進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-0407201916365000
論文名稱(中文) 遙相關月雨量預報模式應用於石門水庫乾旱預警
論文名稱(英文) Application of Teleconnection-based Monthly Rainfall Forecasting Model for Drought Warning in Shihmen Reservoir Watershed
校院名稱 成功大學
系所名稱(中) 水利及海洋工程學系
系所名稱(英) Department of Hydraulics & Ocean Engineering
學年度 107
學期 2
出版年 108
研究生(中文) 陳弘
研究生(英文) Horng Chen
電子信箱 hung30115@gmail.com
學號 N86064115
學位類別 碩士
語文別 中文
論文頁數 120頁
口試委員 口試委員-陳憲宗
口試委員-蕭政宗
口試委員-陳昭銘
口試委員-徐國錦
指導教授-游保杉
中文關鍵字 乾旱預警模式  乾旱指標  遙相關  雨量預報模式  支撐向量機  隨機森林 
英文關鍵字 drought warning model  drought indices  teleconnection  rainfall forecasting model  support vector machine  random forest 
學科別分類
中文摘要 研究旨在以兩種機器學習法–支撐向量機與隨機森林發展石門水庫乾旱預警模式,預測枯水期未來一至三個月之水庫蓄水量。模式輸入變量除現況乾旱預警指標(如標準化降雨指標與標準化水庫蓄水量指標),本研究亦導入未來乾旱預警指標(如未來之標準化降雨指標),以提升模式預測之精確性。而未來乾旱預警指標之取得,主要以遙相關分析為基礎,採用機器學習法(支撐向量機與隨機森林)進行學習NCEP (National Centers for Environmental Prediction) 大氣因子資料或ENSO (El Nino Southern Oscillation) 指標與枯水期水庫集水區雨量間之關聯性,以建置枯水期雨量預報模式,預測未來一至三個月雨量。最後整合所發展之乾旱預警模式與遙相關月雨量預報模式,進行未來水情燈號之研判,以供乾旱來臨前之超前布署相關決策參考。
結果顯示,於乾旱預警模式中,兩機器學習法之預報能力相近;於遙相關月雨量預報模式中,隨機森林於模式率定期間預報能力略勝支撐向量機,且兩種機器學習法預報結果皆有高值低估、低值高估之傾向,故本研究導入雨量預報誤差修正模式。結果顯示:於率定期間,模式之誤差能被有效修正,但驗證期間則否。
最後整合兩模式,將遙相關月雨量預報模式之預報結果應用至乾旱預警模式,結果顯示,乾旱預警模式於未來第一個月對未來雨量較不敏感,因此整合模式預報結果與使用歷史觀測值相差無幾;於未來第三個月,多數月份對未來雨量皆敏感,因此預報結果受雨量預報準確度影響較大,雨量預報之準確與否便反映在蓄水量預報上。最後將蓄水量預報結果應用至水情燈號之判別,結果顯示,未來第一、第二個月燈號預測之準確率較佳,於未來第三個月仍須加強。
英文摘要 This study aims to develop the drought warning models based on two machine learning (ML) methods, i.e., support vector machine (SVM) and random forest (RF). The Standardized Reservoir Storage Index (SRSI) was forecasted for a period of 1 to 3 months in advance during the dry period using the aforementioned models (i.e., SVM and RF) and then transformed into the corresponding regime lights.
The 37-years rainfall and reservoir storage data obtained from the Shihmen reservoir watershed of Taiwan were used as the study dataset. The first 29-years and the later 8-years data were selected as the calibration period and validation period, respectively. In the drought warning model, the results indicate that adding the future rainfall as the model inputs could improve the accuracy of predicted regime lights. Therefore, a teleconnection-based monthly rainfall forecasting (TMRF) model was built for simulating the future rainfall as the model input of the drought warning model.
The performances of all models were compared using the criteria including root mean square error and modified coefficient of efficiency, and the scattering plot. The results reveal that both the drought waring models based on SVM and RF could perform well, while the TMRF model based on RF outperformed the TMRF model based on SVM. Besides, the performances of the TMRF models were pretty good during the calibration period but not good during the validation period. It was also found that the precision of simulated rainfall would affect the performance of reservoir storage prediction a lot if the future rainfall is a sensitive factor to the drought warning model, especially in the 3rd month prediction. However, the accuracy of the 3rd-month rainfall prediction should be improved in future work.
