Abstract
The sampling adjustment factor (SAF) can correct the underestimation of fixed time interval maxima (F-Maxima) relative to maxima of moving time windows (M-Maxima) as a direct consequence of temporal discretization of time series. Radar data can help to understand the significance of using moving window aggregation rather than fixed window. Here, we investigate SAFs for two gridded radar quantitative precipitation estimates from the German Meteorological Service (RADKLIM-RW and RADKLIM-YW, DWD 2018) with different temporal resolutions (5 min and hourly) and a spatial resolution of 1 km $$\times $$ × 1 km from 2001 to 2016. For each grid cell’s time series within Germany, the overall maximum intensities per chosen duration were derived with (1) moving and (2) fixed window aggregation. Correction factors (SAF) from fixed to moving window were retrieved. The findings of this study partly match previous findings, more for the 5-min product than for the hourly product. No clear dependency on topography or season could be derived from the analysis. It showed that high SAF are usually related to high(er) actual rainfall values. The study emphasises the probabilistic nature of the rainfall maxima correction and shows that it is important to not only consider average SAFs, but also take an in-depth look at the distribution when correcting maxima. As a consequence, design precipitation as used, e.g., in context of flash or urban floods could profit from spatially adjusted or uniform scaling, depending on the characteristics found in the spatial distribution of the SAFs.