本論文以濁水溪沖積扇為研究區域，利用濁水溪沖積扇內第一含水層之地下水位、鄰近之雨量及灌溉用抽水量作為輸入資料，以支撐向量回歸建立能預測未來短期1至3個月之地下水位預報模式，結果顯示在山區提前一個月之地下水位預報之效率係數約落在0.8，提前二至三個月之預報也有不錯的成果，但沿海地區預報結果較山區差，效率係數約落在0.4，而提前二至三個月之預報也較不理想。
除地下水位之短期預報外，本研究也進行氣候變遷下世紀末（2081年至2100年）地下水位之推估；本研究利用臺灣氣候變遷推估資訊與調適知識平台提供之世紀末雨量推估值，以及歷年各月平均地下水位與抽水量作為氣候變遷情境輸入資料，推估在世紀末不同溫室氣體排放情境下，未來地下水位的變化。本研究比較世紀末地下水位與目前地下水位（2007年至2016年）之差異，發現未來豐水期的情境地下水位普遍增高，而枯水期的情境地下水位普遍更低，此與情境雨量的時間分布型態逐漸集中有關。
本研究最後將未來氣候變遷之地下水位推估結果輸入至STAR-BME時間空間分析軟體中，得到濁水溪沖積扇之地下水位分布情形，對於未設地下水位觀測井或資料不足之區域，本研究之地下水位空間分布推估結果將能提供地下水資源應用的參考依據。

The objectives of this study are short-term forecasting and long-term projection of groundwater levels in the Choushui River alluvial fan. Hydrologic data regarding groundwater level, rainfall, and artificial pumping in the study area are collected. First, statistical analysis was applied to understand the basic characteristics of the groundwater level in the Choushui River alluvial fan. Then, four sets of input variables were used in the groundwater level forecasting model developed by the support vector machines. Results of show that using more input variables (including groundwater level, rainfall, and pumping data) achieves more accurate forecasting results. The support vector machine groundwater level forecasting model was applied to perform short-term forecasting with lead times of 1 to 3 months. Forecasting results at point locations were further input into the STAR-BME software to perform spatiotemporal analysis to obtain the spatial pattern of groundwater level. The groundwater level forecasting model was further applied to project long-term groundwater levels under climate change scenarios. Also, the STAR-BME was applied to derive the spatial pattern of groundwater level in the study area under climate change.

Keywords: groundwater level forecasting, support vector regression, climate change, spatiotemporal analysis

