Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Feliks M.

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: procesy hydrometeorologiczne

Sortuj według:

Ogranicz wyniki do:

Tradycyjne metody detekcji cykliczności procesów hydrometeorologicznych na przykładzie rzeczywistych ciągów pomiarowych

Feliks M.

Czasopismo Techniczne. Środowisko

2002

R. 99, z. 5-Ś

13--22

Intensive increase of computational power available to process data, being collected and produced, caused that a large numberr of inwestigation method was constructed, and a large amount of different cycle of the climatic system variations was found. Some of them have strong physical support or natural explanation but major kinds of this periodicity have no confirmation, neither physical nor environmental. Astronomic cycles cause periodicity in various physical processes within the atmosphere-ocean-cryosphere system. If the interactions within the components of the climatic system combine properly and balance each other, they can give roughly cyclic variations. But therre is other possible way that can give rise to periodicity. For short time series, cyclicity may often be the result of sampling fluctuation. In that case it is called pseudo or sampling cyclicity. Determining that these pseudo-cycles are not statistically significant poses the main problem for analysis. The objective of this text is to present four methods considered as classical and relevant tools to solve problems of cyclic climatic variations. Various historical time series was put under analyze by applied described methods. It was shown how could the results for particular method differ from each other. No physical explanation for found varrious cycles is provided in the paper. The methods employed are: regression, autocorrelation function, periodogram and spectral density function. The data used are: monthly water level of the Baltic sea in Świnoujście (S) in the period 1986-1997, precipitation totals (P mm) in July at Warsaw over 1811-1960 and mean air temperature (7°C) in July at Warsaw over 1811-1960, annual flows of the River Odra at Słubice for the period 1951-1989 and annual mean Wolf numbers for the period 1749-1996. Figures 1-2 shows plotted time series for every phenomenon. Plot of the raw data set versus time makes possible recognition of the patterns how process evaluate can. It is the best starting point for further investigation. Another pictures present the way how particular method works.

Porównanie wybranych metod detekcji cykliczności procesów losowych na przykładzie różnych szeregów czasowych. Cz. 1, obszar czasowy

Feliks M.

Czasopismo Techniczne. Budownictwo

2001

R. 98, z. 3-B

114--132

Ower the last decades, there has been a great increase of methods and techniques used for investigation of periodic structure of climatologic processes, caused by vital importance of determining climatic changes. The conventional methods and their modification as new computational techniques, differ between one another on the effectiveness and accuracy of results. In the paper there are presented methods applied for detection of periodicity in time series, and suitable statistical procedures for verification of periods detected. Regression and autocorrelation function methods were shown in their applications to various real data sets.