本研究以新店溪為對象，探討氣候變遷與都市污染負荷變化下之河川水質變化趨勢。研究整合2014至2023年水質、氣象、流量、人口、生活用水量、污水下水道接管率及污水處理廠進流水質濃度等資料，建立月尺度資料庫，並以生化需氧量（BOD5）、化學需氧量（COD）、氨氮（NH₃-N）及懸浮固體（SS）作為主要預測標的。為反映生活污水對河川水質之影響，本研究依人口、用水量、接管率及進流水質濃度推估生活污染負荷量，並結合SSP2-4.5與SSP5-8.5情境，建立2024至2043年未來輸入資料。本研究比較多元線性回歸、隨機森林及CatBoost三種模型，並設計氣象水文、污染負荷及綜合輸入組合進行分析。結果顯示，CatBoost於綜合輸入組合下表現較穩定，驗證期BOD5、COD、NH₃-N及SS之推估結果的納許效率係數分別為0.52、0.57、0.62及0.65，百分比誤差分別為25.81、17.3%、26.5%及37.7%。特徵重要度分析顯示，BOD5、COD及NH₃-N主要受流量與污染負荷量影響，反映河川稀釋能力與生活污水輸入之重要度；SS則較受降雨影響，顯示其與地表沖刷及顆粒物輸送關係較密切。未來情境推估結果顯示，SSP2-4.5與SSP5-8.5下BOD5、COD及NH₃-N之長期平均濃度差異有限，高溫室氣體排放情境未必造成水質參數濃度明顯升高，推測與未來人口下降及污水下水道接管率提升，使污染負荷量減少有關。相較之下，SS於SSP5-8.5情境下略高，且可能於特定高降雨月份出現尖峰，顯示未來水質風險較可能集中於極端降雨事件下之懸浮固體與非點源污染問題。整體而言，本研究證實整合氣象、水文與污染負荷量之機器學習模式，可應用於都市河川水質預測與未來情境分析。未來新店溪水質管理除應持續推動污水下水道接管與生活污染源削減外，亦應加強枯水期污染累積管理，以及高降雨事件下之懸浮固體與都市逕流監測。

This study investigates water quality trends in the Hsintien River under climate change and urban pollution load changes. Monthly data from 2014 to 2023, including water quality, meteorological data, streamflow, population, domestic water consumption, sewerage connection rate, and wastewater treatment plant influent quality, were integrated. Biochemical oxygen demand (BOD5), chemical oxygen demand (COD), ammonia nitrogen (NH₃-N), and suspended solids (SS) were selected as prediction targets. Domestic pollution loads were estimated using population, water consumption, sewerage connection rate, and influent water quality concentrations. Future input data from 2024 to 2043 were established under SSP2-4.5 and SSP5-8.5 scenarios. Multiple linear regression, random forest, and CatBoost models were compared using different input combinations. The results show that CatBoost performed most stably under the integrated input combination, with validation Nash-Sutcliffe efficiency coefficients of 0.52, 0.57, 0.62, and 0.65 for BOD5, COD, NH₃-N, and SS, respectively. Feature importance analysis indicated that BOD₅, COD, and NH₃-N were mainly affected by streamflow and pollution loads, while SS was more influenced by rainfall. Future projections suggest limited changes in BOD₅, COD, and NH₃-N, but SS may increase during high-rainfall months.

Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32.
Calvin, K., Dasgupta, D., Krinner, G., Mukherji, A., Thorne, P. W., Trisos, C., Romero, J., Aldunce, P., Barrett, K., & Blanco, G. (2023). IPCC, 2023: Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland.
Chang, C.-H., Cai, L.-Y., Lin, T.-F., Chung, C.-L., Van der Linden, L., & Burch, M. (2015). Assessment of the impacts of climate change on the water quality of a small deep reservoir in a humid-subtropical climatic region. Water, 7(4), 1687–1711.
Chang, C., Kuan, W., Lui, P., & Hu, C. (2008). Relationship between landscape characteristics and surface water quality. Environmental monitoring and assessment, 147(1), 57–64.
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,
Chetty, S., & Pillay, L. (2019). Assessing the influence of human activities on river health: a case for two South African rivers with differing pollutant sources. Environmental monitoring and assessment, 191(3), 168.
Eggimann, S., Truffer, B., & Maurer, M. (2015). To connect or not to connect? Modelling the optimal degree of centralisation for wastewater infrastructures. Water Research, 84, 218–231.
