Publications related to the GRACE Missions (no abstracts)

• Sorted by Date • Sorted by Last Name of First Author •

Hydrological variations revealed by GNSS vertical displacement inversion combined with probabilistic principal component analysis

Sun, Yinxiao, Chang, Guobin, Feng, Yong, Qian, Nijia, Huan, Yueyang, and Cao, Yu, 2026. Hydrological variations revealed by GNSS vertical displacement inversion combined with probabilistic principal component analysis. Measurement Science and Technology, 37(25):256306, doi:10.1088/1361-6501/ae781b.

Downloads

• from the NASA Astrophysics Data System  • by the DOI System  •

@ARTICLE{2026MeScT..37y6306S,
       author = {{Sun}, Yinxiao and {Chang}, Guobin and {Feng}, Yong and {Qian}, Nijia and {Huan}, Yueyang and {Cao}, Yu},
        title = "{Hydrological variations revealed by GNSS vertical displacement inversion combined with probabilistic principal component analysis}",
      journal = {Measurement Science and Technology},
     keywords = {GNSS, probabilistic principal component analysis, GRACE, equivalent water height},
         year = 2026,
        month = jun,
       volume = {37},
       number = {25},
          eid = {256306},
        pages = {256306},
     abstract = "{Elastic crustal deformation from surface mass transfer can be accurately
        measured by high precision GNSS, providing a new way to invert
        regional terrestrial water storage. This study proposes a
        vertical displacement inversion method based on probabilistic
        principal component analysis (PPCA). PPCA uses a probabilistic
        generative model to explicitly characterize noise, enabling
        robust missing-data interpolation and signal feature extraction.
        The method is applied to GNSS observations in California and
        Nevada, USA, from 2012 to 2022. The results indicate that the
        first nine principal components account for 82.33\% of the data
        variance, revealing interannual modes and seasonal cycles of
        regional water storage change. The inverted equivalent water
        height shows high consistency with GRACE and GLDAS products in
        seasonal variation, with correlation coefficients of 0.74 and
        0.67, respectively. Across four missing data scenarios (random
        deletion, station continuous interruption, seasonal missing, and
        extreme event), the root mean square error of data
        reconstruction via PPCA (3.430 mm, 3.396 mm, 4.726 mm, 3.731 mm)
        is consistently and significantly lower than that of the
        conventional PCA-ALS method (14.368 mm, 22.455 mm, 21.382 mm,
        15.295 mm). Moreover, the GNSS inversion results exhibit higher
        spatial resolution. The annual mean amplitude derived from GNSS
        (112 mm) is larger than those from GRACE (66 mm) and GLDAS (59
        mm), enabling better capture of local hydrological signals. In
        addition, the hydrological drought indices established based on
        GNSS and GRACE inversion results are in high agreement (with a
        correlation coefficient of 0.74), jointly revealing that the
        response of hydrological processes to meteorological drought
        exhibits a characteristic lag of approximately two months. This
        study demonstrates that PPCA provides an effective solution for
        robustly reconstructing data from flawed GNSS observations and
        inverting highly reliable spatiotemporal variations in
        hydrological conditions.}",
          doi = {10.1088/1361-6501/ae781b},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2026MeScT..37y6306S},
      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

Generated by bib2html_grace.pl (written by Patrick Riley modified for this page by Volker Klemann) on Mon Aug 24, 2026 17:14:25

---

Mon Aug 24, F. Flechtner