Estimation of precipitation preceding precipitation extremes of different lengths

For water management purposes, the design rainfall intensities with different durations are an essential data basis. However, the actual runoff response of such a rainfall event is substantially influenced by other factors, including the time structure of the rainfall event and the previous saturation of the landscape. In our paper, we focus on the latter of these factors because the design of small-scale water management structures is usually based on the assumption that the saturation of the catchment at the beginning of a rainfall event of a given duration is on average. The aim of our research is to test the validity of this assumption in different parts of the country and to trace any general patterns of the relation between precipitation extremes and antecedent precipitation totals in relation to relief.

We use data from 60 Czech rain-gauge stations for which short-term rainfall intensity data for several decades are at our disposal. At each station, we select annual maximum rainfall totals for time windows ranging from half an hour to one day. We estimate the magnitude of antecedent saturation using the antecedent precipitation index over a 30-day period (API30); for sub-daily extreme totals, we also consider any precipitation on that day from the morning to the start of the precipitation episode. Because the index exhibits almost the same seasonal distribution as the daily precipitation totals, we express the magnitude of API30 relative to the normal for a given calendar day. For each station, we calculate relative saturation values before each annual maximum rainfall episode of a given length. Considering the episodes with decreasing weight according to rainfall magnitude, we evaluate average relative saturation values. We find that already before episodes with a length of 150 minutes we have to take into account slightly increased saturation, especially in mountainous regions in the east of the Czech Republic.