public health perspective - with implications for urban planning. Science of The Total Environment, 796, 148605. https://doi.org/10.1016/j.scitotenv.2021.148605

Eeftens, M., Meier, R., Schindler, C., Aguilera, I., Phuleria, H., Ineichen, A., Davey, M., Ducret-Stich, R., Keidel, D., Probst-Hensch, N., Künzli, N., & Tsai, M. Y. (2016). Development of land use regression models for nitrogen dioxide, ultrafine particles, lung deposited surface area, and four other markers of particulate matter pollution in the Swiss SAPALDIA regions. Environmental Health: A Global Access Science Source, 15(1). https://doi.org/10.1186/s12940-016-0137-9

Gauss, M., Ellingsen, K., Isaksen, I. S. A., Dentener, F. J., Stevenson, D. S., Amann, M., & Cofala, J. (2007). Changes in nitrogen dioxide and ozone over Southeast and East Asia between year 2000 and 2030 with fixed meteorology. Terrestrial, Atmospheric and Oceanic Sciences, 18(3), 475–492. https://doi.org/10.3319/TAO.2007.18.3.475(EA)

Han, S., Bian, H., Feng, Y., Liu, A., Li, X., Zeng, F., & Zhang, X. (2011). Analysis of the relationship between O3, NO and NO2 in Tianjin, China. Aerosol and Air Quality Research, 11(2), 128–139. https://doi.org/10.4209/aaqr.2010.07.0055

He, S., Dong, H., Zhang, Z., & Yuan, Y. (2022). An Ensemble Model-Based Estimation of Nitrogen Dioxide in a Southeastern Coastal Region of China. Remote Sensing, 14(12). https://doi.org/10.3390/rs14122807

Hsu, C. Y., Zeng, Y. T., Chen, Y. C., Chen, M. J., Lung, S. C. C., & Wu, C. (2020). Kriging-based land-use regression models that use machine learning algorithms to estimate the monthly btex concentration. International Journal of Environmental Research and Public Health, 17(19), 1–14. https://doi.org/10.3390/ijerph17196956

Janhäll, S. (2015). Review on urban vegetation and particle air pollution – deposition and dispersion. Atmospheric Environment, 105, 130–137. https://doi.org/10.1016/j.atmosenv.2015.01.052

Jin, L., Berman, J. D., Warren, J. L., Levy, J. I., Thurston, G., Zhang, Y., Xu, X., Wang, S., Zhang, Y., & Bell, M. L. (2019). A land use regression model of nitrogen dioxide and fine particulate matter in a complex urban core in Lanzhou, China. Environmental Research, 177. https://doi.org/10.1016/j.envres.2019.108597

Kendrick, C. M., Koonce, P., & George, L. A. (2015). Diurnal and seasonal variations of NO, NO2 and PM2.5 mass as a function of traffic volumes alongside an urban arterial. Atmospheric Environment, 122, 133–141. https://doi.org/10.1016/j.atmosenv.2015.09.019

Lee, C.-S., Chang, K.-H., & Kim, H. (2018). Long-term (2005–2015) Trend Analysis of PM2.5 precursor gas NO2 and SO2 concentrations in Taiwan. Environmental Science and Pollution Research, 25(22), 22136–22152. https://doi.org/10.1007/s11356-018-2273-y
Liu, M., Lin, J., Wang, Y., Sun, Y., Zheng, B., Shao, J., Chen, L., Zheng, Y., Chen, J., Fu, T. M., Yan, Y., Zhang, Q., & Wu, Z. (2018). Spatiotemporal variability of NO2 and PM2.5 over Eastern China: Observational and model analyses with a novel statistical method. Atmospheric Chemistry and Physics, 18(17), 12933–12952. https://doi.org/10.5194/acp-18-12933-2018

Meena, G. S., & Jadhav, D. B. (2007). Study of diurnal and seasonal variation of atmospheric NO2, O3 , H2O and O4 at Pune, India. In Atmósfera (Vol. 20, Issue 3).

