| Summary: | 碩士 === 國立中興大學 === 土木工程學系所 === 105 === Global climate change has leading to extreme precipitations worldwide. Rainfall distribution of Taiwan is characterized as non-uniform in time and space, and the slope of most rivers are very steep. That results in the watersheds become difficult to store rain water and to use water resources efficiently. Therefore, water resource is always a critical issue in Taiwan and to conjunctive use of surface and ground water is getting important. From the point view of hydrological cycle, the surface flow is more sensitive than groundwater. To better understand water flow characteristics and groundwater level variabilities, a surface and groundwater coupled model, WASH123D, is applied in this study. And the Pingtung Plain is selected as study site. The WASH123D model is a physical based computational model which can simulate surface and groundwater flows with fully interactions. The study used the seasonal outlook with Weather Generator models (WGEN), mesoscale weather model of Weather Research and Forecasting Model (WRF), and watershed model of WASH123D to conduct simulation of Pingtung Plain. The local hydrological impacts to watershed scale hydrology for the upcoming 3 to 6 months of groundwater levels are discussed.
The study used seasonal outlook from Central Weather Bureau (CWB) to incorporate WGEN simulation to produce rainfalls. The results indicated seasonal outlooks of rainfall have tended to hit rainfall range in raining season, but more easily to obtain rainfall amounts in dry season. Besides, the seasonal outlook of rainfall is not easy to predict extreme rainfall scenario both in wet and dry seasons. The sequent groundwater simulation showed the groundwater level in low-lying area of wet season have average errors of 0.85 m as compared to observed wells. The simulations indicated that seasonal outlook integrated WGEN and WASH123D model to implement groundwater levels is feasible. The study also used WRF weather model to generate regional precipitations and used WASH123D for watershed simulations. Integrated simulations showed a reasonable response of groundwater levels are obtained except for some mountainous regions. The inaccuracy is supposedly from the mountainous precipitation generated by WRF or hydraulic conductivities used in WASH123D. Average errors of simulated groundwater level and observations are 1.67 m. Parameters can be further improved in future study. In short, the study has successfully incorporated seasonal outlook with WGEN models, WRF weather model, and WASH123D watershed model, to investigate the possible groundwater levels. This approach is identified useful for possible conjunctive use of surface and ground water resources in Pingtung Plain.
|
|---|
