Basic principles of the TEVY index for the quantification of temperature variability within a year

Abstract

In the context of climate change, there is a need for the determination of appropriate indexes for the quantification of temperature variability. A new index (TEVY index) is proposed in this work. This index uses the deviation of the observed temperature values from those estimated from a Fourier harmonic analysis. For this purpose, a nearly 50-year time series data from 4 stations in Greece, with very different climatic conditions, are used. One station is located in the colder northern region of Greece, another one is in the warmest southern part, while the 2 other stations are representative of continental and Mediterranean climatic features. A Fourier harmonic analysis is carried out to obtain the Fourier series which simulates the observed data time series. Fourier harmonic analysis, which is relied on the Fourier transform, is a well-established method for time series analysis, particularly for modelling periodic data. Using this procedure, an index of temperature variability is proposed, as the sum of the divergence of the above-mentioned Fourier series from the observed data. The index results are analysed as a function of the different climatic features of each station.