Daily precipitation data from 38 stations in Iran are analyzed for the period of 1960-2001 to evaluate the possible long term trend in annual and seasonal precipitation. Statistical analysis using the run test shows that the annual and seasonal precipitation series are homogeneous at most of the stations, with the exception being the summer precipitation in the arid regions. The Mann-Kendall method was used to test the existence of trends, and the least square method to estimate the values of trend. The values of trend are expressed in terms of the percentage of mean annual and seasonal precipitation over the study period. There is a significant trend (at 90% level) in the annual precipitation 10 stations and in the number of rainy days at 21 stations. The results show that the annual precipitation has a decreasing trend at stations in the west, northwest and southeast and an increasing trend at most of the stations in the other parts of the country. The winter precipitation shows a trend very similar to that of the annual precipitation. Most of the stations have shown a reduced precipitation during the spring and increased precipitation in autumn. Large scale study of air mass trajectories and local evaluations through station meta data are required for exploring the possible causes of such changes. The results show that for the stations (seasons) where total precipitation has a positive, large and often significant trend, heavy precipitation shows an increasing trend and light precipitation shows a decreasing trend. For some other stations (seasons), where total precipitation has a large negative trend, there is negative trend in the upper intensity quantiles and a positive trend in the lower intensity quantiles.
)