Correction annual maximum discharge based on appropriate probability distribution function in south of Iran

In this research for choosing the best distribution function for the annual maximum discharge (AMD) in the southern provinces of Iran; The daily discharge of 108 hydrometric stations (1983-2012) were used; Data were fitted with 65 probability distribution functions. After the goodness of fit tests using the statistical calculations, the best distributions function for the AMD was determined and eventually the discharge amounts were calculated with different return periods and compared with the result of the common distribution functions like log- Pearson (III), log- normalIII and Weakby. The result shown, the Waekby distribution functions with the 2.43% and the first rank, the log- PearsonIII with the frequency of 6.13% and the second rank and log- normal III with the frequency of the 5.6% and the third rand, gained the best statistical distribution. The MBE in AMD estimation showed that in 2.5 and 10- year return period, the Weakby statistical distribution and in the 25.5 and 100-year return period. The log- Pearson III statistical distribution has a better estimation. Comparing the RMSE with MAPE in both Weakby and log PearsonIII statistical distribution, it is found that Weakby statistical distribution has a better estimation in the different return periods in this index.