National Statistics Taiwan. (2022). Total Population. National Statistics, Republic of China (Taiwan). Retrieved September 13, 2023, from https://eng.stat.gov.tw/point.asp?index=9
National Oceanic and Atmospheric Administration. (2010). Nitrogen Dioxide. Science On a Sphere. Retrieved December 3, 2022, from https://sos.noaa.gov/catalog/datasets/nitrogen-dioxide/
Özkaynak, H., Baxter, L. K., Dionisio, K. L., & Burke, J. (2013). Air pollution exposure prediction approaches used in air pollution epidemiology studies. Journal of Exposure Science and Environmental Epidemiology, 23(6), 566–572. https://doi.org/10.1038/jes.2013.15

Pancholi, P., Kumar, A., Bikundia, D. S., & Chourasiya, S. (2018). An observation of seasonal and diurnal behavior of O3–NOx relationships and local/regional oxidant (OX = O3 + NO2) levels at a semi-arid urban site of western India. Sustainable Environment Research, 28(2), 79–89. https://doi.org/10.1016/j.serj.2017.11.001

Pouliou, T., Kanaroglou, P. S., Elliott, S. J., & Pengelly, L. D. (2008). Assessing the health impacts of air pollution: A re-analysis of the Hamilton Children's cohort data using a spatial analytic approach. International Journal of Environmental Health Research, 18(1), 17–35. https://doi.org/10.1080/09603120701844290

Rodríguez-Pérez, R., & Bajorath, J. (2020). Interpretation of machine learning models using shapley values: application to compound potency and multi-target activity predictions. Journal of Computer-Aided Molecular Design, 34(10), 1013–1026. https://doi.org/10.1007/s10822-020-00314-0
US EPA. (2022). Nitrogen Dioxide (NO2) Pollution. EPA. Retrieved July 16, 2022, from https://www.epa.gov/NO2-pollution/basic-information-about-NO2
Voiculescu, M., Constantin, D. E., Condurache-Bota, S., Călmuc, V., Roșu, A., & Bălănică, C. M. D. (2020). Role of meteorological parameters in the diurnal and seasonal variation of NO2 in a Romanian urban environment. International Journal of Environmental Research and Public Health, 17(17), 1–15. https://doi.org/10.3390/ijerph17176228

Wagner, P., & Schäfer, K. (2017). Influence of mixing layer height on air pollutant concentrations in an urban street canyon. Urban Climate, 22, 64–79. https://doi.org/10.1016/j.uclim.2015.11.001

Wang, L., Wang, J., Tan, X., & Fang, C. (2020a). Analysis of NOx pollution characteristics in the atmospheric environment in Changchun city. Atmosphere, 11(1). https://doi.org/10.3390/ATMOS11010030

Wang, Z., Anthony, J. L., Erickson, L. E., Higgins, M. J., & Newmark, G. L. (2020b). Nitrogen Dioxide and Ozone Pollution in the Chicago Metropolitan Area. Journal of Environmental Protection, 11(08), 551–569. https://doi.org/10.4236/jep.2020.118033

Wong, P. Y., Lee, H. Y., Zeng, Y. T., Chern, Y. R., Chen, N. T., Candice Lung, S. C., Su, H. J., & Wu, C. da. (2021a). Using a land use regression model with machine learning to estimate ground level PM2.5. Environmental Pollution, 277. https://doi.org/10.1016/j.envpol.2021.116846

Wong, P. Y., Su, H. J., Lee, H. Y., Chen, Y. C., Hsiao, Y. P., Huang, J. W., Teo, T. A., Wu, C. da, & Spengler, J. D. (2021b). Using land-use machine learning models to estimate NO2 concentration variations in Taiwan. Journal of Cleaner Production, 317. https://doi.org/10.1016/j.jclepro.2021.128411

Wu, C. da, Chen, Y. C., Pan, W. C., Zeng, Y. T., Chen, M. J., Guo, Y. L., & Lung, S. C. C. (2017). Land-use regression with long-term satellite-based greenness index and culture-specific sources to model PM2.5 spatial-temporal variability. Environmental Pollution, 224, 148–157. https://doi.org/10.1016/j.envpol.2017.01.074

Wu, C. da, Zeng, Y. T., & Lung, S. C. C. (2018). A hybrid kriging/land-use regression model to assess PM2.5 spatial-temporal variability. Science of the Total Environment, 645, 1456–1464. https://doi.org/10.1016/j.scitotenv.2018.07.073

Zhang, L., Guo, X., Zhao, T., Xu, X., Zheng, X., Li, Y., Luo, L., Gui, K., Zheng, Y., & Shu, Z. (2022). Effect of large topography on atmospheric environment in Sichuan Basin: A climate analysis based on changes in atmospheric visibility. Frontiers in Earth Science, 10. https://doi.org/10.3389/feart.2022.997586