A self-organizing map (SOM) neural network was developed from Argo gridded datasets in order to estimate a subsurface temperature anomaly (STA) from remote sensing data. The SOM maps were trained using anomalies of sea surface temperature (SST), height (SSH), and salinity (SSS) data from Argo gridded monthly anomaly datasets, labeled with Argo STA data from 2005 through 2010, which were then used to estimate the STAs at different depths in the North Atlantic from the sea surface data. The estimated STA maps and time series were compared with Argo STAs including independent datasets for validation. In the Gulf Stream path areas, the STA estimations from the SOM algorithm show good agreement with in situ measurements taken from the surface down to 700-m depth, with a correlation coefficient larger than 0.8. Sensitivity of the SOM, when including salinity, shows that with SSS anomaly data in the SOM training process reveal the importance of SSS information, which can improve the estimation of STA in the subtropical ocean by up to 30%. In subpolar basins, the monthly climatology SST and SSH can also help to improve the estimation by as much as 40%. The STA time series for 1993–2004 in the midlatitude North Atlantic were estimated from remote sensing SST and altimetry time series using the SOM algorithm. Limitations for the SOM algorithm and possible error sources in the estimation are briefly discussed.

Ocean heat content (OHC) evolutions calculated from the data sets (WOA, MyOcean, ORAS4, and SODA) were examined at different depth ranges in this study. According to the OHC changes, the subsurface and deeper ocean (SDO, 300–2000 m) heat content rapidly increased over the world's ocean basins during 1998–2013, indicating significant warming in the SDO during the recent global surface warming hiatus. Almost all the ocean basins warmed up, but with various contributions to the global SDO warming tied to the recent hiatus. The role of the Indian Ocean is particularly important as it has accounted for about 30% of global SDO heat uptake during the hiatus. The combined use of multiple data sets can reveal inconsistencies in SDO warming analysis results, and improve our understanding of the role of the SDO in the recent hiatus. The heat uptake in global SDO during the hiatus was about 2.37, 5.44, 3.75, and 2.44 × 10²² joules with trends of 0.40, 0.70, 0.77, and 0.48 W m⁻² according to WOA, MyOcean, ORAS4, and SODA respectively, presenting obviously inconsistent SDO warming signals. MyOcean shows OHC overestimates in different ocean basins, while ORAS4 presents more reliable SDO OHC analysis. In general, the global SDO has sequestered a significant amount of heat—about 3.50 × 10²² joules with trends of 0.59 W m⁻² on average among the four data sets—during the recent hiatus, demonstrating widespread and significant warming signals in the global SDO. There remain substantial uncertainties and discrepancies, however (especially in the PO and SO), in the available SDO warming information due to insufficient subsurface observation coverage and variations in the data set generation techniques used among different researchers.

Two sets of more than 850 days of mooring records and satellite altimeter data are used to explore the features and spatiotemporal evolution of near-inertial waves (NIWs) near Nansha Island in the southern South China Sea (SCS). The observed NIWs are dominated by clockwise (downward energy propagation) motions and show a clear blue shift with a distinct peak frequency of 1.09 f during two large NIW events. The near-inertial kinetic energy (NIKE) is primarily concentrated in the upper layer and radiated downward. The largest value of depth-integrated NIKE reaches 3.5 KJ/m². Besides, the NIWs are dominated by the first three modes, which account for 80% of the total NIKE. Moreover, the depth-integrated NIKE exhibits an apparent seasonal variation, with the largest NIKE in winter, which is almost three times larger than that in other seasons. Every large NIKE event is attributed to the passage of storms and is dominated by mode-2 NIWs. The dominance of the mode-2 NIWs is likely caused by the interaction between NIWs and mesoscale eddies.