利用GPS季節性垂直位移探討臺灣儲水量變化

Abstract

GPS可以精準測量地殼變形，因此被廣泛用於板塊運動與斷層滑移的研究中。在臺灣，GPS的時間序列中有明顯的季節性訊號，且與水循環有密切的關係。本研究主要分為兩個部分：首先以GPS測站覆蓋率佳的西南部為研究區域，探討地表變形和地下水位之間的關係，接下來再以GPS的垂直位移反演陸上儲水量變化。在多數區域，季節性地表高程變化為乾溼季轉變造成的荷重變化之彈性變形，在大量抽取地下水的區域則是受到孔隙水壓主控。在西部麓山帶，GPS記錄的地表高程變化與地下水位高度相關，且2010和2015年初兩次嚴重的乾旱事件，可以從前一年較小的季節性變化看出。本研究移除平原的測站資料，並以PREM地球層狀模型計算之格林函數建立圓盤荷重模型，再將臺灣以0.2度為間隔劃分隔點，將GPS的垂直位移反演成陸上儲水量變化。平均而言，南部的季節性儲水量變化量約為北部的兩倍，此為梅雨與颱風帶來降水大量集中於夏、秋兩季的結果。與以衛星觀測計算的GLDAS-Noah資料相比，GPS反演的結果可以反應較多不同因素造成的荷重變化。因此，由密集GPS構成的觀測網可以用來反演儲水量變化的空間分布。

Global Positioning System (GPS) is widely used in studying seismic cycle deformation and plate tectonics. In Taiwan, we discover significant seasonal variation in GPS position time series and the seasonality greatly corresponds to hydrological cycle. In this study, we discuss the relation between the surface motion and seasonal water loading in southwestern Taiwan taking advantage of a dense spatial coverage of continuous GPS network. The annual GPS vertical deformation is mostly due to the elastic response to variations of surface loads in the wet and dry seasons, while some plain areas with massive water withdrawal are primary influenced by pore pressure effect. The seasonal vertical deformation on foothills is highly correlated to groundwater level, and is able to detect the occurrence of drought in the early 2010 and 2015 beforehand. We remove stations located in alluvial fan and estimate terrestrial water storage variation using a disk-load model with Green’s functions computed from an elastic earth model, PREM. We divide Taiwan into 0.2 by 0.2 grids and use seasonal GPS vertical displacements to invert the terrestrial water storage. In average, the inverted seasonal water variation is about 2 times larger in southern Taiwan compared to northern Taiwan due to heavy rainfalls during monsoons and typhoons in summer. Comparing soil moisture seasonal variation from GLDAS-Noah, GPS records integrated water storage variation including soil moisture, groundwater, reservoir etc. Consequently, GPS data from a dense array could be used as a tool to map the spatial variation of terrestrial water storage.