圖書典藏及影音數位平台

堤防與護岸為防洪安全之最重要防線，而水利建造物檢查則為維繫堤防護岸安全之重要措施，惟對流域整體防洪課題而言，現行檢查措施僅係單點式針對建造物本身是否已受損進行相關檢查工作，並進行建造物補強與修復，以預防建造物損壞所造成之洪水災害，並無法全面性檢視流域內潛在可能引發洪水災害之風險來源，例如觸發防水洩水構造物喪失功能之結構性因素或未逐堤河段所面臨之問題等。有鑑於此，流域整體防洪宜導入風險管理之思維，依據流域特性進行洪水風險評估，以釐清各河段所面臨之風險程度及背後之風險來源，並擬定可行之因應策略。 經濟部水利署已於民國107年6月20日召開「研商河川水系風險評估相關事宜」會議，經決議風險之定義為危險度乘脆弱度之概念公式，並擬定危險因子及脆弱因子資料庫。緣此，本計畫將以此資料庫為基礎，針對東港溪與四重溪等，依其流域特性探討影響洪水風險之關鍵因子，進行洪水風險評估，並系統性地研擬洪水災害防範對策及計畫，以達減低災害發生之目標。

Due to the fact that hydraulic structure checkup only implemented at the suspicious points in the past, potential risk cannot be entirely examined. For the purpose to strengthen disaster prevention, watershed management is currently directed to advanced conception of risk management. By evaluating watershed risks that is caused by various disaster factors, more solid and applicable strategies or plans can be implemented under limited resources. The second chapter mainly presented the collection of basic data sets relevant to basins of Donggang River and its tributaries. The river geomorphologic characteristics between Chaozhou Bridge and Wanluan bridge was reformed by river remediation structures built in 2012. According to previous studies, Donggang River channel beds show deposition tendency, while Niujiowan creek and Wanan creek beds minor degradation or aggregation trends As for precipitation station distribution evaluation, the Thiessen Polygon analysis shows that the representative area of each station ranges from 25 km2 to 40 km2 in average. In terms of flow/stage station distribution evaluation, by checking up the installation recommendation lists in the previous study report, all currently existing stations mainly satisfy most regulations. The risk assessment presented in Chapter 3 is the core study of this project and is mainly followed the concept of “Risk=Hazard × Vulnerability”. Vulnerability and risk levels are graded individually. The pre-defined factors linked to hazard and vulnerability have to be determined by evaluating the hydrologic /geomorphological/ geographical/geological/structural/prevention plans/evacuation plans/ population/ demographic data sets and historical flooding or remediation engineering collected that interpret the degree of each factor. By giving the factor weights preliminarily suggested by AHP or expert questionnaires, as well as hosting communication meetings, the factors and weights were finally determined. The hazard and vulnerability degrees are examined by the weighted factors and are graded also into high, medium and low for reference. The overall disaster-prone risk of each river channel cross-section then is calculated and classified into high, medium and low levels by given hazard and vulnerability degrees and the concept of “Risk=Hazard × Vulnerability” Risk identification At risk indentification, risk factor was preliminarily suggested by early report and finally determined by expert questionnaires, as well as hosting communication meetings. Final pick 21 risk factor is flood level, hydraulic erosion, soil and sand influence, fault distribution, soil liquefaction, beach width, curve influence, river bed siltation, building standard status, disaster repair frequency, levee form, levee protection bank current height, water gate status, river crossing structure, population, evacuation site, land use, disaster prevention autonomy, early warning system, evacuation plan, flood control. Hazard qualitative analysis After the above-mentioned procedures, Donggang River in terms of hazard qualitative has 52 cross sections classified into low, 82 cross sections into medium, and zero cross section into high level. As for the main tributaries, Wanan creek has 39/15/0 cross sections belong to low/medium/high levels respectively, and .Niujiowan creek on the other hand has 63/23/0 cross sections classified into low/medium/high levels respectively. Vulnerability qualitative analysis Donggang River in terms of Vulnerability qualitative has none cross section classified into low, 11 cross sections into medium, and 123 cross section into high level. As for the main tributaries, Wanan creek has 51/3/0 cross sections belong to low/medium/high levels respectively, and .Niujiowan creek on the other hand has 76/10/0 cross sections classified into low/medium/high levels respectively. In geomorphologic quantitative analysis, the results show overflow could occur at the right bank of cross section #8.1 and #23 at Donggang River due to the numerical experiments with Manning`s n uncertainty. The chance of overflow at #8.1 and #23 are 68% and 24% respectively. In bed degradation/aggregation quantitative analysis, the results show overflow could occur at the left bank of cross section #4.1 and #23 at Donggang River due to the numerical experiments with bed form uncertainty. The chance of overflow at #4.1 is 46%. In hydrologic uncertainty analysis with consideration of climate change scenarios bringing more (110%) design flows for the channels , the results show overflow probabilities of Donggang River overall rise from 0.9%-11.6%, various due to the location of local drainages merging into main channels The tributaries however shows minor addition of overflow probabilities. In general, the increment portions are less than 0.52%. In hydrologic uncertainty analysis with current design flow and levee heights, the results show that the left bank of cross section #4 and #4.1 of Donggang River has highest risk (4.2%) of overflow chance under the design flow of 3561 cms..Other potential high flooding risk locations are left bank of cross section #6 and#11 of Niujiowan creek under the design flow of 168 cms, though the overflow chance is way lower (1.7%). HEC-RAS 2D flooding simulations implemented for the case with 50-yr occurrence flow (Q50) starting from cross section #8.1 display no overall incident occurs.As the results, no economic/life/ecological loss or cultural property damage resulted from the Q50 flooding scenario. To conclude, this report illustrates 25 cross sections with medium risk level along Donggang River. Currently no cross section with medium or high risk level is evaluated along Wanan creek and Niujiowan creek. The overflow risk treatment and remediation strategies could generally count on engineering and non-engineering approaches. We suggest for those treatment-required cross sections, concurrently applying engineering methods such as dredging, sort out, embankment protect and non-engineering methods with disaster prevention, flooding area control, and promoting disaster resistant communities could be more piratical and effective. Nevertheless, the applied approaches should be still considered case by case under various location sensitive characteristics and engineering feasibility.