Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/34373
標題: Complex System Analysis and Management Strategies in Eutrophication
優養化複雜系統分析與管理策略研究
作者: 蔡大偉
David, D-W.Tsai
關鍵字: 模式預測
Model prediction
單因子設計
優養動力
多變量分析
優養化管理
總量管理
泥砂產量
德基水庫
水里溪集水區
One-factor design
Power of eutrophication
Multivariate statistic analysis
Eutrophication management
Management of TMDL
Sediment TMDL
Te-Chi reservoir
Shuili-stream watershed
出版社: 水土保持學系所
引用: 參考文獻 中文文獻: 1.丁安迪 (2004),「自來水配水管網三鹵甲烷模式模擬之研究」,碩士論文,中興大學。 2.中央氣象局日月潭觀測站(1998~2005),逐日最高最低溫觀測報告書。 3.王智益(2000),「BASINS及CE-QUAL-RIV1應用於非點源污染傳輸及設計流量之研究」,碩士論文,台灣大學。 4.台灣電力公司明潭發電廠(1998~2003),明潭發電廠運轉期間環境監測計畫。 5.行政院農委會水土保持局(2004),「水土保持技術規範」,行政院農委會水土保持局,南投市。 6.行政院農委會水土保持局(2005),「水土保持手冊」,中華水土保持學會,台北市。 7.李建忠(1996),「HSPF應用於瑪家水庫優養潛勢分析之研究」,碩士論文,成功大學。 8.李佳叡 (2000),「實施隔週休二日對到訪森林遊樂區人數改變之分析—時間數列介入分析模型之應用」,碩士論文,台灣大學。 9.李向宇 (2002),「貝氏估計在時間序列上的應用」,碩士論文,交通大學。 10.李珮璇(2002),「暴雨初期沖刷對水源水質衝擊之評估」,碩士論文,台北大學。 11.李惠妍(2003),「類神經網路與迴歸模式在台股指數期貨預測之研究」,碩士論文,成功大學。 12.吳輝龍、張文詔、黃俊德(1995),「坡地檳榔園試區水土流失量第一年成果初步探討」,中華水土保持學報,第26卷,第3期,第197-209頁。 13.吳嘉俊、盧光輝、林俐玲(1996),「土壤流失量估算手冊」,國立屏東技術學院,屏東縣。 14.吳俊逸(2001a),「自組非線性系統應用於地層下陷之預測」,碩士論文,成功大學。 15.吳俊穎(2001b),「衛星影像監測永和山水庫水質之研究」,碩士論文,中華大學。 16.吳淑姿(2002),「海洋單細胞真菌-Schizochytrium sp. S31生產多元不飽和脂肪酸-DHA」,博士論文,台灣大學。 17.吳仁杰 (2003),「時間序列及橫斷面評價模式應用之比較」,碩士論文,中原大學。 18.汪三華 (2001),「台股期貨與現貨之價格及報酬率長短期關聯性探討」,碩士論文,中國文化大學。 19.林茂文 (1992),「時間數列分析與預測」,華泰出版社,台北市。 20.林紀男 (2000),「結合類神經網路與分位數回歸估計多期報酬率之風險值」,碩士論文,台北大學。 21.林俊全(2001),「南投地區塊體崩移之研究-以水里溪集水區為例」,國科會整合型計畫。 22.林慧姿、張嘉容、黃春松、廖苹邁 (2002),「統計學」,新科技書局出版,台北縣。 23.林雍富(2002),「應用BASINS模式於非點源污染傳輸之模擬-以霧社水庫為例」,碩士論文,台灣大學。 24.林俐玲、杜怡德、何宜娟(2002),「應用農業非點源污染模式(AGNPS)評估坡地開發泥砂產量」,水土保持學報,第三十四卷,第三期,第165~176頁。 25.林文賜(2002),「集水區空間資訊萃取及坡面泥砂產量推估之研究」,中興大學博士論文。 26.林師模、陳范欽 (2003a),「多變量分析~管理上的應用」,雙葉書廊有限公司,台北市。 27.林葦旻 (2003b),「土壤之物理性質與力學參數相關性分析 --- 台灣地區山坡地土壤」,碩士論文,中華大學。 28.林鎮洋 (2003),「集水區水質總量管理模式與控制技術之研究」,期末報告,國科會委託研究計畫。 29.林志融 (2004),「棲地適宜性分析應用於生態廊道規劃之研究 --- 以山羌及有勝溪流域為例」,碩士論文,東華大學。 30.林炤映 (2004),「以水質自動監測系統與統計方法分析日月潭水庫之水質變化趨勢」,碩士論文,大葉大學。 31.林俊宇(2004),「應用迴歸分析方法預測軟體發展時程」,碩士論文,成功大學。 32.林昭遠、陳昱豪、林家榮(2006),「集水區泥砂產量推估之研究」,水土保持學報。 33.邱昱嘉(2003),「應用HSPF模式與銫-137技術於集水區產砂量推估之研究」,碩士論文,台灣大學。 34.邱乙哲 (2004),「結構性時間序列在匯率分析的應用」,碩士論文,銘傳大學。 35.周天穎、葉美伶(1997),「水里溪集水區檳榔種植對土壤沖蝕之影響及其經濟分析」,中華水土保持學報28(2):87-97。 36.周夢柏(2002),「應用財務比率分析我國商業銀行獲利能力之實證研究」,碩士論文,朝陽大學。 37.范楓旻(2002),「放山雞養雞場非點源污染模式之研究」,碩士論文,成功大學。 38.俞其海 (1990),「實用統計學」,昭人出版社出版,台中市。 39.洪慧鈞 (2002),「水庫優養化評估指標與優養化水體三鹵甲烷生成潛勢之探討」,碩士論文,中興大學。 40.夏禹九、陳萓蓉(1999),「農業非點源模式應用於集水區分析上之探討」,88年度水土保持及集水區經營研究計畫成果彙編。 41.孫嘉鴻 (2004),「灰預測與演化式類神經網路應用於台指選擇權之研究」,碩士論文,朝陽科技大學。 42.張維泰 (2002),「空氣污染之線性趨勢分析檢定方法的比較」,碩士論文,中正大學。 43.張秉元 (2004),「花蓮地區棲地組成與黃嘴角鴞數量之關係」,碩士論文,東華大學。 44.張玉姍(2004),「翡翠水庫集水區非點源污染整治區域優先順序之評估」,碩士論文,台北大學。 45.張秀琴(2004),「利用QUAL2E水質模式模擬淡水河系興建污水下水道之水質影響」,碩士論文,中原大學。 46.涂珮琪 (2003),「二角多甲藻水層萃取物之毒性研究」,碩士論文,台灣大學。 47.康健廷 (2002),「我國商業銀行風險值 (VaR) 評價模型之比較分析」,碩士論文,台北大學。 48.陳登源、楊錦洲、林茂文、蔡豐清 (1997a),「管理數學」,國立空中大學發行,臺北縣。 49.陳鴻烈、鄭慧玲 (1997b),「台中縣政府大甲溪及德基水庫水質監測研究方案」,台中縣環境保護局委託研究計畫,pp.94~117。 50.陳鴻烈、鄭慧玲 (1998),「水庫優養化之時間數列分析研究」,水土保持學報,第三十卷,第四期,第331~337頁。 51.陳樹群、簡如宏、馮智偉、巫仲明(1998),「本土化土壤沖蝕指標模式之建立」,中華水土保持學報,第二十九卷,第三期,第233~247頁。 52.陳鴻烈、梁家柱、鄭慧玲、王久泰 (1999),「利用相加性季節變動模式之德基水庫優養化時間數列分析」,水土保持學報,第三十一卷,第三期,第139~144頁。 53.陳奕宏、王怡諭 (1999b),「環境微生物」,中華民國環境工程學會編印,臺北市。 54.陳伶姮 (2000),「河川旬流量時間序列研究~以石門水庫入流量為例」,碩士論文,海洋大學。 55.陳鴻烈、梁家柱、王久泰、鄭慧玲 (2000b),「以多變量統計主因子分析法探討德基水庫水質之變異性研究」,水土保持學報,第三十二卷,第一期,第5~10頁。 56.陳鴻烈、梁家柱、王久泰、鄭慧玲 (2000c),「德基水庫優養化之空間分析研究」,水土保持學報,第三十二卷,第三期,第117~124頁。 57.陳琪婷 (2003),「以二氧化錳催化降解水中氨氮之研究」,碩士論文,中山大學。 58.陳正昌、程炳林、陳新豐、劉子鍵 (2003),「多變量分析方法:統計軟體應用」,五南圖書出版股份有限公司,台北市。 59.陳錦嫣 (2003),「GIS技術與實務應用ARCVIEW 3.X & 8.X」,文京圖書有限公司,台北市。 60.陳鴻傑(2003),「曾文水庫集水區污染物傳輸及水庫水質模擬」,碩士論文,台灣大學。 61.陳威宏(2003),「溫泉廢水對於水環境之影響-以南勢溪流域為例」,碩士論文,台北大學。 62.陳鴻烈、梁家柱、羅惠芬、鄭慧玲 (2004),「水庫優養化時間序列模式分析比較研究」,水土保持學報,第三十六卷,第二期,第169~178頁。 63.陳文俊(2004),「礦場廢水BMPs除污效率之研究-以瑪鋉溪瓷土礦場為例」,碩士論文,台北大學。 64.陳鴻烈、蔡大偉 (2005a),「不同模式之預測能力研究」,水土保持學報,第三十七卷,第二期。 65.陳鴻烈、蔡大偉 (2005b),「不同預測時距對模式預測能力影響之研究」,水土保持學報,第三十七卷,第三期,第237~250頁。 66.陳鴻烈、蔡大偉 (2005c),「德基水庫優養水質研究」,水土保持學報,第三十八卷,第一期。 67.陳鴻烈、蔡大偉 (2005d),「優養化模式預測能力之最佳化研究」,農林學報,第五十五卷,第一期。 68.陳鴻烈、蔡大偉(2006a),「最佳集水分區模擬之研究」,水土保持學報。 69.陳鴻烈、蔡大偉 (2005b),「以複迴歸分析法探討水質因子與優養化全模式之研究」,水土保持學報。 70.陳鴻烈、蔡大偉 (2006c),「不同優養水質共線性分析及模式選擇之研究」,水土保持學報。 71.陳鴻烈、蔡大偉 (2006d),「不同模式時距對模式預測能力影響之研究」,水土保持學報,第三十八卷,第二期。 72.黃鈺真(2001),「HSPF模式應用於曾文水庫集水區非點源污染負荷之推估」,碩士論文,成功大學。 73.黃信源 (2002a),「台灣地區行動電話需求預測模式之建構與評估~時間序列之應用」,碩士論文,台北大學。 