一般而言，常流量時期高山區域河川流量之量測工具主要有下列兩種，分別為流速計與ADCP。利用這兩種量測工具進行接觸式的河川流速量測之後，該時間之河川流量可由流速量測結果推算而得知，搭配該時間之現地水位記錄可繪製出一條屬於現地的水位-流量率定曲線，此曲線被大量應用於洪水事件高水位時期洪水流量之推估。然而利用率定曲線進行極端洪水事件之流量推算，可確保人員及量測器材之安全，但由於該率定曲線先天特性所致，當洪水流量越大於實際量測期間所測得之最大值，其誤差會隨之而大幅增加。因此，一個安全、可行、可靠及準確的流量量測方法對於位於這些高山區域之河溪研究與調查是非常需要的。
本研究嘗試利用影像分析之方法進行高山區域河川流量之推估，洪泛時期流量推估應用之可行性尤為重點。影像流速分析之方法可簡稱為PIV，為一項藉由影像分析進行目標區域流速分佈推估之技術，在實驗室內為一項技術，廣泛被應用於分析水槽流速流場之分佈狀況，但應用於高山區域溪流等野外現地之相關研究在台灣甚為稀少。本研究所採用之影像分析技術最大之特點為無人造追蹤顆粒，應用於研究室內水槽之PIV技術，通常會添加鋁粉做為追蹤顆粒，但由於現地條件限制，本研究所採用之PIV技術主要追蹤目標為水面上之泡沫或天然顆粒。待目標區域之流速分佈分析完成後，流速結果將被轉換成流量並與現地實際量測和率定曲線推估而得之流量進行分析與比較進而求的現場之速度指數。
本研究以位為台灣雪霸國家公園之內之七家灣溪為研究區域。由比較結果可得知，常流量期間，利用影像分析推估流量之誤差低於利用率定曲線進行流量推估。另外由於高水位期間尚無實測資料，本研究利用同時間不同相片進行流量推估，其結果顯示同時間不同照片所推估之流量值非常相近，由此可知本研究所採用之PIV技術可提供一個可靠且一致的結果。本研究進一步比較由PIV、率定曲線、流速計及ADCP所推估或實測之流量並轉換成指標流速，由結果可知，利用影像分析之流量推估結果，常流量與洪峰流量之速度指數結果截然不同，常流量之速度指數大於1，此結果代表常流量時期表面流速低於平均流速；而於洪峰流量時期的PIV推估值與率定曲線推估值相比，其速度指數小於1，此結果代表表面流速大於平均流速。前者結果與理想的速度剖面並不一致，顯示於常流量低水位時期，其表面流速易受限地底床條件影響；而後者結果顯示，於高水位時期，其結果與理想之速度剖面狀況較為吻合。綜合上述分析可得之，無人造追蹤顆粒之PIV技術對於高山區域流量量測可提供一個安全、準確且可靠之額外選擇，但對於未來實際之應用，仍需幾場大型的洪水事件之現地實際量測結果進行速度指數之率定。

Flow discharge in mountainous area is traditionally measured by contact equipment such as flow current meter or ADCP under baseflow condition. In flood event, the high flow discharge is estimated by rating curve generally. Although the discharge estimated by rating curve is a safe way for operator and measurement equipment under the extreme flow condition, the estimated discharge is inaccurate when the value beyond the maximum value of direct measured discharge. Noncontact measurement methods are thus needed for estimating high flow.
This study utilized the technique of image analysis, PIV, to find a reliable and available way for high flow discharge estimation. While it is commonly applied in the lab setting, PIV has also recently been applied for field studies of measurements of river-surface flows. However, calibration of the PIV instruments is an important issue. The importance of calibration is especially high when applying PIV technique in mountainous areas of Taiwan where river flow is usually shallow during baseflow condition and discharge cannot be easily measured by ADCP. On the other hand, under the high flow condition, safety concern is challenging.
Chijiawan creek was chosen as the study area. The results appeared that the discharge estimated by PIV was more accurate than which estimated by rating curve under the baseflow condition. Due to no direct measured data under the high flow condition, this study estimated and compared the flow discharge by different images at first. From the estimated results, the different images had a high consistency estimated value. This result indicated that the unseeded PIV offered a reliable estimated value during the flood event. The study also calculated the velocity-index by those estimated results and found out the index was 1.17 for base flow and smaller than 0.7 for high flow condition.
In summary, the results appeared that the unseeded PIV offered an alternative to ADCP or rating curve for flood discharge estimation in mountainous area and the index value of direct measured data was useful for future application in this study area. However, this method still needs the direct measured data during the flood event to further calibration, which is suggested for future studies.

1.Adrian, R. J., Particle-imaging techniques for experimental fluid mechanics, Annu. Rev. Fluid Mech., 23, 261– 304, 1991.
2.Aya, S., I. Fujita., and M. Yagyu., Field observation of flood in a river by video image analysis, Proc. Hydraul. Eng., 39, 447–452, 1995.
