The Taiwan Water Resources Agency uses four techniques to monitor subsidence in Taiwan, namely data from leveling, global positioning system (GPS), multi-level compaction monitoring wells (MCMWs), and interferometry synthetic aperture radar (InSAR). Each data type has advantages and disadvantages and is suitable for different analysis tools. Only MCMW data provide compaction information at different depths in an aquifer system, thus they are adopted in this study. However, the cost of MCMW is high and the number of MCMW is relatively low. Leveling data are thus also adopted due to its high resolution and accuracy. MCMW data provide compaction information at different depths and the experimental data from the wells provide the physical properties. These data are suitable for a physical model. Leveling data have high monitoring density in spatial domain but lack in temporal domain due to the heavy field work. These data are suitable for a black-or grey-box model. Poroelastic theory, which is known to be more conscientious than Terzaghi's consolidation theory, is adopted in this study with the use of MCMW data. Grey theory, which is a widely used grey-box model, is adopted in this study with the use of leveling data. A fusion technique is developed to combine the subsidence predicted results from poroelastic and grey models to obtain a spatially and temporally connected two-dimensional subsidence distribution. The fusion model is successfully applied to subsidence predictions in Changhua, Yunlin, Tainan, and Kaohsiung of Taiwan and obtains good results. A good subsidence model can help the government to make the accurate strategies for land and groundwater resource management.
|Number of pages||7|
|Journal||Proceedings of the International Association of Hydrological Sciences|
|Publication status||Published - 2015 Nov 12|
|Event||9th International Symposium on Land Subsidence, NISOLS 2015 - Nagoya, Japan|
Duration: 2015 Nov 15 → 2015 Nov 19
All Science Journal Classification (ASJC) codes
- Earth and Planetary Sciences(all)