74.黃意茹 (2002b),「都市綠覆率與氣溫之相關研究~衛星影像類神經網路分類法之應用」,碩士論文,逢甲大學。 75.黃建智(2002),「流域集水區非點源污染模式之研究」碩士論文,成功大學。 76.黃佳慧(2005),「以HSPF營養鹽模組討論農業對水庫非點源污染負荷的貢獻」,成功大學碩士論文。 77.彭昭英(1998),「SAS與統計分析」,儒林圖書有限公司,台北市。 78.湯健文 (2003),「類神經網路於因果關係模型與時間數列模型之應用」,碩士論文,中華大學。 79.葉怡巖 (2001),「水庫水質之變異對於淨水廠中三鹵甲烷生成之影響」,碩士論文,逢甲大學。 80.葉宗育(2001),「懸浮固體濃度即時監測暨集水區水質模式NPSM應用於翡翠水庫」,碩士論文,台灣大學。 81.經濟部水利署德基水庫集水區管理委員會(1983~1996),「德基水庫集水區整體治理計畫水質監測與管理模式研究」。 82.經濟部水利署德基水庫集水區管理委員會 (2002),「德基水庫集水區第四期整體治理計畫水質監測與管理工作成果摘要總報告」。 83.楊景婷 (2001),「時間序列分類分析方法:技術發展與評估」,碩士論文,中山大學。 84.楊雅梅 (2001),「台灣水庫集水區水質指標與管理系統建立之研究」,碩士論文,台灣大學。 85.萬鑫森、黃俊義(1989),「台灣坡地土壤沖蝕」,中華水土保持學報,第二十卷,第二期,第17-45頁。 86.廖秀華(1990),「應用地理資訊系統推估土壤沖蝕潛能」中興大學碩士論文。 87.劉錦蕙 (1999),「乾、濕季時間序列分析」,碩士論文,逢甲大學。 88.農委會水土保持局(2004),水土保持規範。 89.廖文蓓 (2002),「翡翠水庫中藻類種群消長之動態模擬」,碩士論文,台灣大學。 90.蔡裕春 (2001),「台灣地區營造工程物價指數預測之研究~以類神經網路與ARIMA模式」,碩士論文,輔仁大學。 91. 蔡志宏 (2004),「匯率預測模型之檢測,結合時間序列與總體經濟模型」,碩士論文,暨南國際大學。 92. 潘南飛編譯 (2003),「工程統計」,全威圖書有限公司,台北縣。 93. 盧光輝(1996),「水里溪集水區土壤流失量之推估」,85年度水土保持及集水區經營研究計畫成果彙編。 94. 盧昭堯、蘇志強、吳藝均(2005),「台灣地區年等降雨沖蝕指數圖之修訂」,中華水土保持學報,第三十六卷,第二期,第159~172頁。 95. 環保署(1998),「地面水體分類及水質標準」,環保署,台北市。 96. 環保署(2006),「水質評估指標」(網址:www.epa.gov.tw/b/b0100.asp?Ct_Code=05X0000747X0001137)。 97. 謝斌暉(1999),「暴雨期間河川水質模式設計流量之研究」,碩士論文,台灣大學。 98. 謝孟洲,2004「Aspergillus Carneus M34 聚木醣酶與植酸酶同步生產之評估及聚木醣酶最適化生產」,碩士論文,中興大學。 英文文獻: 1. Arheimer, B. and R. Lidén (2000), “Nitrogen and Phosphorus Concentrations from Agricultural Catchments ― Influence of Spatial and Temporal Variables”, Journal of Hydrology, Vol.227, pp.140-159. 2. A. S. Donigian, Jr. J. T. Love(2002), “Sediment calibration procedures and guidelines for watershed modeling”, AQUA TERRA Consultants. 3. A. S. Donigian, Jr. J. T. Love(2002), “WATERSHED MODEL CALIBRATION AND VALIDATION THE HSPF EXPERIENCE”, AQUA TERRA Consultants. 4. Antikainen, R., R. Lemola, J. I. Nousiainen, L. Sokka, M. Esala, P. Huhtanen and S. Rekolainen (2005), “Stocks and Flows of Nitrogen and Phosphorus in the Finnish Food Production and Consumption System”, Agriculture, Ecosystems and Environment, Vol.107, pp.287~305. 5. Aaron Katz, Matthias vom Hau, James Mahoney(2005), “EXPLAINING THE GREAT REVERSAL IN SPANISH AMERICA: FUZZY-SET METHODS VERSUS STATISTICAL METHODS”, Department of Sociology, Brown University. 6. Ben-Gal, I. and L. B. Levitin (2001), “An application of information theory and error-correcting codes to fractional factorial experiments”, Journal of Statistical Planning and Inference. Vol.92, p.p.267~287. 7. Bolger, F. and D. Onkal-Atay (2004), “The Effects of Feedback on Judgmental Interval Predictions”, International Journal of Forecasting, Vol.20, pp.29~39. 8. Barbeau, M. A., K. Durelle and R. B. Aiken(2004) “A design for multifactorial choice experiments: an example using microhabitat selection by sea slugs onchidoris bilamellata (L.).”, Journal of Experimental Marine Biology and Ecology, Vol. 307, p.p.1~16. 9. Bechmanna, M. E., D. Bergeb, H. O. Eggestada and S. M. Vandsemb (2005), “Phosphorus Transfer from Agricultural Areas and Its Impact on the Eutrophication of Lakes — Two Long-Term Integrated Studies from Norway”, Journal of Hydrology, Vol. 304, p.p.238~250. 10. Cheng, W. P. and F.-H. Chi (2003), “Influence of Eutrophication on the Coagulation Efficiency in Reservoir Water”, Chemosphere, Vol.53, pp.773~778. 11. Corradi, V. and N. R. Swanson (2004), “Some Recent Developments in Predictive Accuracy Testing with Nested Models and (Generic) Nonlinear Alternatives”, International Journal of Forecasting, Vol.20, pp.185~199. 12. C. Carlona, M. Dalla Valle, A. Marcomini(2004), "Regression models to predict water–soil heavy metals partition coefficients in risk assessment studies", Environmental Pollution, Vol.127, p.p.109-115. 13. C. Sarbu, H.F. Pop(2005), “Principal component analysis versus fuzzy principal component analysis A case study: the quality of danube water (1985–1996)”, Talanta, Vol.65, p.p.1215-1220. 14. Danilov, R. and N. G. A. Ekelund (1999), “The Efficiency of Seven Diversity and One Similarity Indices Based on Phytoplankton Data for Assessing the Level of Eutrophication in Lakes in Central Sweden”, The Science of the Total Environment, Vol.234, pp.15~23. 15. Deana, A. M. and S. M. Lewis(2002), “Comparison of group screening strategies for factorial experiments”, Computational Statistics & Data Analysis, Vol.39, p.p.287~297. 16. Donev, A. N.