Robust daily satellite sea surface salinity reconstruction using deep learning in low-salinity coastal regions.

Mar Pollut Bull

Department of Civil, Urban, Earth, and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, South Korea; Graduate School of Artificial Intelligence, UNIST, Ulsan, South Korea; Graduate School of Carbon Neutrality, UNIST, Ulsan, South Korea. Electronic address

Published: July 2025


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

Spatiotemporally continuous sea surface salinity (SSS) is essential for monitoring rapid changes in the physical and biogeochemical characteristics of oceans and plays a crucial role in effective coastal environment management. Traditional physics-based SSS methods often oversmoothed salinity variations and struggle to capture sharp gradients, leading to reduced accuracy in river-dominated areas. In addition, the long revisit times of satellite data limit real-time monitoring. To address these challenges, we propose the multi-scale aware interpolation network (MAIN), a self-supervised deep neural network designed to generate near-real-time, gap-free daily SSS data without relying on future observations. The input data consisted of soil moisture active passive (SMAP) L2B swath data and the bias-corrected SMAP data that were enhanced using in situ measurements. Validation against in situ measurements and existing satellite-derived SSS products demonstrates that MAIN significantly improves salinity estimation accuracy. In the Amazon region, operational products showed a root mean square error (RMSE) of 1.04 psu, while MAIN reduced this to 0.71 psu. In the East Asian region, the RMSE for operational products was 0.64 psu, whereas MAIN improved it to 0.51 psu. Feature representation analysis revealed that MAIN effectively captured the spatial expansion of low-salinity water over time and significantly reduced the salt-and-pepper noise typically associated with L-band measurements. These findings demonstrate that MAIN is a robust and scalable framework that enhances gap-free SSS estimation, offering new opportunities for monitoring transient oceanic processes in diverse marine environments.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.marpolbul.2025.118462DOI Listing

Publication Analysis

Top Keywords

sea surface
8
surface salinity
8
situ measurements
8
operational products
8
main
6
sss
5
data
5
robust daily
4
daily satellite
4
satellite sea
4

Similar Publications

This study investigates the vertical profiles, pollution status and ecological risks of heavy metal(loid)s contamination in three sediment cores (N21, N03, and 38002) from the North Yellow Sea (NYS), with a focus on the influence of grain size effects on sedimentary profiles. The results revealed distinct vertical distribution patterns of heavy metal(loid)s content among the three sediment cores. Enrichment Factor (EF) and Geo-accumulation Index (I) assessments identified Sb as significantly enriched, indicating anthropogenic influence, whereas Co, Cr, Cu, Ni, and Zn primarily originated from natural weathering.

View Article and Find Full Text PDF

Regulation of phytoplankton community by mesoscale eddies and the East Indian Coastal Current in the southwestern Bay of Bengal during the northeast monsoon.

Mar Environ Res

September 2025

Key Laboratory of Marine Ecosystem Dynamics, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou, 310012, China; State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou, 310012, China; Key Laborator

Sri Lanka is uniquely situated at the junction of the Bay of Bengal (BOB) and the Arabian Sea (AS), where phytoplankton community may be strongly influenced by ocean dynamical processes, particularly mesoscale eddies and the East Indian Coastal Current (EICC). Here, to explore these regulatory mechanisms, phytoplankton and physicochemical parameters were collected from the top 200 m water column in the eastern and southern seas of Sri Lanka during the winter monsoon. Results showed higher concentrations of nutrients and phytoplankton abundance within the regions affected by EICC and cyclonic eddy (CE) compared to anticyclonic eddy (ACE).

View Article and Find Full Text PDF

Impact of wettability heterogeneity on methane hydrate growth kinetics in partially water-saturated sediments.

J Colloid Interface Sci

August 2025

Research Centre of Ecology & Environment for Coastal Area and Deep Sea, Guangdong University of Technology, Guangzhou 510006, China; Guangdong Basic Research Center of Excellence for Ecological Security and Green Development, Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou),

Hypothesis: Gas hydrate formation in sediments is influenced by the availability of gas-water interfacial areas, which governs gas-water interactions. The surface wettability of sediment particles is expected to affect the spatial distribution of water within the pore space, thereby altering the extent of gas-liquid contact. Consequently, by tuning the wettability heterogeneity of the sediment, the spatial distribution of pore water can be regulated, which in turn influences the gas-water interactions and the kinetics of gas hydrate formation.

View Article and Find Full Text PDF

Sea surface temperature of the Red Sea has increased by up to 0.45 °C per decade over the last 30 years, and coral bleaching events are becoming more frequent. A reef bleaching event was observed in October 2020, whereby some parts of the Red Sea experienced more than 12 °C-weeks.

View Article and Find Full Text PDF

The distribution characteristics of marine litter in the intertidal zone of Lvhua Island and its impact on the macrobenthic community structure.

Mar Environ Res

September 2025

College of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, 201306, China; Engineering Technology Research Center of Marine Ranching, Shanghai Ocean University, Shanghai, 201306, China; Comprehensive Workstation for Marine Ranching in the East China Sea Region, Expert Consul

Marine litter typically originates from human discards at sea or enters the ocean through land-based pathways such as surface runoff and natural disasters. The extensive accumulation of plastic litter poses severe threats to marine life. In August 2024, a specialized survey was conducted to investigate the distribution characteristics of marine litter and macrobenthic communities across four intertidal zones on Lvhua Island (XIAO'AO, DA'AO, FANGANG, and SHIZIKENG).

View Article and Find Full Text PDF