3.Coleman, H.W., Steele Jr., W.G., Experimentation and Uncertainty Analysis for Engineers, second ed. John Wiley & Sons, Inc., New York, 1999.
4.Clarke, R. T., Uncertainty in the estimation of mean annual flood due to rating-curve in definition. Journal of Hydrology 222, 185–190, 1999.
5.Costa, J.E., Spicer, K.R., Cheng, R.T., Haeni, F.P., Melcher, N.B., Thurman, E. M., Plant, W.J., Keller, W.C., Measuring stream discharge by noncontact methods: a proof-of-concept experiment. Geophysical Research Letters 27 (4), 553e556, 2000.
6.Fread, D.L., Computation of stageedischarge relationships affected by unsteady flows. Water Resources Bulletin 2 (2), 213227, 1975.
7.Fujita, I., and S. Komura., Application of video image analysis for measurements of river-surface flows (in Japanese), Proc. Hydraul. Eng., 38, 733– 738, 1994.
8.Fujita, I., M. Muste, and A. Kruger., Large-scale particle image velocimetry for flow analysis in hydraulic applications, J. Hydraul. Res., 36(3), 397–414, 1998.
9.Fujita, I., S. Aya, and T. Deguchi., Surface velocity measurement of river flow using video images of an oblique angle, paper presented at 27th Congress, Int. Assoc. of Hydraul. Eng. and Res., San Francisco, Calif, 1997.
10.Fenton, J.D., Keller, R.J., The Calculation of Streamflow from Measurement of Stage. Technical Report Cooperative Research Centre for Catchment Hydrology and Centre for Environmental Applied Hydrology. Department of Civil and Environmental Engineering, The University of Melbourne, Australia, 2001.
11.Fujita, I., and T. Hino., Unseeded and seeded PIV measurements of river flows videotaped from a helicopter, J. Visualization, 6(3), 245–252, 2003.
12.Giuliano Di Baldassarre., Pierluigi Claps., A hydraulic study on the applicability of flood rating curves, Hydrology Research, 42.1, 2011.
13.Guillaume Dramais , Jérôme Le Coz, Benoît Camenen, Alexandre Hauet., Advantages of a mobile LSPIV method for measuring flood discharges and improving stageedischarge curves, Journal Hydrology-environment Research,5 (2011) 301-312, 2010.
14.Henderson, F.M., Open Channel Flow. The Macmillan Company, New-York, NY, 1966.
15.Kuczera, G., Correlated rating curve error in flood frequency inference. Water Resour. Res. 32(7), 2119–2127, 1996.
16.Muste, M., Fujita, I., Hauet, A. Large-scale particle image velocimetry for measurements in riverine environments. Special Issue on Hydrologic Measurements. Water Resources Research 44, W00D19. doi: 10.1029/2008WR006950, 2008
17.Muste. M., Ho. H-C., Kim. D., Considerations on direct stream flow measurements using video imagery: Outlook and research needs, J Hydro-environment Research, Volume 5, Issue 4, 289-300, 2011.
18.Rantz, S.E., Measurement and Computation of Streamflow, Vol 1: Measurement of Stage and Discharge. Water Supply Paper 2175. U.S.Geological Survey, Washington, D.C, 1982.
19.Raffel, M., C. Willert, and J. Kompenhans, Particle Image Velocimetry: A Practical Guide, Springer, New York, 1998.
20.Surendra Kumar Mishra1., Vijay P. Singh, Hysteresis-based flood-wave analysis using the concept of strain, Hydrol. Process. 15, 1635–1651, 2001.
21.World Meteorological Organisation, Guide to Hydrological Practice, WMO Pubbl. No. 168, 1994.
22.Weitbrecht. V., Muste M., Creutin J. D., and Jirka G. H., Gesch -windigkeitsmessungen mit Particle-Image-Velocimetry: Labor- und Feldmessungen, in Wasserbauliches Versuchswesen, BAWMitt, 2007.
23.William.th, PIVLAB software, http://pivlab.blogspot.tw/, 2010
24.王筱雯，七家灣溪一號壩壩體及棲地改善工程-泥砂衝擊物理模型及數值分析，雪霸國家公園管理處，2010。
25.林國暉，土石流表面速度影像估算方法之研究國立成功大學水利及海洋工程研究所碩士論文，2006。
26.李明靜，河川表面流速與流量非接觸式量測方法之發展及應用，國立成功大學水利及海洋工程研究所博士論文，2003。
27.柯雅卿，通用型影像量測系統應用研究，國立成功大學水利及海洋工程研究所碩士論文，2004。
28.黃稚桓，具雙重模式光學三維量測系統之建構，國立成功大學水利及海洋工程研究所碩士論文，2009。
29.劉定宣，3D電腦視覺之攝影機校正，國立成功大學航空太空工程研究所碩士論文，2003。
30.繆紹綱，數位影像處理，台灣培生教育出版股份有限公司，新北市，2007