黃宏斌臺灣大學:生物環境系統工程學研究所呂冠德Liu, Guan-DeGuan-DeLiu2010-05-052018-06-292010-05-052018-06-292009U0001-2306200914282000http://ntur.lib.ntu.edu.tw//handle/246246/181164摘要門水庫為我國重要之水利工程之ㄧ，但其上游集水區因先天之地質條件不佳，加上後天颱風豪雨之侵襲，造成大量崩塌及土石流等災害，將影響水資源之規劃與利用，其集水區治理工作已刻不容緩。中，對於集水區內降雨特性之掌握多寡，乃是影響治理工程成功與否之關鍵，因此，本研究之主要目的為找出適用於石門水庫上游集水區之雨量推估模式，進ㄧ步建議集水區內雨量監測站網之最佳佈置。評估各種類型之雨量推估模式之適用性，本研究以集水區內14個雨量測站於1986至2007年間之颱風降雨資料，以一般克利金法、非連結克利金法及去除高程之平均趨勢法評估各該模式於集水區颱風降雨之適用性。結果顯示以一般克利金法最能準確推估集水區內之颱風降雨特性，為最適用於石門水庫上游集水區之雨量推估模式。外，以結合克利金法與最大熵值法之站網評估模式來進行集水區內雨量監測站網之評估，最後對於石門水庫上游集水區內颱風降雨之雨量監測站網設置提出一最佳之站網設置。Abstracthe Shih-men Reservoir is one of the most important reservoirs for water supply in northern Taiwan. Owing to the characteristics of Shih-men Reservoir watershed (weak geologic condition, steep slopes, high mountains, high-intensity rainfall, and frequently occurred typhoon), its watershed management is urgent and necessary.he understanding of the characteristics of rainfall is the key for successful engineering design so that a reasonable raingauge network capable of providing reliable rainfall data is important to engineering design. The aim of this study is to show the results of different geostatistic methods for estimating the rainfall variation of Shih-men Reservoir watershed and to propose the optimum number and spatial distribution of raingauge network in the watershed.n order to evaluate the effectiveness of the geostatistic methods and the existing raingauge network, a network of 14 raingauges in Shih-men Reservoir watershed and the rainfall date between 1986 and 2007 were chosen for analysis. Three different geostatistic approach (Ordinary Kriging, Disjunctive Kriging and Detrend method) have been used as estimators and their results are compared and discussed. In this study, Ordinary Kriging seems to give the optimum results.urthermore, we use the method composed of Kriging and the maximum joint entropy approach to determine the optimum number and spatial distribution of raingauge network in the watershed. The result shows that only five raingauges could provide the enough information of Typhoon rainfall in Shih-men Reservoir watershed.目錄要 Ibstract II錄 III目錄 V目錄 VI錄 VII、前言 1、文獻回顧 3-1 集水區雨量之推估 3-2 雨量站網之設計 5-2-1 克利金法於雨量站網之設計 5-2-2 熵理論於雨量站網之設計 5、理論分析 7-1 克利金法(Kriging Method) 7-1-1 一般克利金法(Ordinary Kriging , OK) 8-1-2 區域化變數 8-1-3 非連結克利金法(Disjunctive Kriging , DK) 10-1-4 去除平均趨勢法(Detrend Method) 10-2 熵(Entropy) 11-2-1 邊際熵(Marginal Entropy) 11-2-2 聯合熵(Joint Entropy) 11、研究方法 13-1 研究區域概況 13-2 雨量資料之選用與雨量測站之選取 15-3 雨量推估模式之驗證 19-4 雨量站網之規劃設計 20-5 研究步驟 22、結果與討論 24-1 集水區內颱風降雨與高程之關係 24-2 推估模式之適用性 26-3 集水區雨量站網之設置 30、結論 36、參考文獻 37錄 39application/pdf1714334 bytesapplication/pdfen-US石門水庫上游集水區克利金法熵雨量監測站網Shih-men Reservoir WatershedKrigingEntropyRaingauge Networkthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/181164/1/ntu-98-R95622019-1.pdf