Investigation of Chaos and Dynamical Phase Space Reconstruction of Precipitation in Daily, Weekly and Monthly Scales (Case Study: Qarah-Soo Watershed in Kermanshah)

In this research the chaocity of precipitation time series in daily, weekly and monthly scales in the Pole-Kohneh and Ghourbaghestan Stations located in Qarah-Soo watershed is investigated. In order to reassure about the absence of stochastic behavior of the time series, method of surrogate data is applied. Thereafter, the aperiodicity of the time series is investigated using power spectrum analysis. The results of chaocity investigation indicates that according to correlation dimension method and Lyapunov exponent, the daily scale is not chaotic but weekly and monthly scales are chaotic (existence of the largest Lyapunov exponent and correlation dimension between 4 to 7). Also, the phase space reconstruction is done using time delay and embedding dimension method (embedding dimensions of 7 to 10 and time delay of 2 to 4 for Pole-Kohneh Station; embedding dimensions of 9 to 19 and time delay of 2 to 3 for Ghourbaghestan Station). Findings of the research indicate the stochasticity of short term precipitation time series (daily) and chaosity of medium term precipitation time series (weekly and monthly).