| Summary: | 博士 === 國立成功大學 === 資源工程學系碩博士班 === 99 === In groundwater hydrology, transmissivity (T) and storativity (S) are important parameters in groundwater recharge estimation. This study develops an analytical approach to estimating T in basins with the use of the stream hydrographs of water level and flow records. Our physical model describes T values in unconfined aquifers as large-scale basin T averages. The proposed analytical approach is useful when data on basins are scarce.
In Taiwan, precipitation varies across locations and seasons. The annual mean rainfall is approximately 2,500 mm, 80% of which falls during the wet season (May–October) and 20% falls during the dry season (November–April). Therefore, this study combines the instantaneous recharge theory, the master recession curve (MRC), and the recession-curve-displacement method to verify estimates of mean basin values. We use stream hydrographic records obtained during the wet and dry seasons. Moreover, statistical methods are applied to estimate groundwater recharge.
We select hydrographic data on daily mean streamflow and water table obtained from three streamflow gauging stations in southern Taiwan. Mean absolute error (MAE) evaluation criteria are used to select the most appropriate season for T estimation. MAE values in the Kaoping, Dongkang, and Linbian basins during the wet (dry) season are 0.48 (0.82), 0.41 (0.61), and 1.08 (1.71), respectively.
Three case studies, which compare field records obtained from a pumping test, demonstrate the feasibility of the proposed analytical approach to estimating T. The MRC of the wet season is appropriate in three basins, indicating that a recharge episode can evaluate aquifer reliably based on stream hydrographic records. However, using the recession index formula of Rorabaugh and Simons (1966), some S estimates seem too small.
In the groundwater recharge study, the case of the Kaoping River is classified as an all-basin case; the Kaoping River basin holds groundwater recharge from the Chishan River basin, Laonong River basin, Ailiao River basin, and Kaoping River. The null hypothesis of normal distribution cannot be rejected at a significance level of 0.05. Furthermore, through the method of moments estimation, we estimate the annual groundwater recharge for the all-basin case and the Kaoping basin at 54.7±13.6 and 46.6±16.9 hundred million m3, respectively. Moreover, a comparison of the error percentages of annual groundwater recharge between the all-basin case and the Kaoping basin reveals that 34% of all samples have 20% or more errors. The analytical results demonstrate that the annual groundwater recharge estimates differ between the all-basin case and the Kaoping basin.
The analytical results show a positive correlation between basin area and recession index. The recession index and the recharge associated with a single rainfall recharge episode are positively related to the basin area. Specifically, larger basins usually have higher recession indices and recharge from a single rainfall recharge episode. Additionally, in estimating groundwater recharge, we must multiply the recharge depth with the basin area using the recession-curve-displacement method. Integration with a larger area and a higher recession index results in a multiplication effect on groundwater recharge. Thus, groundwater recharge is overestimated in the present research cases.
|
|---|