(2004), “Design of experiments in the presence of errors in factor levels”, Journal of Statistical Planning and Inference, Vol.92, p.p.267~282. 17. de Gooijer, J. G., A. Vidiella-i-Anguera (2004), “Forecasting Threshold Cointegrated Systems”, International Journal of Forecasting , Vol.20, pp.237~253. 18. Diewert, W. E., W. F. Alterman and R. C. Feenstra (2004), “Time Series Versus Index Number Methods of Seasonal Adjustment”, Discussion Paper No.04-06, The University of British Columbia Vancouver, Canada. 19. Dejaegher, B., J. Smeyers-Verbeke and Y. V. Heyden(2005), “The variance of screening and supersaturated design results as a measure for method robustness”, Analytica Chimica Acta, Vol. 544, p.p.268~279. 20. Edoardo Reisenhofer, Alessio Picciotto, Dongfang Li(1995), “An factor analysis approach to the study of eutrophication of a shallow, temperate lake(San Daniele, North Eastern Italy)”, Analytica Chimica Acta, Vol.306, p.p.99-106. 21. Franses, P. H. and D. van Dijk (2005), “The Forecasting Performance of Various Models for Seasonality and Nonlinearity for Quarterly Industrial Production”, International Journal of Forecasting, Vol.21, pp.87~102. 22. G. R. Foster, D. K. McCool, K. G. Renard, W. C. Moldenhauer(1981), “Conversion of the universal soil loss equation to SI metric units”, Journal of Soil and Water Conservation, Vol., pp.355~359. 23. G. G. Pyle, S. M. Swanson, D. M. Lehmkuhl(2001), “Toxicity of Uranium Mine-Receiving Waters to Caged Fathead Minnows, Pimephales promelas”, Ecotoxicology and Environmental Safety, Vol.48, p.p.202-214. 24. Gardner Jr., E. S., J. Diaz-Saiz (2002), “Sea- asonal Adjustment of Inventory Demand Series: A Case Study”, International Journal of Forecasting, Vol.18, pp.117~123. 25. Guy Perrie`re, Jean Thioulouse(2003), “Use of correspondence discriminant analysis to predict the subcellular location of bacterial proteins”, Computer Methods and Programs in Biomedicine, Vol.70, p.p.99-105. 26. Goupy, J. (2005), “What kind of experimental design for finding and checking robustness of analytical methods?”, Analytica Chimica Acta, Vol.544, p.p184~190. 27. Gruau, G., M. Legeas, C. Riou, E. Gallacier, F. Martineau and O. Henin (2005), “The Oxygen Isotope Composition of Dissolved Anthropogenic Phosphates: A New Tool for Eutrophication Research?”, Water Research, Vol.39, pp.232~238. 28. Gillen, D. (2005), Lecture 6, Statistical Methods II Lecture Notes, Department of Statistics, University of California, Irvine. 29. Henk J. L. Heessen, Niels Daan(1996), “Long-term trends in ten non-target North Sea fish species”, ICES Journal of Marine Science, Vol.5, p.p.1063–1078. 30. Hiscock, J. G., C. S. Thourot and J. Zhang (2003), “Phosphorus Budget ― Land Use Relationships for the Northern Lake Okeechobee Watershed, Florida”, Ecological Engineering, Vol.21, pp.63-74. 31. John L. Curnutt(2000), “Host-area specific climatic-matching :similarity breeds exotics”, Biological Conservation, Vol.94, p.p.341-351. 32. Jeong, Y. (2000), “Asymptotically optimal adaptive designs in factorial experiments”, Journal of Statistical Planning and Inference. Vol.84, p.p.263~283. 33. J.P.Shepard, H. Riekerk(2002), “A comparison of the watershed hydrology of coastal forested wetland and the mountainous uplands in the Southern US”, Journal of hydrology, Vol.263, p.p.92~104. 34. John E. Richards(2004), “Recovering dipole sources from scalp-recorded event-related-potentials using component analysis: principal component analysis and independent component analysis”, International Journal of Psychophysiology, Vol.54, p.p.201-220. 35. John C. Imhoff, Jonathan Clough, Richard A. Park, Andrew Stoddard, Earl Hayter(2004), “Evaluation of chemical bioaccumulation models of aquatic ecosystems final report”, national exposure research laboratory office of research and development U.S. environmental protection agency Athens, Georgia 30605. 36. Jason Love and Anthony Donigian, Jr.(2004), “DERIVING MODEL INPUT FOR A LUMPED PARAMETER WATERSHED MODEL”, AWRA. 37. Jarvie, H. P., M. D. Jürgens, R. J. Williams, C. Neal, J. J. L. Davies, C. Barrett and J. White (2005), “Role of River Bed Sediments as Sources and Sinks of Phosphorus Across Two Major Eutrophic UK River Basins: The Hampshire Avon and Herefordshire Wye”, Journal of Hydrology, Vol. 304, p.p.51~74. 