Trend analysis and change point detection of seasonal and annual precipitation in Iran

Abstract The aim of this study was to analyse spatial and temporal homogeneity and trends in seasonal and annual precipitation and their fluctuations during 1957–2016 in Iran. For testing homogeneity, the Pettitt‐Whitney–Mann, the Standard Normal Homogeneity or Alexandersson's SNHT, Buishand's and the Von Neumann tests were used and for trend analysis, Mann‐Kendall and Sequential Mann‐Kendall tests were used. Homogeneity tests showed that most of the seasonal and annual precipitation time series were homogeneous and the change point was detected only at a few stations. Results indicated that in general, there was an upward trend during 1957–1986, followed by a significant downward trend during 1987–2016. The sequential Mann‐Kendall test detected several possible change points over time, although most of them were found statistically non‐significant at 95% confidence level. Upward and downward trends started in annual and seasonal precipitation after 1960 and 1990, respectively, at most stations, except in spring. In spring, an upward trend started after 1990 at some stations which showed a shift in precipitation from winter to spring. Also, the Sen's slope estimator indicated a higher downward trend in the northern parts of Iran. Two findings are worthy of note based on the results of this study. First, in general, a significant downward trend occurred in the whole country after1990. Second, a shift in precipitation from winter to spring was detected at some stations in different homogeneous regions.