論文目次 目錄
摘要 I
Extended Abstract II
誌謝 IX
目錄 XI
表目錄 XIII
圖目錄 XV
第一章 緒論 1
1-1 研究動機與目的 1
1-2 文獻回顧 3
1-2-1乾旱 3
1-2-2 氣候變化之遙相關 4
1-2-3 支撐向量機之應用 8
1-2-4 隨機森林之應用 9
1-3 本文組織架構 10
第二章 研究區域與資料概述 12
2-1 研究區域概述 12
2-2 水文資料概述 12
2-3乾旱指標 13
2-3-1標準化降雨指標 14
2-3-2標準化水庫蓄水量指標 18
第三章 研究方法 20
3-1 支撐向量機 20
3-2 隨機森林 26
3-3 模式參數設定 29
3-4 評鑑指標 30
第四章 乾旱預警模式 31
4-1 模式建置 31
4-2 不同機器學習法之預報結果 32
4-3 重要變量之探討 49
第五章 遙相關月雨量預報模式 52
5-1 遙相關指標之介紹與建置 53
5-1-1 遙相關指標之介紹 53
5-1-2 遙相關機制指標建置 56
5-2模式輸入變量選取 62
5-3模式建置流程 63
5-4 不同機器學習法之預報表現 65
5-5誤差修正模式 75
5-5-1 RF誤差修正模式 75
5-5-2 誤差修正後結果 76
5-6模式驗證 80
5-7小結與討論 94
第六章 整合乾旱預警模式與遙相關月雨量預報模式 95
6-1 模式架構 95
6-2 不同機器學習法之預報結果 96
6-3 討論 107
6-4 水情燈號之預測 108
第七章 結論與建議 112
7-1 結論 112
7-2 建議 113
參考文獻 114
附錄一 水庫系統模式 附1-1
附錄二 乾旱預警水情燈號 附2-1
附錄三 乾旱預警模式有無加入未來雨量指標之比較 附3-1
附錄四 遙相關機制建置 附4-1
附錄五 SVM應用於以模擬雨量預測未來水情燈號 附5-1

參考文獻 1. 王盼(2014)。基於隨機森林模型的需水預測模型及其應用。水資源保護,第30卷,第1期,第34-37頁。
2. 朱錦紅、柯文雄、許晃雄(2004)。台灣降水變化及其季節性異常的環流機制。環境保護學會學刊,第27卷,第1期,第72至89頁。
3. 吳定澄、許友貞、李忠潘(2015)。聖嬰–南方震盪現象對台灣氣溫及雨量之影響。海洋工程學刊,第15卷,第一期,第57-65頁。
4. 吳明進、陳圭宏(1996)。聖嬰/南方振盪(ENSO)現象對台灣短期氣候變化影響之研究-春雨部分。八十五年度委託計畫研究報告第CW85-2M-04號,交通部中央氣象局。
5. 呂季蓉(2006)。台灣南部地區長期乾旱趨勢分析之研究。國立成功大學水利及海洋工程學系,碩士論文。
6. 宋嘉文(2003)。氣候變遷對台灣西半部地區降雨及乾旱影響之研究,國立成功大學水利及海洋工程學系,碩士論文。
7. 李明安(2015)。聖嬰現象對台灣天氣系統的影響。台灣氣候變遷調適科技整合研究計畫電子報,第12期。
8. 周家慶(2008)。水庫乾旱風險預警及水庫操作決策支援系統之建置研究。國立台灣海洋大學河海工程學系,博士論文。
9. 林國龍、張毓堯、范光龍(2008)。2002-2003年聖嬰現象對台灣雨量的影響。環境保護學會學刊,第31卷,第1期。
10. 林焜詳(2016)。支撐向量機與隨機森林應用於颱風時雨量預報之比較。國立成功大學水利及海洋工程學系,碩士論文。
11. 袁倫欽(2005)。水庫供水操作與乾旱預警系統之研究。國立台灣海洋大學河海工程學系,博士論文。
12. 袁舴(2016)。時雨量系集預報之即時誤差修正。國立成功大學水利及海洋工程學系,碩士論文。
13. 許晃雄(2010)。乾旱環境下的莫拉克颱風。台大校友雙月刊,第69期,第39至42頁。
14. 許晃雄、陳圭宏(1994)。台灣地區低頻大氣變化與熱帶海洋及大氣的關係。氣象科技中心技術報告彙編,第014號,第33至58頁,交通部中央氣象局。
15. 郭志豪(2012)。應用 ENSO 對台灣南部乾旱之機率預報。國立成功大學水利及海洋工程學系,碩士論文。
