三維水庫水理與優養模式之發展與應用

Abstract

豐枯季節分明的特性，使得台灣地區大部分的水庫水面高程變化劇烈；再加上地形坡度大，使得水庫水面在昇降的過程中，水庫邊界也跟著劇烈的改變。本研究以翡翠水庫為研究對象，發展一個能處理邊界急遽變化下的三維水理水質模式，同時並模擬不同藻類在水庫中不同季節中的變化。 模式以台北翡翠水庫管理局於1999年與2000年的水理、水質與藻類資料，進行模式的檢定與驗證。根據實測資料顯示，這兩年中，水庫的最高與最低的水位差，約為20公尺，變化十分劇烈。於水庫中，夏季水庫中有分層效應，冬季則有翻轉現象。由於上游區域的人為活動，造成翡翠水庫的水質在某些季節有優養化的情形產生。在不同環境因子下，主要生長的藻類種類也隨之改變。 模擬結果顯示，本模式對於劇烈變化的地形模擬程度良好，計算網格隨著水位的不同而變化，入流的位置也隨著調整。於1999年與2000年分別為枯水與豐水期的情形下，本模式依然能夠完整的描述模擬區域的水位變化。與大壩附近的溫度剖面線相比，可以再現水庫中的分層效應。為了模擬水位變化，網格必須能在乾濕兩種狀態進行切換，同時為了符合地形的幾何形狀，網格必須精細，再加上因為網格乾濕狀態的切換，水理與水質的計算時距必須相等，造成整個模式需要的運算時間大幅增加。 本研究所發展的水質模式，除了模擬有機磷、正磷酸鹽、氨氮、硝酸氮、有機氮、生化需氧量與溶氧之外，尚包含可模擬不同藻種的變數。在藻類的模擬中，溫度限制因子的計算方式，與一般模擬單一藻類的模式不同，以鐘型分佈來計算溫度限制因子。模擬結果與實測資料皆顯示水庫中的水質隨著時間改變而有不同的變化。而水庫的主要藻類中，綠藻主要約在夏秋兩季生長，藍綠藻則為春夏，而矽藻的主要生長期間為冬季至春初。本研究同時也發現，於翡翠水庫中，除了矽藻之外，其餘藻類主要營養鹽限制因子為磷系統；矽藻則是氮系統與磷系統交互限制生長。將不同藻類換算回葉綠素a的結果顯示，模擬時距中，水庫於夏末秋初呈現優養化的情形，其餘大部分時間處於中養階段，與實測資料相符。

In this study, a three-dimensional hydrodynamic and water quality model was developed to simulate the circulation pattern and the trophic level of reservoirs with highly variable bathymetry in Taiwan. As reservoirs in Taiwan are characterized by the rapid changes in bathymetry and the transient variations of the storage volumes (i.e. surface elevations), this presents various challenges for the modelers. Preliminary model results obtained for the Feitsui Reservoir show that the transient variation of the storage volume can be reproduced by the model, whereby the dynamic fluctuation of the surface elevation at the dam site is replicated by the model for a two-year period (from 1999 to 2000). Subsequent tests conducted using the model include hydrothermal simulations to ensure the accurate predictions of the spatial and temporal variations of temperature in the reservoir, with particular focus on matching the thermocline structure during the summer stratification period. Results of the temperature simulation reveal that a stratification phenomenon occurred during summer and early autumn in 1999 and 2000, and subsequently led to an overturn phenomenon. The hydrodynamic results derived were then used to run the water quality modeling subprogram. An eutrophication model that can simulate eight water quality variables was also developed in this study. Biological variables were incorporated, including four groups of phytoplankton such as cyanobacteria, green algae, diatom, and all the others. The hydrodynamic and water quality simulation uses the same grids and time steps in order to handle the complexity of the geometry of Feitsui reservoir. Simulation results indicate that the temperature, light, and nutrient are the growth limiting factors for phytoplankton. A new temperature function used in CE-QUAL-ICM was applied in this model. The results shown that this function is suitable for temperature-limiting phytoplankton simulation in Feitsui Reservoir. Both the field data and model simulation results also show that each algae has its own growing period. It was found that phosphorus is the nutrient limiting factor for most phytoplankton, except for diatom, which is controlled by both phosphorus and nitrogen. Lastly, the carbon-phosphorus-nitrogen ratio for the four groups of phytoplankton was made distinct so that the interaction of nutrients could be emphasized.