Frequency Analysis of the Monthly Rainfall Data at Sulaimania Region, Iraq

摘要

Different frequency distributions models were fitted to the monthly rainfall data in Sulaimania region, north Iraq. Three rainfall gauging stations data were used, Sulaimania city, Dokan Dam, and Derbendikhan Dam metrological stations, for the period (1984-2010). The distributions models fitted are of Normal, Log-normal, Wiebull, Exponential and Two parameters Gamma type. The Kolmogorov-Smirnov test was used to evaluate the goodness of fit. The fittings were done for the overall data and for each month separately. The Gamma, Exponential and Weibull distributions were found as the best fits for the three stations respectively for the overall models, while for the monthly models different distribution type was found as the best fit for each month and each station, however the Gamma distributions was found to have the highest percent of best fit. The best fitted distributions were used to forecast three sets of monthly rainfall data for each station and compared to the observed ones for the last 7- years of data. The t-test,F-test and Kolmogorov- Smirnov test indicate the capability of these models to produce data that has the same frequency distribution of the observed one. Comparison between the performances of the overall and periodic models reveals that there no distinguishable improvement of the monthly model over the overall one