Education

Moscow Institute of Physics and Technology
2015 — 2017, Master, Phystech-School of Applied Mathematics and Computer Science (FPMI)
Moscow Institute of Physics and Technology
2011 — 2015, Bachelor, Phystech-School of Applied Mathematics and Computer Science (FPMI)

Dissertations

2022, Candidate , Математическое моделирование, численные методы и комплексы программ, 05.13.18
Fadeev R.Y., Goyman G.S., Tolstykh M.A.
2024-07-01 citations by CoLab: 0 Abstract  
The paper considers a version of SLAV atmosphere model for medium-range weather prediction. The main focus of the paper is on a new software for handling input-output operations with the file system, that has been implemented into SLAV model. The use of this software has reduced the elapsed time for ten-day weather forecast by about 15 $$\%$$ . This result is obtained for two technologies: the deterministic weather forecast with high spatial resolution, and the ensemble technology with a twice coarser horizontal and vertical resolution.
Shashkin V.V., Goyman G.S., Tretyak I.D.
2024-07-01 citations by CoLab: 0 Abstract  
The next-generation atmospheric dynamics model for numerical weather prediction and climate studies is developed at INM RAS and Hydrometcentre of Russia. The model uses a cubed-sphere grid with quasi-uniform resolution over the sphere. Given the multiple areas of probable model applications, we decided to enhance its flexibility as compared to the previous models. Actually, our model is a set of basic building blocks (differential operators, time-integration schemes, parallel exchanges, etc.) that can be used to construct a wide class of atmospheric equations solvers. Currently, we have developed a non-hydrostatic solver for high-resolution global weather forecasting and a solver using hydrostatic approximation in vertical for climate simulations. Both solvers were tested using generally accepted idealized problems and showed promising results. In this article, we motivate the basic concepts of our model and describe its software and numerical design. Parallel efficiency is examined. The model scales well at least up to 6912 cores of Roshydromet Cray XC-40 system.
Tolstykh M.A., Fadeev R.Y., Shashkin V.V., Zaripov R.B., Travova S.V., Goyman G.S., Alipova K.A., Mizyak V.G., Tischenko V.A., Kruglova E.N.
2024-07-01 citations by CoLab: 1 Abstract  
A long-range forecast system based on the improved version of the SLAV072L96 global atmosphere model has been verified at the Hydrometcenter of Russia. The model has the horizontal resolution of 0.9°  $$\times$$  0.72° in longitude and latitude and 96 vertical levels and includes modern parametrizations for the subgrid-scale processes in the atmosphere and active soil layer. Main features and particularities of this model version are presented along with a brief description of ensemble long-range meteorological forecast technology using this model. Some verification scores for long-range forecasts based on the archive data from ERA5 reanalysis for 1991–2015 and on the data for 2023 are presented.
Alipova K.A., Mizyak V.G., Tolstykh M.A., Goyman G.S.
2024-02-01 citations by CoLab: 1 Abstract  
Abstract An algorithm for stochastic perturbation of the semi-Lagrangian trajectories is implemented in the ensemble weather prediction system based on the global atmosphere model SL-AV20 with a horizontal resolution of approximately 20 km, 51 vertical levels, and Local Ensemble Transform Kalman Filter (LETKF). The combined use of methods for stochastic perturbation of trajectories and the parameters and tendencies of subgrid-scale processes parameterizations allows to generate ensembles with a larger spread compared to ensembles without stochastic perturbations of trajectories. An improvement in probabilistic estimates of the ensemble forecasts for various variables is shown. The comparison of two versions of ensemble prediction system is presented.
Tretyak I.D., Goyman G.S., Shashkin V.V.
2023-12-01 citations by CoLab: 1 Abstract  
Abstract We present spatial approximation for shallow water equations on a mesh of multiple rectangular blocks with different resolution in Cartesian geometry. The approximation is based on finite-difference operators that fulfill Summation By Parts (SBP) property – a discrete analogue of integration by parts. The solution continuity conditions between mesh blocks are imposed in a weak form using Simultaneous Approximation Terms (SAT) method.We show that the resulting discrete divergence and gradient operators are anti-conjugate. The important consequences are the discrete analogues for mass and energy conservation laws along with the proof of stability for linearized equations. The numerical shallow water equations model based on the presented spatial approximation is tested using problems with meteorological context. Test results prove high-order accuracy of SBP-SAT discretization. The interfaces between mesh blocks of different resolution produce no significant noise. The local mesh refinement is shown to have positive effect on the solution both locally inside the refined region and globally in the dynamically coupled areas.
Shashkin V.V., Goyman G.S., Tolstykh M.A.
2023-02-01 citations by CoLab: 4 Abstract  
Summation-by-Parts Finite Differences (SBP-FD) is an approach for approximation of differential operators satisfying a discrete analogue of integration by parts analytic property. SBP-FD allows one to build provably stable high-order spatial approximations. SBP-FD methods are widely used for approximation of partial differential equations in multiblock domains with logically-rectangular curvilinear mesh inside. The gnomonic cubed-sphere grid is an example of such a domain. However, applications of the SBP-FD approximation for the cubed-sphere grid in meteorological context are not a widespread research area. We present a SBP-FD based shallow water model using a non-staggered grid. The model is total-energy conserving, mass-conservative and has discrete analogues of other mimetic properties such as curl-free gradient property. The shallow water model is tested with the commonly used Williamson test suite supplemented with the Galewsky barotropic instability case. High-order convergence is shown for tests with analytic solutions. The SBP-FD shallow water model is more accurate than low-order mimetic finite-element counterparts, but slightly less accurate than high-order finite-volume, spectral-elements and discontinuous Galerkin schemes.
Tolstykh M., Goyman G., Biryucheva E., Shashkin V., Fadeev R.
2023-01-01 citations by CoLab: 2 Abstract  