38. Kuersteiner, G. M. (2002a), Time Series Analysis Lecture Note,NO.1, Massachusetts Institute of Technology, U.S.A. 39. Kuersteiner, G. M. (2002b), Time Series Analysis Lecture Note,NO.4, Massachusetts Institute of Technology, U.S.A. 40. Kleijnen, J. P. C. (2005), “An overview of the design and analysis of simulation experiments for sensitivity analysis”, European Journal of Operational Research, Vol.164, p.p287~300. 41. Katz, A., M. V. Hau and J. Mahoney (2005), Explaining the Great Reversal in Spanish America: Fuzzy-Set Methods Versus Statistical Methods, Department of Sociology, Brown University. 42. Laura C. Schneider, R. Gil Pontius Jr.(2001), "Modeling land-use change in the Ipswich watershed, Massachusetts, USA", Agriculture, Ecosystems and Environment, Vol.85, p.p.83-94 43. Lau, S. S. S. and S. N. Lane (2002), “Biological and Chemical Factors Influenc-ing Shallow Lake Eutrophication: A Long-Term Study”, The Science of the Total Environment, Vol.288, pp.167~181. 44. Lawrence, M. and M. O’Connor (2005), “Judgmental Forecasting in The Presence of Loss Functions”, International Journal of Forecasting, Vol.21, pp.3~14. 45. Mississippi Department of Environmental Quality Office(2003), “Total Maximum Daily Load for Sediment/Siltation and Organic Enrichment/Low Dissolved Oxygen”. 46. Navarro-Esbri, J., E. Diamadopoulos and D. Ginestar (2002), “Time Series Analysis and Forecasting Techniques for Municipal Solid Waste Management”, Resources, Conservation and Recycling, Vol.35, pp.201~214. 47. Nijboer, R. C. and P. F. M. Verdonschot (2004), “Variable Selection for Modelling Effects of Eutrophication on Stream and River Ecosystems”, Ecological Modelling, Vol.177, pp.17~35. 48. Nicolaos Lambrakis, Andreas Antonakos, George Panagopoulos(2004), “The use of multicomponent statistical analysis in hydrogeological environmental research”, Water Research, Vol.38, p.p.1862-1872. 49. Ord, K. (2004), “Charles Holt’s Report on Exponentially Weighted Moving Averages: An Introduction and Appreciation”, International Journal of Forecasting, Vol.20, pp.1~3. 50. Pehlivanoglu, E. and D. L. Sedlak (2004), “Bioavailability of Wastewater-Derived Organic Nitrogen to the Alga Selenastrum Capricornutum”, Water Research, Vol.38, pp.3189~319 51. Pedro R. Peres-Neto, Donald A. Jackson, Keith M. Somers(2004), “How many principal components? stopping rules for determining the number of non-trivial axes revisited”, Computational Statistics & Data Analysis, Vol.49, p.p.974-997. 52. Paul R. Hummel, John L. Kittle, Jr., Paul B. Duda, Avinash Patwardhan(2004), “Calibration of a Watershed Model for Metropolitan Atlanta”, AQUA TERRA Consultants. 53. R.X. Liu , J. Kuang, Q. Gong, X.L. Hou(2003), “Principal component regression analysis with SPSS”, Computer Methods and Programs in Biomedicine, Vol.71, p.p.141-147. 54. Ryo Saegusaa , Hitoshi Sakano , Shuji Hashimoto(2004), “Nonlinear principal component analysis to preserve the order of principal components”, Neurocomputing, Vol.61, p.p.57-70. 55. R.V. Smith, C. Jordan, J.A. Annett(2005), “A phosphorus budget for Northern Ireland: inputs to inland and coastal waters”, Journal of Hydrology, Vol.304, p.p.193-202. 56. Riina Antikainen, Riitta Lemola, Jouni I. Nousiainen, Laura Sokka, Martti Esala, Pekka Huhtanen, Seppo Rekolainen(2005), “Stocks and flows of nitrogen and phosphorus in the Finnish food production and consumption system”, Agriculture, Ecosystems and Environment, Vol.107, p.p.287~305. 57. S.K. Jenson and J. O. Domingue(1988), “Extracting Topographic Structure from Digital Elevation Data for Geographic Information System Analysis”, PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, Vol.54, No.11, p.p.1593~1600. 58. SAS Institute Inc., Cary(1999a), “The CLUSTER Procedure”, SAS/STAT User’s Guide, Version 8, chapter 23. 59. SAS Institute Inc., Cary(1999b), “The PRINCOMP Procedure”, SAS/STAT User’s Guide, Version 8, chapter 52. 60. S. J. Steel, N. Louw(2001), “Variable selection in discriminant analysis: measuring the influence of individual cases”, Computational Statistics & Data Analysis, Vol.37, p.p.249-260. 61. Shijin Ren(2002), “Predicting three narcosis mechanisms of aquatic toxicity”, Toxicology Letters, Vol.133, p.p.127-139. 62. Smith, R. V., C. Jordan and J. A. Annett (2005), “A Phosphorus Budget for Northern Ireland: Inputs to Inland and Coastal Waters”, Journal of Hydrology, Vol.304, pp.193-202. 