Probabilistic investigation of soil moisture conditions over Greece for use in hydrological design

In hydrological design, the most widely used, in Greece and worldwide, procedure for the estimation of active rainfall is based on the runoff curve number method (SCS-CN). This is an empirical estimation method of active rainfall, which has been developed by the Soil Conservation Service after experiments in specific types of watersheds in the United States of America. Therefore, when applying the method in Greek basins deviations from reality or misleading results may occur. The uncertainty is involved in selecting the appropriate curve number (CN), which is the main parameter of the method. According to the classical concept of SCS, CN depends on the hydrological soil type (defined groups with different classes of soil permeability and land use) and the antecedent moisture conditions, (AMC). The antecedent moisture conditions are classified into three categories, dry (AMC1), average (AMC2), and wet (AMC3), and is selected according to the antecedent precipitation. Most of the times in hydrological design in Greece, the soil moisture conditions are considered average (AMC2) . In this paper, we attempt a probabilistic approach of soil moisture conditions in Greece, in order to choose the most appropriate CN and therefore have more reliable results when applying the method in Greek basins. For this reason, we collected rainfall time series from 215 stations across Greece. From the analysis of the samples we assigned a probability of occurrence to each moisture condition, both in continuous time (i.e. for a random day of the year) and during the day of the annual maximum rainfall. We also made spatial analyses, the results of which are depicted on maps for practical use in flood studies. Finally, an application of hydrologic design was made considering the probabilistic approach of soil moisture conditions. The results are compared with those derived by considering each type of antecedent moisture condition separately.