當前全球範圍內，空氣污染問題日益嚴重，對健康和社會發展構成重大威脅。其中，細懸浮微粒(PM2.5)被認為是最嚴重的污染物之一，與多種健康問題密切相關。同時，氣候變遷是相當重要的環境議題，溫室氣體排放增加導致氣候變暖，影響生態系統平衡和人類社會型態。空氣污染與氣候變遷之間存在的複雜交互作用，增加了未來空氣品質預測的困難和不確定性，這兩個議題已成為21世紀重要的環境挑戰。本研究旨在了解氣候變遷下PM2.5濃度的未來變化趨勢。為了推估未來長期的PM2.5濃度，本研究蒐集了臺灣本島1994年至2019年期間的日平均PM2.5濃度數據進行建模。方法上，本研究整合了土地利用迴歸模型（Land Use Regression, LUR）的概念和機器學習的優勢，以捕捉PM2.5與周邊土地利用種類及其他相關變數之間的非線性關係。五種機器學習演算法，包括eXtreme Gradient Boosting (XGBoost)、Gradient Boosting Machine (GBM)、Light Gradient Boosting Machine (LGBM)、Categorical Boosting (CatBoost)和Random Forest，被用於預測PM2.5濃度。模型中重要變數以SHAP（SHapley Additive exPlanations）值進行篩選。透過多種驗證方法，包括資料分割（data splitting）方法、十折交叉驗證（10-fold cross-validation）以及其他時間與空間驗證方法來評估模型表現，並以特定指標選出表現最佳的模型。為了推估在IPCC第六次評估報告(The 6th Assessment Report, AR6)中所提出新溫室氣體排放情境下的長期PM2.5濃度變化，本研究納入了AR6氣象資料與未來人口資料。選用由大氣環流模式(Global Climate Model, GCM) MIROC6、MPI-ESM1-2-LR及TaiESM1模擬的推估氣象資料，用以推估在AR6 SSP1-2.6、SSP2-4.5、SSP3-7.0及SSP5-8.5四種情境下，2021年至2040年、2041年至2060年及2081年至2100年的PM2.5濃度變化趨勢。最終選擇GBM模型作為表現最佳的模型。其在訓練集上的R2達到0.81，在測試集和十折交叉驗證中的R2也分別為0.60，顯示出良好的預測能力。在驗證成果方面， GBM模型在不同時間和空間驗證中都表現出穩健性，除了證實其在不同研究時期和地區的預測準確性外，還展示了該模型用於推估未來PM2.5濃度的可行性。與基期年平均PM2.5濃度(15.18μg/m3)相比，未來所有時段及情境下的年平均PM2.5濃度預計將減少，範圍介於12.41μg/m3至13.70μg/m3之間。然而，對比不同情境和時段後發現，PM2.5濃度在各情境間和時段內都沒有顯著的線性變化趨勢，顯示出未來PM2.5濃度變化的不確定性。這些未來PM2.5濃度變化的預測結果可廣泛應用於環境政策制定、健康風險評估以及城市規劃等方面。然而，雖然預測顯示PM2.5濃度會有所減少，但變化趨勢的不確定性需要引起重視，並在相關領域進行進一步研究和應對措施。

Air pollution has emerged as a critical concern, which PM2.5 stands out as a rimary pollutant, posing significant threats to both human health and the sustainable evelopment of society. Meanwhile, climate change is a crucial environmental issue, with increased greenhouse gas emissions leading to global warming, affecting cosystem balance and human society. The complex interactions between air quality and climate change complicate the prediction of future PM2.5 levels. The aim of this study is to understand the future trends in PM2.5 concentrations under climate change. This research integrated the concept of land use regression and leverages the advantage of the machine learning algorithm. To predict PM2.5 changes of 2021 to 2040, 2041 to 2060 and 2081 to 2100 under the climate scenarios outlined in the IPCC Sixth Assessment Report (AR6), future meteorological and population data were considered as variables. The Gradient Boosting Machine (GBM) was selected as the final model based on various kinds of model validation. The training R² was 0.81, while the testing R² and 10-fold cross-validation R² were both 0.60, demonstrating the model's predictive capability. Additionally, the model performed well in different temporal and spatial validations, indicating its robustness and ability to predict future PM2.5 levels. Compared with the annual mean PM2.5 concentration of base period (1994 to 2019), the annual mean PM2.5 concentration is projected to decrease across all future time periods and scenarios. However, upon comparing different scenarios and time periods, it was found that PM2.5 concentrations do not exhibit significant linear trends across scenarios or periods, highlighting the uncertainty in future changes in PM2.5 levels. However, while the predictions indicate a reduction in PM2.5 levels, the uncertainty in the trends should be noted, and further research and measures are needed in relevant fields. These predicted changes in future PM2.5 concentrations can be widely applied to environmental policy-making, health risk assessments, and urban planning.