16. 郭家妏(2014)。隨機森林在河川水位即時預報之應用。國立成功大學水利及海洋工程學系,碩士論文。
17. 陳思尹(2016)。應用機器學習法於QPESUMS即時雨量預報。國立成功大學水利及海洋工程學系,碩士論文。
18. 陳昭銘、汪鳳如、呂芳川、郭漱泠(2002)。聖嬰現象與1998年台灣異常氣候:全年偏暖與春雨偏多現象。大氣科學期刊,第30期,第331至350頁。
19. 陳憲宗(2006)。支撐向量機及模糊推理模式應用於洪水水位之即時機率預報。國立成功大學水利及海洋工程學系,博士論文。
20. 陳薇伊(2011)。克利金法即時修正大甲溪雷達估計降雨。國立成功大學水利及海洋工程學系,碩士論文。
21. 童慶斌、陳嘉和、劉子明(2002)。結合長期氣象預測資料建立乾旱預警系統。水資源管理2002研討會。
22. 黃文政、程俊銘(2003)。石門水庫集水區降雨量乾旱指標之建立。天氣分析與預報研討會,136-140。
23. 黃文揚、徐南蓉(2005)。以全球海水表面溫度預測台灣附近之夏季降雨量。中國統計學報,第43卷,第4期,第371至386頁。
24. 楊富堤(2001)。水庫供水決策支援系統之研究。國立台灣海洋大學河海工程學系,博士論文。
25. 廖志翔(1995)。台灣地區春季降雨分析及聖嬰/南方振盪(ENSO)關係之探討。八十四年度研究報告第CW84-1A-04號,交通部中央氣象局。
26. 劉雅慈(2018)。石門水庫乾旱預警指標之研究。國立成功大學水利及海洋工程學系,碩士論文。
27. 劉鑌鋈(2009)。利用機器學習修正QPESUMS雷達估計降雨。國立成功大學水利及海洋工程學系碩士論文。
28. 盧孟明(2000)。聖嬰現象與台灣異常氣候關係之探討。大氣科學,第28卷,第2期,第91至114頁。
29. 蕭政宗(2000)。聖嬰現象與台灣地區降雨量之相關性分析。農業工程學報,第46卷,第1期,第93至109頁。
30. 羅萬倫(2015)。短期氣候預報在石門水庫梅雨期之水資源管理應用。國立中央大學水文與海洋科學研究所,碩士論文。
31. Belayneh, A., Adamowski, J., Khalil, B., & Quilty, J. (2016). Coupling machine learning methods with wavelet transforms and the bootstrap and boosting ensemble approaches for drought prediction. Atmospheric Research, 172-173, 37-47.
32. Bjerknes, J. (1969). Atmospheric teleconnection from the equatorial Pacific. Monthly Weather Review, 97, 163–172.
33. Breiman, L., & Cutler, A. (2001). Random Forests. Machine Learning, 45(1), 5-32.
34. Chen, J., Li, M., & Wang, W. (2012). Statistical Uncertainty Estimation Using Random Forests and Its Application to Drought Forecast. Mathematical Problems in Engineering, 2012, 1-12.
35. Chen, J., Li, T., & Shih, C. (2008). Asymmetry of the El Niño-Spring Rainfall Relationship in Taiwan. Journal of the Meteorological Society of Japan, 86(2), 297-312.
