Satellite L-band radiometry has transformed open-ocean salinity monitoring, but coastal retrieval remains difficult. Bright land signals leak into ocean measurements, while side lobes, imaging artifacts, radio frequency interference, and calibration errors distort brightness temperature near shore. Existing forward models also assume fully developed, wind-driven seas and often neglect fetch limits, wave age, shallow-water effects, and wave–current interactions. Current products provide 40–100 km effective resolution, with coastal uncertainty of 0.5–1.0 practical salinity units (psu) and substantial data loss within 50–100 km of land. Based on these challenges, in-depth research on integrated, physically consistent coastal salinity retrieval systems is needed.
Researchers from Ocean University of China, the National Satellite Ocean Application Service, and the Institute of Oceanography, Chinese Academy of Sciences, presented (DOI: 10.34133/remotesensing.1058) the framework in the Journal of Remote Sensing on July 10, 2026. Their Perspective addresses the persistent inability of spaceborne L-band radiometers to provide stable, high-resolution SSS estimates near coasts, where land contamination, imaging artifacts, and incomplete representations of sea-state and current effects reduce data quality and limit operational monitoring of freshwater processes in practice.
The framework links two separated improvement routes. Pathway I cleans the measurement chain by combining visibility-domain corrections with brightness temperature (TB)-domain corrections, which is particularly critical for IMRs, reducing land–sea contamination (LSC) while protecting detail. Pathway II deepens the forward model by incorporating wave development, fetch, wave age, foam, shallow-water effects, and current-induced roughness changes. Rather than proposing one universal algorithm, the authors organize methods by Technology Readiness Level (TRL) and connect them to a roadmap. The goal is to move coastal products from 40–100 km resolution toward 10–20 km while achieving accuracy better than 0.3 psu within 100 km of shore in practice.
Quantitative evidence reviewed in the paper shows why a system-level solution is required. In coastal zones, SSS uncertainty typically reaches 0.5–1.0 psu, while biases in river plumes can exceed 0.5 psu. Land-induced TB contamination may extend hundreds of kilometers offshore, and masking or windowing often sacrifices coverage and resolution. For the measurement pathway, the authors highlight visibility phase adjustment (VPA) to suppress Gibbs oscillations and antenna-pattern-based corrections to estimate residual land leakage. For the physics pathway, they recommend adding wave age, fetch, significant wave height, peak period, directional spreading, and current fields to TB models. The roadmap separates near-term standardization and testbeds, mid-term co-design of instruments and retrieval systems, and long-term integration with data assimilation and coastal freshwater observing networks. Physics-aware artificial intelligence (AI) is proposed for structured residual correction and hybrid modeling, but learned components should remain anchored in transparent physical constraints and robust error statistics.
“The central challenge is to close the loop between a ‘clean’ measurement chain and a ‘deep’ physical forward model,” the authors wrote. They emphasized that the framework does not claim one universally optimal correction. Instead, it coordinates instrument teams, retrieval developers, and coastal oceanographers around measurable coastal-performance goals and transferable physical principles.
This work is a Perspective rather than a new experimental study. The authors synthesized findings from satellite missions, instrument studies, radiative-transfer research, wave and current modeling, and recent coastal correction methods. They compared approaches across the interferometric microwave radiometer (IMR) measurement chain and the brightness temperature (TB) forward-model chain, classified maturity using Technology Readiness Levels, and organized needs into near-, mid-, and long-term priorities. No new field dataset or laboratory experiment was reported, and the article states that no data are associated with the research.
The framework could guide future L-band satellite design, coastal product reprocessing, and operational assimilation systems. Better coastal SSS maps would strengthen monitoring of river discharge, estuarine mixing, extreme rainfall, ecosystem stress, and freshwater transport. Over the next decade, mission teams could jointly optimize antennas, calibration, land-contamination control, sea-state modeling, and current-aware retrieval. Longer term, salinity, sea surface height, currents, and wave state could be estimated together through satellite, radar, model, and in situ observations, creating a reliable coastal freshwater observing system.
###
References
DOI
10.34133/remotesensing.1058
Original Source URL
https://spj.science.org/doi/10.34133/remotesensing.1058
Funding Information
NNational Natural Science Foundation of China, Grant 42406175; Hainan Key Research and Development Program, Grant ZDYF2023SHFZ089.
About Journal of Remote Sensing
The Journal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.