63. Takao Kunimatsu, Miki Sudo, Takeshi Kawachi(1999), “Loading rates of nutrients discharging from a golf course and a neighboring forested basin”, Wat. Sci. Tech., Vol.39, p.p.99-107. 64. Turner-Fairbank HRC (2000), Speed Prediction for Two-Lane Rural Highways, Research, Development, and Technology, Turner-Fairbank Highway Research Center, Publication No.99-171, VA. 65. Terui, N. and Herman K. van D. (2002), “Combined Forecasts from Linear and Nonlinear Time Series Models”, International Journal of Forecasting, Vol.18, pp.421~438. 66. U.S. EPA(1992), EPA Compendium of Watershed-scale Models for TMDL Development, EPA 841-R-92-002. 67. U.S. EPA (1999), “Draft Guidance for Water Quality-based Decisions:The TMDL Process(Second Edition)”, EPA-841-D-99-001 . 68. U.S. EPA(2004), BASINS User Manual. 69. U.S. EPA(2004), BASINS Exercises. 70. U.S. EPA(2004), BASINS Lectures. 71. U.S. EPA(2004), BASINS Appendix. 72. U.S. EPA(2004), BASINS Technical Notes. 73. U.S. EPA(2004), BASINS Case Study. 74. U.S. EPA(2004), HSPF User Manual. 75. U.S. EPA(2004), WinHSPF 2.0 User Manual. 76. U.S. EPA(2004), WDMUtil Version 2.0 Users Manual. 77. Vivienne H. McNeil, Malcolm E. Cox, Micaela Preda(2005), “Assessment of chemical water types and their spatial variation sing multi-stage luster analysis, Queensland, Australia”, Journal of Hydrology, Vol.310, p.p.181–200. 78. WUNDERLIN DANIEL ALBERTO, DI´ AZ MARI´A DEL PILAR, AME´ MARI´A VALERIA, PESCE SILVIA FABIANA, HUED ANDREA CECILIA, BISTONI MARI´A DE LOS A ´ NGELES(2001), “PATTERN RECOGNITION TECHNIQUES FOR THE EVALUATION OF SPATIAL AND TEMPORAL VARIATIONS IN WATER QUALITY. A CASE STUDY: SUQUI´A RIVER BASIN (CO´ RDOBA–ARGENTINA)”, Wat. Res. Vol. 35, pp. 2881–2894. 79. Wang, Z. and D. A. Bessler (2004), “Fore- casting Performance of Multivariate Time Series Models with Full and Reduced Rank: An Empirical Examination”, International Journal of Forecasting, Vol.20, pp.683~695. 80. Xu, F.-L., S. Tao, R. W. Dawson, and B.-G. Li (2001), “A GIS-Based Method of Lake Eutrophication Assessment”, Ecological Modelling, Vol.144, p.p.231~244. 81. Xiuzhen Li, Duning Xiao, Rob H. Jongman, W. Bert Harms, Arnold K. Bregt(2003), "Spatial modeling on the nutrient retention of an estuary wetland", Ecological Modelling, Vol.167, p.p.33-46 82. Yu, S. L(2000).,“Techniques for Source Water Protection:TMDL Analysis and Best Management Practices,”In Proceeding of The 6th International Workshop on Drinking Water Quality Managenent and Treatment Technology, Taiwan, R.O.C., May 28-29, 2000. 83. Zahran, A. and C. M. Anderson-Cook(2003), “A general equation and optimal design for a 2-factor restricted region”, Statistics & Probability Letters. Vol.64, p.p9~16. 84. Zou, H., Y. Yang (2004), “Combining Time Series for Forecasting”, International Journal of Forecasting, Vol.20, pp.69~84.
摘要: This research was to analysis the complex system in eutrophication, and our research place were Te-Chi reservoir and Shuili-stream watershed. We based on results of a series of analysis to discuss the management strategies in eutrophiction. We wished the result of this study could build a standard method to analysis a complex system, and be as a reference for management. This study could taken apart to three parts, the first part was to optimize the prediction ability of time series models. We used the one factor design method to reach the target. We used additive and multiplicative time series models to predict the eutrophication condition. The result showed that we could reduce 64.14% error in best situation. We also found that the additive model had better predition ability for eutrophication. Although the multiplicative model worked worse, it might be used to find out outliers to improve prediction ability in models. The second part of this study was to find out the key parameter for eutrophication, and the research method we used was multivariate statistical analysis. The statistical methods were including of descriptive statistics and simple regression, multiple regression, regression model selection methods, principle component analysis, discriminant analysis, cluster analysis. (1) descriptive statistics and simple regression: the result of the study showed that water in the Te-Chi reservoir had a trend to be