Baccarelli, A., Wright, R., Bollati, V., Litonjua, A., Zanobetti, A., Tarantini, L., Sparrow, D., Vokonas, P., & Schwartz, J. (2010). Ischemic heart disease and stroke in relation to blood DNA methylation. Epidemiology, 21(6), 819-828 %@ 1044-3983.
Bayraktar, H., & Turalioglu, F. S. (2005). A Kriging-based approach for locating a sampling site—in the assessment of air quality. Stochastic Environmental Research and Risk Assessment, 19, 301-305 %@ 1436-3240.
Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., & Courchamp, F. (2012). Impacts of climate change on the future of biodiversity. Ecology letters, 15(4), 365-377 %@ 1461-1023X.
Briggs, D. J., Collins, S., Elliott, P., Fischer, P., Kingham, S., Lebret, E., Pryl, K., Van Reeuwijk, H., Smallbone, K., & Van Der Veen, A. (1997). Mapping urban air pollution using GIS: a regression-based approach. International Journal of Geographical Information Science, 11(7), 699-718 %@ 1365-8816.
Chen, G., Li, S., Knibbs, L. D., Hamm, N. A. S., Cao, W., Li, T., Guo, J., Ren, H., Abramson, M. J., & Guo, Y. (2018). A machine learning method to estimate PM2. 5 concentrations across China with remote sensing, meteorological and land use information. Science of the Total Environment, 636, 52-60 %@ 0048-9697.
Chen, J., de Hoogh, K., Gulliver, J., Hoffmann, B., Hertel, O., Ketzel, M., Bauwelinck, M., Van Donkelaar, A., Hvidtfeldt, U. A., & Katsouyanni, K. (2019). A comparison of linear regression, regularization, and machine learning algorithms to develop Europe-wide spatial models of fine particles and nitrogen dioxide. Environment international, 130, 104934 %@ 100160-104120.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system.
Chen, W., Lu, X., Yuan, D., Chen, Y., Li, Z., Huang, Y., Fung, T., Sun, H., & Fung, J. C. H. (2023). Global PM2. 5 prediction and associated mortality to 2100 under different climate change scenarios. Environmental science & technology, 57(27), 10039-10052 %@ 10013-10936X.
Chen, Z., Chen, D., Zhao, C., Kwan, M.-p., Cai, J., Zhuang, Y., Zhao, B., Wang, X., Chen, B., & Yang, J. (2020). Influence of meteorological conditions on PM2. 5 concentrations across China: A review of methodology and mechanism. Environment international, 139, 105558 %@ 100160-104120.
Chiu, H.-F., Tsai, S.-S., Chen, P.-S., Liao, Y.-H., Liou, S.-H., Wu, T.-N., & Yang, C.-Y. (2011). Traffic air pollution and risk of death from gastric cancer in Taiwan: petrol station density as an indicator of air pollutant exposure. Journal of Toxicology and Environmental Health, Part A, 74(18), 1215-1224 %@ 1528-7394.
Clougherty, J. E., Kheirbek, I., Eisl, H. M., Ross, Z., Pezeshki, G., Gorczynski, J. E., Johnson, S., Markowitz, S., Kass, D., & Matte, T. (2013). Intra-urban spatial variability in wintertime street-level concentrations of multiple combustion-related air pollutants: the New York City Community Air Survey (NYCCAS). Journal of exposure science & environmental epidemiology, 23(3), 232-240 %@ 1559-1064X.