36. Chen, S., & Zha, X. (2018). Effects of the ENSO on rainfall erosivity in the Fujian Province of southeast China. Science of The Total Environment, 621, 1378-1388.
37. Das, S. (2018). Goodness-of-Fit Tests for Generalized Normal Distribution for Use in Hydrological Frequency Analysis. Pure And Applied Geophysics, 175(10), 3605-3617.
38. Deng, S., Chen, T., Yang, N., Qu, L., Li, M., & Chen, D. (2018). Spatial and temporal distribution of rainfall and drought characteristics across the Pearl River basin. Science of The Total Environment, 619-620, 28-41.
39. Edwards, D. C. (1997). Characteristics of 20th century drought in the United States at multiple time scales. Atmospheric Science Paper No. 634, May 1–30.
40. Fang, X., & Kuo, Y.-H. (2013). Improving Ensemble-Based Quantitative Precipitation Forecasts for Topography-Enhanced Typhoon Heavy Rainfall over Taiwan with a Modified Probability-Matching Technique. Monthly Weather Review, 26, 3908–3932.
41. Ganguli, P., Reddy, MJ. (2014). Ensemble prediction of regional droughts using climate inputs and the SVM-copula approach. Hydrological Processes, 28(19), 4989-5009.
42. Genuer, R., Poggi, J. M., & Tuleau-Malot, C. (2010). Variable Selection Using Random Forests. Pattern Recognition Letters, 31(14), 2225-2236.
43. Haktanir, T. (1992). Comparison of various flood frequency distributions using annual flood peaks data of rivers in Anatolia. Journal of Hydrology, 136(1-4), 1-31.
44. Haktanir, T., & Horlacher, H. B. (1993). Evaluation of various distributions for flood frequency analysis. Hydrological Sciences Journal, 38(1), 15-32.
45. Hayes, M. J., Svoboda, M. D., Wilhite, D. A., Vanyarkho, O. V. (1999). Monitoring the 1996 drought using the standardized precipitation index. Bulletin of the American Meteorological Society, 80(3), 429-438.
46. Huang, KD., Ye, L., Chen, L., Wang, QS., Dai, L., Zhou, JZ., Singh, VP., Huang, MT., Zhang, JH. (2018). Risk analysis of flood control reservoir operation considering multiple uncertainties. Journal of Hydrology, 565, 672-684.
47. Jalalkamali, A., Moradi, M., & Moradi, N. (2014). Application of several artificial intelligence models and ARIMAX model for forecasting drought using the Standardized Precipitation Index. International Journal of Environmental Science and Technology, 12(4), 1201-1210.
48. Komasi, M., Sharghi, S., Safavi, HR. (2018) Wavelet and cuckoo search-support vector machine conjugation for drought forecasting using Standardized Precipitation Index (case study: Urmia Lake, Iran). Journal of Hydrounformatics, 20(4), 975-988.
49. Krause, P., Boyle, D. P., & Bäse, F. (2005). Comparison of different efficiency criteria for hydrological model assessment. Advances in Geosciences,5, 89-97.
50. Kumar, A., & Chander, S. (1987). Statistical Flood Frequency Analysis — An Overview. Hydrologic Frequency Modeling, 19-35.
51. Landwehr, J. M., Matalas, N. C., & Wallis, J. R. (1979). Estimation of parameters and quantiles of Wakeby Distributions: 1. Known lower bounds. Water Resources Research, 15(6), 1361-1372.
52. Landwehr, J. M., Matalas, N. C., & Wallis, J. R. (1979). Probability weighted moments compared with some traditional techniques in estimating Gumbel Parameters and quantiles. Water Resources Research, 15(5), 1055-1064.
53. Lee, J. H., & Julien, P. Y. (2016). Teleconnections of the ENSO and South Korean precipitation patterns. Journal of Hydrology, 534, 237-250.
54. Legates, D. R., & Mccabe, G. J. (1999). Evaluating the use of “goodness-of-fit” Measures in hydrologic and hydroclimatic model validation. Water Resources Research,35(1), 233-241.
55. Legates, D. R., & Mccabe, G. J. (2012). A refined index of model performance: A rejoinder. International Journal of Climatology,33(4), 1053-1056.