SL-AV is a global atmosphere model used for operational medium-range and long-range forecasts. Following previous successful implementation of single precision in some parts of the model dynamics solver, such computations are introduced into some parts of parameterization block of the model. Also, some memory optimizations are implemented. As a result, the elapsed time needed to compute 24-h forecast for SL-AV version with 10 km horizontal resolution and 104 vertical levels is reduced by 22% while using 2916 Cray XC40 processor cores, without detrimental effects on forecast accucacy.
Alipova K.A., Goyman G.S., Tolstykh M.A., Mizyak V.G., Rogutov V.S.
2022-12-01 citations by CoLab: 5 Abstract  
Abstract Algorithms for stochastic perturbation of parameters and tendencies of physical parameterizations for subgrid-scale processes are implemented into the ensemble prediction system. This system is based on the global semi-Lagrangian atmospheric model SL-AV with the resolution of 0.9 × 0.72 degrees in longitude and latitude, respectively, 96 vertical levels, and our implementation of the Local Ensemble Tranform Kalman Filter (LETKF). The use of stochastically perturbed parameterizations allows to generate ensembles with a significantly larger spread compared to one obtained with the method of static parameter perturbation. An improvement in the probabilistic estimates of the ensemble forecast for different seasons is shown.
Goyman G.S., Shashkin V.V.
2021-06-01 citations by CoLab: 3 Abstract  
• Horizontal discretizations on a reduced lat-lon grid with C-type Arakawa variables staggering. • Conservation properties analysis of the schemes at staggered and unstaggered reduced grids. • Longitudinal interpolation design allowing to construct conservative discretizations. • Analysis of the numerical waves dispersion characteristics. In this work, we consider discretization of the linearized shallow water equations on a reduced latitude-longitude grid with an analogue of Arakawa C-type variables staggering. The resulting schemes are based on the use of longitudinal interpolation procedures and can be of arbitrary order of accuracy. We also present the analysis of conservation and wave propagation properties of these schemes for the linearized shallow water model. This analysis reveals constraints on the interpolation procedures that ensure mass and total energy conservation. The presented approach for construction and analysis of the schemes is also applicable for the unstaggered reduced latitude-longitude grid case.
Goyman G., Shashkin V.
2021-01-01 citations by CoLab: 1 Abstract  
One of the most important aspects that determine the efficiency of an atmospheric dynamics numerical model is the time integration scheme. It is common to apply semi-implicit integrators, which allow to use larger time steps, but requires solution of a linear elliptic equation at the every time step of a model. We present implementation of linear solvers (geometric multigrid and BICGstab) within ParCS parallel framework, which is used for development of the new non-hydrostatic global atmospheric model at INM RAS and Hydrometcentre of Russia. The efficiency and parallel scalability of the implemented algorithms have been tested for the elliptic problem typical for numerical weather prediction models using semi-implicit discretization at the cubed sphere grid.
Shashkin V.V., Goyman G.S.
2021-01-01 citations by CoLab: 3 PDF Abstract  
Abstract Next generation weather prediction atmospheric models will have horizontal resolution of about 3-5 km. The problem dimension will be 1010. One will need to use efficiently 104-105 computational cores to make a practical operational forecast. This leads to the need for the deep revision of numerical methods and algorithms used in atmospheric models. One of the problems to be solved is the horizontal discretization of atmospheric dynamics equations on the quasi-uniform spherical grids. This problem can be investigated using shallow water model that is much computationally cheaper than the use of full atmosphere model. We are developing an atmospheric dynamics solver for the next generation numerical weather prediction model at the Institute of Numerical Mathematics and Hydrometeorological center of Russia. Within this work, the solver for the shallow water equations using gnomonic cubed-sphere grid has been developed. The solver is verified using standard shallow water test cases. The accuracy of the presented solver is analysed. The good agreement to the reference solutions is achieved, when 4-th order spatial approximations are used.
Tolstykh M.A., Goyman G.S., Fadeev R.Y., Travova S.V., Shashkin V.V.
2021-01-01 citations by CoLab: 3 PDF Abstract  
Abstract The global atmosphere model SL-AV is applied for operational numerical weather prediction in Russia at the time scales from days to month and also for simulation of modern climate. It is important to achieve the best wall-clock time for all of the model applications with different grid sizes. In this paper, we analyse the effect of switching computations and parallel exchanges from double to single precision in the semi-Lagrangian advection and elliptic equations solver blocks on the quality of the medium-range forecast and evaluate the change in wall-clock time. We have also optimized OpenMP parallelization in the computations of the right-hand sides of meteorological prognostic equations. The effect of these changes is studied for SL-AV model versions with different resolutions.
Shashkin V.V., Goyman G.S.
2020-12-16 citations by CoLab: 6 Abstract  
AbstractThis paper proposes the combination of matrix exponential method with the semi-Lagrangian approach for the time integration of shallow water equations on the sphere. The second order accuracy of the developed scheme is shown. Exponential semi-Lagrangian scheme in the combination with spatial approximation on the cubed-sphere grid is verified using the standard test problems for shallow water models. The developed scheme is as good as the conventional semi-implicit semi-Lagrangian scheme in accuracy of slowly varying flow component reproduction and significantly better in the reproduction of the fast inertia-gravity waves. The accuracy of inertia-gravity waves reproduction is close to that of the explicit time-integration scheme. The computational efficiency of the proposed exponential semi-Lagrangian scheme is somewhat lower than the efficiency of semi-implicit semi-Lagrangian scheme, but significantly higher than the efficiency of explicit, semi-implicit, and exponential Eulerian schemes.
Tolstykh M., Goyman G., Fadeev R., Shashkin V.
2020-12-05 citations by CoLab: 2 Abstract  
