Answers are grounded in this paper and cite their sources.
The capability of the Chinese HY-2B/C/D altimeter constellation for gridded sea level anomaly (SLA) reconstruction using only HY-2 observations remains insufficiently evaluated. This study presents a physics-guided framework designed for near-real-time (NRT) SLA reconstruction in the Northwest Pacific and retrospectively evaluates it using 2022−2025 HY-2B/C/D observations. The framework combines an OI-derived coarse background with deep-learning refinement, with target-period reconstruction driven solely by HY-2B/C/D along-track data. Validation against an identical set of 1255 independent tide-gauge observations yields mean RMSE of 0.051 m and a mean correlation coefficient (r) of 0.83 over 12 tide-gauge stations. Inter-product comparisons during 2022−2025 show closer agreement with the DUACS DT and NRT products than with NSOAS, although differences in SLA reference conventions and geophysical correction standards among the products should be considered when interpreting regional discrepancies. Scale-dependent evaluation using SARAL/AltiKa along-track observations indicates a mean effective spatial resolution of approximately 300 km. Dynamical consistency is further assessed against surface-drifter observations using geostrophic velocity anomaly magnitude, yielding mean RMSE of 0.181 m s−1 and r of 0.54. These results demonstrate that spatially coherent and dynamically plausible daily SLA fields can be reconstructed with HY-2B/C/D as the sole altimetric observations during the target period, supporting the potential of the proposed framework for NRT SLA reconstruction under sparse observational constraints.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: