區域水文分析為整合區域內水文觀測資料延伸至未量測地點之方法，係水資源工程規劃與管理於未量測地點推估雨量或流量之重要工具。本研究旨在發展一套適合於臺灣南部地區（從北港溪至高屏溪流域）之區域流量延時曲線推估模式。由於短紀錄之水文站較不具代表性，研究中選取紀錄年限大於10年之49個水文觀測站作為分析樣本，並統計1963年至2023年之長期年均雨量作為關鍵氣象變量。為使推估架構具備水文同質性，參考前人區域化成果將研究區域劃分為嘉南上游、下游及高屏溪流域等三個水文均一區，以符合南部地區特有之降雨與地文特性。為建立具備高度穩健性之區域推估模式，分別採用傳統統計學之單變量線性回歸（simple linear regression）與多變量線性回歸（multiple linear regression）分析，以及隨機森林（random forest）、支撐向量機（support vector machine）與極端梯度提升法（extreme gradient boosting）等三種機器學習演算法，發展以集水面積與年均雨量為輸入變量之區域推估模式。為模擬未量測地點應用情境並評估不同區域推估模式之表現，採用留一驗證（leave-one-out cross-validation）進行準確度檢驗。分析結果顯示，相較於傳統線性回歸分析，利用機器學習演算法所發展之區域推估模式，在處理南部豐枯懸殊之流況特徵上表現較優，能有效捕捉集水區氣象、地文因子與流量指標間之複雜映射關係。本研究所發展之區域流量延時曲線推估模式，可提供於未量測地點從事水資源工程規劃與管理快速評估潛能流量之參考依據。

RRegional hydrological analysis extends regional observations to ungauged locations and is an important tool for estimating rainfall or streamflow in water resources planning and management. This study develops a regional flow duration curve estimation model for southern Taiwan, covering the river basins from the Beigang River to the Gaoping River. A total of 49 hydrological stations with records longer than 10 years were selected, and long-term mean annual rainfall from 1963 to 2023 was calculated as a key meteorological variable. Based on previous regionalization results, the study area was divided into three homogeneous regions: the upstream Chianan region, the downstream Chianan region, and the Gaoping River Basin. Simple linear regression, multiple linear regression, random forest, support vector machine, and extreme gradient boosting were applied using catchment area and mean annual rainfall as input variables. Leave-one-out cross-validation was used to simulate ungauged-site conditions and evaluate model accuracy. The results show that machine learning models outperformed traditional linear regression methods in capturing the highly variable flow characteristics of southern Taiwan and the complex relationships among catchment factors and flow indices. The proposed model can provide a reference for rapid potential streamflow assessment at ungauged locations.

Archfield, S. A., Steeves, P. A., Guthrie, J. D., & Ries Iii, K. G. (2013). Towards a publicly available, map-based regional software tool to estimate unregulated daily streamflow at ungauged rivers. Geoscientific Model Development, 6(1), 101–115.
Bawa, A., Mendoza, K., Srinivasan, R., O'Donchha, F., Smith, D., Wolfe, K., Parmar, R., Johnston, J. M., & Corona, J. (2025). Enhancing Hydrological Modeling of Ungauged Watersheds through Machine Learning and Physical Similarity-based Regionalization of Calibration Parameters. Environ Model Softw, 186.
Beck, H. E., Pan, M., Lin, P., Seibert, J., van Dijk, A. I. J. M., & Wood, E. F. (2020). Global Fully Distributed Parameter Regionalization Based on Observed Streamflow From 4,229 Headwater Catchments. Journal of Geophysical Research: Atmospheres, 125(17).
Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32.
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,
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine learning, 20(3), 273–297.
Hrachowitz, M., Savenije, H. H. G., Bl?schl, G., McDonnell, J. J., Sivapalan, M., Pomeroy, J. W., Arheimer, B., Blume, T., Clark, M. P., Ehret, U., Fenicia, F., Freer, J. E., Gelfan, A., Gupta, H. V., Hughes, D. A., Hut, R. W., Montanari, A., Pande, S., Tetzlaff, D.,…Cudennec, C. (2013). A decade of Predictions in Ungauged Basins (PUB)—a review. Hydrological Sciences Journal, 58(6), 1198–1255.
Kan, G., He, X., Ding, L., Li, J., Hong, Y., Ren, M., Lei, T., Liang, K., Zuo, D., & Huang, P. (2017). Daily streamflow simulation based on the improved machine learning method. Tecnolog?a y ciencias del agua, 8(2), 51–60.
Keshtegar, B., Allawi, M. F., Afan, H. A., & El-Shafie, A. (2016). Optimized River Stream-Flow Forecasting Model Utilizing High-Order Response Surface Method. Water Resources Management, 30(11), 3899–3914.
Khandelwal, A., Xu, S., Li, X., Jia, X., Stienbach, M., Duffy, C., Nieber, J., & Kumar, V. (2020). Physics guided machine learning methods for hydrology. arXiv preprint arXiv:2012.02854.
Parajka, J., Merz, R., & Bl?schl, G. (2005). A comparison of regionalisation methods for catchment model parameters. Hydrology and earth system sciences, 9(3), 157–171.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., & Dubourg, V. (2011). Scikit-learn: Machine learning in Python. the Journal of machine Learning research, 12, 2825–2830.
Shiau, J.-T., & Hsu, H.-T. (2016). Suitability of ANN-Based Daily Streamflow Extension Models: a Case Study of Gaoping River Basin, Taiwan. Water Resources Management, 30(4), 1499–1513.
Singh, L., Mishra, P. K., Pingale, S. M., Khare, D., & Thakur, H. P. (2022). Streamflow regionalisation of an ungauged catchment with machine learning approaches. Hydrological Sciences Journal, 67(6), 886–897.
Solanki, H., Vegad, U., Kushwaha, A., & Mishra, V. (2025). Improving Streamflow Prediction Using Multiple Hydrological Models and Machine Learning Methods. Water Resources Research, 61(1).
Yu, P.-S., & Yang, T.-C. (1996). Synthetic regional flow duration curve for southern Taiwan. Hydrological Processes, 10(3), 373–391. <373::Aid-hyp306>3.0.Co;2-4
Zhang, Y., Ye, A., Li, J., Nguyen, P., Analui, B., Hsu, K., & Sorooshian, S. (2025). Improve streamflow simulations by combining machine learning pre-processing and post-processing. Journal of Hydrology, 655.
Booker, D., & Snelder, T. (2012). Comparing methods for estimating flow duration curves at ungauged sites. Journal of Hydrology, 434, 78–94.
Golian, S., Murphy, C., & Meresa, H. (2021). Regionalization of hydrological models for flow estimation in ungauged catchments in Ireland. Journal of Hydrology: Regional Studies, 36, 100859.
Kratzert, F., Klotz, D., Herrnegger, M., Sampson, A. K., Hochreiter, S., & Nearing, G. S. (2019). Toward improved predictions in ungauged basins: Exploiting the power of machine learning. Water Resources Research, 55(12), 11344–11354.
Shu, C., & Ouarda, T. B. (2012). Improved methods for daily streamflow estimates at ungauged sites. Water Resources Research, 48(2).
邱慶誠，1998，濁水溪流域區域合理化公式之研究，國立成功大學水利及海洋工程學系，台南市。
楊承道、王俊寓、林士堯，2023，網格化觀測雨量V2版資料生產履歷，https://tccip.ncdr.nat.gov.tw/upload/data_profile/20220706104221.pdf。
水利署，2025，中華民國一一三年臺灣水文年報總冊。