Studying spatio-temporal patterns of vertical displacements caused by groundwater mass changes observed with GPS

The sensitivity of Global Positioning System (GPS) technique to detect displacements caused by Total Water Storage (TWS) changes has already been analyzed, while its sensitivity to individual TWS components, such as groundwater, is still in question. We use the probabilistic Principal Component Analysis (pPCA) to examine the spatio-temporal variations of vertical displacements from 98 GPS stations located in 9 regions of the world recognized as those where changes in groundwater masses are the most significant. To study the Earth's crust displacements induced by changes in groundwater masses, we remove from GPS displacements those induced by changes in masses of other water storage components using the WaterGAP Global Hydrology Model (WGHM). For validation purposes, displacements predicted by the Gravity Recovery and Climate Experiment (GRACE) missions from which the WGHM compartments other than groundwater are subtracted are also used. We estimate an average global trend of vertical displacements induced by groundwater mass changes equal to 0.4 +/- 0.17 mm/yr for GRACE-WGHM differences. We observe that this trend is overestimated by groundwater-induced displacements estimated for GPS-WGHM differences for several regions around the world and mostly underestimated by WGHM. Spatio-temporal patterns are very coherent between GPS-WGHM differences and GRACEWGHM differences with significant signatures observed for periods of intense natural and/or human-induced groundwater mass changes. We show that the determined GRACE-WGHM differences and GPS-WGHM differences capture most of the wet and dry periods reflected by the Standardized Precipitation Evapotranspiration Index (SPEI). We also find that groundwater-induced displacements estimated from the three datasets we use, show a prominent 6-year cycle. Finally, we demonstrate that GPS displacements can add very essential information to study and monitor local displacements due to groundwater mass changes and may successfully contribute to the climate-related research.

Automated summary

This study uses probabilistic Principal Component Analysis (pPCA) to analyze the spatio-temporal variations of vertical displacements from 98 GPS stations to investigate the impact of groundwater mass changes on crustal displacements. The results show that the trend of vertical displacements induced by groundwater mass changes is overestimated by GRACE-WGHM data but underestimated by GPS-WGHM data. There are significant spatio-temporal patterns and correlations between GPS-WGHM displacements and GRACE-WGHM displacements with periods of intense groundwater mass changes. This study demonstrates that GPS displacements can provide essential information for studying and monitoring local displacements caused by groundwater mass changes.