In this study, an attention U-net network was proposed to reconstruct the subsurface temperature (ST) field with high temporal and spatial resolution in the South China Sea (SCS) from sea surface parameters observed by satellites. In addition to sea surface temperature, sea-level anomaly, and sea surface wind (SSW) field, the wind stress curl, which influences the 3-D structure of temperature through the induced Ekman pumping and transport, was also input into the model. The five-day average vertical temperature profiles with a spatial resolution of 0.5° from simple ocean data assimilation (SODA) reanalysis were used for training and evaluating the network. The results show that the attention U-net model performs quite well in ST reconstruction in the upper 100-m layers of the SCS. The additional input of wind stress curl helps to improve the model accuracy. The average root-mean-square error (RMSE)/bias of ST decreases from 1.08°C/−0.21°C to 1.01°C/−0.05°C. In particular, the RMSE near the thermocline is reduced significantly by up to 10.9%. The estimation error of the attention U-net model is much smaller than that of some linear and tree models in the SCS, especially in shallow waters and regions with complex dynamic processes. The case study also shows that our model is capable of capturing the evolution of mesoscale processes in the SCS. The combination of satellite observations with a high-precision ST reconstruction model will help us comprehensively understand the fine structure and variation of temperature and circulation in the marginal seas and open oceans.

The Chinese Ocean Salinity Satellite is dedicated to global sea surface salinity (SSS) mapping with two payloads onboard, which are the interferometric microwave radiometer (IMR) and the microwave imager combined active and passive (MICAP), both using the interferometric aperture synthesis radiometry. One of the payloads is an L-band interferometric radiometer system with a Y-shaped antenna array. The other one, MICAP, is a 1-D passive/active combined system, where the L-, C-, and K-band interferometric radiometers’ antenna arrays are arranged in a line, and the active device is an L-band scatterometer. Based on the payload configurations, a series of simulations is applied to analyze the payloads performance, including the brightness temperature (TB) characteristic, the SSS accuracy, and the effects of Sun and land contamination. The TB simulation results show that the IMR possesses a finer spatial resolution compared to the MICAP L-band radiometer, whereas the latter achieves a better TB radiometric resolution. With the assistance of the C- and K-band radiometers and the L-band scatterometer, the combined retrieval accuracy improves compared with the SSS retrieved using only L-band TB and auxiliary parameters. The best accuracy of SSS is achieved when considering all the available measurements from two payloads together in a combined retrieval. Finally, the results related to Sun and land contamination that endanger the interferometric radiometer measurements are presented and discussed. The performance simulation in this article has been used as a reference for the design phase of two payloads and will be updated with the development of payload manufacture.

This study compares wind speeds derived from five satellite microwave radiometers with those directly observed by buoy-mounted anemometers and the global analyses produced by the European Center for Medium-Range Weather Forecasts (ECMWF) model. Buoy comparisons yield wind speed root mean square errors of 0.82 m/s for WindSat, 1.45 m/s for SSMIS F16, 1.39 m/s for SSMIS F17, 1.43 m/s for AMSR-E, and 1.45 m/s for AMSR2. The overall mean bias for each satellite is typically <0.25 m/s when averaged over all selected buoys for a given study time. The satellite wind speeds are underestimated with respect to the buoy observations at a band of the tropical Pacific Ocean from −8°S to 4°N. The mean buoy–satellite difference as a function of year is always <0.4 m/s, except for SSMIS F16. The selected satellite wind speeds show an obvious seasonal characteristic at high latitudes. In comparison with the ECMWF data, some obviously positive differences exist at high southern latitudes in January and at high northern latitudes in July.

Haiyang-2B (HY-2B) is the second marine dynamic environment satellite of China. Sea surface temperature (SST) products from the scanning microwave radiometer (SMR) onboard HY-2B satellite are evaluated against in situ measurements. Approximately, ten months of data are used for the initial evaluation, from January 15, 2019 to November 15, 2019. The temporal and spatial windows for collocation are 30 min and 25 km, respectively, which produce 450 416 matchup pairs between HY-2B/SMR and in situ SSTs. The statistical comparison of the entire data set shows that the mean bias is −0.13 °C (SMR minus buoy), and the corresponding root-mean-square error (RMSE) is 1.06 °C. Time series of collocations for the SST difference shows that a good agreement is found between HY-2B/SMR and in situ SSTs after June 15, revealing a mean bias and an RMSE of only 0.09 °C and 0.72 °C, respectively. A three-way error analysis is conducted between the SMR, Global Precipitation Measurement Microwave Imager (GMI), and in situ SSTs. Individual standard deviations are found to be 0.41 °C for the GMI SST, 0.15 °C for the in situ SST, and 1.03 °C for the SMR SST. The results indicate that the HY2B/SMR SST products need to be improved during the period from January 15, 2019 to June 15, 2019.

High-resolution salinity information is of great significance for understanding the marine environment. We here propose a deep learning model denoted the “Attention U-net network” to reconstruct the daily salinity fields on a 1/4° grid in the interior of the South China Sea (SCS) from satellite observations of surface variables including sea surface salinity, sea surface temperature, sea level anomaly, and sea surface wind field. The vertical salinity profiles from the GLORYS2V4 reanalysis product provided by Copernicus Marine Environment Monitoring Service were used for training and evaluating the network. Results suggest that the Attention U-net model performs quite well in reconstructing the three-dimensional (3D) salinity field in the upper 1000 m of the SCS, with an average root mean square error (RMSE) of 0.051 psu and an overall correlation coefficient of 0.998. The topography mask of the SCS in the loss function can significantly improve the performance of the model. Compared with the results derived from the model using Huber loss function, there is a significant reduction of RMSE in all vertical layers. Using sea surface salinity as model inputs also helps to yield more accurate subsurface salinity, with an average RMSE near the sea surface being reduced by 16.4%. The good performance of the Attention U-net model is also validated byin situmooring measurements, and case studies show that the reconstructed high-resolution 3D salinity field can effectively capture the evolution of underwater signals of mesoscale eddies in the SCS. The resolution and accuracy of sea surface variables observed by satellites will continue to improve in the future, and with these improvements, more precise 3D salinity field reconstructions will be possible, which will bring new insights about the multi-scale dynamics research in the SCS.

Detailed information about mangroves is crucial for ecological and environmental protection and sustainable development. It is difficult to capture small patches of mangroves from satellite images with relatively low to medium resolution. In this study, high-resolution (0.8–2 m) images from Chinese GaoFen (GF) and ZiYuan (ZY) series satellites were used to map the distribution of mangroves in coastal areas of Guangdong Province, China. A deep-learning network, U²-Net, with attention gates was applied to extract multi-scale information of mangroves from satellite images. The results showed that the attention U²-Net model performed well on mangrove classification. The overall accuracy, precision, and F1-score values were 96.5%, 92.0%, and 91.5%, respectively, which were higher than those obtained from other machine-learning methods such as Random Forest or U-Net. Based on the high-resolution mangrove maps generated from long satellite image time series, we also investigated the spatiotemporal evolution of the mangrove forest in Shuidong Bay. The results can provide crucial information for government administrators, scientists, and other stakeholders to monitor the dynamic changes in mangroves.