The Kuroshio Extension region is well known for its strong eddy activity. In this paper, using satellite altimetry‐measured sea surface height anomaly data from 1993 to 2012 in an extended Kuroshio Extension region (140–180°E, 25–45°N), we analyze eddy characteristics: eddy size, polarity, lifetime, intensity, trajectory, and spatial and temporal distributions. Using temperature and salinity vertical profiles measured by Argo floats, we examine the eddy impact on vertical stratification. During the 20‐year period, 7,574 eddies are identified (based on following complete eddy trajectories) with a lifetime equal to or longer than 4 weeks. The numbers of cyclonic and anticyclonic eddies are found to be approximately the same. The distribution of eddy sizes peaks at a radius of about 40 km. The radius at the peak is at the same order as the first baroclinic deformation radius or the horizontal shear scale of the Kuroshio flow. The normalized eddy statistical characteristics show that eddies have different characteristics at different stages of their lifetimes. Among eddies with lifetimes longer than 50 weeks, more anticyclonic (cyclonic) eddies are found north (south) of 35°N. In contrast, among eddies with lifetimes shorter than 20 weeks, more cyclonic (anticyclonic) eddies are found north (south) of 35°N. The asymmetric distribution of eddies suggests two different eddy generation mechanisms: (1) the development of meanders in the Kuroshio path leading to the pinch off of eddies with longer lifetime (larger size) and (2) horizontal shear instability (barotropic instability) leading to eddies of shorter life (smaller size). We further apply an eddy‐resolved numerical product to quantitatively investigate the eddy generation mechanisms.

Oceanic eddies play an important role in global energy and material transport, and contribute greatly to nutrient and phytoplankton distribution. Deep learning is employed to identify oceanic eddies from sea surface height anomalies data. In order to adapt to segmentation problems for multi-scale oceanic eddies, the pyramid scene parsing network (PSPNet), which is able to satisfy the fusion of semantics and details, is applied as the core algorithm in the eddy detection methods. The results of eddies identified from this artificial intelligence (AI) method are well compared with those from a traditional vector geometry-based (VG) method. More oceanic eddies are detected by the AI algorithm than the VG method, especially for small-scale eddies. Therefore, the present study demonstrates that the AI algorithm is applicable of oceanic eddy detection. It is one of the first few of efforts to bridge AI techniques and oceanography research.

Mesoscale eddies are important to ocean circulation due to their roles in the transport of mass, energy, and heat. This study employs a combination of data sources to initiate a statistical analysis of eddy spatiotemporal characteristics in the Bay of Bengal (BOB) to elucidate the sea surface and vertical structures of the eddies and their impacts on sea surface chlorophyll (Chl) distributions. The results suggest that 1237 cyclonic eddies (CEs) and 1121 anticyclonic eddies (AEs) were detected in 26 years. The number of two eddy polarities was almost the same, and most of them spread to the west or southwest direction. The vertical change of temperature (T) and salinity (S) caused by the eddies is studied and the anomalous eddies, i.e., a CE (AE) eddy with warm (cold) water at the center, are mainly distributed on the northeast side of the Island of Sri Lanka. Furthermore, CEs are found to increase Chl concentration in the surrounding sea by approximately 11.15%, while AEs decrease concentrations also by approximately 11.25%. Changes in Chl concentrations occur most rapidly during the mature and intensification eddy phases. Observations also indicate that the strong local current and wind fields are the primary mechanisms in eddy generation.

Oceanic mesoscale eddies typically span tens to hundreds of kilometers and last several weeks to months. They are a key dynamic process influencing the distribution of oceanic energy, affecting energy budgets, heat distributions, and material transport. Advances in satellite remote sensing and in situ observation over the past half-century have substantially enhanced our understanding of the statistical properties, energetics, and dynamic mechanisms of mesoscale eddies, establishing this field as a major focus in physical oceanography. This article presents a comprehensive review of recent advancements in 6 key areas of mesoscale eddy research: (a) eddy detection methods and data products (leading automated detection methods and the corresponding eddy datasets and data products currently available), (b) eddy generation and dissipation (eddy generation, development, and dissipation mechanisms), (c) energy cascades (energy cascades between eddies, large-scale circulation and submesoscale processes, the role of these processes in modulating the energy transfer of near-inertial internal and surface waves, and research progress in related parameterization), (d) air–sea interaction (eddy-induced interactions, especially the interaction between eddies and typhoons), (e) interdisciplinary studies (impacts of eddies on biogeochemistry, acoustics, and other related fields), and (f) observation studies (programs dedicated to studying mesoscale eddies through in situ observation). This article also discusses research prospects on mesoscale eddies.

Sea surface temperature (SST) is one of the most important parameters in air–sea interaction, and its accurate prediction is of great significance in the study of global climate change. However, SST is affected by heat flux, ocean dynamic processes, cloud coverage, and other factors, which means it contains linear and nonlinear components. Existing prediction models, especially single prediction models, cannot effectively handle these linear and nonlinear components in the meantime, degrading their accuracy concerning the prediction of SST. To remedy this weakness, this paper proposes a novel prediction model by the Lagrange multiplier method to combine the auto-regressive integrated moving average (ARIMA) model and the back propagation (BP) neural network model, where these two models have superior prediction performance for linear and nonlinear components, respectively. Moreover, the genetic algorithm is exploited to construct the genetic algorithm BP (GABP) neural network to further improve the performance of the proposed model. To verify the effectiveness of the proposed model, experiments predicting the SST based on historic time-series data are performed. The experiment results indicate that the mean absolute error (MAE) of the ARIMA-GABP model is only 0.3033 ${}^{°}$C and the root mean square error (RMSE) is 0.3970 ${}^{°}$C, which is better than the ARIMA model, BP neural network model, long short-term memory (LSTM) model, GABP neural network model, and ensemble empirical model decomposition BP model among various datasets. Therefore, the proposed model has superior and robust performance concerning predicting SST.

High‐resolution sea surface velocity (SSV) is crucial for advancing our understanding of ocean sub‐mesoscale processes, energy cascades, etc. The recently launched Surface Water and Ocean Topography (SWOT) satellite measures sea surface height with a sub‐mesoscale resolved resolution. Based on geostrophic balance, the so‐called geostrophic velocity in SWOT is estimated. Although the SWOT‐derived velocity is not true geostrophic velocity as it does not consider the separation of balanced and unbalanced motions, it offers valuable insights into both geostrophic and ageostrophic velocities. Here we propose a machine learning‐based model to infer the SSV using SWOT and drifter data. The result demonstrates the error between the geostrophic velocities from SWOT and the total velocities from the drifter are reduced by about 50%. Furthermore, the kinetic energy of inferred velocities aligns more closely with reanalysis data, particularly at low latitudes. This study thus presents a promising approach for inferring global SSV using SWOT data.