Adamowski, J., & Chan, H. F. (2011). A wavelet neural network conjunction model for groundwater level forecasting. Journal of Hydrology, 407(1-4), 28-40.
Allard, D., d'Or, D., & Froidevaux, R. (2011). An efficient maximum entropy approach for categorical variable prediction. European Journal of Soil Science, 62(3), 381-393.
Bogaert, P., & D'Or, D. (2002). Estimating soil properties from thematic soil maps: the Bayesian Maximum Entropy approach. Soil Science Society of America Journal, 66(5), 1492-1500.
Christakos, G. (1990). A Bayesian/maximum-entropy view to the spatial estimation problem. Mathematical Geology, 22(7), 763-777.
Christakos, G. (1991). Some applications of the Bayesian, maximum-entropy concept in geostatistics. In Maximum entropy and Bayesian methods (pp. 215-229). Springer, Dordrecht.
Christakos, G., & Li, X. (1998). Bayesian maximum entropy analysis and mapping: a farewell to kriging estimators. Mathematical Geology, 30(4), 435-462.
Christakos, G., & Serre, M. L. (2000). BME analysis of spatiotemporal particulate matter distributions in North Carolina. Atmospheric Environment, 34(20), 3393-3406.
Christakos, G., Olea, R. A., Serre, M. L., Wang, L. L., & Yu, H. L. (2005). Interdisciplinary public health reasoning and epidemic modelling: the case of black death (p. 320). New York: Springer.
D'Or, D., Bogaert, P., & Christakos, G. (2001). Application of the BME approach to soil texture mapping. Stochastic Environmental Research and Risk Assessment, 15(1), 87-100.
Daliakopoulos, I. N., Coulibaly, P., & Tsanis, I. K. (2005). Groundwater level forecasting using artificial neural networks. Journal of hydrology, 309(1-4), 229-240.
Dudik, M., Phillips, S. J., & Schapire, R. E. (2004). Performance guarantees for regularized maximum entropy density estimation. In International Conference on Computational Learning Theory (pp. 472-486). Springer, Berlin, Heidelberg.
Fletcher, R. (2013). Practical methods of optimization. John Wiley & Sons.
Huang, F., Huang, J., Jiang, S. H., & Zhou, C. (2017). Prediction of groundwater levels using evidence of chaos and support vector machine. Journal of Hydroinformatics, 19(4), 586-606.
Jaynes, E. T. (1957). Information theory and statistical mechanics. Physical review, 106(4), 620.
Rajaee, T., Ebrahimi, H., & Nourani, V. (2019). A review of the artificial intelligence methods in groundwater level modeling. Journal of hydrology, 572, 336-351.
Shannon, C.E. & Weaver, W. (1949) The Mathematical Theory of Communication. University of Illinois Press, Urbana.
Suryanarayana, C., Sudheer, C., Mahammood, V., & Panigrahi, B. K. (2014). An integrated wavelet-support vector machine for groundwater level prediction in Visakhapatnam, India. Neurocomputing, 145, 324-335.
Yoon, H., Jun, S. C., Hyun, Y., Bae, G. O., & Lee, K. K. (2011). A comparative study of artificial neural networks and support vector machines for predicting groundwater levels in a coastal aquifer. Journal of hydrology, 396(1-2), 128-138.
Yu, H. L., Chen, J. C., Christakos, G., & Jerrett, M. (2009). BME estimation of residential exposure to ambient PM10 and ozone at multiple time scales.
Yu, H. L., & Chu, H. J. (2010). Understanding space–time patterns of groundwater system by empirical orthogonal functions: a case study in the Choshui River alluvial fan, Taiwan. Journal of Hydrology, 381(3-4), 239-247.
Yu, H. L., Wang, C. H., Liu, M. C., & Kuo, Y. M. (2011). Estimation of fine particulate matter in Taipei using landuse regression and Bayesian maximum entropy methods. International journal of environmental research and public health, 8(6), 2153-2169.
Yu, H. L., Yang, S. J., Yen, H. J., & Christakos, G. (2011). A spatio-temporal climate-based model of early dengue fever warning in southern Taiwan. Stochastic Environmental Research and Risk Assessment, 25(4), 485-494.
Yu, H. L., & Lin, Y. C. (2015). Analysis of space–time non-stationary patterns of rainfall–groundwater interactions by integrating empirical orthogonal function and cross wavelet transform methods. Journal of Hydrology, 525, 585-597.
Yu, H. L., Lee, C. H., & Chien, L. C. (2016). A spatiotemporal dengue fever early warning model accounting for nonlinear associations with hydrological factors: a Bayesian maximum entropy approach. Stochastic Environmental Research and Risk Assessment, 30(8), 2127-2141.
Yu, H. L., Wu, Y. Z., & Cheung, S. Y. (2020). A data assimilation approach for groundwater parameter estimation under Bayesian maximum entropy framework. Stochastic Environmental Research and Risk Assessment, 34(5), 709-721.
Yu, P. S., Chen, S. T., & Chang, I. F. (2006). Support vector regression for real-time flood stage forecasting. Journal of Hydrology, 328(3-4), 704-716.
方熙廷、陳柏嘉、余化龍、譚義績（2017）貝氏最大熵法應用於雲林地區地層下陷之時空分析，工程環境會刊，第36卷，第32至48頁。
吳孟庭（2015）應用貝氏最大熵法於臺北盆地水文地質推估，國立臺灣大學生物環境系統工程學系碩士論文。
吳郁璋（2019）貝式最大熵法與資料同化方法之整合，國立臺灣大學生物環境系統工程學系博士論文。
李傳煒（2008）濁水溪沖積扇地區地下水之探討與預測研究，國立中興大學土木工程學系碩士論文。
林李耀等（2010）臺灣氣候變遷推估與資訊平台建置計畫第一年期末報告（精簡報告）。
林李耀等（2011）臺灣氣候變遷推估與資訊平台建置計畫第二年期末報告。
林政偉（2018）彰雲地區灌溉地下水抽水量推估之研究，國立成功大學水利及海洋工程學系博士論文。
翁叔平、楊承道（2012）臺灣地區月降雨及溫度 1 公里網格資料庫之建立（1960-2009）及其在近未來（2015-2039）的氣候推估應用，大氣科學，第40卷，第4期，第349至369頁。
張少華（2018）應用貝氏最大熵法推估土壤有效深度，國立臺灣大學生物環境系統工程學系碩士論文。
陳聖傑（2012）地下水水位之主成分分析-以濁水溪沖積扇為例，國立成功大學資源工程學系碩士論文。
國立台灣大學（2009）地下水補注機制水力特性調查分析先驅研究，經濟部水利署。
黃嘉琦、林琨達、葉信富（2017）近年臺灣濁水溪流域地下水儲存量變化趨勢探討，中華水土保持學報，第48卷，第1期，第36至43頁。
葉信富（2016）流域動態儲水量與水文特性時空變遷之研究，科技部研究計畫報告。
楊亞欣（2013）濁水溪沖積扇地下水與地層下陷聯合運用模擬與分析，國立成功大學資源工程學系碩士論文。
經濟部水利署（2020）109年度彰化與雲林地區地層下陷監測及分析。
賴典章、費立沅、江崇榮（2003）臺灣地區地下水分區特性，水文地質調查與應用研討會論文集，第1至24頁。
顧尚真（2010）貝氏最大熵法於 Quantum GIS 上之建構–應用於台灣地區對空氣懸浮粒子長期暴露之研究，國立臺灣大學生物環境系統工程學系碩士論文。