Thu Oct 08, F. Flechtner
• Sorted by Date • Sorted by Last Name of First Author •
Feng, Bo, Xu, Yan, Jiang, Nan, Guo, Biaobiao, Yin, Peng, Guo, Ao, Xu, Tianhe, and Schuh, Harald, 2026. Reconstructing terrestrial water storage and quantifying drought evolution in Australia using LSTM networks. Journal of Hydrology, 679:136335, doi:10.1016/j.jhydrol.2026.136335.
• from the NASA Astrophysics Data System • by the DOI System •
@ARTICLE{2026JHyd..67936335F,
author = {{Feng}, Bo and {Xu}, Yan and {Jiang}, Nan and {Guo}, Biaobiao and {Yin}, Peng and {Guo}, Ao and {Xu}, Tianhe and {Schuh}, Harald},
title = "{Reconstructing terrestrial water storage and quantifying drought evolution in Australia using LSTM networks}",
journal = {Journal of Hydrology},
keywords = {GNSS, GRACE, Terrestrial water storage, LSTM, Drought monitoring},
year = 2026,
month = oct,
volume = {679},
eid = {136335},
pages = {136335},
abstract = "{High-precision monitoring of Terrestrial Water Storage (TWS) is critical
amid climate-driven increases in drought severity. To overcome
the inherent limitations of the Gravity Recovery and Climate
Experiment (GRACE)---coarse spatial resolution and data gaps and
the Global Navigation Satellite System (GNSS)---uneven station
distribution and high-frequency noise, we introduce a Long
Short-Term Memory (LSTM) framework. This framework reconstructs
high-quality TWS across Australia by integrating Equivalent
Water Height (EWH) time series, meteorological data, and
temporal markers to effectively capture GNSS-GRACE
discrepancies. Joint inversion results demonstrate substantial
spatiotemporal consistency with Global Land Data Assimilation
System (GLDAS), GRACE, and precipitation records. Notably, the
LSTM-enhanced EWH significantly improves agreement with
benchmarks: the correlation with GRACE reached 0.854 in the
Northern Territory and surged from 0.290 to 0.656 in station-
sparse Western Australia. Crucially, the model successfully
bridges the GRACE mission gap, ensuring TWS continuity.
Furthermore, the quantification of drought evolution (2011--
2023) using a derived Drought Severity Index (LSTM-DSI) and
NscPDSI revealed a correlation of 0.663, demonstrating robust
consistency in capturing extreme hydrological signals. By
integrating the temporal resolution of GNSS with the physical
stability of GRACE, this multi-source fusion approach provides a
robust foundation for investigating hydroclimatic interactions
in Australia.}",
doi = {10.1016/j.jhydrol.2026.136335},
adsurl = {https://ui.adsabs.harvard.edu/abs/2026JHyd..67936335F},
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 16:44:00
GRACE-FO
Thu Oct 08, F. Flechtner