The paper presents a spatial distribution of changes of air temperature (T) in the Arctic. Estimates of their spatial relations in the study region were based on a correlation analysis. T in the Arctic is most strongly correlated spatially in winter and spring, and least in summer. The radius of extent of statistically significant correlation coefficients of changes of T at the stations Svalbard Lufthavn, Ostrov Kotelny and Resolute A is equal to 2000-2500 km in winter and 1500-2000 km in summer. An attempt was done to delimit the regions of consistent occurrence of the anomalies T with respect to the signs and magnitudes, as well as of the regions with the most coherent T. The Wroclaw dendrite method was used to solve this problem. Relations of the mean areał T of the climatic regions and of the Arctic as a whole, with the northern hemisphere of temperature and selected climatic factors are presented.

Small sample properties of unrestricted and restricted canonical correlation estimators of cointegrating vectors for panel vector autoregressive process are considered when the cross-sectional dependencies occur in the process generating nonstationary panel data. It is shown that the unrestricted Box-Tiao estimator is slightly outperformed by the unrestricted Johansen estimator if the dynamic properties of the underlying process are correctly specified. The comparison of performance of the restricted canonical correlation estimator of cointegrating vectors for the panel VAR and for the classical VAR applied independently for each cross-section reveals that the latter performs better in small samples when the cross-sectional dependence is limited to the error terms correlations, even though it is inefficient in the limit, but it falls short in comparison to the former when there are cross-sectional dependencies in the short-run dynamics and/or in the long-run adjustments.

In the dissertation the data modeling has been shown for the data that regards the damages, which value is above zero. With the use of

Weibull distribution, with prior regression and correlation analysis chosen parameters that defines the life time and failure level of two

populations of AlSi17Cu5 were defined. The calculation sheet of reliability allows to create so called survival diagram, and on the basis of

durability data the average warrantee can be determined, on the pre-exploitation period.

In this paper, the recent ice regime variations in the Kara Sea have been described and quantified based on the high-resolution remote sensing database from 2003 to 2017. In general, the Kara Sea is fully covered with thicker sea ice in winter, but sea ice cover is continuously declining during the summer. The year 2003 was the year with the most severe ice conditions, while 2012 and 2016 were the least severe. The extensive sea ice begins to break up before May and becomes completely frozen at the end of December again. The duration of ice melting is approximately twice than that of the freezing. Since 2007, the minimum ice coverage has always been below 5%, resulting in wide open-waters in summer. Furthermore, the relevant local driving factors of external atmospheric forcing on ice conditions have been quantitatively calculated and analyzed. Winter accumulated surface air temperature has been playing a primary role on the ice concentration and thickness condition in winter and determining ice coverage index in the following melt-freeze stage. Correlation coefficients between winter accumulated temperature and ice thickness anomaly index, the ice coverage anomaly index, duration of melt-freeze stage can approach -0.72, -0.83 and 0.80, respectively. In summer, meridional winds contribute closely to summer ice coverage anomaly index, with correlation coefficient exceeding 0.80 since 2007 and 0.90 since 2010.