trophic. We looked for the most relative factors with eutrophication by using simple regression. When we considered all data of watershed, the most important eutrophication factors including total phosphorous, suspend solid, water transparency and chlorophyll-a could be obtained. This result was corresponding to Carlson's eutrophication index. (2) multiple regression: The results showed that the R2 value in the multiple regression model was 0.9116 that could highly explain the model variance. And we found that there was collinearity condition between suspending solid, COD, and chlorophyll-a. Otherwise, the results of analysis of variance showed that there were 5 possible outlets. After calibrating the outlets, we could get the total model of eutrophication. This study result showed the model could explain the trend of eutropjication completely, and could be the foundation of reservoir management in the future. (3) regression model selection methods: Three regression methods including forward, backward, and stepwise were used to analyze the important water quality factors that had been screened by simple regression analysis. The results showed that all standardized R2 values of three methods were better than that of the orginal multiple regression model, and the best method was forward regression analysis. Consequently, regression methods could screen efficiently the important water quality factors of eutrophication to promote the explanatory ability of models. (4) principle component analysis: The results showed that the first three principal components were nutrition index of phosphorous, nitrogen, sodium and concentration of alga. And those principal components could explain 75.92% variance of the model in our study. This results proved that the major reason of eutrophication in the Te-Chi reservoir was the nutrient. After comparing principal components with original variables by regression model selection method, we proved the model formed by original variables had better ability to explain the variance of the model, but the principal component could solve the colinearity problem in the model. But if we looked the colinearity like an important subject, we still suggested to use the principal component regression model. (5) discriminant analysis: We used the TSI to be the basis of classification , and three kinds of methods including of discriminant analysis, canonical discriminant analysis and stepwise discriminant analysis. The results showed that the dicrrminant analysis classification had 63.16% correct rate and 0.03% stratified error rate after cross-validating. The best result of stepwise discriminant analysis was using forward or stepwise selection, and its result had 57.89% correct rate . The first canonical factor in the canonical discriminant analysis could properly classified the original data, and could explain 98.06% variance of the model. The first canonical factor was major formed by water temperature, suspending sand, ammonia nitrogen, total phosphorus, cod, chlorophyl-a, and secchi disk. After using discriminant analysis, the results proved that the major water quality factor of eutrophication we chosen in the Te-Chi reservoir could properly classified the data. And the classification result could have the same with the TSI, it also showed that the water quality factors we used could explain the trend of eutrophication. (6) cluster analysis: We used the TSI to be the standard of classification , and two kinds of methods including of hierarchical clustering and disjoint clustering. The results showed that the The best result of hierarchical clustering was using McQuitty's method, and its result best fitted for the trend of eutrophication . The McQuitty's method suggested we could classify the data into 4 clusters. The outliers would affect the result of cluster analysis, and it would have better result after calibrating outliers. In addition the result of disjoint clustering could properly classified the original data when we used 4 clusters as the correct clusters. Variables in the model were arranged by R2 and the order was as follows: phosphate, chlorophyl-a, cod, suspending sand, turbidity, total phosphorus, water temperature, secchi disk, nitrate, dissolved oxygen, ammonia nitrogen, and sodium. The most important variable of all is