圖書典藏及影音數位平台

阿公店水庫(以下稱本水庫)集水區大部分為青灰泥岩地形，質地鬆軟且吸水易化，每遇颱風豪雨即造成庫區大量沉滓運移入庫，自民國42年水庫開始運轉以來，每年約有50萬立方公尺之淤砂量淤積於庫底，至民國80年時，水庫淤積率已達71%。民國86年起辦理「阿公店水庫更新工程計畫」，進行水庫浚渫清淤1,160萬m3，庫區有效容量回復至1,837萬立方公尺，並改建溢洪管及灌溉管，實施空庫防淤操作，期大量減少洪水時之沉滓落淤，以維持水庫之有效容量。自民國95年更新改善完成操作迄今，每年約有20萬立方公尺之淤積，已較更新前減少許多。由過去本水庫空庫防淤操作之觀察，蓄水高程、溢洪管出流量等因子對排砂效率有明顯的影響。藉由現地觀測並整合歷年防淤操作之觀測及淤積測量成果，評估排砂成效，進而提出阿公店水庫之防淤策略，做為水庫後續運轉管理之參考依據。 為進一步減少水庫淤積量，擬透過河道放淤方式增加水庫排砂量。然因水庫下游河道緊鄰岡山地區，增加排砂量恐對下游河防安全及水質造成影響。爰此提出本計畫，除分析水庫空庫防淤成效外，並評估下游河道容受能力，以量化阿公店水庫之放淤量，提供水庫永續操作規劃之參考。

This project has observed empty flushing operation of Agondian Reservoir and evaluated desiltation efficiency. Then the strategy of siltation prevention was suggested. It is also to raise desiltation amount through applying dredged material in river for further reduction of sediment deposition in the reservoir. The carrying capacity of the downstream channel was evaluated to determine the desiltation amount. This project is carried out from March 29th to December 31th in 2019. The results of each main work are describe briefly below. 1. Updating general data (1) General data of reservoir catchment. Images of remote sensing, rainfall records, in- and outflow data of the reservoir were collected. The bare area in the upper catchment has increased 6.74 ha. In comparison to that of 2018 due to the expansion of the existed bare area. Sediment deposition near the confluence of the Jhuoshui Stream had hindered sediment transpot . The administration excavate a new channel to lead the inflow to the spill shaft. The river bank structure, which got damaged in 2018, at the upstream side of the Siaogangshan Bridge has been restored and the river bank has been reinforced. (2) Data of desiltation operation. The concentration-flow calibration curve has been updated by integrating this year’s data.The turbidity-sediment concentration calibration curve has been also updated. There were 6 heavy rain events this year. The total inflow sediment during the 6 events is 105×103 m3 and outflow sediment 122.2×103 m3. (3) A database was constructed to store observation materials and related images. (4) The reference material of reservoir desilting was collected for evaluating the feasibility of applying dredged material in downstream channel. 2. Collecting and analyzing data of automated hydrological observation system (1) Samples for measuring sediment concentration have been taken. Totally 404 samples were taken during the 6 heavy rain events. The inflow sediment concentration of the Jhuoshui Stream reached 20,000 mg/L, and the observed sediment concentration at the downstream of spill shaft was over 5,000 mg/L. (2) The administration had lowered the water storage elevation to 30 m before desiltation operation and kept it between 29-31 m until September 10th. The auxiliary desiltation channel was kept open in this period. And the spillway was open by heavy rain. 3. Analyzing sediment distribution according to the observation results of sediment transport This project using numerical models to estimate inflow flood and sediment, analyzing sediment distribution in the reservoir and assess the impact on the downstream channel respectively. (1) The inflow was calculated by using rainfall-runoff model. The inflow sediment could then be estimated by flow-sediment rating curve. The simulation result is similar to the observation results. (2) The spatial variation of reservoir bed was analyzed by using a 2-d model. The result shows that the sediment transport capacity will reduce if the reservoir water level elevation is higher than 30 m. (3) The bed variation of the downstream channel after heavy rain events was simulated by a 1-d model. The results show a larger variation between spillway and the Siaogangshan Bridge. The erosion depth is about 0.8 m and the deposition is about 0.5 m. The variation downstream the Siaogangshan Bridge is about 0.2 m. The simulation results are similar to that of the channel servey. (4) The sediment transport in the downstream channel was simulated by a 1-d model. The result shows that sediment deposits at the outlet of spillway. It is similar to the observation result. 4. Analyzing the desiltation efficiency of empty flushing (1) The reservoir was operated in the empty flushing period according to the principles: to keep reservoir water level elevation at low stage, to keep the auxiliary desiltation channel open and perturb the sediment, and to open the gate of the spillway if heavy rain occurred in the catchment. The sediment concentration of the outflow is apparently higher. (2) The reservoir water level elevation was kept 29-31 m. The sediment concentration of outflow during heavy rain events is about 5,600-8,900 ppm. (3) In the period of empty flushing operation, the total inflow sediment is estimated to be 105×103 m3 and outflow sediment 122.2×103 m3. In addition to the inflow sediment, 17×103 m3 reservoir sediment was transported out the reservoir. (4) The reservoir sediment survey shows an increase of the storage capacity about 140×103 m3 between Sep. 2018 and Sep. 2019. The increased storage capacity and the desiltation by 6 heavy rain events make the total desiltation amount of 260×103 m3. 5. evaluating the feasibility of applying dredged material in downstream channel Applying the dredged material in the downstream channel is taken into consideration to reduce reservoir sediment. It is also a way of final disposal. A numerical model was used to assess the impact on the river channel and flood prevention. The monitoring data of EPA were introduced to describe the influence on the water quality. (1) Applying dredged material in the downstream channel could enhance the efficiency of sediment reduction. (2) The dredged material is suggested to lay out in the main channel of Agondian River near the outlet of the spill shaft according to the results of shear stress analysis, channel survey. (3) The reservoir outflow during empty flushing period is about 8 cms. 2.5-3.5×103 m3 sediment could be transported by this flow so that 50×103 m3 sediment could be transported into the sea. The residual sediment could be transported by increasing flow. (4) The impact on the downstream channel, flood prevention and water quality was taken into consideration by the feasibility evaluation. The simulation results show that there is no direct impact on the downstream channel and the flood prevention. The water level is under the threshold value. The channel bed has a variation about 0.5 m. The removed sediment could cause high concentration of suspended matters in Agondian River during the scouring process.