Earth Surface Deformation in the North China Plain Detected by Joint Analysis of GRACE and GPS Data

Abstract

Mass redistribution of the Earth causes variable loading that deforms the solid Earth. While most recent studies using geodetic techniques focus on regions (such as the Amazon basin and the Nepal Himalayas) with large seasonal deformation amplitudes on the order of 1–4 cm due to hydrologic loading, few such studies have been conducted on the regions where the seasonal deformation amplitude is half as large. Here, we use joint GPS and GRACE data to investigate the vertical deformation due to hydrologic loading in the North China Plain, where significant groundwater depletion has been reported. We found that the GPS- and GRACE-derived secular trends and seasonal signals are in good agreement, with an uplift magnitude of 1–2 mm/year and a correlation of 85.0%–98.5%, respectively. This uplift rate is consistent with groundwater depletion rate estimated from GRACE data and in-situ groundwater measurements from earlier report studies; whereas the seasonal hydrologic variation reflects human behavior of groundwater pumping for agriculture irrigation in spring, leading to less water storage in summer than that in the winter season. However, less than 20% of weighted root-mean-squared (WRMS) reductions were detected for all the selected GPS stations when GRACE-derived seasonal deformations were removed from detrended GPS height time series. This discrepancy is probably because the GRACE-derived seasonal signals are large-scale, while the GPS-derived signals are local point measurements.

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

“…1, the cause of the difference between our results and Liu's work (Liu et al, 2014) is we removed atmospheric and non-tidal ocean-loading effects while they did not. However, we found that the amplitude of GPS after removing atmospheric and non-tidal ocean-loading effects is still greater than the GRACE while we added the AOD1B (GAC solution) de-aliasing model to the GRACE solutions (i.e.…”

“…no atmospheric and non-tidal ocean corrected, please see the Table S1). The most obvious difference between our results and Liu's work (Liu et al, 2014) is that they adopt the load Love numbers from Guo et al (2004) to transform these coefficients into vertical surface displacement estimates. We check the two results of Love numbers (ocean load and atmospheric pressure load) from Guo et al (2004), there are significant differences between ocean-load and atmospheric pressure-load Love numbers.…”

“…In contrast to previous focus study (Liu et al, 2014), the most obvious difference between our results and their work is we removed other loading effects (e.g. atmospheric and non-tidal ocean) in order to reflect the seasonal and long-term displacement caused by TWS loads better.…”

“…Previous studies used GRACE data, land surface models, and well observations to provide insight on groundwater depletion in the NCP (Su et al, 2011;Zhong et al, 2009;Feng et al, 2013;Moiwo et al, 2013;Tang et al, 2013;Huang et al, 2015). Liu et al (2014) has discussed loading displacement in the NCP before. Only five GPS stations' (i.e.…”

“…BJFS, BJSH, JIXN, TAIN, and ZHNZ) data are used in their work. Although they calculated the seasonal amplitudes, phases, and trends of vertical displacement from GRACE and GPS, the atmospheric and non-tidal ocean-loading effects were not removed in the Liu's work (Liu et al, 2014); i.e. they added the Atmosphere and Ocean De-aliasing Level-1B (AOD1B) solution (GAC solution) back to the GRACE spherical harmonic solutions.…”