Huge computer resources needed to promptly compute the 24-h global weather forecast dictate the necessity to optimize the numerical algorithms of the model and their parallel implementation. We present some experience gained while implementing the new high-resolution version of the SL-AV global atmosphere model for numerical weather prediction at parallel systems with many thousands of processor cores. Unlike our previous scalability studies, we need to minimize the elapsed time of the forecast at given processor cores number which is currently about 4000. The results of optimizations are shown for two Roshydromet high-performance computer systems.
Kawai Y., Tomita H.
Geoscientific Model Development scimago Q1 wos Q1 Open Access
2025-02-07 citations by CoLab: 0 Abstract   Cites 1
Abstract. Focusing on the future global atmospheric simulations with a grid spacing of O(10–100 m), we developed a global nonhydrostatic atmospheric dynamical core with high-order accuracy by applying a discontinuous Galerkin method (DGM) for horizontal and vertical discretization. Furthermore, considering a global large-eddy simulation (LES), a Smagorinsky–Lilly turbulence model was introduced to the proposed global dynamical core in the DGM framework. By conducting several tests with various polynomial (p) orders, the impact of the high-order DGM on the accuracy of the numerical simulations of atmospheric flows was investigated. To show high-order numerical convergence, a few modifications were made in the experimental setup of existing test cases. In addition, we proposed an idealized test case to verify global-LES models, which is a global extension of an idealized planetary boundary layer (PBL) turbulence experiment performed in our previous studies. The error norms from the deterministic test cases, such as the linear-advection and gravity-wave tests, showed an optimal convergence rate achieved by an approximately p+1-order spatial accuracy when the temporal and round-off errors were sufficiently small. In the climatic test cases, such as the Held–Suarez test, the kinetic energy spectra indicated the advantage of effective resolution when large polynomial orders were used. In the LES experiment, the global model provided a reasonable vertical structure of the PBL and energy spectra because the results under shallow-atmosphere approximation reproduced those obtained in the plane computational domain well.
Fadeev R.Y., Belyaev K.P., Kuleshov A.A., Resnyanskii Y.D., Smirnov I.N., Strukov B.S., Zelenko A.A.
2024-12-16 citations by CoLab: 0 Abstract   Cites 1
This paper presents the results of numerical experiments with the atmosphere, ocean, and sea ice coupled model, which consists of the semi-Lagrangian SLAV atmospheric model and the Nucleus for European Modelling of the Ocean (NEMO) model. The models are coupled using the OASIS3-MCT software. The spatial and temporal variability of the ocean’s characteristics on its surface and in the water column is analyzed.
Gao Y., Xiu Y., Nie Y., Luo H., Yang Q., Zampieri L., Lv X., Uotila P.
2024-11-22 citations by CoLab: 0 Abstract   Cites 1
AbstractIn this study, the subseasonal Antarctic sea ice edge prediction skill of the Copernicus Climate Change Service (C3S) and Subseasonal to Seasonal (S2S) projects was evaluated by a probabilistic metric, the spatial probability score (SPS). Both projects provide subseasonal to seasonal scale forecasts of multiple coupled dynamical systems. We found that predictions by individual dynamical systems remain skillful for up to 38 days (i.e., the ECMWF system). Regionally, dynamical systems are better at predicting the sea ice edge in the West Antarctic than in the East Antarctic. However, the seasonal variations of the prediction skill are partly system‐dependent as some systems have a freezing‐season bias, some had a melting‐season bias, and some had a season‐independent bias. Further analysis reveals that the model initialization is the crucial prerequisite for skillful subseasonal sea ice prediction. For those systems with the most realistic initialization, the model physics dictates the propagation of initialization errors and, consequently, the temporal length of predictive skill. Additionally, we found that the SPS‐characterized prediction skill could be improved by increasing the ensemble size to gain a more realistic ensemble spread. Based on the C3S systems, we constructed a multi‐model forecast from the above principles. This forecast consistently demonstrated a superior prediction skill compared to individual dynamical systems or statistical observation‐based benchmarks. In summary, our results elucidate the most important factors (i.e., the model initialization and the model physics) affecting the currently available subseasonal Antarctic sea ice prediction systems and highlighting the opportunities to improve them significantly.
Xu L., Zhang X., Wu T., Yu H., Du W., Zhang C., Chen N.
Geophysical Research Letters scimago Q1 wos Q1 Open Access
2024-11-04 citations by CoLab: 0 Abstract   Cites 1
AbstractFlash droughts are rapidly developing extreme weather events with sudden onset and quick intensification. Global prediction of flash droughts at sub‐seasonal time scales remains a great challenge. Current state‐of‐the‐art dynamic models subject to large errors and demonstrate low skills in global flash drought prediction. Here, we develop a machine learning‐based framework that uses meteorological forecasts as inputs to predict global root‐zone soil moisture and flash droughts from 1 day to 2 week lead times. The results indicate that 33% and 24% global flash drought onset and termination events can be correctly predicted by machine learning at 7 day lead time, versus 19% and 11% fractions by state‐of‐the‐art dynamic model. The developed machine learning model demonstrates substantial improvements over dynamic model in global soil moisture prediction, and thus enhances global flash drought forecasting skills in space and time. The presented framework may benefit global flash drought prediction and early warning at sub‐seasonal scales.
Rozinkina I.A., Rivin G.S., Bagrov A.N., Blinov D.V., Bykov F.L., Vaskova D.V., Zakharchenko D.I., Bundel A.Y., Kirsanov A.A., Polyukhov A.A., Shatunova M.V., Shuvalova Y.O., Eliseev G.V., Astakhova E.D., Nikitin A.E.
2024-10-23 citations by CoLab: 0 Abstract   Cites 1
