Several series of numerical experiments were conducted to generate a mesoscale eddy in a deep rotating basin (4000 m deep) with stratification typical of mid- and low-latitude of the World Ocean, including an upper mixed layer, a density jump layer, and a main pycnocline. A sea level anomaly driven by Ekman transport convergence was created at the basin’s center by specifying a wind stress configuration typical for a stationary atmospheric anticyclone. Due to geostrophic adjustment, the resulting pressure anomaly generated an anticyclonic baroclinic eddy in the basin, the rotation velocity of which decreased with depth. Numerical experiments varying the horizontal scale of the eddy, the Coriolis frequency, and the stratification parameters yielded a universal relationship between the ratio of the rotation velocity in the surface and bottom layers and the ratio of the horizontal scale of the eddy to the first baroclinic Rossby radius of deformation. Comparing this universal relationship with published estimates of the horizontal scale of mesoscale eddies in the World Ocean from altimetry and shipboard measurements revealed that at low- and mid-latitudes, the rotation velocity typically weakens with depth by more than 80 %. In the presence of a bottom topography disturbance in the form of an abyssal channel under a mesoscale baroclinic eddy, a current arises in the channel whose velocity is significantly higher than the bottom velocity in the undisturbed eddy. The velocity disturbance caused by the abyssal channel is not limited to the channel itself, but reaches the surface, provided that the width of the channel exceeds the first baroclinic Rossby radius of deformation.

Unsteady convection in a two-component medium with significantly different transfer coefficients (e. g., in salt water) is theoretically studied. As a specific example, flows arising in a neutrally stratified medium over a flat inclined surface, from which constant (after switching on) heat and admixture fluxes enter the medium, are considered. A well-known exact solution to the unsteady convection problem near a vertical wall with boundary conditions of the second kind is used, which is generalized to the case of a sloping lower boundary. The complete system of hydrothermodynamic and admixture transport equations is reduced to a linear system due to the symmetry of the problem, without any assumptions about the smallness of the perturbation amplitudes. The linearity of the system allows for the use of the superposition principle — independently considering and summing the dynamic effects of the two components of the medium, which determine the deviations in its density. The solution is expressed in terms of repeated probability integrals. The results, in particular, the direction of the resulting flows, depend significantly on the ratio of the exchange coefficients for different substances. For example, nontrivial situations are possible where positive buoyancy influxes nevertheless lead to the emergence of downward currents. Flow turbulence, leading to abrupt changes in effective exchange coefficients, can, in particular, change the direction of convective flows.

To assess the accuracy of the modeling hydrophysical fields, a comparison of three types of conservative approximation schemes for advective terms in the motion equations was made. These schemes ensure conservation of energy (Experiment 1), potential enstrophy (Experiment 2), and both energy and potential enstrophy (Experiment 3). A numerical simulation method   of the Black Sea circulation was applied by using the Marine Hydrophysical Institute model. Three diagnostic simulations with realistic boundary conditions for 2016 were performed, and the results were compared with each other and with observational data. It was found that based on all three schemes, the experimental results were close to each other and qualitatively corresponded to the observational data. Validation of the reconstructed hydrophysical fields using measurements of the temperature, salinity, and current velocity revealed small quantitative differences from the observational data. An analysis of the circulation energetics showed that the greatest difference occurs in the work of advection and buoyancy forces in areas of intense dynamics (in the periphery of mesoscale eddies and jet currents). From the analysis of the forces balance it follows that the simultaneous observance of the conservation laws of energy and potential enstrophy in the difference problem ensures higher accuracy in describing the Archimedes force work and, consequently, in reproducing the processes of baroclinic instability.

The Bussol Strait is the deepest and second-widest strait in the Kuril Islands chain. It dominates in tidal water transport from the Sea of Okhotsk to the North Pacific Ocean, ventilating its waters in layers at depths of several hundred meters. The strait’s mountainous topography and irregular configuration make it appropriate to model it using curvilinear boundary-fitted coordinates, which map the physical domain of the boundary-value problem onto the canonical computational domain. In this formulation, high-resolution modeling of the strait’s extremely active tidal dynamics was performed in a hydrostatic approximation. At the open boundaries of the domain, sea level fluctuations are specified, caused by a total tide containing 14 harmonics over a synodic month of 29.5 days. The initial and boundary conditions for stratification are based on data from a digital climate atlas. Computed tidal currents are analyzed in different phases of the tidal cycle. The maximum current velocities and average water transport through the strait were estimated for the simulation period. The modeling results indicate the important role of the submarine volcanic massif in the strait’s tidal dynamics. The obtained results are realistic to the extent permitted by the hydrostatic formulation of the boundary value problem. In this regard, this work should be considered a preliminary step toward a more representative and complete (non-hydrostatic) modeling of the Bussol Strait’s dynamics.