56. Lin, G.F., Chen, G.R., Wu, M.C., & Chou, Y.C. (2009). Effective forecasting of hourly typhoon rainfall using support vector machines. Water Resources Research, 45, W08440.
57. Lin, GF., Chen, GR., Huang, PY., Chou, YC. (2009). Support vector machine-based models for hourly reservoir inflow forecasting during typhoon-warning periods. Journal of Hydrology, 372(1-4), 17-29.
58. Liong, S. Y., & Sivapragasam, C. (2002). Flood Stage Forecasting with Support Vector Machines. American Water Resources Association, vol. 38, iss. 1, 173-186.
59. McKee, T. B., Doesken, N. J., and Kleist, J. (1993). The relationship of drought frequency and duration to time scales. Paper presented at the Proceedings of the 8th Conference on Applied Climatology.
60. Mishra, A. K., and Singh, V. P. (2010). A review of drought concepts. Journal of hydrology, 391(1-2), 202-216.
61. Mishra, A., and Desai, V. (2005a). Drought forecasting using stochastic models. Stochastic Environmental Research and Risk Assessment, 19(5), 326-339.
62. Mishra, A., and Desai, V. (2005b). Spatial and temporal drought analysis in the Kansabati river basin, India. International Journal of River Basin Management, 3(1), 31-41.
63. Mishra, A., and Singh, V. P. (2009). Analysis of drought severity‐area‐frequency curves using a general circulation model and scenario uncertainty. Journal of Geophysical Research: Atmospheres, 114(D6).
64. Mishra, A., Desai, V., and Singh, V. (2007). Drought forecasting using a hybrid stochastic and neural network model. Journal of Hydrologic Engineering, 12(6), 626-638.
65. Mishra, A., Singh, V., and Desai, V. (2009). Drought characterization: a probabilistic approach. Stochastic Environmental Research and Risk Assessment, 23(1), 41-55.
66. Nguyen, TH., El Outayek, S., Lim, SH., Nguyen, VTV. (2017). A systematic approach to selecting the best probability models for annual maximum rainfalls – A case study using data in Ontario (Canada). Journal of Hydrology, 553, 49-58.
67. Okoli, K., Breinl, K., Brandimarte, L., Botto, A., Volpi, E., Di Baldassarre, G. (2018). Model averaging versus model selection: estimating design floods with uncertain river flow data. Hydrological Science Journal, 63(13-14), 1913-1926.
68. Oztekin, T. (2007). Wakeby distribution for representing annual extreme and partial duration rainfall series. Meterological Applications, 14(4),381-387.
69. Prasad, A.M., Iverson, L.R., & Liaw, A., (2006). Newer Classification and Regression Tree Techniques-Bagging and Random Forests for Ecological Prediction. Ecosystems, 9(2), 181-199.
70. Rao, A. R., & Arora, P. S. (1987). An Empiricial Study of Probability Distributions of Annual Maximum Floods. Hydrologic Frequency Modeling, 449-466.
71. Ropelewski, C.F. and M.S. Halpert (1987). Global and regional scale precipitation patterns associated with El Niño/Southern Oscillation. Monthly Weather Review, 115, 1606–1626.
72. Vapnik, V. N. (1995). The Nature of Statistical Learning Theory, Springer-Verlag New York.
73. Walker, G. T. (1923). Correlation seasonal variations of weather VIII. Memoirs of the India Meteorological Department, 24, 75–131.
74. Walker, G. T. (1924). World weather IX. Memoirs of the India Meteorological Department, 24, 275–332.
75. Walker, G. T. (1928). World weather III. Memoirs of the Royal Meteorological Society, 17, 97–106.
76. Walker, G. T. and E. W. Bliss (1932). World weather V. Memoirs of the Royal Meteorological Society, 4, 53–84.
77. Walker, G. T. and E. W. Bliss (1937). World weather VI. Memoirs of the Royal Meteorological Society, 4, 119–139.
78. Wȩglarczyk, S. (1998). The interdependence and applicability of some statistical quality measures for hydrological models. Journal of Hydrology,206(1-2), 98-103.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2020-09-01起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2020-09-01起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw