Publications related to the GRACE Missions (no abstracts)

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

Empirical Assessment of Storm–Time Thermospheric Density Inversion Methods From LEO POD Data

Constant, Charles, Brownhall, Indigo, Aruliah, Anasuya, Ziebart, Marek, and Bhattarai, Santosh, 2026. Empirical Assessment of Storm–Time Thermospheric Density Inversion Methods From LEO POD Data. Earth and Space Science, 13(8):e2025EA004914, doi:10.1029/2025EA004914.

Downloads

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

@ARTICLE{2026ESS...1304914C,
       author = {{Constant}, Charles and {Brownhall}, Indigo and {Aruliah}, Anasuya and {Ziebart}, Marek and {Bhattarai}, Santosh},
        title = "{Empirical Assessment of Storm-Time Thermospheric Density Inversion Methods From LEO POD Data}",
      journal = {Earth and Space Science},
     keywords = {thermospheric density, geomagnetic storms, precise orbit determination, energy dissipation rate, low Earth orbit, data assimilation},
         year = 2026,
        month = aug,
       volume = {13},
       number = {8},
          eid = {e2025EA004914},
        pages = {e2025EA004914},
     abstract = "{Thermospheric mass density is one of the largest sources of operational
        uncertainty for spacecraft in low Earth orbit, particularly
        during geomagnetic storms. The growing population of Global
        Navigation Satellite System--equipped satellites presents a data
        set of opportunity: precise orbit determination (POD) data
        streams can be used to estimate thermospheric density along
        their paths. We present a multi--timescale benchmark of the two
        main cooperative density inversion methods, the Energy
        Dissipation Rate method and POD--accelerometry, against
        accelerometer--derived effective densities from the CHAMP and
        GRACE--FO--A satellites over 49 geomagnetic storms. At orbit--
        effective cadence, both methods achieve ${r}^{2}\ge 0.97$ and a
        log--normal scatter $\text{SD\%}\le 22\%$, and are statistically
        indistinguishable in typical--orbit scatter. However, the Energy
        Dissipation Rate method tracks storm--time variations more
        faithfully and produces fewer large--error orbits (${r}^{2}$
        0.992 vs. 0.985; RMS error 12.0\% vs. 13.5\%), at roughly
        $6\times $ lower computational cost. Increasing the fit--span to
        three orbits reduces $\text{SD\%}$ to 13\%--14\% for both
        methods against matched--interval accelerometer effective
        densities, but ${r}^{2}$ against unaveraged (10 s) accelerometer
        data drops sharply beyond about half an orbit. Optimizing the
        arc length below one full orbit offers an attractive compromise,
        achieving ${r}^{2}=0.93$--0.94 and $\text{SD\%}\approx
        42$--$46\%$ against unaveraged accelerometer data. Retrieval
        accuracy depends strongly on drag--acceleration magnitude, with
        RMS errors below 10\% where drag exceeds $4\times
        1{0}^{-7}\,\mathrm{m}\,{\mathrm{s}}^{-\mathrm{2}}$ but growing
        above 30\% below
        $1{0}^{-8}\,\mathrm{m}\,{\mathrm{s}}^{-\mathrm{2}}$. These
        results provide empirical observation--error statistics for
        storm--time POD--derived density retrievals, directly relevant
        to next--generation assimilative models.}",
          doi = {10.1029/2025EA004914},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2026E&SS...1304914C},
      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 Thu Oct 08, 2026 16:44:01

---

Thu Oct 08, F. Flechtner