Hydrometeorological Research Centre of Russian Federation

A technology for operational complex forecasting (postprocessing)1 of squall wind gusts at the points of weather station location is described. The technology was implemented at the Hydrometeorological Research Center of Russia. It uses the results of forecasts of several forecast models and the complex forecast of average wind. The tests have shown an increase in forecast accuracy as compared to original models.

Abstract This study investigates the association between adolescents' digital literacy (DL) and cognitive control using ecologically valid, scenario‐based tasks developed within an evidence‐centred design (ECD) framework alongside a modified Stroop‐like verbal–spatial test. The research was conducted at scale with 2860 students in years 7–8 (aged 13–14) from 102 schools across four regions of the Russian Federation, as part of a school‐based administration. DL was assessed with performance tasks emulating authentic digital activities; scores were analysed in the Rasch framework. Cognitive control was indexed by response correctness speed (RCS) – a composite of accuracy and time – from incongruent trials of the verbal–spatial test. Associations between RCS and DL (overall and subconstruct scores) were estimated using Spearman's ρ with 10,000‐iteration permutation tests and Bonferroni adjustment. Overall, DL showed a significant positive correlation with cognitive control ( ρ = 0.407, p Bonf <0.01). Stronger correlations were observed for cognitively demanding subconstructs – computational literacy and information creation – followed by information analysis and search, whereas weaker or non‐significant relationships were found for domains reflecting procedural or socio‐normative knowledge (e.g., digital security, digital communication). These results indicate that executive functions selectively underpin DL performance under realistic cognitive demands, supporting the use of scenario‐based, process‐sensitive assessments for both measurement and instruction. The findings have implications for the design of DL assessments that minimize construct‐irrelevant variance and for pedagogies that combine explicit digital strategies with the fostering of attentional regulation and inhibitory control.

This paper examines selection of river flow forecasting methods based on factors limiting the lead time and accuracy of forecasts, end users’ requirements, and the results of applying these methods to specific water bodies. Methods that allow widespread automated implementation in operational use are recommended as the most effective. The paper gives some examples of successful implementation of these recommendations for river flow forecasting in Russia, based on methods developed at the Hydrometeorological Center of Russia. The proposed recommendations are aimed at improving the scientific validity and effectiveness of operational decisions regarding water resources use and protection against hazardous hydrological phenomena.

It is proposed to use spline interpolation in shallow water condition to obtain a continuous temperature profile (without density inversion) in a column containing layers of ice and water and a selected layer of sea bottom soil. For the algorithm work, it is necessary to know the temperatures on the ice surface and at the soil layer bottom. The temperature field obtained in this way for the northern part of the Caspian Sea was used as the initial one in the test calculation of a 3-day ice thickness forecast based on the experimental forecasting model. NCEP model data were used at all stages of the work.

3 , -

The paper examines the impact of applying the parameter perturbations for schemes describing sub-grid scale processes in the SL-AV atmosphere model on characteristics of monthly mean atmospheric circulation in ensemble subseasonal forecasts. 25-year retrospective forecasts are computed for different seasons with different perturbation sets, and the results are compared with the forecasts without perturbations. It is shown that perturbing just three parameters reduces forecast errors of some variables as compared to unperturbed forecasts without introducing errors into integral characteristics.

The corrosion of low-carbon steel in solutions of 1 M H2SO4 + 1 M H3PO4 containing Fe(III) salts is studied relative to similar solutions of 2 M H2SO4 and 2 M H3PO4. Steel corrodes in these systems by reacting with the acid solution and salts of Fe(III). Partial reactions of the anodic ionization of iron and the cathodic reduction of H+ and Fe(III) cations occur on steel. The first two are characterized by kinetic control; the last, by diffusion control. The accelerating effect Fe(III) cations have on the corrosion of steel in the studied environments is mainly due to the reduction of Fe(III). A rotating steel disk electrode is used to study the effect the nature of solution convection has on the kinetics of Fe(III) reduction in the considered corrosion systems. Coefficients of diffusion are determined for Fe(III) cations in solutions of H2SO4, H2SO4 + H3PO4, and H3PO4 via the cyclic voltammetry method with using a Pt electrode, allowing the construction of model dependences of the cathode current density of a steel disk on its frequency of rotation. The dependences are then compared to experimental data in order to establish the reasons for the observed differences. It is assumed that the differences are determined by the release of hydrogen gas on steel, which both shields the steel’s surface and changes the flow of the aggressive environment near the electrode from laminar to turbulent. Empirical dependences of the rate of steel corrosion on the intensity of the medium’s flow in the considered environments, obtained using data on the mass loss of metal samples, are described by linear equation k = kst + λw1/2, where kst is the rate of steel corrosion in a static environment, w is the speed of rotation of the propeller mixer that creates the flow of the medium, and λ is an empirical coefficient.

The prediction of the level of air pollution by gaseous and aerosol constituents in cities becomes increasingly significant in view of their serious negative impact on public health and growing ecological risks. The article presents an approach to estimating and adjusting the emission power of anthropogenic sources based on direct and inverse modeling. The WRF-Chem model was used as a direct simulation tool, and the IMDAF system developed by the authors was used for inverse simulation. The results of direct simulation provided data on meteorological fields and the distribution of admixtures necessary for solving adjoint problems. The use of the adjoint problem method made it possible to calculate a correction factor that determines how much the power of sources that fall into the sensitivity zone should be changed to achieve the best agreement with measurements. Our approach can be used to improve the prediction of air quality, refine the inventories of anthropogenic emissions, and develop the strategies for reducing the ecological risks on global and regional scales.

In 2022–2023, the Hydrometeorological Research Center of Russia successfully tested a new version of the SLAV10 global atmosphere model designed for medium-range deterministic weather forecast. The model has a horizontal resolution of about 10 km and 104 vertical levels. The tests have shown that the SLAV10 model allows preparing more accurate weather forecasts compared to the previous version of the SL-AV model for deterministic prediction (variable grid spacing of 16–26 km, 51 vertical levels), both due to its higher spatial resolution and more detailed description of subgrid-scale processes. The increased demand of the new model version for computer resources requires a deeper code optimization, as well as input/output algorithms.