Dawson, J. P., Racherla, P. N., Lynn, B. H., Adams, P. J., & Pandis, S. N. (2009). Impacts of climate change on regional and urban air quality in the eastern United States: Role of meteorology. Journal of Geophysical Research: Atmospheres, 114(D5 %@ 0148-0227).
de Hoogh, K., Gulliver, J., van Donkelaar, A., Martin, R. V., Marshall, J. D., Bechle, M. J., Cesaroni, G., Pradas, M. C., Dedele, A., & Eeftens, M. (2016). Development of West-European PM2. 5 and NO2 land use regression models incorporating satellite-derived and chemical transport modelling data. Environmental research, 151, 1-10 %@ 0013-9351.
Di, Q., Amini, H., Shi, L., Kloog, I., Silvern, R., Kelly, J., Sabath, M. B., Choirat, C., Koutrakis, P., & Lyapustin, A. (2019). An ensemble-based model of PM2. 5 concentration across the contiguous United States with high spatiotemporal resolution. Environment international, 130, 104909 %@ 100160-104120.
Farida, Y., Farmita, M., Ulinnuha, N., & Yuliati, D. (2022). Forecasting Population of Madiun Regency Using ARIMA Method. CAUCHY: Jurnal Matematika Murni dan Aplikasi, 7(3), 420-431 %@ 2477-3344.
Fiore, A. M., Naik, V., & Leibensperger, E. M. (2015). Air quality and climate connections. Journal of the Air & Waste Management Association, 65(6), 645-685 %@ 1096-2247.
Forouzanfar, M. H., Afshin, A., Alexander, L. T., Anderson, H. R., Bhutta, Z. A., Biryukov, S., Brauer, M., Burnett, R., Cercy, K., & Charlson, F. J. (2016). Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. The lancet, 388(10053), 1659-1724 %@ 0140-6736.
Friedman, J. H. (2001). Greedy function approximation: a gradient boosting machine. Annals of statistics, 1189-1232 %@ 0090-5364.
Habre, R., Moshier, E., Castro, W., Nath, A., Grunin, A., Rohr, A., Godbold, J., Schachter, N., Kattan, M., & Coull, B. (2014). The effects of PM2. 5 and its components from indoor and outdoor sources on cough and wheeze symptoms in asthmatic children. Journal of exposure science & environmental epidemiology, 24(4), 380-387 %@ 1559-1064X.
Hamra, G. B., Guha, N., Cohen, A., Laden, F., Raaschou-Nielsen, O., Samet, J. M., Vineis, P., Forastiere, F., Saldiva, P., & Yorifuji, T. (2014). Outdoor particulate matter exposure and lung cancer: a systematic review and meta-analysis. Environmental health perspectives.
Ho, T. K. (1998). The random subspace method for constructing decision forests. IEEE transactions on pattern analysis and machine intelligence, 20(8), 832-844 %@ 0162-8828.
Hoek, G., Beelen, R., De Hoogh, K., Vienneau, D., Gulliver, J., Fischer, P., & Briggs, D. (2008). A review of land-use regression models to assess spatial variation of outdoor air pollution. Atmospheric environment, 42(33), 7561-7578 %@ 1352-2310.
Hong, C., Zhang, Q., Zhang, Y., Davis, S. J., Tong, D., Zheng, Y., Liu, Z., Guan, D., He, K., & Schellnhuber, H. J. (2019). Impacts of climate change on future air quality and human health in China. Proceedings of the national academy of sciences, 116(35), 17193-17200 %@ 10027-18424.
Hu, X., Belle, J. H., Meng, X., Wildani, A., Waller, L. A., Strickland, M. J., & Liu, Y. (2017). Estimating PM2. 5 concentrations in the conterminous United States using the random forest approach. Environmental science & technology, 51(12), 6936-6944 %@ 0013-6936X.
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30.
Kioumourtzoglou, M.-A., Schwartz, J. D., Weisskopf, M. G., Melly, S. J., Wang, Y., Dominici, F., & Zanobetti, A. (2016). Long-term PM2. 5 exposure and neurological hospital admissions in the northeastern United States. Environmental health perspectives, 124(1), 23-29 %@ 0091-6765.
Leelőssy, Á., Molnár, F., Izsák, F., Havasi, Á., Lagzi, I., & Mészáros, R. (2014). Dispersion modeling of air pollutants in the atmosphere: a review. Open Geosciences, 6(3), 257-278 %@ 2391-5447.
