|
| |
|
|
Changes in occurrence of climate types in the simulations of future climate
Mejsnar, Jan ; Mikšovský, Jiří (advisor) ; Halenka, Tomáš (referee)
The thesis compares climate classification of the Czech Republic based on two regional climate models (ALADIN-Climate and RegCM3) and on two model period (1961-1990 and 2021-2050). First, a short summary of climate classifications suitable for the Czech Republic is introduced. Above all are mentioned the classifications of Köppen, Kurpelová and Konček. The advantages of these classification are presented, especially their relation to occurrence of vegetation. Further, this thesis mentions today's regional climate models. The data of monthly values of precipitation and temperature were analysed by single classifications criteria and consequently visualized by the program SURFER to show individual climatic subtypes. The data represented averaged values for the periods 1961-1990 and 2021-2050. Finally the distribution and the veracity of both models are discussed.
|
|
|
Nonlinear modelling of meteorological data
Pop, Lukáš ; Raidl, Aleš (advisor) ; Mikšovský, Jiří (referee)
Thesis deals with possibilities of application of nonlinear methods based on the theory of dynamical systems, for meteorological problems. In the first part there is explored description of the dynamical systems, namely deterministic chaotic systems, for them this methods were introduced. There is described the concept of attractor and its typical invariants. In the next part there are summarized different ways of reconstruction of phase space of system by means of time series produced by it. A modification of Cao's method is proposed here. Possibilities of prediction of meteorological quantities are proposed here. In the next part there is described influence of different sorts of signal noise and inaccuracies of measuring to the dynamical system that is described by logistic equation. That is a simple example of deterministic chaotic system. In the last part there are analysed single types of errors, of them nonlinear methods of prediction are encumbered. There are described typical properties and specialities of meteorological data. There are tried modifications of local linear models also, what is applied to data of NCEP/NCAR reanalysis.
|
|
| |
|
| |
|
| |
|
|
Attribution of European temperature variability during 1882–2010: A statistical perspective
Mikšovský, Jiří ; Pišoft, P.
Gridded monthly temperature data (GISTEMP and Berkeley Earth) covering the European region were investigated for the presence of components attributable to climate forcings, both anthropogenic and natural, and to major modes of internal climate variability. Effects of individual predictors were separated by multiple linear regression applied to time series over 1882–2010. It was shown that the presence of a warming trend correlated with greenhouse gases concentration was generally strong in European temperatures and typically combined with mild cooling ascribable to anthropogenic aerosols. Components attributable to variations in solar activity were rather weak and not statistically significant in most locations, as were the imprints of large volcanic eruptions. A strong association between North Atlantic Oscillation phase and temperature was confirmed for much of Europe, while temperature oscillations synchronized with the El Niño Southern Oscillation were quite limited in magnitude and displayed low levels of statistical significance. The influence of the Atlantic Multidecadal Oscillation was noticeable particularly in the western-most parts of Europe, whereas the Pacific Decadal Oscillation’s significant impact extended to Scandinavia.
|
guest ::
RSS feed.