Abstract

Surface air temperature time series, observed in 42 Iranian synoptic stations, are analyzed in frequency domain. We used the least squares spectral analysis method, which is able to take into account nonequally spaced data series. One can distinguish the statically meaningful harmonic fluctuations in the time series using this method as well. The minimum length of data is 44 years. The aim of spectral analysis is to detect the periodicities longer than one year. Synoptic stations are distributed almost homogenously in Iran. The result of spectral analysis are as follows: 1. All the stations experience periodicities of 2 to 5 years, with amplitudes from 0.1ºC to 0.5 ºC. The corresponding maps are derived. These variations are attributed to El-Nino effect by pervious researches. 2. All the stations show 9 to 12 years periodicities with 0.1 ºC to 0.7 ºC variations. These periods may be attributed to 11 years solar activities as well as to the first harmonic of 18.6 variations of the earth spin axis due to nutations effect. 3. In 36% of stations we observe 17 to 20 years periodicities with amplitude varying from 0.0 ºC to 0.6 ºC. They are probably due to nutations effect of the Earth's rotation axis, which causes the change of the altitude of the sun. 4. In 60% of stations periodicities between 20 to 25 years with amplitudes from 0.2 ºC to 0.7 ºC are observed. They are attributed to the solar activities known as Hale period. 5. In some stations we observe harmonics with more than 25 years periods. But the length of our data sets is not sufficient in order to explain with certitude about their cause(s).