坡地崩塌與土層含水量息息相關。本研究使用水筒模式(Tank Model)模擬降雨過程集水區地表逕流及土層內水量(滲漏、中間出流及貯水量)之變化，參照日本學者將模擬所得的土層貯水量定義為土壤水分指數(Soil Water Index , SWI)，並將此指數作為坡地崩塌或土石流發生潛勢的判斷指標。然而，水筒模式的分析結果與模式的水文及地文相關參數有關。
本研究以高屏溪流域上游集水區為研究對象，使用粒子群最佳化演算法(Particle Swarm Optimization, PSO)及三場颱風降雨事件來進行水筒模式的參數率定，並將模擬結果與直接借用參考參數之模擬結果進行比較。結果發現水筒模式搭配 PSO 法率定參數所得之模擬結果較能貼切反映集水區地表逕流及土層內水量之變化。本研究並探討 PSO 法加上參數的限制條件對模擬結果的影響，結果顯示增加參數合理的限制條件，可增加參數率定的效率及提升模擬的結果。研究也顯示不同場降雨事件率定所得的參數略有不同，多場降雨事件率定所得參數的平均值可作為該集水區的代表參數。最後，本研究利用高雄集來DF072 土石流潛勢溪流集水區的土砂災害事件對土壤水分指數進行模擬，研究顯示引用 Ishihara and Kobatake (1979)依地質類別提出之參數與應用 PSO 法率定的參數，兩者對第一層水筒貯水深度(S1)與第二層水筒貯水深度(S2)的模擬結果相似。因此，本研究認為若僅須對 S1 或 S2 進行模擬與討論，則可直接引用 Ishihara and Kobatake (1979)提出之參數作為水筒模式的參數。

Landslide is closely related to soil moisture content. In this study, Tank Model was used to simulate the changes in surface runoff and water depth in the soil layer (percolation, intermediate outflow, and storage) in the watershed during the rainfall process. According to Japanese scholars, the simulated storage in the soil layer is defined as Soil Water Index (SWI), and this index is used as an indicator of the potential for landslide or debris flow. Furthermore, the analysis results of Tank Model are related to the model's hydrological and geological related parameters. In this study, the upstream of the Gaoping River Basin is taken as this research object. Particle Swarm Optimization (PSO) and three typhoon rainfall events were used to calibrate the parameters of Tank Model, and compare the simulation results with the simulation results of the reference parameters. It was found that the simulation results obtained by using Tank Model and PSO calibration parameters can better reflect the changes in surface runoff and water depth in the soil layer in the watershed. In this study, the influence of using PSO with parameter constraints on the simulation results is also discussed. The results show that adding reasonable constraints on the parameters can increase efficiency of parameter calibration and improve simulation results. This study also shows that the parameters of the different rainfall event calibration are not the same, and the average value of the parameters obtained from the calibration of multiple rainfall events can be used as the representative parameters of this watershed. Finally, this study uses the Kaohsiung Jilai DF072 potential debris flow torrent watershed to simulate SWI. The simulation results of S1 and S2 are similar between the parameters suggested by Ishihara and Kobatake (1979) and the parameters calibrated by PSO.Therefore, this study noted that if only S1 or S2 needs to be simulated and discussed, the parameters suggested by Ishihara and Kobatake (1979) can be directly used as the parameters of Tank Model.

[1] Kennedy, James, and Russell Eberhart. “Particle swarm optimization.” Proceedings of ICNN'95-international conference on neural networks. Vol. 4. IEEE, 1995.
[2] R. C. Eberhart, and J. Kennedy, “A new optimizer using particle swarm theory”,Proc. Sixth International Symposium on Micro Machine and Human Science, Nagoya, Japan, 1995, pp.39-43.
[3] Y. Shi, and R. C. Eberhart, “A modified particle swarm optimizer”, Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence, pp. 69-73, May 1998.
[4] Y. Shi and R. C. Eberhart, “Empirical Study of Particle Swarm Optimization”,IEEE ,Intl. Congress on Evolutionary Computation, pp. 101-106, Jul. 1999.
[5] Y. Shi, and R. C. Eberhart, “Parameter Selection in Particle Swarm Optimization”, Proceedings of the Seventh Annual Conference on Evolutionary Programming, pages 591-600, 1998
[6] R. C. Eberhart and Y. Shi, “Comparing inertia weights and constriction factors in particle swarm optimization”, Proc. 2000 Congr. Evolutionary Computation, pp. 84-88, 2000-July.
[7] Duan, Q, Sorooshian, S and Gupta, V.K, 1992, “Effective and efficient global optimization for conceptual rainfall-runoff models” Water Resources Research, Vol.28, No.4, pp.1015-1031.
