水庫集水區極端水文對水質影響之研究

A Study on the Impact of Extreme Hydrological Scenarios on Reservoir Watershed Water Quality

徐宗宏(Tsung-Hung Hsu)

指導教授 ： 林鎮洋

國立臺北科技大學/工程學院/工程科技研究所/博士(2010年)

https://doi.org/10.6841/NTUT.2010.00281

摘要

本研究採用以暴雨事件為基礎之極端流量-暴雨方法(CFS)並設計三類型降雨情境(包括中小型暴雨、大型暴雨以及極端暴雨)以模擬國內魚逮魚崛溪集水區之臨界水質條件，對象包括濃度、單位時間負荷與總負荷等。研究所採用之集水區模式為HSPF以模擬河川流量與水質。流量模擬係由一輔助之專家系統軟體HSPEXP協助流量之率定與驗證，水質之率定與驗證則藉助於一些適合性指標，如判定係數(R2)、效率係數(ENS)、Willmott’s D 指標、以及HSPF應用手冊中所建議的誤差標準等。模擬值與觀測值在率定期間(2005-2006)與驗證期間(2007)均有良好之吻合度。情境模擬的結果，懸浮固體(SS)、總磷(TP)與氨氮(NH3-N)的最大濃度分別為129mg/L(在極端暴雨情境)、0.633mg/L以及0.271mg/L(皆在中小型暴雨)，以上極端濃度又分別為國內甲類水體標準之5.2、31.6以及2.7倍。所以在探討之污染物中以總磷(TP)最為關鍵。此外，本研究也探討了各污染物之最大單位時間負荷與最大總負荷。結果顯示懸浮固體之最大單位時間發生於極端暴雨情境，而總磷與氨氮則發生於中小型暴雨之情境下。模擬結果亦顯示當流量隨暴雨逐漸增加之際，暴雨中後段期間的單位時間負荷與總負荷輸出將相當可觀。而各污染物之最大總負荷輸出皆發生阿里山-莫拉克之情境。例如，懸浮固體在阿里山-莫拉克之暴雨條件下之總負荷高達1,864,803公斤，此一場降雨之負荷值與文獻中相似區域的年負荷上限值相當，故在極端降雨下之總負荷輸出，又以懸浮固體最為關鍵。此外，本研究採用初始沖刷分析以進一步探討上述模擬之結果。結果顯示，在中小型暴雨情境下各污染物皆無明顯之初始沖刷現象，而在大型與極端暴雨境下，總磷與氨氮具有明顯之初始沖刷現象，甚至在某些大雨情境下具有強烈之初始沖刷現象。此亦表示非都會區如本研究之魚逮魚崛溪集水區仍存在有初始沖刷之條件。本研究也探討了國內魚逮魚崛溪集水區、基隆河流域與美國Muddy creek集水區之極端水質濃度之比較。

關鍵字

Critical flow-storm (CFS) Approach ； HSPF ； HSPEXP ； goodness-of-fit measure ； first flush analysis

並列摘要

This study discusses the critical water qualities, including concentration, load flux, and total load, of the Dy-yu creek watershed by applying the storm event-based critical flow-storm (CFS) approach along with three groups of devised storm patterns including, small to medium storms, large storms and extreme storms. The Hydrological Simulation Program-FORTRAN (HSPF) has been used to simulate stream flow and water qualities. The flow simulation was well calibrated with the aid of software, expert system for calibration of HSPF (HSPEXP), while the water quality simulation was calibrated/ validated by means of some goodness-of-fit measures such as the coefficient of determination (R2), Nash-Sutcliffe efficiency (ENS), Willmott’s D index, and mean error criteria recommended by the HSPF application guide. The simulated results are in good agreement with the observed field data in both the calibration (2005-2006) and validation (2007) periods. By applying the CFS approach, the maximum concentration of suspended solid (SS), total phosphorus (TP), and ammonia nitrogen (NH3-N) were found to be 129 mg/L(in extreme storm scenario), 0.633 mg/L, and 0.271 mg/L(both in small to medium storm scenario), respectively, which are 5.2, 31.6, and 2.7 times the allowable concentrations as stipulated by local law. TP is the key pollutant among the discussed constituents. Moreover, the critical load fluxes and total loads of each constituent were also discussed. The critical load flux of SS occurred at extreme storm scenario, while those of nutrients TP and NH3-N happened at small to medium storm scenario. The simulated results also showed that the load fluxes and total loads of pollutants will become noticeable at middle and late stage of the storm as stream flow building up gradually. For example, the maximum total load of SS produced in extreme storm scenario can reach as high as 1,864,803 kg which is on the upper range of the annual loads in similar area. For further quantitative and qualitative analyses, the first flush analysis was applied to the above-simulated results. Results showed that the first flush of discussed pollutants was not significant under most small to medium storms with shorter durations. However, first flush phenomena were prominent for nutrients TP and NH3-N under large and extreme storm scenarios, while some of them, in some large storm scenarios, can be classified as strong first flush. Results also revealed that first flush phenomenon is also applicable to the non-urban areas, such as the Dy-yu creek watershed, under some specific storm conditions. Several case studies about critical concentrations of water body, including Dy-yu creek watershed (Taiwan) and Muddy creek watershed (Virginia), were also discussed.

並列關鍵字

Critical flow-storm (CFS) Approach ； HSPF ； HSPEXP ； goodness-of-fit measure ； first flush analysis

參考文獻

[35] Lee, P.H. (2002). Assessment of First Flush Impact on Source Water Quality, Master thesis, National Taipei University of Technology. (in Chinese)

[2] TWEPA. (2010). “Regulations for controlling Nonpoint Source Pollution.” web site, http://www.epa.gov.tw/.

[3] Chang, N.B., Chen, H.W., Jeng, K.Y., Ning, S.K, and Lee, C.S. (2001). “Evaluation of Non-point Source Loads in the Reservoir Watershed using the GIS/GPS/RS Information Technologies and Numerical Models.” Water International, 26(2): 239–251.

[4] Lin, C.Y., Lin, W.Z. and Chou, W.C. (2002). “Soil erosion prediction and sediment yield estimation: the Taiwan experience.” Soil and Tillage Research, 68(2): 143-152.

[6] Li, K.C. and Yeh M.C. (2004). “Nonpoint source pollution potential index: A case study of the Feitsui Reservoir watershed, Taiwan.” Journal of Chinese Institute of Engineers, 27(2): 253-259.