The article presents the values of methane concentration in the near-surface layer of atmospheric air directly measured from the board of “Akademik Mstislav Keldysh” research vessel in the Kara and Barents Seas in the summer of 2024. It is the first time when the marine expeditionary data for the Russian part of the Arctic have been compared with the data obtained by “Videomodule” towed unmanned underwater vehicle engineered at the Shirshov Institute of Oceanology of the Russian Academy of Sciences. The analysis shows that bacterial mats in areas of methane seepage in the Arctic have a limited contribution to the increase in CH₄ concentration in the near-surface layer of the atmosphere. Their influence is likely to be noticeable only in shallow areas (up to 50 m deep), where methane can reach the water surface without being fully oxidized in the ocean. At the same time, short-lived peaks (lasting up to several minutes) of elevated CH₄ concentrations are observed, reaching 3 ppm or higher. The main factor influencing   the variability of atmospheric CH₄ concentrations is the direction of air mass transport: when air masses are advected from the north and west, minimum CH₄ concentrations (1.98 ppm) are recorded; conversely, when air masses are carried away from the   mainland, methane concentrations increase (up to 2.10 ppm when from the Yamal Peninsula).

The procedure and results of constructing an acoustic map for water area containing single, multiple, and composite underwater objects using an underwater hydroacoustic vision system are considered in the article. The study was conducted in a scale model of a reservoir implemented in the test tank. The objects were located in the Fresnel zone of an extended linear acoustic array model implemented as a hydroacoustic receiver linearly movable according to a predetermined program. The acoustic reflectivity of the underwater objects was ensured by a single acoustic illumination source. The method for localizing underwater objects is based on wavefront curvature in the Fresnel zone. The acoustic reflectivity of the suspected object location was determined by integrating the spatio-temporal signal field along generating line determined taking into account the delays of the signal wavefront reflected from the object. The feasibility of constructing an acoustic map of the water area and subsequent objects recognition (single, composite, multiple) as well as separating objects from group within the view field of the underwater hydroacoustic vision system are demonstrated. An increase in the contrast of objects acoustic images due to the accumulation of wave energy across the entire signal field has been experimentally confirmed.

The aim of the present study is to develop a method for calculating the diffuse component of the bidirectional reflectance of a strongly absorbing and highly-forward scattering medium like seawater. The relevance of this issue is due to the widespread use of the bidirectional reflectance in the processing of ocean color measurements. Explicit expressions for the first two terms of the expansion are derived. The first term reproduces the well-known result of quasi-single-scattering approximation, and the second corresponds to the contribution of quasi-double scattering (the small-angle multiple scattering before, between, and after two events of large-angle scattering). To validate our analytical results, we carry out numerical integration of the radiative transfer equation for the Henyey-Greenstein phase function, as well as for the two-term Henyey-Greenstein function that models light scattering in seawater. It is shown that for optical parameters typical to seawater, the first two terms of the expansion (i. e. the sum of quasi-single- and quasi-double-scattering contributions) prove to be sufficient to describe the bidirectional reflectance with high accuracy over a wide range of illumination/viewing angles.

A scanning hyperspectrometer is an instrument that simultaneously records radiation in hundreds of narrow spectral channels. When studying the water surface, it is commonly used to determine impurity concentrations and detect oil films. At the same time, optical methods for measuring sea wave characteristics usually require contact calibration or operate only in the sun-glitter zone. This study demonstrates that a scanning hyperspectrometer can be additionally used for wave diagnostics under diffuse sky illumination, with its spectral channel enabling simultaneous detection of surface pollutants that affect wave dynamics. Processing was performed on field data from the Oka River and the Black Sea. Surface slopes are retrieved from relative fluctuations of spectral radiance, which provides absolute slope values without contact measurements and compensates for spectral absorption features. Spectral analysis allowed separation of wind waves and slicks, as well as estimation of current velocity. Analysis of slope histograms at different distances from the observer revealed the wave-slope shadowing effect. Two slope retrieval models are proposed, including a new refined model that accounts for sky brightness gradients. This refined model yields a wave spectrum consistent with contact wave-gauge data. The technique is promising for integrated monitoring of water areas, including the detection of surfactant films.

The beaching of large patches of natural and anthropogenic debris following storm events is a phenomenon observed along numerous marine shorelines. Today, such wash-outs are becoming increasingly contaminated with plastic litter, which poses a significant threat to coastal ecosystems. A 22-month continuous video sequence, recorded by an autonomous stationary camera, allowed the observation the wash-outs on the northern shore of the Sambian Peninsula (the Baltic Sea). Hydrophysical and meteorological variables from reanalysis data are analyzed to predict the timing of marine litter beaching using machine learning models. The performance of multiple machine learning models was evaluated to assess their ability to predict the time at which marine litter would be beached on the shore. The accuracy of artificial neural network, random forest classifier and gradient boosting classifier of machine learning models are compared. The random forest classifier model (83.4 ± 7.6 % for F1-score) and the artificial neural network model (81.7 ± 3.9 % for F1-score) appear to be the most efficient models for predicting the wash-out formation and beaching. Sea level, wave direction and steepness are the most significant parameters in training models.