| 摘要: | 地下水對於支持生態系統和確保人類面對重大與不可預測的氣候變遷的調適非常重要，尤其是當地面水系統因人口快速膨脹和氣候變遷而變得無法永續供給時。因此，地下水成為彈性與調適水資源利用的重要標的。然而，國際上目前針對水資源有效管理的研究相對較少，尤其在探討氣候變遷對區域性地面水和地下水聯合運用的衝擊面向上。因此，本研究應用耦合SWAT-MODFLOW數值模式，其中包括土壤水評估工具(SWAT)和三維有限差分地下水流模式(MODFLOW-NWT)，以估算河川流量、地下水補注和濁水溪沖積扇的地面水與地下水交換量;之後依據補注區的空間分佈與面積比例，採用未來氣候變遷情境探討其對地下水補注的影響。本研究採用TCCIP提供的 5公里的氣候變遷情境空間解析度，共100年具有該流域特徵的未來氣候條件資料，進行相關模擬與分析。模式於河川流量和地下水位率定與驗證結果相當好，顯示模式可描述觀測資料而具有相當可靠度。本研究分別使用Nash–Sutcliffe模型效率係數、 R2、偏差百分比、均方根誤差和平均絕對誤差等方法，量化模式率定與驗證結果。結果顯示，濁水溪流量在SWAT模式的率定和驗證的 NSE 分別為 0.920 和 0.846;北港溪的 NSE 值分別為 0.549 和 0.548。高相關性(NSE = 0.98)和較小誤差(3 m)的地下水位率定結果顯示，利用 SWAT 模式提供的地下水補注量與研究區的地下水流模式結果相當，驗證了耦合SWAT-MODFLOW的適用性。而氣候變遷於對地下水補注時空變異評估中，採用2005-2100年基期與四種二氧化碳排放濃度路徑(RCP)情境做模擬。研究結果顯示，地下水補注區主要在扇頂區，與文獻中劃定的地下水補注地質敏感區相似。而在乾旱年份，河川對淺層含水層的補注率低於地下水向河川的出流量;此外，氣候變遷影響有明顯集中在某幾年的現象。本研究採用排名最佳的GCM (MIROC5)進行氣候變遷對地下水補注衝擊的評估。結果顯示，研究區2020至2100年，氣候變遷在不同RCP中對地下水補注的最大和最小影響率分別為RCP2.6 (66.36%，-41.92%)、RCP4.5 (51.86%，-39.48%)、RCP6.0 (56.11%，-40.13%)，或極端氣候 RCP8.5 (48.93%, -39.85%)。結果顯示，即使地下水補注率主要受到地質和土壤特性影響，氣候變遷情境仍然對地下水補注產生顯著影響。本研究建立耦合SWAT-MODFLOW數值模式，並成功應用於現地的模擬，該模式可作為未來有效與可行的水資源管理工具，協助解決日漸枯竭的地面水和地下水資源。研究成果將有助於決策者和利益相關者，制定永續的水資源政策。;Groundwater is important for supporting ecosystems and assuring human resilience to significant and unpredictable climatic change, especially as surface water systems become more unsustainable due to fast population expansion and climate change. Therefore, groundwater use has become a flexible and adaptive feature for usage demands. However, there are currently fewer studies evaluating the effective management of water resources to be aware of the possible impact of climate change on the combination between the land surface and subsurface in local areas. The objectives of this study were to apply the coupled SWAT-MODFLOW models, which include the Soil Water Assessment Tool (SWAT) and Modular Three-Dimensional Finite-Difference Groundwater Flow (MODFLOW-NWT), to estimate streamflow discharge, groundwater recharge, and water exchange between surface water and groundwater in the Choushui River Alluvial Fan, Taiwan. For further comprehensive strategies, the research assessed the distribution and proportion of recharge areas and the impact of future climate change scenarios influence on groundwater recharge. The finest practical spatiotemporal resolutions of five kilometres over 100 years were selected to accommodate the future climatic conditions of catchment features provided by TCCIP. A strong association exists between simulation and actual observations, as shown by the input′s calibration and verification of the output parameters (streamflow and groundwater level) results. Confidence in the calibrated model was enhanced by validation through generally good statistical performance for the temporal pattern of streamflow and groundwater level, with the Nash–Sutcliffe model efficiency coefficients, R2, percent bias, root mean squared error, and mean absolute error, respectively, which helps achieve a reliable simulation of the watershed response. The results showed that the calibrated and validated SWAT model in the Choushui river were 0.920 and 0.846, respectively, for NSE. For the case of the Pei-Kang river, the NSE values for calibration and validation were 0.549 and 0.548, respectively. Computing the groundwater head of the high correlation (NSE = 0.98) and small error (3 m) revealed that the groundwater recharge from the SWAT model utilized is consistent with the hydrogeological model in the study area and provided the background for coupled SWAT-MODFLOW. The spatiotemporal variability of groundwater recharge for 2005-2100 was estimated under the baseline and four representative concentration pathways (RCPs) scenarios. The findings showed that recharge mainly occurs in the proximal fan area, catching up with some high potential recharge locations with previously delineated sensitive areas for groundwater recharge in literature. The results also showed that during the dry years, the recharge rate from the streams to the shallow aquifer was lower than the groundwater portion discharge to the streams. Besides, the climate change signal predominates the annual variability, resulting in a more pronounced pattern of greater recharge concentrated in a few years. This study expressed the procedure for assessing the impact of climate change on groundwater recharge based on the top one ranking (MIROC5) projection of the GCMs. The maximum and minimum impact rates of climate change on groundwater recharge in the study area from the 2020s to the 2100s were RCP2.6 (66.36%, -41.92%), RCP4.5 (51.86%, -39.48%), RCP6.0 (56.11%, -40.13%), or extreme climate RCP8.5 (48.93%, -39.85%), respectively. The result suggests that even when groundwater recharge lies in geology and soil properties, the effects of climate change still substantially influence it. The well-tested coupled model would be a valuable tool for evaluating a wide variety of realistic scenarios in order to determine the most efficient and workable water resource management plans for replenishing the critically depleted surface water and groundwater supplies. These findings help decision-makers and stakeholders devise sustainable water resource strategies. |