This paper presents an outline of short-term regional forecast system based on the COSMO-Ru2By configuration (with a mesh size of 2.2 km) of the COSMO model, which provides numerical weather forecasts up to 48 hours for the European part of Russia and for the Republic of Belarus. The description is accompanied by analysis of verification results. The COSMO-Ru2By system has been implemented at the Hydrometcenter of Russia through the Collaboration Program with Belhydromet and now it is an integral part of the COSMO-Ru operational limited-area numerical weather prediction system running on the CRAY XC40-LC supercomputer of Roshydromet. The COSMO-Ru2By technology has several features: (1) an explicit simulation of deep convection, (2) a vast integration domain, (3) integrated assimilation of the Doppler meteorological radar data and (4) a coupled visualization system that provides a multitude of charts, including maps for different regions that cascadingly increase the details of images. All these aspects are important for forecasting rapidly developing weather processes over the coming hours. The operational trials conducted in 2020–2021 under the supervision of the Roshydromet Central Methodological Commission on Forecasts demonstrated high quality of near-surface weather parameter forecasts with the COSMO-Ru2By, which were significantly more accurate than products from other numerical weather forecast models available to forecasters at the Hydrometcenter of Russia. The comparison with non-hydrostatic models with coarser resolution (e.g., the COSMO-Ru6ENA with a mesh size of 6.6 km) as well with configurations with the same mesh size running in other technologic scenario was made. This analysis was focused on the model capacity of predicting of highly variable weather parameters (such as wind gusts, precipitation, and weather parameters in mountainous areas). For this purpose, the authors employed several additional approaches: assessments over geographically homogeneous areas, the use of special metrics for verifying forecasts of rare events, and comparisons of forecasts with radar data.
Fadeev R.Y.
2024-10-23 citations by CoLab: 0 Abstract   Cites 1
The paper discusses some refinements to the SL-AV atmospheric model, which have made it possible to substantially reduce the systematic errors in reproducing the averaged characteristics of the atmospheric circulation. The main focus of the paper is the wind gustiness parameterization and its role in the calculation of the turbulence fluxes in the atmosphere boundary layer. The statistical significance of the influence of the considered changes on the accuracy of atmospheric circulation simulation is evaluated based on the analysis of the quality of retrospective long-range forecasts with a lead time up to four months (for different seasons) and on the study of averaged characteristics of the model atmosphere in comparison with the ERA5 reanalysis.
Lima G.S.
2024-08-27 citations by CoLab: 0 Abstract   Cites 2
To develop a numerical method for global geophysical fluids, we usually need to choose a spherical grid and numerical approximations to represent the partial derivative equations. Some alternatives include the use of finite differences or finite volumes with latitude–longitude or reduced grids. Each of these cases has some advantages and also some limitations. This paper presents a comparison between two methods and describes a composite model using them side by side. The first is a well-known method for latitude–longitude grids and was used from 75[Formula: see text][Formula: see text]S until 75[Formula: see text][Formula: see text]N. The second is a recently developed scheme for reduced grids and was used only in the polar regions. The similarity between the two methods allows the use of small adaptations in their approximations to obtain consistency and mass conservation also in the transition between the two regions. The composite model combines advantages of the other two schemes and has a smaller computational cost. Numerical tests indicated order 2 of convergence, prevention of grid-imprinting errors, and avoidance of nonlinear instability. This model has numerical properties that may lead to efficient implementations with massive parallel computation.
Kulikova I.A., Vilfand R.M., Khan V.M., Kruglova E.N., Tishchenko V.A., Emelina S.V., Kaverina E.S., Nabokova E.V., Subbotin A.V., Sumerova K.A., Tolstykh M.A.
2024-08-01 citations by CoLab: 0 Abstract   Cites 1
The main advantages of ensemble prediction systems and probabilistic climate forecasts as compared to deterministic ones are considered. Quality assessment has been performed for retrospective probabilistic forecasts of air temperature and precipitation on subseasonal timescales (the forecasts that were obtained with a new version of the SLAV072L96 global semi-Lagrangian atmospheric model developed in the Hydrometcenter of Russia and the Marchuk Institute of Numerical Mathematics of the Russian Academy of Sciences). Skill scores of probabilistic forecasts indicate the presence of a useful signal in the context of the forecasts of two extreme gradations of air temperature and precipitation at weekly and monthly integration intervals. It is concluded that the use of probabilistic approaches makes it possible to expand the time interval of the “usefulness” of forecasts from a week to a month, as well as to obtain estimates of uncertainty of a forecast and its potential economic value. The results of the study are expected to be used in the operational practice of the Hydrometcenter of Russia/North Eurasia Climate Centre, as well as in the preparation of consensus forecasts during sessions of regional climate forums.
Fadeev R.Y.
2024-08-01 citations by CoLab: 0 Abstract   Cites 1
Abstract SLNE is the coupled model, that was developed in 2023. SL and NE here are the first two letters from SLAV (Semi-Lagrangian, based on Absolute Vorticity equation) model of the atmosphere and NEMO (Nucleus for European Modelling of the Ocean) ocean model that have been coupled using OASIS3-MCT software. The initial conditions for SLAV and NEMO are specified from an atmospheric and ocean analyses produced in Hydrometcentre of Russia. The 2010–2021 hindcast accuracy study shows, that SLNE has comparable errors to the operational SLAV model on a sub-seasonal time scale. The SLNE model has improved prediction skill of the 2010 heatwave features in comparison to SLAV, that is a motivation for further work to improve the coupled model.
Day J.J., Svensson G., Casati B., Uttal T., Khalsa S., Bazile E., Akish E., Azouz N., Ferrighi L., Frank H., Gallagher M., Godøy Ø., Hartten L.M., Huang L.X., Holt J., et. al.
Geoscientific Model Development scimago Q1 wos Q1 Open Access
2024-07-24 citations by CoLab: 2 Abstract   Cites 2
