Laboratory for Ocean Modelling

Scientific results:

1. The Laboratory has created a hierarchy of models of the ocean general circulation  that is based on the NEMO numerical model (Barnier et al. 2006) that is developed and supported by the European consortium of research infrastructure. We developed global configurations of the model with resolutions of 1/4° and 1/12°, which were later adapted to conduct computations using a parallel computing cluster of the Shirshov Institute of Oceanology of Russian Academy of Sciences, as well as on the Lomonosov-2 supercomputer of Moscow State University. Later we conducted long-period global numerical experiments (1979–2018),in which special attention was paid to the precise reproduction of the state of the ocean and its climate variability. In these experiments we analyzed the roles of the meso-scale and sub-meso-scale dynamics of the ocean in the formation of the ocean’s own variability at the global scale. Moreover, we assessed their influence on global changes in the ocean, including the average sea level, the global thermohaline circulation, the dynamics of the main ocean currents.
2. We have assessed the renewable energy related to ocean currents on the basis of global numerical high-resolution experiments. For this purpose, we for the first time used an eddy-resolving numerical model of the ocean circulation in the global configuration. In these experiments we implemented virtual turbine power stations (TPS) in a global eddy-resolving model of the ocean circulation and for the first time we modeled changes of currents related to TPS, thus quantitatively improving the assessment of the potential effect of TPS on local modes of the ocean circulation, as well as increasing the precision of the assessment of the energy potential. The effect of a multitude of turbines (i. e. virtual TPSs) were parameterized by introducing an additional resistance coefficient to equations of motion for nodes of the computation grid in 42 regions along the trajectories of the Gulf Stream and the Kuroshio. This allowed for the first time to determined regions  with the highest and the lowest decline in the theoretically available power (TAP), which are optimal for the placement of TPSs. Moreover, for the first time we conducted a quantitative assessment of relations between TPS and the dynamics of ocean currents.
3. The Laboratory has developed a model of the ocean circulation for the North Atlantic with a horizontal resolution of 1/12° and 75 vertical levels. This configuration (referred to in the following as NNATL12) reproduces the dynamics of the sub-Polar cycle with a resolution of about 4.5 km. Subsequently for the first time for this region we developed configurations with a higher horizontal resolution (1/36° and 1/60° with 150 and 300 vertical levels). For all these configurations we proposed a new parameterization and a numerical scheme for vertical mixing that accounts for the strengthening of the intensity of mixing related to the step-wise representation of the topography of the floor. The use of the parameterization improves the reproduction of deep-water currents, which is confirmed by comparison with data of observations. This ensured more precise reproduction of deepwater currents in the Danish strait, which is an important component of circulation in the Atlantic. On the basis of the developed models we conducted long-period numerical experiments covering the period from 1992 to now. We validated model computations with the use of vessel-based measurements of the floor on a section along the 60^th parallel north. For this purpose, we had developed methods of diagnostics ensuring optimal correspondence between the results of the modeling and spot observations, as well as factoring in the scaling of the modeled and observed characteristics.
4. Conducting model experiments would not have been possible without creating high-resolution atmospheric boundary conditions. To this end, for the first time we conducted numerical experiments for the northern part of the Atlantic ocean with high resolution (about 11 km, 50 vertical levels) on the basis of a regional nonhydrostatic atmospheric model (WRF–ARW 3.8.1) for the period from 1979 to 2018 with lateral boundary conditions from the ERA–Interim atmospheric reanalysis. To obtain the optimal configuration of «spectral nudging», for this experiment we conducted a number of tests of sensitivity to various wave lengths and height of its use. This unprecedented numerical experiment was given the name «North Atlantic Atmospheric Downscaling». An experiment with low spatial resolution that was conducted concurrently allowed to assess the role of meso-scale atmospheric processes in the interaction between the ocean and  the atmosphere and the ocean circulation. We have conducted an assessment of the quality of reproduction of the characteristics of the interaction between the ocean and the atmosphere, including extreme heat currents, which bears critical importance for reproducing convection, wind disturbance and the vertical mixing related to them, as well as the interaction between the ocean and the atmosphere in the tropics.
5. The results of the numerical modeling of meso-scale processes in the ocean and their role in the formation of large-scale circulation has also been analyzed in a number of theoretical studies of the vortex activity of the ocean. We researched how an arbitrary excitation is split into a slow quasi-geostrophic and a fast ageostrophic components. In a barotropic model the slow component was described by a two-dimensional hydrodynamics equation  for the geostrophic function of current. This component was determined by two related equations of potential vorticity in the quasi-geostrophic approximation in the lower and the upper layer. This allowed us to determine asymptotic solutions for averaged fields that exist in all the regions (including boundary layers) over long periods of time. Precise decompositions for averaged fields were compared with asymptotic solutions. We also reviewed geostrophic adaptation in the model of a rotating ocean, when the direction of angular velocity of rotation does not match gravity. To analyze these effects, we reviewed two models: a barotropic model and a model of a stable neutrally stratified fluid. The obtained results turned out to be important for the understanding of numerical experiments in reproducing the dynamics of meso-scale processes in the ocean.
6. The Laboratory has developed a regional configuration of a high-resolution numerical ocean model for the Black Sea and the Sea of Azov (BSAS12), and we conducted more than 20 tests of sensitivity, including to analyze the water exchange through the Kerch Strait. To conduct the experiments for the Black Sea region on the WRF, atmospheric model we accumulated a regional array of high-resolution boundary conditions for the period from 1979 to 2018. To validate the BSAS12 model experiments, we developed a database of the oceanographic parameters of the Black Sea on the basis of instrumental measurements and satellite observations that will be widely used in the future. The completed numerical experiments for the first time allowed to conduct an analysis of the dynamics of small river plumes in the Black Sea, as well as a validation of its propagation on the basis of satellite data. Later the modeled characteristics of river plumes were linked to the river runoff in the Black Sea. Additional regional research was conducted for other closed and semi-closed basins. In particular, experiments in modeling the ocean circulation and wind disturbance were conducted for the Barents Sea, for which we developed unstructured and regional conformal grids. We also conducted regional and local research of chlorophyll (Chl a) and suspended matter on the basis of a numerical modeling combined with an analysis of satellite and rediolocation images.
7. Our researchers have developed a unified configuration of the NEMO global model with a resolution of 0.5º and a model of the general atmospheric circulation of the Main Geophysical Observatory. Relying on the produced configuration, we performed several control experiments, and in this manner the model was prepared for further implementation of other components of the climate system and conducting historical and scenario-based experiments. We are currently working to optimize computational resources to perform these experiments. To analyze the reliability of modeling, we analyzed of the characteristics of temperature on the surface in the existing experiments with climate models.
8. We have created a regional configuration of the model with ultra-high resolution that unites the dynamics of the ocean, ice, waves and the atmosphere. This configuration is able to forecast the state of of the environment in high-latitude regions in real time. This system is currently developed as part of an agreement with «Gazprom Neft».
9. We have performed global computations of wind disturbance with the use of a spectral wave model, which is a part of the model of the dynamics of the ocean. This allowed the Laboratory to enter the COWCLIP international collaboration.