Lindsey, R., & Dahlman, L. (2020). Climate change: Global temperature. Climate. gov, 16.
Liu, H., Yan, G., Duan, Z., & Chen, C. (2021). Intelligent modeling strategies for forecasting air quality time series: A review. Applied Soft Computing, 102, 106957 %@ 101568-104946.
Lu, X., Li, R., & Yan, X. (2021). Airway hyperresponsiveness development and the toxicity of PM2. 5. Environmental Science and Pollution Research, 28(6), 6374-6391 %@ 0944-1344.
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, 30.
Mao, W., Wang, W., Jiao, L., Zhao, S., & Liu, A. (2021). Modeling air quality prediction using a deep learning approach: Method optimization and evaluation. Sustainable Cities and Society, 65, 102567 %@ 102210-106707.
Nyoni, T., & Mutongi, C. (2019). Prediction of total population in Togo using ARIMA models.
O'Neill, B. C., Tebaldi, C., Van Vuuren, D. P., Eyring, V., Friedlingstein, P., Hurtt, G., Knutti, R., Kriegler, E., Lamarque, J.-F., & Lowe, J. (2016). The scenario model intercomparison project (ScenarioMIP) for CMIP6. Geoscientific Model Development, 9(9), 3461-3482 %@ 1991-3959X.
O’Gorman, P. A. (2015). Precipitation extremes under climate change. Current climate change reports, 1, 49-59.
O’Neill, B. C., Kriegler, E., Ebi, K. L., Kemp-Benedict, E., Riahi, K., Rothman, D. S., Van Ruijven, B. J., Van Vuuren, D. P., Birkmann, J., & Kok, K. (2017). The roads ahead: Narratives for shared socioeconomic pathways describing world futures in the 21st century. Global environmental change, 42, 169-180 %@ 0959-3780.
Orru, H., Ebi, K. L., & Forsberg, B. (2017). The interplay of climate change and air pollution on health. Current environmental health reports, 4, 504-513.
Peng, L.-C., Lin, Y.-P., Chen, G.-W., & Lien, W.-Y. (2019). Climate change impact on spatiotemporal hotspots of hydrologic ecosystem services: A case study of Chinan catchment, Taiwan. Water, 11(4), 867 %@ 2073-4441.
Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2018). CatBoost: unbiased boosting with categorical features. Advances in neural information processing systems, 31.
Reid, C. E., Jerrett, M., Petersen, M. L., Pfister, G. G., Morefield, P. E., Tager, I. B., Raffuse, S. M., & Balmes, J. R. (2015). Spatiotemporal prediction of fine particulate matter during the 2008 northern California wildfires using machine learning. Environmental science & technology, 49(6), 3887-3896 %@ 0013-3936X.
Riva, D. R., Magalhães, C. B., Lopes, A. A., Lanças, T., Mauad, T., Malm, O., Valença, S. S., Saldiva, P. H., Faffe, D. S., & Zin, W. A. (2011). Low dose of fine particulate matter (PM2. 5) can induce acute oxidative stress, inflammation and pulmonary impairment in healthy mice. Inhalation toxicology, 23(5), 257-267 %@ 0895-8378.
Shapley, L. S. (1953). A value for n-person games.
Shinde, P. P., & Shah, S. (2018). A review of machine learning and deep learning applications.
Tai, A. P. K., Mickley, L. J., Jacob, D. J., Leibensperger, E. M., Zhang, L., Fisher, J. A., & Pye, H. O. T. (2012). Meteorological modes of variability for fine particulate matter (PM 2.5) air quality in the United States: implications for PM 2.5 sensitivity to climate change. Atmospheric Chemistry and Physics, 12(6), 3131-3145 %@ 1680-7324.
Thangavel, P., Park, D., & Lee, Y.-C. (2022). Recent insights into particulate matter (PM2. 5)-mediated toxicity in humans: an overview. International journal of environmental research and public health, 19(12), 7511 %@ 1660-4601.
Trenberth, K. E. (2011). Changes in precipitation with climate change. Climate research, 47(1-2), 123-138 %@ 0936-0577X.
Turnock, S. T., Reddington, C. L., West, J. J., & O’Connor, F. M. (2023). The Air Pollution Human Health Burden in Different Future Scenarios That Involve the Mitigation of Near‐Term Climate Forcers, Climate and Land‐Use. GeoHealth, 7(8), e2023GH000812 %@ 002471-001403.