Abstract. Although the quality of weather forecasts in the polar regions is improving, forecast skill there still lags behind lower latitudes. So far there have been relatively few efforts to evaluate processes in numerical weather prediction systems using in situ and remote sensing datasets from meteorological observatories in the terrestrial Arctic and Antarctic compared to the mid-latitudes. Progress has been limited both by the heterogeneous nature of observatory and forecast data and by limited availability of the parameters needed to perform process-oriented evaluation in multi-model forecast archives. The Year of Polar Prediction (YOPP) site Model Inter-comparison Project (YOPPsiteMIP) is addressing this gap by producing merged observatory data files (MODFs) and merged model data files (MMDFs), bringing together observations and forecast data at polar meteorological observatories in a format designed to facilitate process-oriented evaluation. An evaluation of forecast performance was performed at seven Arctic sites, focussing on the first YOPP Special Observing Period in the Northern Hemisphere (NH-SOP1) in February and March 2018. It demonstrated that although the characteristics of forecast skill vary between the different sites and systems, an underestimation in boundary layer temperature variability across models, which goes hand in hand with an inability to capture cold extremes, is a common issue at several sites. It is found that many models tend to underestimate the sensitivity of the 2 m air temperature (T2m) and the surface skin temperature to variations in radiative forcing, and the reasons for this are discussed.
Fadeev R.Y., Goyman G.S., Tolstykh M.A.
2024-07-01 citations by CoLab: 0 Abstract   Cites 8
The paper considers a version of SLAV atmosphere model for medium-range weather prediction. The main focus of the paper is on a new software for handling input-output operations with the file system, that has been implemented into SLAV model. The use of this software has reduced the elapsed time for ten-day weather forecast by about 15 $$\%$$ . This result is obtained for two technologies: the deterministic weather forecast with high spatial resolution, and the ensemble technology with a twice coarser horizontal and vertical resolution.
Shashkin V.V., Goyman G.S., Tretyak I.D.
2024-07-01 citations by CoLab: 0 Abstract   Cites 8
The next-generation atmospheric dynamics model for numerical weather prediction and climate studies is developed at INM RAS and Hydrometcentre of Russia. The model uses a cubed-sphere grid with quasi-uniform resolution over the sphere. Given the multiple areas of probable model applications, we decided to enhance its flexibility as compared to the previous models. Actually, our model is a set of basic building blocks (differential operators, time-integration schemes, parallel exchanges, etc.) that can be used to construct a wide class of atmospheric equations solvers. Currently, we have developed a non-hydrostatic solver for high-resolution global weather forecasting and a solver using hydrostatic approximation in vertical for climate simulations. Both solvers were tested using generally accepted idealized problems and showed promising results. In this article, we motivate the basic concepts of our model and describe its software and numerical design. Parallel efficiency is examined. The model scales well at least up to 6912 cores of Roshydromet Cray XC-40 system.
Tolstykh M.A., Fadeev R.Y., Shashkin V.V., Zaripov R.B., Travova S.V., Goyman G.S., Alipova K.A., Mizyak V.G., Tischenko V.A., Kruglova E.N.
2024-07-01 citations by CoLab: 1 Abstract  
A long-range forecast system based on the improved version of the SLAV072L96 global atmosphere model has been verified at the Hydrometcenter of Russia. The model has the horizontal resolution of 0.9°  $$\times$$  0.72° in longitude and latitude and 96 vertical levels and includes modern parametrizations for the subgrid-scale processes in the atmosphere and active soil layer. Main features and particularities of this model version are presented along with a brief description of ensemble long-range meteorological forecast technology using this model. Some verification scores for long-range forecasts based on the archive data from ERA5 reanalysis for 1991–2015 and on the data for 2023 are presented.
Alipova K.A., Mizyak V.G., Tolstykh M.A., Goyman G.S.
2024-02-01 citations by CoLab: 1 Abstract  
Abstract An algorithm for stochastic perturbation of the semi-Lagrangian trajectories is implemented in the ensemble weather prediction system based on the global atmosphere model SL-AV20 with a horizontal resolution of approximately 20 km, 51 vertical levels, and Local Ensemble Transform Kalman Filter (LETKF). The combined use of methods for stochastic perturbation of trajectories and the parameters and tendencies of subgrid-scale processes parameterizations allows to generate ensembles with a larger spread compared to ensembles without stochastic perturbations of trajectories. An improvement in probabilistic estimates of the ensemble forecasts for various variables is shown. The comparison of two versions of ensemble prediction system is presented.
Lam R., Sanchez-Gonzalez A., Willson M., Wirnsberger P., Fortunato M., Alet F., Ravuri S., Ewalds T., Eaton-Rosen Z., Hu W., Merose A., Hoyer S., Holland G., Vinyals O., Stott J., et. al.
Science scimago Q1 wos Q1 Open Access
2023-12-22 citations by CoLab: 289 PDF Abstract  
Global medium-range weather forecasting is critical to decision-making across many social and economic domains. Traditional numerical weather prediction uses increased compute resources to improve forecast accuracy, but does not directly use historical weather data to improve the underlying model. Here, we introduce “GraphCast,” a machine learning-based method trained directly from reanalysis data. It predicts hundreds of weather variables, over 10 days at 0.25° resolution globally, in under one minute. GraphCast significantly outperforms the most accurate operational deterministic systems on 90% of 1380 verification targets, and its forecasts support better severe event prediction, including tropical cyclones tracking, atmospheric rivers, and extreme temperatures. GraphCast is a key advance in accurate and efficient weather forecasting, and helps realize the promise of machine learning for modeling complex dynamical systems.
Tretyak I.D., Goyman G.S., Shashkin V.V.
2023-12-01 citations by CoLab: 1 Abstract  
Abstract We present spatial approximation for shallow water equations on a mesh of multiple rectangular blocks with different resolution in Cartesian geometry. The approximation is based on finite-difference operators that fulfill Summation By Parts (SBP) property – a discrete analogue of integration by parts. The solution continuity conditions between mesh blocks are imposed in a weak form using Simultaneous Approximation Terms (SAT) method.We show that the resulting discrete divergence and gradient operators are anti-conjugate. The important consequences are the discrete analogues for mass and energy conservation laws along with the proof of stability for linearized equations. The numerical shallow water equations model based on the presented spatial approximation is tested using problems with meteorological context. Test results prove high-order accuracy of SBP-SAT discretization. The interfaces between mesh blocks of different resolution produce no significant noise. The local mesh refinement is shown to have positive effect on the solution both locally inside the refined region and globally in the dynamically coupled areas.