[8] Duan, Q, Sorooshian, S and Gupta, V. K, 1993, “Calibration of rainfall-runoff models: Global Optimiaztion to the Sacramento soil moisture accounting model.” Water Resources Research, Vol.29, No.4, pp.1185-1194.
[9] Duan, Q, Sorooshian, S and Gupta, V.K, 1994, “Optimal use of the SCE-UA global optimization method for calibrating watershed models” J. of Hydrology, Vol.158, No.3, pp.265-284.
[10] Singh, V.P.,“ Computer Models of Watershed Hydrology”, Water Resources Publications, Highlands Ranch, CO, USA, pp.16
[11] Rong-Song Chen, Lan-Chieh Pi and Cheng-Cheng Hsieh (2005), Application of Parameter Optimization Method for Calibrating Tank Model, Journal of the American Water Resources Association, Vol. 41, No. 2, pp. 389-402.
[12]Shih-Kai Chen, Rong-Song Chen, Tzung-Ying Yang, “Application of a tank model to assess the flood-control function of a terraced paddy field.”,Hydrological Sciences Journal, 59 (5) (2014), pp. 1020-1031
[13] Kuok King Kuok, Harun Sobri, and Shamsudin Siti,“Global optimization methods for calibration and optimization of the hydrologic tank model's parameters.”, Canadian Journal of Civil Engineering, 1 (1). pp. 1-14. ISSN 1923-1636 (2010)
[14] Kuok King Kuok, Sobri Harun and Po-Chan Chiu,“Comparison of PSO and SCE for Auto-Calibration of Hourly Tank model’s Parameters.” Int. J. Advance. Soft Comput. Appl., Vol. 3, No. 3 ,(2011)
[15] Singh, V.P., “Computer models of watershed hydrology”, Water Resource Publications, Highlands Ranch, Colorado, USA, 1995
[16] Ishihara Y, Kobatake S, “Runoff model for flood forecasting.” Bulletin of Disaster Prevention Research Institute, Kyoto University, 29(1):27–43 (1979)
[17] Ishihara Y, Kobatake S, “On the storm runoff process in the Ara experimental basin.” Bulletin of the Disaster Prevention Research Institute 26.2 (1976): 83-100
[18] Suryoputro N, Suhardjono, Soetopo W, Suhartanto E,“Calibration of infiltration parameters on hydrological tank model using runoff coefficient of rational method.”, In: AIP conference proceedings, AIP Publishing 1887:020056. , (2017)
[19] Clerc, M., “The swarm and the queen: towards a deterministic and adaptive particle swarm optimization” Proceedings of the IEEE Congress on Evolutionary Compution, Seoul, Korea, 1999.
[20] Eberhart, R. C., and Shi, Y., “Particle Swarm Optimization: Developments, Application and Resources,” Proceedings of the 2001 Congress on Evolutionary Computation, vol.1, 2001, pp. 81-86.
[21] Yokoo and Kazama(2012), Numerical investigations on the relationships between watershed characteristics and water balance model parameters:
[22] Legates DR, McCabe Jr GJ (1999) Evaluating the use of “goodness-of-fit”measures in hydrologic and hydroclimatic model validation. Water Resources
Research 35: 233-241.
[23] 林冠瑋，「陳有蘭溪流域的山崩作用在颱風及地震是建中與河流輸砂量之相對關係」，國立台灣大學地質科學研究所碩士論文，2005
[24] 陳文福、李毅宏，「結合地文與降雨條件以判定土石流發生之研究-以陳有蘭溪集水區為例」，台灣地理資訊學刊，第二期：27-44，2005
[25] 翁秉瑞，「應用統計方法預測邊坡崩塌潛勢」，國立中興大學水土保持學研究所碩士學位論文，2013
[26] 陳樹群、施姵瑜，「極端水文事件土砂量對陳有蘭溪河川型態演變影響分析」，中華水土保持學報，44(4)，311-323，2013
[27] 何春蓀，普通地質學，1994
[28] 李明熹，「土石流發生降雨警戒分析及其應用」，國立成功大學水利及海洋工程研究所博士論文，2006。
[29] 台灣水利論著分類合訂本系列之一:水資源、水文、氣象(5/5)，王如意，pp.3301
[30] 楊國賢(2015)，『不同土地利用型態下降雨-逕流分析之研究』，國立中興大學土木工程學系研究所博士論文。
[31] 小山慎太郎(1976)，”Powell 之共軛方向法於水筒模式參數探索之應用”，日本農業土木學會論文集，第 65 期，pp.42-47
[32] 田中丸治哉：”タンクモデル定数の大域的探索”，農業土木学会論文集，63.4，pp.103-112，1995．
[33] 永井明博，角屋睦：”流出モデル定数の最適化手法”，京都大学防災研究所年報，第 22 号 B-2，pp.209-224，1979．