Abstract
Seasonal signals in satellite geodesy time series are mainly derived from a number of loading sources, such as atmospheric pressure and hydrological loading. The most common method for modeling the seasonal signal with quasi-period is to use the sine and cosine functions with the constant amplitude for approximation. However, due to the complexity of environmental changes, the time-varying period part is very difficult to model by the geometric or physical method. We present a machine learning method with Gaussian process to capture the quasi-periodic signals in the geodetic time series and optimize the estimation of model parameters by means of maximum likelihood estimation. We test the performance of the method using the synthetic time series by simulating the time-varying and quasi-periodic signals. The results show that the fitting residuals of the new model show a better random fluctuation, while the traditional models still leave the clear periodic systematics signals without being fully modeled. The new model illustrates a higher reliability of linear trend estimation, and a lower uncertainty and model fitting RMSE, even in time series with shorter time span. On the other hand, it shows a strong capacity to restore the missing data and predict the future changes in time series. The method is successfully applied to modeling the real coordinate time series of the GNSS site (BJFS) from IGS network, and the equivalent water height (EWH) time series in North China obtained from gravity satellites. Therefore, it is recommended as an alternative for precise model reconstruction and signals extraction of satellite geodesy time series, especially in modeling the complex time-varying signals, estimating the secular motion velocity, and recovering the large missing data.
Similar content being viewed by others
Data availability
The software package used in this paper can be downloaded from the website (https://github.com/SBH08180815/GP_Time_Series_Tool). GNSS data used in this study were obtained from Nevada Geodetic Laboratory (http://geodesy.unr.edu). The data for GRACE are available at http://icgem.gfz-potsdam.de/series. And the data for GLDAS and Precipitation are available at https://disc.gsfc.nasa.gov/.
References
Bao Z, Chang G, Zhang L, Chen G, Zhang S (2021) Filling missing values of multi-station GNSS coordinate time series based on matrix completion. Measurement 183:109862
Bennett RA (2008) Instantaneous deformation from continuous GPS: contributions from quasi-periodic loads. Geophys J Int 174:1052–1064
Bevis M, Brown A (2014) Trajectory models and reference frames for crustal motion geodesy. J Geodesy 88:283–311
Bian Y, Yue J, Ferreira VG, Cong K, Cai D (2021) Common mode component and its potential effect on gps-inferred crustal deformations in Greenland. Pure Appl Geophys 178:1805–1823
Blewitt G, Lavallée D (2002) Effect of annual signals on geodetic velocity. J Geophys Res: Solid Earth 107:ETG 9-1-ETG 9-11
Bogusz J (2015) Geodetic aspects of GPS permanent station non-linearity studies. Acta Geodyn Et Geomater 12(4):180. https://doi.org/10.13168/AGG.2015.0033
Bogusz J, Figurski M (2014) Annual signals observed in regional GPS networks. Acta Geodyn Et Geomater 11:125–131
Bogusz J, Klos A (2016) On the significance of periodic signals in noise analysis of GPS station coordinates time series. GPS Soluti 20:655–664
Bos MS, Fernandes RMS, Williams SDP, Bastos L (2013) Fast Error Analysis of Continuous GNSS Observations with Missing Data. J Geodesy 87:351–360
Chen Q, van Dam T, Sneeuw N, Collilieux X, Weigelt M, Rebischung P (2013) Singular spectrum analysis for modeling seasonal signals from GPS time series. J Geodyn 72:25–35
Didova O, Gunter B, Riva R, Klees R, Roese-Koerner L (2016) An approach for estimating time-variable rates from geodetic time series. J Geodesy 90:1207–1221
Dong D, Fang P, Bock Y, Webb F, Prawirodirdjo L, Kedar S, Jamason P (2006) Spatiotemporal filtering using principal component analysis and Karhunen-Loeve expansion approaches for regional GPS network analysis. J Geophys Res: Solid Earth 111(B03405):1–16
Ghaderpour E, Ghaderpour S (2020) Least-squares spectral and wavelet analyses of V455 andromedae time series: the life after the super-outburst. Publ Astron Soc Pac 132:114504
Ghaderpour E, Pagiatakis SD (2019) LSWAVE: a MATLAB software for the least-squares wavelet and cross-wavelet analyses. GPS Solut 23:50
Ghaderpour E, Vujadinovic T (2020) Change detection within remotely sensed satellite image time series via spectral analysis. Remote Sens 12:4001
Hines TT, Hetland EA (2018) Revealing transient strain in geodetic data with Gaussian process regression. Geophys J Int 212:2116–2130
Horwath M, Rülke A, Fritsche M, Dietrich R (2010) Mass variation signals in GRACE products and in crustal deformations from GPS: a comparison. Springer, Berlin Heidelberg