Jang, C.-S. (2016). Using probability-based spatial estimation of the river pollution index to assess urban water recreational quality in the Tamsui River watershed. Environmental monitoring and assessment, 188(1), 36.
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.
Lin, L.-Y., Lin, C.-T., Chen, Y.-M., Cheng, C.-T., Li, H.-C., & Chen, W.-B. (2022). The Taiwan climate change projection information and adaptation knowledge platform: a decade of climate research. Water, 14(3), 358.
Lin, S., Kim, J., Hua, C., Kang, S., & Park, M.-H. (2023). Comparing artificial and deep neural network models for prediction of coagulant amount and settled water turbidity: Lessons learned from big data in water treatment operations. Journal of Water Process Engineering, 54, 103949.
Lin, Y.-I., Pan, S.-Y., Chang, H.-H., Yu, M.-S., & Lin, W.-L. (2023). Will extreme drought impact the reservoir water quality? A 30-year observational study. Agricultural Water Management, 289, 108574.
Liu, W.-C., & Chan, W.-T. (2016). Assessment of climate change impacts on water quality in a tidal estuarine system using a three-dimensional model. Water, 8(2), 60.
Makumbura, R. K., Mampitiya, L., Rathnayake, N., Meddage, D., Henna, S., Dang, T. L., Hoshino, Y., & Rathnayake, U. (2024). Advancing water quality assessment and prediction using machine learning models, coupled with explainable artificial intelligence (XAI) techniques like shapley additive explanations (SHAP) for interpreting the black-box nature. Results in Engineering, 23, 102831.
Márquez, F. P. G., Al-taie, A. H. S., Zakur, Y. A., Alsadoon, A., Flaih, L. R., & Zakoor, Y. A. (2024). Exploring the Potential of the Machine Learning Techniques in the Water Quality Assessment: A Review of Applications and Performance. International Conference on Advanced Engineering, Technology and Applications,
Metcalf, & Eddy, A. (2014). Wastewater engineering treatment and resource recovery. McGraw-Hill Education.
Moiseenko, T. I. (2022). Surface water under growing anthropogenic loads: from global perspectives to regional implications. Water, 14(22), 3730.
Pilecka, J., Grinfelde, I., Jumite, L. E., Valujeva, K., & Didze, V. L. (2020). The anthropogenic impact on surface water quality: Case study of Latvia. International Multidisciplinary Scientific GeoConference: SGEM, 20(3.1), 329–337.
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.
Singha, C., Bhattacharjee, I., Sahoo, S., Abdelrahman, K., Uddin, M. G., Fnais, M. S., Govind, A., & Abioui, M. (2024). Prediction of urban surface water quality scenarios using hybrid stacking ensembles machine learning model in Howrah Municipal Corporation, West Bengal. Journal of Environmental Management, 370, 122721.
Suwadi, N. A., Derbali, M., Sani, N. S., Lam, M. C., Arshad, H., Khan, I., & Kim, K.-I. (2022). An optimized approach for predicting water quality features based on machine learning. Wireless Communications and Mobile Computing, 2022(1), 3397972.
Tung, C.-P., Ting-Chuan, L., Wei-Ting, L., & Yun-Ju, C. (2012). Climate change impact assessment for sustainable water quality management. TAO: Terrestrial, Atmospheric and Oceanic Sciences, 23(5), 565.
Wang, Y.-B., Liu, C.-W., & Lee, J.-J. (2015). Differentiating the spatiotemporal distribution of natural and anthropogenic processes on river water–quality variation using a self-organizing map with factor analysis. Archives of environmental contamination and toxicology, 69(2), 254–263.
Xia, X., Wu, Q., Mou, X., & Lai, Y. (2015). Potential impacts of climate change on the water quality of different water bodies. J. Environ. Inform, 25(2), 85–98.
Yan, X., Zhang, T., Du, W., Meng, Q., Xu, X., & Zhao, X. (2024). A comprehensive review of machine learning for water quality prediction over the past five years. Journal of Marine Science and Engineering, 12(1), 159.
Zhu, M., Wang, J., Yang, X., Zhang, Y., Zhang, L., Ren, H., Wu, B., & Ye, L. (2022). A review of the application of machine learning in water quality evaluation. Eco-Environment & Health, 1(2), 107–116