An Evaluation of Probability Distributions of Synthetic Storms

The study examines whether 56 data sets consisting of 100 synthetic storms with the same probability distribution (Gumbel) are different than the distribution provided for at the beginning. For this purpose, synthetic synthetic storms of Gumbel distribution with a specific time distribution and random effective durations, of which population averages and variances are known, are being derived with Monte Carlo simulation method. The parameters of the storm values derived were determined using the maximum possibility method for 7 probability distribution widely used in hydrology, and their compliance was examined using Chi-square () and probability plot correlation coefficient tests (PPCC). It was seen that the probability distribution of the precipitation input can be different from the main distribution (Gumbel) provided for. The reason for this is that the precipitation inputs created are in the form of synthetic storms of different periods and the sample statistics of these series are different from the main distribution based on sampling.