Detecting seasonal and long-term vertical displacement in the North China Plain using GRACE and GPS

Abstract

Abstract. In total, 29 continuous Global Positioning System (GPS) time series data together with data from Gravity Recovery and Climate Experiment (GRACE) are analysed to determine the seasonal displacements of surface loadings in the North China Plain (NCP). Results show significant seasonal variations and a strong correlation between GPS and GRACE results in the vertical displacement component; the average correlation and weighted root-meansquares (WRMS) reduction between GPS and GRACE are 75.6 and 28.9 % respectively, when atmospheric and nontidal ocean effects were removed, but the annual peak-topeak amplitude of GPS (1.2-6.3 mm) is greater than the data (1.0-2.2 mm) derived from GRACE. We also calculate the trend rate as well as the seasonal signal caused by the mass load change from GRACE data; the rate of GRACEderived terrestrial water storage (TWS) loss (after multiplying by the scaling factor) in the NCP was 3.39 cm yr −1 (equivalent to 12.42 km 3 yr −1 ) from 2003 to 2009. For a 10-year time span (2003 to 2012), the rate loss of TWS was 2.57 cm yr −1 (equivalent to 9.41 km 3 yr −1 ), which is consistent with the groundwater storage (GWS) depletion rate (the rate losses of GWS were 2.49 and 2.72 cm yr −1 during 2003-2009 and 2003-2012 respectively) estimated from GRACE-derived results after removing simulated soil moisture (SM) data from the Global Land Data Assimilation System (GLDAS)/Noah model. We also found that GRACEderived GWS changes are in disagreement with the groundwater level changes from observations of shallow aquifers from 2003 to 2009, especially between 2010 and 2013. Although the shallow groundwater can be recharged from the annual climate-driven rainfall, the important facts indicate that GWS depletion is more serious in deep aquifers. The GRACE-derived result shows an overall uplift in the whole region at the 0.37-0.95 mm yr −1 level from 2004 to 2009, but the rate of change direction is inconsistent in different GPS stations at the −0.40-0.51 mm yr −1 level from 2010 to 2013. Then we removed the vertical rates, which are induced by TWS from GPS-derived data, to obtain the corrected vertical velocities caused by tectonic movement and human activities. The results show that there are uplift areas and subsidence areas in NCP. Almost the whole central and eastern region of NCP suffers serious ground subsidence caused by the anthropogenic-induced groundwater exploitation in the deep confined aquifers. In addition, the slight ground uplifts in the western region of NCP are mainly controlled by tectonic movement (e.g. Moho uplifting or mantle upwelling).

Vertical Displacements Driven by Groundwater Storage Changes in the North China Plain Detected by GPS Observations

“…As mentioned above, the good seasonal correlation between GPS and GRACE signals indicates that the long-term uplifts revealed by GRACE solutions are probably true and mixed in the GPS measurements. Figure 3 [17] and might imply the unloading of hydrological mass in NCP. However, the trend changes to lightly downward or relatively stable after the end of 2010, indicating the increase of hydrological mass loading in this area.…”

“…The SWS anomalies are also estimated by using the average of simulated soil moisture data (2009-2013) from the four LSMs in GLDAS. The groundwater data adopted in this study can provide more spatial details of shallow and deep GWS variations compared with that used in previous similar study of [17], which contained the monthly groundwater table depth of 20 cities, most of them from a shallow unconfined Remote Sens. 2018, 10, 259 5 of 13 well.…”

“…Previous studies have provided significant insights into the seasonal and long-term vertical displacements in NCP by using GPS and GRACE data. For instance, Liu et al (2014) [16] found that the GPS-and GRACE-derived secular trends and seasonal signals are in good agreement, respectively, with an uplift magnitude of 1-2 mm/year and a correlation of 85.0-98.5%; [17] identified that the average correlation and weighted root-mean-squares (WRMS) reduction between GPS and GRACE are 75.6% and 28.9%.These previous studies have focused on comparison, cross-validation of GPS, and GRACE results.…”

The Assessment of Hydrologic- and Flood-Induced Land Deformation in Data-Sparse Regions Using GRACE/GRACE-FO Data Assimilation

“…At each GPS site, the hourly records of the up (vertical) component are averaged into a monthly averaged time series, consistent with the GRACE temporal resolution. The long-term trend is removed from the time series to reduce the plate tectonic effects [7]. The trend is estimated by fitting the GPS time series using the least-squares adjustment associated with the offset, trend, annual, and semi-annual parameters.…”

“…Such processes cause the Earth's surface to subside or uplift, severely affecting the scientific and industrial sectors [3,4]. Despite being affected by different Earth system drivers, land deformation is dominated by the hydrology cycle at annual and interannual scales [1,[5][6][7]. For instance, vertical motions' annual variations in Japan [8] and the Northwestern United States [9] are induced by snow loadings.…”