Analysis And Application Of GPS Time Series In Antarctica And Greenland

The Global Navigation Satellite System(GPS)time series,which boasts a high accuracy,constitutes an effective approach for studying regional crustal displacements,in addition,GPS velocity solutions can validate and constrain glacial isostatic adjustment(GIA)models,which are always used as important corrections for the movements of tectonic plates,and changes in the sea level.Studying the regional crustal displacement in Antarctica and Greenland has important value as a reference for the formation and evolution of global plate tectonics in addition to creating and maintaining reference frames and monitoring the dynamics of ice and snow in polar region.GPS coordinate time series contain both temporal and spatial correlations,the temporal correlations of GPS coordinates can be calculated and determined using maximum likelihood estimation,and spatially correlated coordinates can be considerably reduced using a postprocessing approach,where the spatial correlations in GPS time series are related to the spatial scale of the network.At the global scale,the predominant source of error is the reference frame,which can be reduced by a 7-parameter or 14-parameter similarity transformation;in contrast,for a regional network,the common mode error(CME)constitutes the greatest source of error.The precision of GPS coordinate time series is influenced by not only the CME but also the noise model.Previous researchers have shown that GPS time series contain not only white noise but also colored noise,e.g.,flicker noise and random walk noise.If the effects of colored noise are ignored,the velocity uncertainty can be overestimated by a factor of 4 or even one order of magnitude higher than the signal amplitude,leading to an incorrect physical interpretation.Antarctica and Greenland GPS time series show long-term trend and interannual variability.The long-term trend of contains GIA over the past thousand years and elastic response caused by present-day variations on and near the ice sheet.Annual variability is caused by surface mass loading,which is purely elastic and synchronous with the surface mass changes(SMB),including periodic movements caused by nonstructural deformation,such as glacier mass loading,atmospheric pressure(ATM),continental water storage(GLDAS),and non-tide(ECCO2).The periodic characteristics of nonstructural deformation will lead to periodic changes of reference frame,which cannot be separated from GPS time series by statistical analysis.Moreover,GPS time series contain some abnormal signals,which usually represents some specific physical signals.Extracting a specific abnormal signal has always been one of the hot and difficult points in the geosciences field.For the above reasons,the theories and methods related to GPS time series were studied systematically,and we used the GPS coordinate time series of Antarctica and Greenland to explore the removing of CME,the optimal noise model,GIA assessment,periodic signal correction and transient signals extraction.The main work and contributions of the dissertation are as follows:In Antarctica,we adopt factor analysis for the first time to explore the applicability of time series recorded by 79 GNSS stations from 2010 to 2018 and removed the CME of the residual time series by ICA filtering.The filtering results derived from PCA and ICA are compared and analyzed,after which the AIC is used to determine the optimal noise model before and after ICA/PCA filtering.Finally,we used high-precision GPS data to assess the 7 GIA models.In Greenland,we model crustal displacements caused by ATM,GLDAS,ECCO2,TEM,and SMB for 44 GNET stations.We extract the ATM and SMB common components from GPS time series using ICA.Then,we quantify the effect of each geophysical mechanism from the entire and local perspectives of Greenland on the GPS time series.Finally,we combine GRACE data and determine the effects of transient signals in GPS time series based on M-SSA.Our findings and conclusions are as follows.1.After PCA filtering,the RMS values of the residual time series are reduced by35.24%,23.95% and 30.41% in the E,N,and U components,respectively.After ICA filtering,the RMS values of the residual time series are reduced by 14.45%,8.97%,and 13.27% in the E,N,U components,respectively.The PCA-extracted CME shows some variation over Antarctica,while the CME extracted using ICA has more obvious spatially clustered patterns,indicating that the CME derived from ICA performs better in Antarctica.Different GNSS time series in Antarctica have different optimal noise models with different noise characteristics in different components.The main noise models are the WN+FN and WN+PN models.Furthermore,the spectrum index of most PN is similar to that of FN.2.The predicted velocities of IJ05-R2 are systematically smaller than the GPS observed velocities;the other 6 GIA model predicted velocities are systematically larger than the GPS observed velocities.The WMs of ICE-5G and Geruo13 are relatively small.WANG has the maximum median and WM values.ICE-6G has the minimum WRMS.Because the upper mantle viscosities in the NAP are lower than in the SAP,the GPS velocities shows the largest vertical deformation in the NAP than SAP.In the FRIS ice shelves,the observed GPS velocities and the predicted GIA model velocities are consistent.In EA,the vertical motion is nonsignificant,and the GIA and ice loading have a small impact in this area.3.From the entire perspective,SMB has the maximum influence.The SR is larger in south than in north.ATM has the second greatest influence.The SR of ATM exhibits an overall consistency.ECCO2,GLDAS,and TEM has the least influence.From the local perspective,the differences of adjacent GPS stations near the glacier area are mainly caued by SMB,followed by ATM.The effects of other loads and TEM are small.However,when SMB is removed,the annual amplitude of GPS time series of some stations becomes larger,which indicates that there should be other factors in the stations that have an important contribution to the vertical seasonal variation of the earth's surface.4.Most of GPS stations in Greenland are generally affected by SMB transient signals in 2012–2015.Thesouthwest of Greenland is the most affected area,and the stations close to the Gr IS and glaciers are more affected than those along the coast.The maximum displacement of GPS time series caused by SMB transient is approximately 20.54 mm,and the mean displacement is 5.97 mm.SMB anomalies are mainly caused by runoff,subsidence anomalies during 2009 to mid-2011 were due to the combined contributions from both SMB and D.