Development of The Procedure for Land Subsidence Simulation and Its Application–A Case Study in SiGan, TaChen Village

• Main Authors: Chien-Ming Chen, 陳建銘
• Other Authors: Jing-Wen Chen
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/58683328010646977240

碩士 === 國立成功大學 === 土木工程學系碩博士班 === 93 === 　　In Taiwan, spatial and temporal distribution of rainfall is not uniform. Building new reservoirs usually face the people’s objections. Due to its relatively low development cost, groundwater has been the main source of water supply for most aquacultural and coastal industries. Land-subsidence due to the groundwater overdraft has become the evaluation index for important public works or large development cases. Many researchers keep on developing the models for land-subsidence simulation. However, the models cannot provide suitable information for decision-making on land-subsidence prevention and reclamation. 　　In this study, the integrated numerical groundwater and land-subsidence model, i.e., MODFLOW developed by USGS and its modular subroutine (i.e., IBS1), was used to simulate land-subsidence. For establishing an applicable process to construct land-subsidence models, the study investigated the model in detail, including model construction, initial condition setting, and parameter estimation. The analytical results reveal that the model has a good performance for land-subsidence simulation in the study area by using observed groundwater and compaction data. Moreover, it’s found that the model can appropriately consider the time delay of consolidation and react the compress variety for clay layers. 　　There are eight model parameters that have to be set in the model, in which four parameters for clay layers, i.e., initial head, pre-consolidation head, vertical hydraulic conductivity, and elastic/inelastic storage coefficient, have high non-linear relationships. This study divided the duration for groundwater utilization into several time periods and then used the multiple regression analysis to estimate groundwater levels for each period. According to the results of soil test in the field and laboratory, the study estimated the initial parameter values of soil layers. The aforementioned method decreased the difficulty of model calibration in the study. 　　Finally, based on the features of industries and groundwater utilization, and government’s policies, 5 hypothetical scenarios of groundwater withdrawal were adopted to evaluate the effects of land-subsidence. The analytical results reveal: (1) land-subsidence in the region may be more serious, if groundwater levels decline in the next decade; (2) if the groundwater levels keep constant in the next decade, the land- subsidence still increases 0.65 m; and (3) groundwater levels in the 3th aquifer need to keep from descending for preventing the land-subsidence from getting worse.