phosphate. Comparing cluster analysis with discriminant analysis, the results proved that the major water quality factor of eutrophication for classification is phosphorus and had better effect between the middle and heavy eutrophication condition. Both results of cluster and discriminant analysis showed that it would have better index for classification of eutrophication. The third part was to try to decide the landuse management strategies by TMDL. Because we cooperated the plan of National Science Council, we chosen the Shuili-stream as our study area. In this study, we will simulate the TMDL in the watershed by BASINS model developed by USEPA. This stud could divided to four parts. (1)The simulation of best basins delineation: Four Digital Elevation Models (DEMs) were chosen to compare simulation differences. Three DEMs came from the simulation of ArcGis, Surfer, and WinGrid system, respectively. The other one was provided by National Central University. The simulation results, real environmental conditions, and research requirements, etc., were used to decide the best subwatershed numbers and the optimal simulation efficiency. The best results of 17 subwatersheds obtained from the DEM of Surfer system were corresponding to the original boundary ranges and the basic research requirements in the rivernet of 300 hectares. Basing on this result we suggest to proceed watershed analysis more deeply. (2)The simulation of meteorology data and calibration of hydrology parameters in the model: We will use the WDMUtil tool in the BASINS model to simulate the weather data we need. It includes of evapotranspiration every hour and rainfall every hour. After building the database that the model needs, we will calibrate the hydrology factor in the model. The results showed that we could calibrate the AGWRC, LZSN, UZSN parameters in the model to improve the flow simulation result. Finally our simulation result show the error in total flow is 10%, and it is fit the standard in the model. But the simulation result also show that the 50% low flow and seasonal flow error are very large. The large error still needs to discuss and try to improve in the future. (3)The simulation of sediment TMDL: The simulation was taken apart into two parts. The first part was to simulate sediment load in hillside field, including of pervious and impervious land. The simulation result of sediment TMDL was 63642(ton/year) on hillside field. This result was proved reasonable after comparing with reference. The second part was to simulate suspended solid in the rivers. Our simulation was based on the sediment load that was delivered to rivers. The simulation results were calibrated and validated by observations. The sum of squared error was 603.617 after calibrating the model. And the validated simulation results were almost fitted the observation. (4)Deciding the strategies of sedimend TMDL: The study processes were taken apart into four parts. The first part was to choose a standard of TMDL of sediment ,and our choices were including of soil loss tolerance and water quality standards used in Taiwan EPA. The second part was to choose management strategies which were suitable for this watershed. After thinking about the characteristic of the watershed and management target, we will focus on landuse management. The third part of this study was to find out point area. We based on the result of detail lanuse analysis to decide to let the dry farmland in NO. 5 and 14 subbasins be our first management target. The fourth part was to simulate the benefit of landuse management strategies. According to the simulation results by BASINS model, we found out the best result will happen when we managed the No. 5 and 14 subbasins at the same time. The efficiency of management for reducing soil loss was 4.79% and 0.325% for suspended load in the river. But the benefit of management was not obvious in the watershed. Above all, we found out the condition of soil loss was close to balance in this watershed. So we have not to implement the management strategies. But the average suspended load was still higher than water quality standard in the river. Because of our strategies had not enough benefit, we still should enhance management for water quality in this watershed with other strategies in the future.
