結合逕流與示蹤劑模式解釋山地集水區暴雨期間之動態排水行為

Dynamic Drainage Behaviors under Rainstorms in Mountainous Catchments Interpreted by Coupling Runoff and Tracer Models

李俊逸(Jun-Yi Lee)

指導教授 ： 黃誌川

共同指導教授 ： 李建堂(Cheing-Tung Lee)

國立臺灣大學/理學院/地理環境資源學研究所/博士(2021年)

https://doi.org/10.6342/NTU202100708

摘要

新水通過時間分布(TTDew)及其時間變異意涵水文的儲存與釋放，對明白生物地球化學過程及汙染物傳輸至為重要；儘管如此，對於山地小型河川的極端事件的知識缺口仍需調查。本研究測試極端事件時變TTDew的假說，並發展比較三個不同TTDew與儲庫組合的降雨逕流模式，以估計新水比例、通過時間與參數敏感性。這些模式被應用於北台灣坪林，中尺度、森林且破碎地形集水區的六場颱風事件。
結果顯示新水比例與平均停留時間分別是0.2-0.8與2-11 hr，略同於其他研究的小集水區，顯示山地集水區的快速降雨逕流轉換過程。降雨因子對新水比例與平均通過時間分別是正向與負向的關係，推測是高降雨強度啟動優勢流路徑，新水經由這些路徑抵達河道。模式結構差異導致率定參數及儲庫間之變動。然而，採用多重儲水擇齡之模式不只可完美重製河水δ18O，亦可闡述水的儲存與釋放。隨集水區濕度增加，未飽和層偏向釋放年輕的新水給地表下逕流及下滲，飽和層則偏向儲存新水。敏感性分析指出參數敏感性是與時間相依的，確認新水TTD的非線性。模式比較解釋非飽和層主導新水TTD的非穩態性變化，且突顯採納多重儲水擇齡於強降雨及破碎地質環境的TTD推定及連帶溶解質之重要性。

關鍵字

新水比例 ； 通過時間 ； 停留時間 ； 儲齡選擇方程

並列摘要

Transit time distribution of event water (TTDew) and its time variance regulate underlying hydrologic stores and fluxes, which are crucial for understanding biogeochemical processes and containment transport; nevertheless, such knowledge gap is yet to be researched in extreme events and small mountainous rivers. This study tested the hypothesis of time-variant TTDew in extreme events and compared 3 hydrologic models with different combinations of TTDew and storages for estimating event water fraction (Few), transit time (TTew) and parameter sensitivity. The 3 models were applied onto six typhoon events in a mesoscale, forested, and fractured catchment at Ping-Lin, north Taiwan.
Results show that the Few varied from 0.2 to 0.8 and the mean TTew from 2 to 11h, which were comparable with other small-scale catchments worldwide, showing a fast transfer characteristic in our catchments. It is found that mean rainfall intensity positively controls on the Few and negatively on the mean TTew. High intensity rainfall likely activates preferential flow paths and quickly transfers event water to the stream. Model structural variations result in differences in calibrated parameters and in storage between reservoirs. However, the model considering multiple storage selection not only sufficiently reproduced the δ18O signal in streamwater but also interpreted hydrologic stores and releases. Unsaturated storage has a tendency of exporting younger event water; in contrast, saturated storage stores event water. Sensitivity analysis suggested that parameter sensitivity is event-depend and time-variant, affirming a nonlinear behavior in event water TTD. The non-stationarity of event water TTD could be dominated by storage selections from the unsaturated reservoir. Model structure changes influence interpreted drainage behaviors and transit time, demonstrating the importance of multiple storage selection in catchment-scale water and solute transport.

並列關鍵字

event water fraction ； transit time ； residence time ； storAge selection function

參考文獻

Asano, Y., and Uchida, T.: Flow path depth is the main controller of mean base flow transit times in a mountainous catchment, Water Resour. Res., 48, https://doi.org/10.1029/2011WR010906, 2012.

Barnes, C. J., and Bonell, M.: Application of unit hydrograph techniques to solute transport in catchments, Hydrol. Process, 10, 793-802, 1996.

Benettin, P., Bailey, S. W., Rinaldo, A., Likens, G. E., McGuire, K. J., and Botter, G.: Young runoff fractions control streamwater age and solute concentration dynamics, Hydrol. Process, 31, 2982-2986, 10.1002/hyp.11243, 2017a.

Benettin, P., Soulsby, C., Birkel, C., Tetzlaff, D., Botter, G., and Rinaldo, A.: Using SAS functions and high-resolution isotope data to unravel travel time distributions in headwater catchments, Water Resour. Res., 53, 1864-1878, 10.1002/2016wr020117, 2017b.

Berghuijs, W. R., and Kirchner, J. W.: The relationship between contrasting ages of groundwater and streamflow, Geophys. Res. Lett., 44, 8925-8935, 2017.