圖書典藏及影音數位平台

濁水溪沖積扇地下水區位於臺灣西部海岸之中段，為臺灣地下水資源蘊藏最豐富的地區之ㄧ，然而，因當地地質條件及產業結構之故，仰賴地下水資源比例偏重，大量抽取地下水的結果，引發嚴重地層下陷問題，引起社會高度的關注，該流域儼然成為地下水資源調配與管理之重要地點。近年來，受全球氣候變遷影響，豐水期與枯水期降雨量差距更大，進而影響地下水的補注，且在當地地下水抽用量尚未能掌控下，為能即時掌握及預測地下水位變動趨勢，即早針對未來可能發生地層下陷之區域，提出因應對策，故本署研提本計畫，擬建置濁水溪全流域上、中、下游的降雨、河川流量、地下水間關係與智慧型預測模式，俾作為中部地區地層下陷防治之參據。

The shortage of water resources is a global problem. Due to the steep slopes and gradients of rivers, rapid flows and the uneven spatial and temporal rainfall distribution in Taiwan, most of rainfalls flow directly into the ocean. Groundwater has become an important water resource because of its low cost and easy extraction. The upstream zone and the proximal alluvial fan of the Zhuoshui River are good natural recharge areas of groundwater. However, the over extraction of groundwater occurs in the coastland of south-western Taiwan, which results in serious land subsidence. To obtain and predict the trend of groundwater level variations for making countermeasures in response to possible land subsidence areas in the future, this project aims to establish the relationship between rainfall, streamflow as well as groundwater level and to construct intelligent groundwater level prediction models for the whole Zhuoshui River basin based on the collected long-term data of streamflow, groundwater level, rainfall, groundwater extraction, evaporation and hydrogeological factors, which can provide valuable information for the prevention and the treatment of land subsidence. This project consists of four main parts. (1) Collect and pre-process the data of surface water, groundwater and physiographic characteristics. (2) Analyze the interaction between hydrological factors and groundwater level variations - investigate the different types of storm patterns affecting groundwater level variations and verify water balance conditions; and apply the self-organizing map (SOM) to obtain the clustering results of the spatial-temporal relationship between hydrological factors and groundwater level variations, by which the correlation between hydrological factors and groundwater level variations can be established in consideration of the topological characteristics of the SOM. Therefore the mechanism of groundwater level variations can be obtained through analyzing the impacts of the hydrological factors on groundwater level variations between wet and dry seasons. (3) Construct regional intelligent groundwater level prediction models – establish the relationship between rainfall, streamflow and groundwater level in a large time scale through statistical methods based on monthly data; adopt the SOM to obtain possible distributions of groundwater level variations and the relationship between these distributions and rainfall as well as streamflow; and apply the recurrent configuration of nonlinear autoregressive with exogenous inputs (R-NARX) to predicting the groundwater levels such that a regional intelligent groundwater level prediction model can be constructed after model calibration based on the comparison results between prediction models and the SOM. Results demonstrate that the development of regional intelligent groundwater level prediction model produces high accuracy and stability, and can also be applied to other basins suitably. (4) Conduct technical transfer and user training – provide a user manual for the constructed intelligent groundwater level prediction models and conduct user training to help operators understand and implement the constructed models for groundwater management.