M. A. Tolstykh

Vorticity-divergence mass-conserving semi-Lagrangian shallow-water model using the reduced grid on the sphere

The semi-Lagrangian semi-implicit shallow water model on the sphere using the reduced latitude-longitude grid is presented. The key feature of the model is the vorticity-divergence formulation on the unstaggered grid. The new algorithm for the reconstruction of wind components from vorticity and divergence is described. The mass-conservative version of the model is developed. The conservative cascade scheme (CCS) by Nair et al. is modified to provide a locally-conservative semi-Lagrangian advection algorithm for the reduced grid. Some numerical advection tests are carried out to demonstrate the accuracy of the CCS with the reduced grid. The CCS-based discretization for the continuity equation and finite-volume Helmholtz problem solver are introduced to guarantee the mass-conservation. The results for shallow water tests on the sphere are presented. The results for different versions of the model are compared. They are also compared with the results for the same tests available in literature. The impact of the reduced grid is analyzed. The mass-conservative version of the model using the reduced grid with up to 20% reduction of grid points number has approximately the same accuracy as its non-conservative counterpart implemented on the regular latitude-longitude grid.

Parallel implementation of the cascade mass-conserving semi-Lagrangian transport scheme

Abstract Modern atmospheric models for climate simulations require accurate and efficient, locally mass-conservative and monotonic numerical schemes for treating the transport of atmospheric constituents. One of the ways to design such schemes is Finite-Volume Semi-Lagrangian approach (FVSL). FVSL schemes are characterised by the computational efficiency advantage due to the possibility of using large time-steps and efficient treatment of multiple transported quantities (tracers). This article presents massively parallel and multi-tracer efficient version of the recently developed 3D cascade FVSL transport scheme. Using hybrid distributed-shared memory parallelism with 1D MPI domain decomposition in latitude and OpenMP computations for longitude loops allows to use efficiently up to 1600 computational cores. We hope this number will grow with the future growth of the number of shared memory cores per computational node. Multi-tracer optimisations of the scheme (mostly, developing multi-tracer efficient monotonic filter) allow to reduce the cost of transporting additional tracer to 18–23% of running the scheme with one tracer.

Development of the multiscale version of the SL-AV global atmosphere model

The global hydrodynamic atmosphere model SL-AV is applied for operational mediumrange weather forecast and as a component of the probabilistic long-range forecast system. The review of the previous development of the model is presented and the model features are noted. The existing model versions are described. The unified multi-scale version of the model is developed on the basis of these versions. This version is intended both for numertcal weather prediction and for modeling of climate changes. The numerical experiments on climate modeling with the developed multi-scale version are carried out according to the protocol of the international AMIP2 experiment. First results are pret ented. The pos tibiltty of application of the unified vertion of the SL-AV model for the met dium-range weather forecast, and, after some development, for modeling of climate changes is shown.

Coupled atmosphere–ocean model SLAV–INMIO: implementation and first results

Abstract Coupled atmosphere–ocean models are widely used for climate change modelling. However, there is now more and more evidence on necessity to use such models in numerical weather prediction at different time scales. A coupled model is developed at the Institute of Numerical Mathematics, Shirshov Institute of Oceanology (Russian Academy of Sciences), and Hydrometeorological Research Centre of Russia. Particularities of program implementation for this model are discussed. The atmosphere model SLAV and the World Ocean model INMIO are coupled using the original program for models coupling. The results of numerical experiments with the coupled model demonstrate an agreement with observation data and show a possibility to use this model for probabilistic weather forecasts at time scales from weeks to year.

Parameterization of mires in a numerical weather prediction model

Mires (peat-accumulating wetlands) occupy 8.1% of Russian territory and are especially numerous in the western Siberian Lowlands, where they can significantly modify atmospheric heat and water balances. They also influence air temperatures and humidity in the boundary layers closest to the earths surface. The purpose of our study was to incorporate the influence of mires into the SL-AV numerical weather prediction model, which is used operationally in the Hydrometeorological Center of Russia. This was done by adjusting the multilayer soil component (by modifying the peat thermal conductivity in the heat diffusion equation and reformulating the lower boundary condition for Richards equation), and reformulating both the evapotranspiration and runoff from mires. When evaporation from mires was incorporated into the SL-AV model, the latent heat flux in the areas dominated by mires increased strongly, resulting in surface cooling and hence reductions in the sensible heat flux and outgoing terrestrial long-wave radiation. Presented results show that including mires significantly decreased the bias and RMSE of predictions of temperature and relative humidity 2 m above the ground for lead times of 12, 36, and 60 h from 00 h Coordinated Universal Time (evening conditions), but did not eliminate the bias in forecasts for lead times of 24, 48, and 72 h (morning conditions) in Siberia. Different parameterizations of mire evapotranspiration are also compared.

LOCAL ENSEMBLE TRANSFORM KALMAN FILTER DATA ASSIMILATION SYSTEM FOR THE GLOBAL SEMI-LAGRANGIAN ATMOSPHERIC MODEL

Abstract Recently, ensemble Kalman filters have come into practical data assimilation for numerical weather prediction models. We give an overview of ensemble Kalman filters and problems that arise with practical implementation of ensemble methods. We present our implementation of the local ensemble transform Kalman filter, one of ensemble square root filters using observation localization. Multiplicative and additive inflations are used to prevent filter divergence and to account for the model error. The implemented assimilation system is tested with the global semi-Lagrangian atmospheric model SL-AV using real observations for 2 months of cyclic assimilation (August and September 2012). The system works stably. Application of the ensemble filter significantly reduces first guess (background) errors and corrects the forecast biases.

Structure and Algorithms of SL-AV Atmosphere Model Parallel Program Complex

We present recent modifications of the SL-AV global atmosphere model parallel structure and algorithms. The modification of the hybridMPI+OpenMP parallelization structure as well as new parallel I/O system is described. The new multigrid algorithm for solving the linear algebraic equations systems arising from discretization at the reduced latitude-longitude grid is introduced and the convergence results for this method are presented.

Further Development of the Parallel Program Complex of SL-AV Atmosphere Model

The SL-AV global semi-Lagrangian atmosphere model is applied to the operational medium-range weather forecast at Hydrometeorological center of Russia. The works on increasing the code scalability and using future computer architectures are described. The scalable parallel multigrid algorithm for solving the linear algebraic equations systems is implemented. It is expected that the multigrid algorithm will be used instead of direct algorithm based on fast Fourier transforms requiring global communications. The results for convergence and strong scalability of the multigrid method are given.

A Simplified Extended Kalman Filter assimilation of soil moisture for the SL-AV global medium-range forecast model

The implementation of the Simplified Extended Kalman Filter (SEKF) for the deep soil moisture initialization in the SL-AV global atmosphere model is described. Special attention is paid to the calculation of the observation operator and analysis increment. SL-AV screen-level parameters forecasts are estimated with SEKF and optimal interpolation initialization methods. It is demonstrated that the implementation of the assimilation algorithm improves the model forecast quality for screen-level temperature and relative humidity.

SL-AV atmospheric model version using σ- p hybrid vertical coordinates

The SL-AV atmospheric model version using hybrid vertical coordinates combies the advantages of sigma and isobaric coordinates. The formulation and discretization of model equations maintain the equivalency of the new model version to the basic sigma version in the special case, when hybrid coordinates coincide with sigma coordinates. The SL-AV model version with hybrid vertical coordinate is verified with medium-range weather forecasts. The decrease in the errors of predicted geopotential height and wind as compared to the sigma model version is demonstrated. The use of hybrid coordinates also leads to a certain increase in forecast skill scores for some meteorological parameters characterizing aviation significant weather.