Please use this identifier to cite or link to this item:
Development and Application of Basin Lumped and Distributed Rainfall-Runoff Models (II)
|Contributors: ||NTOU:Department of Harbor and River Engineering|
|Issue Date: ||2011-06-29T01:38:53Z
|Abstract: ||摘要:台灣地區雨量豐沛，年平均降雨量高達2500 公釐，為世界平均值的2.5 倍。且降雨集中 在每年5 月至10 月，佔全年雨量的四分之三，且大部份為颱風所帶來的豪雨。由1897 年至 1997 年資料統計結果顯示，侵襲台灣之颱風總計350 次，即平均每年3.5 次；除此之外，尚 有上千次暴雨掠奪，更加深本省水患的嚴重性。因此如何開發水情防洪預報平台，提供正確 洪水預報資訊，以供相關單位與民眾進行防洪減災措施，減少洪水所帶來的災害，實乃刻不 容緩之要務。 本研究計畫之目的乃在建立集水區降雨逕流演算模式，以提供河道水理演算與低窪地區 淹水模擬，藉以發展流域水情預報資訊平台，進一步與預警格網系統相聯結。有鑑於傳統水 文黑盒模式無法提供足夠水理計算條件，因此研究中擬利用運動波-地貌瞬時單位歷線理論， 建立半分佈式之降雨逕流演算模式。研究中並將利用零慣性波理論與地形指數模式理論，建 立分佈式之降雨逕流演算模式，以瞭解此二模式對於研究集水區逕流模擬之適用情況。 本研究於現階段，已順利完成整合計畫所需之淡水河流域水文與地文資料蒐集，並建立 以運動-地貌瞬時單位歷線理論為基礎之半分佈式降雨逕流模式，藉由水文紀錄資料較完備之 美國Goodwin 試驗集水區，配合實際降雨資料測試結果，顯示該降雨逕流數值模式已能確實 反應暴雨期間之洪流流況，且與逕流紀錄間具有甚佳之一致性。而對於第二年度擬進行之計 畫，本研究也已初步完成以零慣性波理論為基礎之分佈式地表逕流演算模式，並針對具空間 變異性之地文因子與逕流反應敏感性的關係，進行初步探討。本研究第二年度計畫，將結合 零慣性波地表逕流模式，建立以地形指數為架構之地表下逕流數值模式，以完成分佈式降雨 逕流模式(包含地表逕流與地表下逕流)之建置，並完整進行集水區地文及水文變異參數設定 對逕流演算之影響性分析。 而第三年計畫則擬利用第二年計畫之研究成果，針對地文與水文空間變異較不顯著之次 集水區，利用分佈式模式之逕流分析結果，配合運動波-地貌瞬時單位歷線模式，建立集塊式 逕流演算單元，簡化逕流分析過程。本計畫完成後，可將逕流演算模擬結果提供予總計畫， 藉以發展流域水情及淹水預報資訊平台，即時提供流域防洪預警相關資訊。|
abstract:The annual rainfall in Taiwan reaches 2,500 mm, which is about 2.5 times the world average. Three-fourth rainstorms mainly occured between May and October and resulted from typhoons. Records from 1897 to 1997 show that total 350 typhoon events invaded Taiwan, that is 3.5 typhoons in avarage every year. Since the typhoons and concentrated rainstorms caused severe damages in the island, developing a flooding warning system to provide accuracy flood forecasting information for flooding prevention and disaster reduction is considered important for authority and people. The objective of this study is to develop rainfall-runoff models for real-time runoff simulation. The hydrographs generated from the runoff model will be input to the channel-flow routing model for water level predicting, and the flooding information will be disseminate to authorities and people for flood prevention and disaster reduction measures. Since the conventional black-box model cannot provide distributed information for hydraulic routing models, a kinematic-wave based geomorphologic instantaneous unit hydrograph model (KW-GIUH), which is a semi-distributed model, will be developed in this study firstly. A fully distributed runoff model using diffusion-wave approximation and dynamic TOPMODEL theory will also be developed for simulation results comparison and to realize the capability of the semi-distributed model and the fully-distributed model for runoff simulation. Currently, hydrological and geomorphologic data of Tanshui River watershed have been collected, and the kinematic-wave-based geomorphologic instantaneous unit hydrograph model (KW-GIUH) was developed for runoff simulation. In the model verification stage, Goodwin Experimental Watershed in U.S., which contains detail hydrological record data, was selected to test the KW-GIUH model. Good agreements between the simulated results and records imply that the proposed model is adequate to simulate the rainfall-runoff processes during rainstorms. A preliminary structure of a diffusion-wave model, which is planned to simulate surface runoff in the 2nd-yr project, has also been finished. The influences of spatial-variability of the geomorphologic conditions on runoff response were also implemented by sensitive analysis and have been briefly discussed. The subsequent study will develop a subsurface-runoff model based on dynamic TOPMODEL theory, and then to integrate the surface-runoff model as a fully-distributed surfaceand subsurface- runoff model. Furthermore, a complete analysis to investigate the spatial-variability of geomorphologic and hydrologic conditions on runoff response will be preformed detailed. Since a large amount of computing time is required for runoff simulation using a distributed runoff model, a lumped-computational-unit concept will be developed in the third-year project to reduce computing time for runoff routing. The lumped-computational unit is defined as a region that spatial variability of geography and hydrology is insignificant. It is promising to develop an integrated flood warning system using the predicted watershed runoff and linking with the channel-flow routing model to provide real-time flooding information to authorities and people.
|Appears in Collections:||[河海工程學系] 研究計畫|
Files in This Item:
All items in NTOUR are protected by copyright, with all rights reserved.