Yu X., Liu L., Sun C., Jiang Q., Zhao B., Zhang Z., Yu H., Wang B.
Geoscientific Model Development scimago Q1 wos Q1 Open Access
2023-11-07 citations by CoLab: 1 Abstract  
Abstract. As earth system modeling develops ever finer grid resolutions, the inputting and outputting (I/O) of the increasingly large data fields becomes a processing bottleneck. Many models developed in China, as well as the community coupler (C-Coupler), do not fully benefit from existing parallel I/O supports. This paper reports the design and implementation of a common parallel input/output framework (CIOFC1.0) based on C-Coupler2.0. The CIOFC1.0 framework can accelerate the I/O of large data fields by parallelizing data read/write operations among processes. The framework also allows convenient specification by users of the I/O settings, e.g., the data fields for I/O, the time series of the data files for I/O, and the data grids in the files. The framework can also adaptively input data fields from a time series dataset during model integration, automatically interpolate data when necessary, and output fields either periodically or irregularly. CIOFC1.0 demonstrates the cooperative development of an I/O framework and coupler, and thus enables convenient and simultaneous use of a coupler and an I/O framework.
Orlov K.G., Mingalev I.V., Fedotova E.A., Mingalev V.S.
This paper describes a method for organizing parallel computing on graphics processors using CUDA technology, in relation to the general circulation model of the lower and middle atmosphere of the Earth. This model is being developed at the Polar Geophysical Institute and is based on the numerical integration of a complete system of equations for the dynamics of viscous atmospheric gas on a high-resolution spatial grid in a spherical layer around the Earth's surface. The model includes a block for calculating the transfer of solar and thermal radiation, which also uses parallel calculations based on CUDA technology.
Shashkin V.V., Fadeev R.Y., Tolstykh M.A., Krivolutskii A.A., Banin M.V.
2023-06-01 citations by CoLab: 2 Abstract  
Dynamics of the stratosphere and ozone layer are among important sources of atmospheric circulation predictability at subseasonal-to-seasonal time scales. The simulation of the stratospheric dynamics with the SL-AV atmospheric general circulation model for the SLAV072L96 seasonal weather prediction configuration is analyzed. The configuration is currently under preoperational testing at the Hydrometcenter of Russia. The model simulates both winter and summer averaged distributions of zonal wind and temperature close to the reanalysis data. The quasi-biennial oscillation of equatorial zonal wind is simulated with a realistic period and amplitude. There is a significant reduction of errors as compared with the previous stratosphere-resolving SL-AV model configuration. It is shown that the stratospheric process simulation enhancement is largely due to the reduction of systematic errors in the simulation of troposphere dynamics. The work on the inclusion of the CHARM photochemical model in the SL-AV model is described. The results of first experiments with the coupled model are given.
Kent J., Melvin T., Wimmer G.A.
Geoscientific Model Development scimago Q1 wos Q1 Open Access
2023-02-22 citations by CoLab: 6 Abstract  
Abstract. This paper introduces a mixed finite-element shallow-water model on the sphere. The mixed finite-element approach is used as it has been shown to be both accurate and highly scalable for parallel architecture. Key features of the model are an iterated semi-implicit time-stepping scheme, a finite-volume transport scheme, and the cubed sphere grid. The model is tested on a number of standard spherical shallow-water test cases. Results show that the model produces similar results to other shallow-water models in the literature.
Shashkin V.V., Goyman G.S., Tolstykh M.A.
2023-02-01 citations by CoLab: 4 Abstract  
Summation-by-Parts Finite Differences (SBP-FD) is an approach for approximation of differential operators satisfying a discrete analogue of integration by parts analytic property. SBP-FD allows one to build provably stable high-order spatial approximations. SBP-FD methods are widely used for approximation of partial differential equations in multiblock domains with logically-rectangular curvilinear mesh inside. The gnomonic cubed-sphere grid is an example of such a domain. However, applications of the SBP-FD approximation for the cubed-sphere grid in meteorological context are not a widespread research area. We present a SBP-FD based shallow water model using a non-staggered grid. The model is total-energy conserving, mass-conservative and has discrete analogues of other mimetic properties such as curl-free gradient property. The shallow water model is tested with the commonly used Williamson test suite supplemented with the Galewsky barotropic instability case. High-order convergence is shown for tests with analytic solutions. The SBP-FD shallow water model is more accurate than low-order mimetic finite-element counterparts, but slightly less accurate than high-order finite-volume, spectral-elements and discontinuous Galerkin schemes.
Tarasevich M., Sakhno A., Blagodatskikh D., Fadeev R., Volodin E., Gritsun A.
2023-01-01 citations by CoLab: 2 Abstract  
The paper discusses the parallel scalability of INM RAS Earth system model (INMCM). To study the parallel performance of INMCM as well as its individual components, a custom software toolkit is used. The results of the study made it possible to identify the optimal parallel configurations of the model and therefore accelerate the computations without significant redesign of the program code.
Tolstykh M., Goyman G., Biryucheva E., Shashkin V., Fadeev R.
2023-01-01 citations by CoLab: 2 Abstract  
SL-AV is a global atmosphere model used for operational medium-range and long-range forecasts. Following previous successful implementation of single precision in some parts of the model dynamics solver, such computations are introduced into some parts of parameterization block of the model. Also, some memory optimizations are implemented. As a result, the elapsed time needed to compute 24-h forecast for SL-AV version with 10 km horizontal resolution and 104 vertical levels is reduced by 22% while using 2916 Cray XC40 processor cores, without detrimental effects on forecast accucacy.
Alipova K.A., Goyman G.S., Tolstykh M.A., Mizyak V.G., Rogutov V.S.
2022-12-01 citations by CoLab: 5 Abstract  
