Understanding space-time patterns of groundwater system by empirical orthogonal functions: A case study in the Choshui River alluvial fan, Taiwan

Natural or anthropogenic activities contribute to changes of groundwater levels in space and time. Understanding the major and significant driving forces to changes in space-time patterns of groundwater levels is essential to groundwater management. This study analyzes monthly observations of piezometric heads from 66 wells during 1997-2002 located in the Choshui River alluvial fan of Taiwan, where groundwater has been the important local water resource for myriads of agricultural or industrial demands. Following spatiotemporal estimations of piezometric heads by Bayesian Maximum Entropy method (BME), this work performs rotated empirical orthogonal function (REOF) analysis to decompose the obtained space-time heads into a set of spatially distributed empirical orthogonal functions (EOFs) and their associated uncorrelated time series. Results show that the leading EOFs represent the most significant driving forces to spatiotemporal changes of groundwater levels in the Choshui River aquifer. These include rainfall recharges from upstream Choshui and Pei-Kang River, pumping activities from aquaculture usages in the coastal areas, as well as water exchanges between surface and subsurface flow of Choshui River. In summary, this study shows the strength of the REOF analysis which can effectively provide integrative views of spatiotemporal changes of groundwater, gaining insights of interactions between the groundwater system and other natural and human activities.