本研究旨在應用基於位置、尺度及形狀參數的廣義加法模型理論(GAMLSS)於年一日最大降雨量，發展非定常水文頻率分析模式，除了傳統的定常水文頻率分析之外，還引入了時間及多種氣候因子作為模式變量。
本研究使用石門雨量測站及玉山氣象站，針對西元1948年~2015年的年一日最大降雨量，於水文頻率分析常見的機率分布下，假設位置參數與尺度參數可為一常數、線性函數或三次樣條函數，建立年一日最大降雨量之非定常水文頻率分析模式，於水文頻率分析中，採用三種模式:定常模式、非定常模式(參數隨時間變化)以及非定常模式(參數隨氣候因子變化)，並選用信息準則Akaike information criterion (AIC)、Bayesian information criterion (BIC)作為模式評鑑指標，將所選模式進行適合度分析，檢驗其誤差值是否接近標準常態分布，並進一步以概似比檢定加以檢驗模式之間的差異性，以檢定模式是否有顯著的差異性。比較3種模式對於年一日最大降雨的描述能力、不同訊息準則所挑選的模式，探討其位置參數及尺度參數假設的差異性及代入的氣候因子變量對於測站的影響，最後以三種模式推估重現期100年之降雨量進行比較，比較結果顯示:非定常水文頻率分析方法可適切反應水文非定常特性，但其100年重現期降雨量相較於傳統定常頻率分析方法有較大幅度之變動。可以做為未來風險評估參考，應用於現有的水工結構物設施。

This study aims to perform nonstationary frequency analysis of maximum daily precipitation based on the generalized additive models for location, scale and shape parameters (GAMLSS). In order to investigate the nonstationary behavior, time and various climate factors were used as covariates. The precipitation data were collected from Shimen and Yushan stations for the period of 1948 to 2015. The data were fitted by using several common probability distributions and the location and scale parameters in the distributions were allowed in the following forms: constant, linear function and cubic spline function. For the frequency analysis, the study set up three different models based on (1) stationary approach, (2) nonstationary approach by using time as covariate and (3) nonstationary approach by using climate factors as covariate. Then, the models were evaluated based on Akaike information criterion (AIC) and Bayesian information criterion (BIC). Additionally, the likelihood ratio test was applied to check whether the difference between models is significant or not. Finally, the models were used to estimate 100-year precipitation. The results suggest that the proposed nonstationary models perform better than the stationary model but there is a large variation in 100-year precipitation.

1. 朱容君(2007)。廣義加法模式建構購丁國家公園稀有植物潛在分布。國立屏東科技大學森林系碩士論文。
2. 李凌琪、熊立華，江聰、張洪剛(2015)。氣溫對長江上游八塘站年逕流的影響分析。長江流域資源與環境，第24卷，第7期。
3. 林世豪(2014)。基於三次樣條曲線預估與粒子群聚類之語音頻譜編碼。國立臺北科技大學電機工程學系碩士論文。
4. 邱宗儀(2007)。宜蘭縣南澳溪流域之植群分類與製圖。國立宜蘭大學碩士論文。
5. 陳柏均(2013)。非定常性水文頻率分析方法之比較探討:以台灣地區年最大日降雨為例。國立台灣大學土木工程學系碩士論文。
6. 葉信富、李佳勳、葉振峰、李振誥(2014)。臺灣南部地區長時間河川流量季節性趨勢變化之研究。臺灣水利，第62卷，第4期。
7. 葉振峰、葉信富、李振誥(2015)。以Mann-Kendall及Theil-Sen檢定法評估臺灣地區長期河川流量長期時空趨勢變化。中華水土保持學會年會及學術研討會論文集。
8. 詹大千(2005)。應用地理資訊系統探討台北市空氣污染與氣喘病之關係。國立陽明大學衛生資訊與決策學系碩士論文。
9. 鍾侑達、郭峻菖、陳昶憲(2009)。台灣區域降雨趨勢分析。農業工程學報，第55 卷，第4 期 。
10. 釋昭慧(2003)台灣生態環境的結構性問題 -- 以水資源問題為主軸的幾項考察
11. Aitkin, M. A general maximum likelihood analysis of variance components in generalized linear models. Biometrics, 55, 117–128, 1999.
12. Akaike, H. A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19(6), 716–723, 1974.
13. Cole, T. J. and Green, P. J. Smoothing reference centile curves: the lms method and penalized likelihood. Statistics in Medicine, 11, 1305–1319, 1992.
14. Du, T., Xiong, L., Xu, C. Y., Gippel, C. J., Guo, S. & Liu, P. Return period and risk analysis of nonstationary low-flow series under climate change. Journal of Hydrology, 527, 234-250,2015.
15. Famiglietti, J. NASA Study Sees Earth's Water Cycle Pulse Quickening. Jet Propulsion Laboratory California Institute of Technology, 2010.
16. Filliben, J. J. The Probability Plot Correlation Coefficient Test for Normality. Technometrics, 17(1), 111-117, 1975.
17. Hastie, T. J. & Tibshirani, R. J. Generalized Additive Models. Chapman and Hall, London.
18. Kemdall, M.G. Rank correlation methods, Charles Griffin, London, 1975.
19. Lima,C. H. R., Lall, U., Troy, T. J. & Devineni, N. A climate informed model for nonstationary flood risk prediction: Application to Negro River at Manaus, Amazonia. Journal of Hydrology, 522, 594-602, 2015.
20. Lin, X. & Zhang, D. Inference in generalized additive mixed models by using smoothing splines. J. RStatist. Soc. B, 61, 381–400, 1999.
21. Mann, H.B. Non-parametric test against trend. Econometrica, 13, 245-259, 1945.
22. Rigby, R. A. & Stasinopoulos, D. M. A semi-parametric additive model for variance heterogeneity. Statist. Comput, 6, 57–65, 1996a.
23. Rigby, R. A. & Stasinopoulos, D. M. Generalized additive models for location, scale and shape, (with discussion). Appl. Statist, 54, 507–554, 2005.
24. Schwarz, G. Estimating the dimension of a model. Ann. Statist., 6, 461–464, 1978.
25. Sen, P.K. Estimates of the regression coefficient based on Kendall’s tau. Journal of the American Statistical Association, 63, 1379-1389, 1968.