洪水為國內天然災害之一，為取得較長之應變時間以減輕災害損失，發展合適的洪水預報模式為非工程防洪措施中極重要之項目。在洪水預報作業中，實務單位通常採用水位做為警報發布之依據，而非流量，因此本研究結合率定曲線與修正型Muskingum模式來建立即時水位預報模式，應用於蘭陽大橋站一至四小時之水位預報。
　
　　首先採用實測水位–流量資料進行分析，比較十種率定曲線之擬合及延伸方式，優選出擬合較佳之率定曲線型式，並分析應用此率定曲線可能存在之不確定性；如此可獲得在某實測水位下，透過此率定曲線轉換得到之流量的可能分布範圍。接著利用重覆取樣之方法，在現有實測水位情況下考慮其流量之不確定性，結合修正型Muskingum模式來估算蘭陽大橋站可能發生之水位分布情況，並進一步預報前置時間一至四小時水位之95%信賴區間。

　　由模式驗證結果可發現，預報水位區間能掌握實際水位變化趨勢，但預報值有些微低估之情形；因此本研究採用誤差修正來做為即時更新預報之方法，經更新後之水位預報值雖有微幅的振盪現象，但整體而言，顯示水位預報歷線有改善之現象且預報準確率亦有提升。

　　Flood forecasting is an important non-structural means of flood mitigation. In practice, the flood stage, rather than the discharge, is often used as the criterion for issuing a flood warning. Therefore, this study proposed a method, which combines the rating curve and the modified Muskingum model, to forecast the flood stages in Lan-yang Bridge.

　　First, ten extrapolation methods of rating curve were investigated using measured stage-discharge data, and the one with both good performance and practical use was chosen. This study analyzed the discharge errors of this kind of rating curve, and derived the uncertainty of a discharge conditional on a certain water stage. One can use the information to estimate the confidence interval of discharge converted from a specific water stage using the rating curve. The uncertainty of discharge was embedded into a modified Muskingum model, which accounts for the lateral flow in real time, to provide flood stage forecasts with 95% confidence interval one to four hours ahead.

　　The proposed model shows slight underestimation on the validation events. Therefore, an updating scheme was conducted by establishing an error prediction model. The updated stage forecasts have a minor drawback of oscillation in forecasted hydrograph. Nevertheless, the updated forecasts with 95% confidence interval illustrate improvement and have higher accuracy.

1. 王如意、易任，「應用水文學」，國立編譯館，1983。
2. 李戎威，「適用於淡水河水係之水位流量率定曲線研究」，第十四屆水利工程研討會，2004。
3. 林芳明，「率定曲線之繪製、修正及各種延伸法應用之比較研究」，屏東農專學報，第153頁至第163頁，1998。
4. 林茂文，「時間數列分析與預測」，華泰書局，1992 。
5. 許盈松、劉傑立、張國強，「流量觀測不確定度評估-以濁水以濁水溪溪洲大橋站為例」，台灣水利，第52卷，第1期， 2004。
6. 陳憲宗，「灰色預測在逕流預報之應用」，國立成功大學水利及海洋工程學系研究所碩士論文，1998。
7. 曾財益，「洪水即時預報模式及不確定性分析之研究」，國立成功大學水利及海洋工程學系研究所碩士論文，1995。
8. 蔡長泰、陳薔諾、蔡智恆與郭律君，「水位-流量率定曲線外插延伸之研究」，第十四屆水利工程研討會，2004。
9. 賴建信、許盈松，「普萊氏流速儀率定關係之研究」，台灣水利，第52卷，第1期， 2004。
10. 賴建信，「以指數平滑法擬合流量率定曲線之研究」，第十四屆水利工程研討會，2004。
11. Box, P.O. , “Accounting for Heteroscedasticity in Rating Curve Estimates”, Journal of Hydrology, Vol. 292, pp.173-181, 2004.
12. Bathurst, J.C. , “Sensivity Analysis of the System Hydrologique Europeen for an Upland Catchment”, Journal of Hydrology, Vol. 87, pp.103-123, 1986.
13. Birkhead, A.L. and C.S.James, “Muskingum River Routing with Dynamic Bank Storage”, Journal of Hydrology, Vol. 264, pp.113-132, 2002.
14. Birkhead, A.L. and C.S.James, “Synthesis of Rating Curves from Local Stage and Remote Discharge Monitoring Using Nonlinear Muskingum Routing”, Journal of Hydrology, Vol. 205, pp.52-65, 1998.
15. Clarke, R.T., “Uncertainty in the Estimation of Mean Annual Flood Due to Rating-Curve Indefinition”, Journal of Hydrology, Vol. 222, pp.185-190, 1999.
16. Clarke, R.T. and E.M.Mendiondo and L.C.Brusa, “Uncertainties in Mean Discharges from two Large South American Rivers due to Rating Curve Variability”, Hydrological Sciences-Journal-des Sciences Hydrologiques, Vol. 45, No. 2, 2000.
17. Franchini, M. and P.Lamberti, “A Flood Routing Muskingum Type Simulation Forecasting Model on Level Data Alone”, Water Resources Research, Vol. 30, No. 7, pp. 2183-2196, 1994.
18. Khan, M.H., “Muskingum Flood Routing Model for Multiple Tributaries”, Water Resources Research, Vol. 29, No. 4, pp. 1057-1062, 1993.
19. Kevin, D.G.. and P.S.Slobodan, “A Computer-Based System for Modeling the Stage-Discharge Relationships in Steady State Conditions”, Hydrological Sciences-Journal-Des Sciences Hydrologiques, Vol. 39, No. 5, 1994.
20. Kshirsagar, M.M. and B.Rajagopalan and U.Lall, “Optimal Parameter Estimation for Muskingum Routing with Ungauged Lateral Inflow”, Journal of Hydrology, Vol. 169, pp. 25-35, 1995.
21. Karen, S.K. and K.Roman, “Precipitation Uncertainty Processor for Probabilistic River Stage Forecasting”, Water Resources Research, Vol. 36, No. 9, pp. 2643-2653, 2000.
22. Liang, G.C. and J.E.Nash, “Linear Models for River Flow Routing on Large Catchment”, Journal of Hydrology, Vol. 103, pp. 157-188, 1988.
23. Leonard, J. and M.Mietton and H.Najib and P.Gourbesville, “Rating Curve Modeling With Manning Equation to Manage Instability and Improve Extrapolation”, Hydrological Sciences-Journal-Des Sciences Hydrologiques, Vol. 45, No. 5, 2000.
24. O’donnell, T., “A Direct Three-Parameter Muskingum Procedure Incorporation Lateral Flow”, Hydrological Sciences-Journal-Des Science Hydrologiques,Vol. 30, No. 4, 1985.
25. Ponce, M. and A.Lugo, “Modeling Looped Rating in Muskingum-Cunge Routing”, Journal of Hydrologic Engineering, Vol. 6, No. 2, 2001.
26. Sivapragasam, C. and N.Muttil, “A New Approach to Extension of Rating Curve ”, Proceedings of the 6th International Conference on Hydroinformatics,2004.
27. Westphal, J.A. and D.B.Thompson and G.T.Stevens and C.N.Strauser, “Stage-Discharge Relations on the Middle Mississippi River”, Journal of Water Resources Planning and Management , Vol. 125, No. 1, 1999.
28. Yu﹐B., “A Systematic Over-Estimation of Flows ”, Journal of Hydrology, Vol. 233, pp. 258-262, 2000.