Abstract Algorithms for stochastic perturbation of parameters and tendencies of physical parameterizations for subgrid-scale processes are implemented into the ensemble prediction system. This system is based on the global semi-Lagrangian atmospheric model SL-AV with the resolution of 0.9 × 0.72 degrees in longitude and latitude, respectively, 96 vertical levels, and our implementation of the Local Ensemble Tranform Kalman Filter (LETKF). The use of stochastically perturbed parameterizations allows to generate ensembles with a significantly larger spread compared to one obtained with the method of static parameter perturbation. An improvement in the probabilistic estimates of the ensemble forecast for different seasons is shown.
Zängl G., Reinert D., Prill F.
Geoscientific Model Development scimago Q1 wos Q1 Open Access
2022-09-23 citations by CoLab: 9 Abstract  
Abstract. This article describes the implementation of grid refinement in the ICOsahedral Nonhydrostatic (ICON) modeling system. It basically follows the classical two-way nesting approach known from widely used mesoscale models like MM5 or WRF, but it differs in the way feedback from fine grids to coarser grids is applied. Moreover, the ICON implementation supports vertical nesting in the sense that the upper boundary of a nested domain may be lower than that of its parent domain. Compared to the well-established implementations on quadrilateral grids, new methods had to be developed for interpolating the lateral boundary conditions from the parent domain to the child domain(s). These are based on radial basis functions (RBFs) and partly apply direct reconstruction of the prognostic variables at the required grid points, whereas gradient-based extrapolation from parent to child grid points is used in other cases. The runtime flow control is written such that limited-area domains can be processed identically to nested domains except for the lateral boundary data supply. To demonstrate the functionality and quality of the grid nesting in ICON, idealized tests based on the Jablonowski–Williamson test case (Jablonowski and Williamson, 2006) and the Schär mountain wave test case (Schär et al., 2002) are presented. The results show that the numerical disturbances induced at the nest boundaries are small enough to be negligible for real applications. This is confirmed by experiments closely following the configuration used for operational numerical weather prediction at DWD, which demonstrate that a regional refinement over Europe has a significant positive impact on the forecast quality in the Northern Hemisphere.
Giorgetta M.A., Sawyer W., Lapillonne X., Adamidis P., Alexeev D., Clément V., Dietlicher R., Engels J.F., Esch M., Franke H., Frauen C., Hannah W.M., Hillman B.R., Kornblueh L., Marti P., et. al.
Geoscientific Model Development scimago Q1 wos Q1 Open Access
2022-09-16 citations by CoLab: 15 Abstract  
Abstract. Classical numerical models for the global atmosphere, as used for numerical weather forecasting or climate research, have been developed for conventional central processing unit (CPU) architectures. This hinders the employment of such models on current top-performing supercomputers, which achieve their computing power with hybrid architectures, mostly using graphics processing units (GPUs). Thus also scientific applications of such models are restricted to the lesser computer power of CPUs. Here we present the development of a GPU-enabled version of the ICON atmosphere model (ICON-A), motivated by a research project on the quasi-biennial oscillation (QBO), a global-scale wind oscillation in the equatorial stratosphere that depends on a broad spectrum of atmospheric waves, which originates from tropical deep convection. Resolving the relevant scales, from a few kilometers to the size of the globe, is a formidable computational problem, which can only be realized now on top-performing supercomputers. This motivated porting ICON-A, in the specific configuration needed for the research project, in a first step to the GPU architecture of the Piz Daint computer at the Swiss National Supercomputing Centre and in a second step to the JUWELS Booster computer at the Forschungszentrum Jülich. On Piz Daint, the ported code achieves a single-node GPU vs. CPU speedup factor of 6.4 and allows for global experiments at a horizontal resolution of 5 km on 1024 computing nodes with 1 GPU per node with a turnover of 48 simulated days per day. On JUWELS Booster, the more modern hardware in combination with an upgraded code base allows for simulations at the same resolution on 128 computing nodes with 4 GPUs per node and a turnover of 133 simulated days per day. Additionally, the code still remains functional on CPUs, as is demonstrated by additional experiments on the Levante compute system at the German Climate Computing Center. While the application shows good weak scaling over the tested 16-fold increase in grid size and node count, making also higher resolved global simulations possible, the strong scaling on GPUs is relatively poor, which limits the options to increase turnover with more nodes. Initial experiments demonstrate that the ICON-A model can simulate downward-propagating QBO jets, which are driven by wave–mean flow interaction.
Sophocleous K., Christoudias T.
Atmosphere scimago Q2 wos Q4 Open Access
2022-09-02 citations by CoLab: 1 PDF Abstract  
Modelling atmospheric composition and climate change on the global scale remains a great scientific challenge. Earth system models spend up to 85% of their total required computational resources on the integration of atmospheric chemical kinetics. We refactored a general atmospheric chemical kinetics solver system to maintain accuracy in single precision to alleviate the bottleneck in memory-limited climate-chemistry simulations and file input/output (I/O) and introduced vectorisation by intrinsic functions to increase data-level parallelism exposure. The application was validated using seven standard chemical mechanisms and evaluated against high-precision implicit methods. We reduced required integration steps by ×1.5–3-fold, in line with double precision, while maintaining numerical stability under the same conditions, accuracy to within 1%, and benefiting from halving the required memory and reducing overall simulation time by up to a factor two. Our results suggest single-precision chemical kinetics can allow significant reduction of computational requirements and/or increase of complexity in climate-chemistry simulations.
Total publications
22
Total citations
112
Citations per publication
5.09
Average publications per year
2.75
Average coauthors
2.95
Publications years
2017-2024 (8 years)
h-index
5
i10-index
3
m-index
0.63
o-index
12
g-index
10
w-index
2
Metrics description

