A study and review of peak discharge transition in mountain river area – a case study of the Talowan river tributary of the Jhuoshuei river

Abstract

本研究探討面對全球氣候變遷環境下山區河川洪峰流量變遷之趨勢，擇定濁水溪支流塔羅灣溪做為案例，並以經濟部水利署「台灣地區主要河川流域水文與水理設計分析系統」軟體分析，暴雨頻率分析係採雲海、廬山及奧萬大(台電)等 3 站為代表站，以徐昇氏多邊形法推求上游集水區平均最大之一日、二日暴雨量及最大連續 24 小時、48 小時暴雨量。利用 K-S 檢定(Kolmogorov-Smirnov test) 與 χ2 檢定(Chi-square test)進行機率分佈之適合度檢定，並以標準誤差(SE)及平方差和(SSE)最小值者為最佳分佈，得塔羅灣溪匯流口控制點最大 24 小時暴雨及最大 48 小時暴雨皆以皮爾遜 III 型為最佳之機率分佈。設計雨型採同位序法，並利用三角形單位歷線法、無因次單位歷線法及運動波-地貌瞬時單位歷線法等三種方法推求各重現期距之洪峰流量。成果顯示，連續最大 24 小時暴雨 100 年重現期距之洪峰流量增加值為 19 至 50% 之情形；連續最大 48 小時暴雨 100 年重現期距之洪峰流量增加值為 36 至 50% 之情形，現行公告之洪峰流量值皆較本研究值小，相關非工程措施包含洪水預警系統、緊急避難路線、教育宣導、預防措施及民眾參與等，或可為本溪未來重要之課題。

The goal of this research is to explore on peak discharge transition in mountain river area under global climate change environment. The peak flow discharges of the Talowan river, a tributary of the Jhuoshuei river, are chosen as an example to show that the effect on peak flow discharges under global climate change environment. A software, Hydrological and Hydraulic Analysis System for Integrated Basin Management of Major Rivers in Taiwan, is used to do analysis of peak flow discharges in this study. That sofeware is developed and used by Water Resources Planning Institute, WRA. The rainfall data is measured from 3 rainfall gauge stations, Yunhai, Lushan, and Auwanda, respectively. The Thiessen’s method is adopted to estimate mean maximum 1-day and 2-day rainfalls on the river basin. Pearson Type III, log-Pearson Type III, 2-parameter log-normal, 3-parameter lognormal, and Gumbel’s extreme Type I distributions are used for rainfall frequency analysis. By means of K-S (Kolmogorov-Smirnov) test and Chi-squared test and using minimum of standard error and minimum sum of squared error as criteria, the results of rainfall frequency analysis show that Pearson Type III distribution is the best goodness of fit for maximum 24-hour and 48-hour rainfall data of the Tarowan river basin. With given design rainfall hyetograph, triangular unit hydrograph, dimensionless unit hydrograph, and kinematic wave based geomorphic instantaneous unit hydrograph method, respectively, are used to estimate peak flow discharges for specified return periods. The results show that 100-year peak flow discharges are about 19 to 50%, and 36 to 50% greater than that of government announcement for maximum 24-hour and maximum 48-hour rainfall, respectively. Therefore, some non-engineering actions should be considered, such as establishing flood warning systems, emergency evacuation plans, education, etc.