因受到降雨時空分布不均及氣候變遷影響,水文不確定性相對提高,故如何有效利用及永續經營有限的水資源,便是非常重要的課題。本文將降雨及地下水水位進行小波分析後,藉由小波時頻能量、小波分解及小波相關分析彼此的關係,並依此關係透過多元線性迴歸來建立降雨及地下水水位的關係式,可了解因降雨所導致的地下水水位變動為何。透過小波分析,得到降雨與地下水的大致關為拆解出來的d4(第四段細節)、d5、d6、d7、d8、d9、d10及a10(第十段近似)波段。其中此區域的四個雨量站因為波段型態非常相似,因此可以使用平均雨量來代替四個不同雨量站資料。而迴歸分析時,則要跳脫一般對於降雨地下水水位的直覺認識,將不同頻率段資料各自看成一個不同的變數來進行迴歸,能夠更好的將結果貼近每個地下水水位的觀測資料。;Due to the uneven spatial and temporal distribution of rainfall and the impact of the climate change, hydrological uncertainty has been increasing. Therefore, how to manage limited water resources effectively and sustainably is a crucial issue. This paper analyzes the relationship between rainfall and groundwater level through wavelet transform, wavelet decomposition and wavelet coherence. And then establishes the multiple linear regression between rainfall and groundwater level based on this relationship was established. Through wavelet analysis, the approximate relationship between rainfall and groundwater is identified, which are the disassembled d4 (fourth detail), d5, d6, d7, d8, d9, d10 and a10 (tenth approximate) bands. Among them, the four rainfall stations in this area present similar patterns, therefore the average rainfall was utilized to replace the dataset of four different rainfall stations. When it comes to regression analysis, we need to get rid of the general intuition about the relationship between rainfall and groundwater level. The data in different frequency bands as regarded as different variables for regression, the results show a better fit to the observation data of each groundwater level.