Van Vuuren, D. P., Edmonds, J., Kainuma, M., Riahi, K., Thomson, A., Hibbard, K., Hurtt, G. C., Kram, T., Krey, V., & Lamarque, J.-F. (2011). The representative concentration pathways: an overview. Climatic change, 109, 5-31 %@ 0165-0009.
Vizcaino, P., & Lavalle, C. (2018). Development of European NO2 Land Use Regression Model for present and future exposure assessment: Implications for policy analysis. Environmental Pollution, 240, 140-154 %@ 0269-7491.
Wang, H., Tian, M., Chen, Y., Shi, G., Liu, Y., Yang, F., Zhang, L., Deng, L., Yu, J., & Peng, C. (2018). Seasonal characteristics, formation mechanisms and source origins of PM 2.5 in two megacities in Sichuan Basin, China. Atmospheric Chemistry and Physics, 18(2), 865-881 %@ 1680-7316.
Weagle, C. L., Snider, G., Li, C., Van Donkelaar, A., Philip, S., Bissonnette, P., Burke, J., Jackson, J., Latimer, R., & Stone, E. (2018). Global sources of fine particulate matter: interpretation of PM2. 5 chemical composition observed by SPARTAN using a global chemical transport model. Environmental science & technology, 52(20), 11670-11681 %@ 10013-11936X.
Wong, P.-Y., Lee, H.-Y., Chen, Y.-C., Zeng, Y.-T., Chern, Y.-R., Chen, N.-T., Lung, S.-C. C., Su, H.-J., & Wu, C.-D. (2021). Using a land use regression model with machine learning to estimate ground level PM2. 5. Environmental Pollution, 277, 116846 %@ 110269-117491.
Wu, C.-D., 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 %@ 0048-9697.
Xiao, Q., Chang, H. H., Geng, G., & Liu, Y. (2018). An ensemble machine-learning model to predict historical PM2. 5 concentrations in China from satellite data. Environmental science & technology, 52(22), 13260-13269 %@ 10013-13936X.
Xie, G., Sun, L., Yang, W., Wang, R., Shang, L., Yang, L., Qi, C., Xin, J., Yue, J., & Chung, M. C. (2021). Maternal exposure to PM2. 5 was linked to elevated risk of stillbirth. Chemosphere, 283, 131169 %@ 130045-136535.
Yang, Q., Yuan, Q., Li, T., Shen, H., & Zhang, L. (2017). The relationships between PM2. 5 and meteorological factors in China: seasonal and regional variations. International journal of environmental research and public health, 14(12), 1510 %@ 1660-4601.
Zakria, M., & Muhammad, F. (2009). Forecasting the population of Pakistan using ARIMA models. Pakistan Journal of Agricultural Sciences, 46(3), 214-223.
Zhang, X., Xiao, X., Wang, F., Yang, Y., Liao, H., Wang, S., & Gao, M. (2023). Discordant future climate-driven changes in winter PM2. 5 pollution across India under a warming climate. Elementa: Science of the Anthropocene, 11(1).
Zou, B., Xu, S., Liu, N., Li, S., Liu, X., Guo, Y., & Zhan, F. B. (2023). PM2. 5 exposure and associated premature mortality to 2100 in China under climate and socioeconomic change scenarios. Earth's Future, 11(9), e2022EF003416 %@ 002328-004277.
國科會臺灣氣候變遷推估資訊與調適知識平台. (2023a). AR6 統計降尺度雨量資料資料生產履歷. https://tccip.ncdr.nat.gov.tw/upload/data_profile/20220718101540.pdf
國科會臺灣氣候變遷推估資訊與調適知識平台. (2023b). AR6 統計降尺度溫度資料資料生產履歷. https://tccip.ncdr.nat.gov.tw/upload/data_profile/20221228143303.pdf
傅宗襁, 劉宇倫, & 張益嘉. (2020). 氣候變遷下未來臺灣埃及斑蚊分布變化趨勢. 國家科學及技術委員會.
劉子明, 鄧澤宇, & 鄭克聲. (2022). 水資源領域應用統計降尺度日資料之轉換研究. 國家科學及技術委員會.