2.2.2. Quantile Mapping Based on Gamma + Generalized Pareto Distribution

Quantile mapping based on gamma + generalized Pareto distribution is a quantile mapping method similar to eQM but assumes that the observed and simulated precipitation density distribution are correctly approximated by gamma, and the temperature density distribution is correctly approximated by Gaussian distribution. Therefore, it uses theoretical distribution in the quantile mapping instead of empirical distribution. Due to the fact that gamma distribution is a light-tailed distribution, it is combined with a general Pareto distribution [39]. The observed and simulated quantiles were interpolated by inverse distance weighting. The 1 mm threshold cut off was also applied to precipitation in this approach. The gpQM bias correction with a 90-day moving window in the case of precipitation was used. Owing to the fact that the seasonal temperature density distribution cannot be approximated by Gaussian distribution in some places in Europe [43], the temperature gpQM

correction produces a large number of infinitive values with a 90-day moving window. Therefore, gpQM was applied only in the case of precipitation.