• A methodology has been developed for assessing the potential of ocean currents for producing environmentally-friendly power.
• Computations have been performed using the high-resolution model that allowed to demonstrate the efficiency of the use of ocean surface currents to produce power. 

Three Candidate of Sciences dissertations and one Doctor of Sciences have been prepared and defended.

The Laboratory conducted the International youth school and conference in computation and information technologies for the environmental studies CITES-2017.

We have developed lecture courses: 

• «Boundary conditions for models of the ocean circulation: methods and prospects»,
• «Numerical modeling of the ocean circulation»,
• «Machine learning in the Earth studies ». 

On the basis of the Laboratory a nulti-processor high-performance computing cluster  has been created that can serve as a unique facility. 

Other results: 

The Laboratory organized the conference «Ocean–atmosphere interaction in the Arctic and the northern part of the Pacific ocean: the key to predictability of climate variability in the middle latitudes» and the seminar «Ocean energy, non-linear processes and the coastal zone». 

Collaborations:

• Research and Computing Center of Moscow State University (Russia): joint research, using the «Lomonosov» and «Lomonosov-2» supercomputers, conducting long-term computations of the circulation in the North Atlantic with high resolution.
• Voeikov Main Geophysical Observatory (Russia): building a unified model of the ocean and the atmosphere for the research of the climate, building a testing configuration of the unified model, conducting pilot experiments.