A Trend Analysis of Annual Precipitation over the Recent Half Century Based on Data of Synoptic Stations in Iran

One of the most important conditions in time series analysis is that time series have to be stationary. A time series is stationary in its statistical parameters when its expected value is time independent. Most time series are non-stationary. Thus it is necessary to recognize the factors causing non-stationariness and eliminate them from the time series. One of the important factors causing non-stationariness is the time series trend. In this research, the annual precipitation data set of 31 synoptic stations in the captials of the provinces in Iran during the period of 1961 to 2010 were used to investigate possible trends. The results of this study indicated that among 31 stations in the country, approximately 32 percent of the stations had a non-significant upward trend at the level of 5 percent, 9 percent of the stations had a non-significant downward trend at the 1 percent level and almost 48 percent of the stations in the country had a non-significant downward trend at the 5 percent level in the recent half-century.