Top-100

Fields of science

1
2
3
4
5
6
Numerical Analysis, 6, 27.27%
Modeling and Simulation, 6, 27.27%
Computer Science Applications, 3, 13.64%
General Physics and Astronomy, 2, 9.09%
Physics and Astronomy (miscellaneous), 2, 9.09%
Computational Mathematics, 2, 9.09%
Applied Mathematics, 2, 9.09%
General Mathematics, 1, 4.55%
Hardware and Architecture, 1, 4.55%
Computational Theory and Mathematics, 1, 4.55%
Information Systems, 1, 4.55%
Computer Networks and Communications, 1, 4.55%
Software, 1, 4.55%
Water Science and Technology, 1, 4.55%
Atmospheric Science, 1, 4.55%
Fluid Flow and Transfer Processes, 1, 4.55%
1
2
3
4
5
6

Journals

1
2
3
4
5
6
1
2
3
4
5
6

Citing journals

5
10
15
20
25
Journal not defined, 8, 6.96%
5
10
15
20
25

Publishers

1
2
3
4
5
6
7
1
2
3
4
5
6
7

Organizations from articles

2
4
6
8
10
12
14
16
18
Organization not defined, 5, 22.73%
2
4
6
8
10
12
14
16
18

Countries from articles

5
10
15
20
Russia, 20, 90.91%
Country not defined, 3, 13.64%
5
10
15
20

Citing organizations

5
10
15
20
25
30
Organization not defined, 26, 23.21%
Show all (5 more)
5
10
15
20
25
30

Citing countries

10
20
30
40
50
60
Russia, 57, 50.89%
Country not defined, 14, 12.5%
France, 4, 3.57%
USA, 4, 3.57%
United Kingdom, 4, 3.57%
Canada, 4, 3.57%
Germany, 3, 2.68%
China, 3, 2.68%
Brazil, 3, 2.68%
Norway, 3, 2.68%
Finland, 2, 1.79%
Sweden, 2, 1.79%
Ireland, 1, 0.89%
Japan, 1, 0.89%
10
20
30
40
50
60
  • We do not take into account publications without a DOI.
  • Statistics recalculated daily.