在全球尺度的氣候變遷影響之下，臺灣的降雨型態在時間與空間的分布上愈加集中、不均。近幾年，旱災、極端強降雨事件頻傳，對於臺灣一直以來面臨的缺水問題更是雪上加霜。地下水一直以來為穩定的水資源來源之一，如何妥善利用地下水一直是我們面臨水資源短缺一個非常重要的課題。我們必須先了解地下水的補注情況，再加以評估如何維持此資源以達永續平衡。 本研究(以屏東平原為例)以水文大數據分析之資料探勘方式，從往年的歷史觀測資料去了解研究區之降雨行為；首先以經驗正交函數分析將降雨之時空特徵，再建立並歸納出不同型態的降雨事件對於地下水補注的影響評估。最終結果顯示，長延時強降雨的補注效果最佳、其次依序為長延時弱降雨、短延時弱降雨、短延時強降雨；另外以地下水補注成效之空間分布而言，屏東平原之沖積扇扇頂對降雨補注最為敏感。希望藉由此研究對降雨以及其對地下水補注行為有更深了解，達到水資源永續利用、防範洪旱災害等目標。 ;Under the influence of the global climate change, the rainfalls in Taiwan become more concentrated and uneven whether in temporal or spatial pattern. In recent years, droughts and extreme heavy rainfall events have frequently happened, and it makes the water shortage problem that Taiwan has always faced even worse. Groundwater is one of stable water resources, and how to properly use groundwater is an important issue when we are facing the shortage of water resources. We should realize the recharge behavior first, then assess or find the way for conservation and sustainable development. This study (takes Pingtung plain for example) uses hydrological big data from historical observation data in the past years. First, we pick up rainfall events, and apply Empirical Orthogonal Function (EOF) to extract the spatial and temporal characteristics of rainfalls, then classify the events into four types (base on the intensity and duration). We can determine the relationship between the rainfall characteristics and groundwater recharge. The results show that the long-duration and high-intensity rainfalls have the best performance on groundwater recharge; and the top of alluvial fan in Pingtung Plain is most sensitive to rainfall replenishment. Hope this study can reveal a deeper understanding of rainfall and its behavior of groundwater recharge.
