Original paper

Assessment and projection of high and low precipitation extremes simulated by COSMO-CLM4.8 for four small river catchments in Hesse, Germany

Huebener, Heike; Stoffels, Benno; Brahmer, Gerhard

Meteorologische Zeitschrift Vol. 25 No. 5 (2016), p. 563 - 575

50 references

published: Oct 31, 2016
published online: Sep 8, 2016
manuscript accepted: Feb 15, 2016
manuscript revision received: Dec 9, 2015
manuscript revision requested: Sep 2, 2015
manuscript received: May 12, 2015

DOI: 10.1127/metz/2016/0708

BibTeX file

O

Open Access (paper can be downloaded for free)

Download paper for free

Abstract

The applicability of the regional climate model COSMO-CLM (CCLM, Version 4.8) at 7 km horizontal resolution is assessed for hydrological research questions, using the driving conditions from the global model ECHAM5, in a bias adjusted (or bias corrected) version. The analysis considers relevant precipitation indices for the reference period 1971 to 2000 both for the hydrological summer- and winter-half-year for four small river catchments in Hesse. Besides the mean values, the main focus here is on several indices used in hydrological assessment of flood and low-flow risks i.e. maximum precipitation of one or more consecutive days (up to 21 days), minimum precipitation of consecutive days (30–120), maximum numbers of consecutive dry or wet days and maximum precipitation of consecutive wet days. The direct model output already represents the observed spatial patterns well but precipitation is generally overestimated (in the order of 10–50 %) for most indices. Values calculated from the bias adjusted data reproduce the mean values for all those quantities much better. However, even in the bias adjusted data the year-to-year variability in the indices is often underestimated, resulting in an underestimation of extremes for the dry as well as for the wet end of the distribution in most catchments. In addition to the reference period, a projection under the scenario A1B for 2021 to 2050 is analyzed. The results show increasing mean precipitation in the winter half year while the precipitation sum in the summer half year stays approximately constant. In most of the other indices we can identify small trends, but most of them are not significant probably due to the unfavorable signal-to-noise ratio for the short time horizon of the projections used here.

Keywords

precipitationbias adjustmentsmall river catchmentsGermany