Hyvärinen A, Oja E (1997) A fast fixed-point algorithm for independent component analysis. Neural Comput 9:1483–1492
Jiang W, Deng L, Zhao L, Zhou X, Liu H (2014) Effects on noise properties of GPS time series caused by higher-order ionospheric corrections. Adv Space Res 53:1035–1046
Klos A, Bos MS, Bogusz J (2018) Detecting time-varying seasonal signal in GPS position time series with different noise levels. GPS Solut 22(1). https://doi.org/10.1007/s10291-017-0686-6
Klos A, Bos MS, Fernandes RMS, Bogusz J (2019) Noise-dependent adaption of the wiener filter for the GPS position time series. Math Geosci 51:53–73
Kondrashov D, Ghil M (2006) Spatio-temporal filling of missing points in geophysical data sets. Nonlin Process Geophys 13:151–159
Koulali A, Clarke PJ (2021) Modelling quasi-periodic signals in geodetic time series using Gaussian processes. Geophys J Int 226:1705–1714
Kreemer C, Blewitt G (2021) Robust estimation of spatially varying common-mode components in GPS time series. J Geodesy. https://doi.org/10.1007/s00190-020-01466-5
Liu N, Dai W, Santerre R, Kuang C (2017) A MATLAB-based Kriged Kalman Filter software for interpolating missing data in GNSS coordinate time series. GPS Solut 22:25
Mao A (1999) Noise in GPS coordinate times series. J Geophys Res. https://doi.org/10.1029/1998JB900033
Matthias S (2008) Gaussian Processes for Machine Learning. Int J Neural Syst 14(02):69–106. https://doi.org/10.1142/S0129065704001899
Ren AK, Xu KK, Shao ZH, Liu XQ, Wang XY (2023) Effect of the 2011 Tohoku-Oki earthquake on continuous GNSS station motions. GPS Solut 27:50
Shen Y, Li W, Xu G, Li B (2014) Spatiotemporal filtering of regional GNSS network’s position time series with missing data using principle component analysis. J Geodesy 88:1–12
Tesmer V, Steigenberger P, Dam TV, Mayer-Guerr T (2011) Vertical deformations from homogeneously processed GRACE and global GPS long-term series. J Geodesy 85:291–310
Tregoning P, Watson C, Ramillien G, Mcqueen H, Zhang J (2009) Detecting hydrologic deformation using GRACE and GPS. Geophys Res Lett. https://doi.org/10.1029/2009GL038718
Webb FH, Zumberge JF (1993) An introduction to the GIPSY-OASIS-II. JPL Publ. D-11088. In
Wei N, Shi C, Liu JN (2015) Annual variations of 3-D surface displacements observed by GPS and GRACE data:a comparison and explanation. Chin J Geophys 58:3080–3088
Williams Simon DP (2004) Error analysis of continuous GPS position time series. J Geophys Res Solid Earth. https://doi.org/10.1029/2003JB002741
Williams S (2003) The effect of coloured noise on the uncertainties of rates estimated from geodetic time series. J Geodesy 76:483–494
Wu H, Li K, Shi W, Clarke KC, Zhang J, Li H (2015) A wavelet-based hybrid approach to remove the flicker noise and the white noise from GPS coordinate time series. GPS Solut 19:511–523
Xu C, Yue D (2015) Monte Carlo SSA to detect time-variable seasonal oscillations from GPS-derived site position time series. Tectonophysics 665:118–126
Xu KK, Gan WJ, Wu JC (2019) Pre-seismic deformation detected from regional GNSS observation network: a case study of the 2013 Lushan, eastern Tibetan Plateau (China), M_s 70 earthquake. J Asian Earth Sci. https://doi.org/10.1016/j.jseaes.2019.05.004
Xu KK, He R, Li KZ, Ren AK, Shao ZH (2022) Secular crustal deformation characteristics prior to the 2011 Tohoku-Oki earthquake detected from GNSS array, 2003–2011. Adv Space Res 69:1116–1129
Xu KK, Wu JC, Wu WW (2015) Detection of transient aseismic slip signals from GNSS spatial-temporal data. Chin J Geophys 58:2330–2338
Zhang N, Xiong J, Zhong J, Leatham K (2018) Gaussian process regression method for classification for high-dimensional data with limited samples. 358–363. https://doi.org/10.1109/ICIST.2018.8426077
Acknowledgements
The research is supported by the National Natural Science Foundation of China (41774041). We are grateful to International GNSS Service (IGS) for providing GNSS data.
Author information
Authors and Affiliations
Contributions
KX proposed this study and wrote the manuscript. SH conducted the numerical computation and experiments. SJ modified the manuscript. JL and JW wrote the program code. WZ reviewed and modified the manuscript. YZ and AR analyzed and validated the method. KL and YL revised the manuscript. KX and All authors were involved in discussions throughout the development.
Corresponding authors
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Xu, K., Hu, S., Jin, S. et al. Reconstruction of geodetic time series with missing data and time-varying seasonal signals using Gaussian process for machine learning. GPS Solut 28, 79 (2024). https://doi.org/10.1007/s10291-024-01616-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10291-024-01616-8