Abstract
We proposed three analogous extremity indices based on the estimated return periods at individual sites and spatial averaging of the values; we optimized both the areal extent and the duration of individual events. The weather extremity index (WEI), the 5 weather abnormality index (WAI), and the flood extremity index (FEI) were applied to the original precipitation data, the seasonally transformed precipitation data, and the runo data to identify extreme precipitation events (EPEs), abnormal precipitation events (APEs), and extreme flood events (EFEs), respectively.
We present 50 events of each type from the period of 1961-2010 in the Czech Republic and compare their inter-annual and seasonal distributions. Most of EFEs were produced by an EPE in warmer half-years, whereas fewer than half of the EFEs were produced by an APE in the remainders of the years because thawing can substantially enhance the discharge at those times.
Most significant EPEs occurred in July and the first half of August, although their hydrological responses were also significantly influenced by the antecedent saturation and other factors. As a result, the accumulation of precipitation extremes during the 1977-1986 period produced less significant flooding than another accumulation after 1996.
In general, the primary discrepancies between the magnitudes of EPEs and EFEs occurred in May and September, when consequent floods were usually much larger and smaller in relation to the WEI, respectively. The 20 hydrological response to APEs was usually strong in December, whereas another accumulation of EFEs in March was usually not due to APEs.
Neither precipitation nor flood extremes occurred from early April through early May. This study confirms that variations in the frequency and/or magnitude of floods can be due not only to variations in the magnitude of precipitation events but also to variations in their seasonal distribution and other factors, primarily the antecedent saturation.