本研究以自然優養複雜系統為研究對象,以大甲溪流域之德基水庫及南投縣水里溪流域為研究區位進行分析,研究中透過一系列分析之結合,以優養化管理為主題進行探討,希望可藉由研究成果建立一複雜系統分析標準流程,並供未來管理工作之依據及參考。研究中共可分為三大部份進行,第一部份為優養化模式預測能力之最佳化研究,此部份研究希望透過單因子試驗設計分析方法,進行優養模式預測之最佳化,使預測誤差能將到最低,研究中主要採用相加性與相乘性時間序列模式進行預測,初步研究成果將預測誤差最大降低64.14%;而在預測模式比較中發現以相加性模式表現較佳,較符合優養變化之趨勢;但相乘性模式具放大誤差之特性,適合進行優養變化之變異點探討,可助於未來尋求離群值進而降低誤差之研究。 第二部份研究為優養化之動力分析研究,研究中希望藉由多變量統計分析方法進行動力分析,若以所採用之分析法來分,內容又可分為(1)敘述統計與單迴歸分析、(2)複迴歸分析、(3)迴歸模式選擇分析、(4)主成份分析、(5)鑑別分析、(6)群集分析等六部份。(1)敘述統計與單迴歸分析部份,,而敘述統計分析結果發現德基水庫有傾向優養化的趨勢;在透過單迴歸分析結果顯示,以整體水庫為考量時,最能掌握與優養化相關的重要水質因子,包括總磷、懸浮固體、透明度與葉綠素a,與卡爾森優養指標理論相符合。(2)複迴歸分析結果得知複迴歸模式解釋能力達0.9116高解釋力,而各個水質因子中以懸浮固體、COD及葉綠素a有較嚴重的共線性;另外殘差分析結果表示存有5個可能的離群值,在經過離群值修正後,即可得到最後的優養全模式,研究顯示此模式可充分解釋優養化變化情形,可作為未來水庫管理的參考依據。 (3)迴歸模式選擇分析研究中共使用三種迴歸分析選擇法,包括順向、逆向選擇迴歸分析與逐步迴歸分析,分別針對先經過單迴歸分析篩選的優養化重要因子進行分析,以挑選出最適宜的因子。研究結果顯示,三種分析法的標準化R2值均較原始複迴歸表現好,其中又以順向選擇分析最佳,表示使用迴歸分析選擇法可有效選出與優養化最相關的因子,進而提升模式的解釋能力。(4)主成份分析結果顯示前3個主成分分別為營養指標磷、營養指標氮、營養指標鈉與藻類濃度,此3個主成分共可解釋模式之75.92%的變異度,表示德基水庫之優養化主要原因為營養鹽所造成。而將主成分分析結果與原始變數透過迴歸模式選擇法來相互比較後,可證明原始變數之迴歸模式選擇結果對模式變異度解釋能力較高,而主成分分析則是可解決模式變項間之共線問題。但若以共線性消除為重要考量下,仍建議使用主成分迴歸模式為較佳的模式。(5)鑑別分析中,共採用(A)分類鑑別分析、(B)典型鑑別分析、(C)逐步鑑別分析等3種方法來區分資料。研究結果顯示,分類鑑別分析經過交叉驗證後,分類正確率可達63.16%,分層後錯誤率估算為0.03%;而逐步鑑別分析中以順向或逐步選擇法分類效果最佳,經交叉驗證後正確率達57.89%。另外,典型鑑別分析結果中的第1鑑別因素可有效的分類原始資料,可解釋98.06%的變異度,而其主要組成為水溫、懸浮固體、氨氮、總磷、COD、葉綠素a、透明度等因子。而最後鑑別結果發現以優養全模式可有效鑑別優養變化,達到與TSI同樣效果,這表示所選用參數可完成解釋優養變化趨勢。(6)群集分析共採用(A)階層群集分析、(B)非階層群集分析等2種方法來分群資料。研究結果顯示,階層分析結果中以馬氏法最符合優養指標TSI的變化趨勢,而該方法建議將觀測組分為4群;另外,非階層分析結果顯示以4群為準時,其分群效果尚稱顯著,其中變數以磷酸鹽為最重要。而若將群集分析結果與鑑別分析比較,結果顯示分群以磷因子最為重要,中高度優養的分群效果較為明顯,但兩者分析結果的差異亦指出應該有更佳的分類指標存在。 第三部份研究為土地利用總量管制策略制定之研究,因配合國科會計畫之執行,而選用水里溪集水區為研究對象,研究中主要使用美國環保署所發展之整合型整體集水區管理模式BASINS來進行模擬,研究流程共可分為(1)最佳集水分區模擬、(2)氣象資料模擬與模式水文校正、(3)泥砂總量模擬、(4)泥砂總量管理策略研擬等4個部份進行。(1)最佳集水分區模擬部份,在數值等高線模型 (DEM) 方面,除了由中央大學提供外,還選用了ArcGis、Surfer及WinGrid模式模擬的DEM來進行分析,以比較不同DEM所模擬出來的結果。最後,再加上配合現場實際情形與研究所需等指標,來決定最佳集水分區數目,以達到最佳模擬效率之目的。研究顯示,以Surfer模式模擬的DEM為基礎,在門檻值為300公頃所決定的水系下,17個子集水區的分區結果最符合原始邊界範圍與研究之基本需求,因此,建議以此為基礎,進一步進行集水區分析工作。(2)氣象資料模擬與模式水文校正研究中,使用BASINS模式下之WDMUtil工具進行氣象資料模擬,模擬項目包括(A)逐時蒸發散量、(B)逐時降雨量。水文校正結果證實調校模式中AGWRC、LZSN、UZSN等3個變數有助於提升模式模擬之精確度,而最後模式模擬結果總逕流量誤差為10%,尚在模式標準之內,而50%最低流量與季節流量誤差等2個指標則是誤差過大,需進一步的研究探討。(3)泥砂總量模擬方面,模擬分為兩部份進行,第一部份首先針對坡地泥砂產量進行模擬,模擬過程中又將坡地分為可滲透區與非滲透區兩區,最後計算結果為63642(ton/year),與相關研究比較後證實為合理。第二部份進行河道懸浮質之模擬,以坡地泥砂產量運移至河道部份為依據進行,最後模擬結果與實測值比較,校正後其誤差平方和為603.617,而驗證後模擬值亦大部分符合實測值。(4)泥砂總量管理策略研擬部份,研究中共可分為四部份進行,(A)選擇適當總量管理目標,本研究決定以容許土壤流失量與環保署水質相關指標為依據。(B)管理策略之選擇,在考量集水區與管理目標之特性後,決定以土地管理為重點。(C)選擇管理重點區,藉由細部土地利用分析,最後決定以第5、14號子集水區之農地中旱田耕種為管理目標。(D)管理效益之評估,透過模式模擬結果,發現以同時治理第5、14號子集水區之效果最好,泥砂產量控制效益為4.79%,河道懸浮質則為0.325%,但成效並不顯著。最後整合模擬結果,在泥砂產量方面集水區已接近土砂平衡狀況,因此不需管理策略之執行;而懸浮質方面則因未達總量管理標準,且管理策略成效不顯著,未來仍需配合其他管理辦法,加強此區之水質管理。
URI: http://hdl.handle.net/11455/34373
其他識別: U0005-1608200615113600
Appears in Collections:水土保持學系

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.