Dynamical and statistical downscaling of COSMO-CLM simulations for climate change adaption of transport infrastructure in Germany

14:00 Tuesday 28 May

SS009 • OC053

Room S10

 

Michael Haller (Germany) 1; Susanne Brienen (Germany) 1; Stefan Krähenmann (Germany) 1; Barbara Früh (Germany) 1

1 - German Meteorological Service, Frankfurter Stra¤e 135, 63067 Offenbach

The knowledge about frequencies and intensities of extreme events, triggered by climate change is one of the key issues in the project ‘Network of experts – Adapting transport infrastructure to climate change and extreme weather events’, launched in 2016 by the German Federal Ministry of Transport and Digital Infrastructure. Several federal agencies are working closely together to develop strategies for dealing with challenges of future traffic infrastructure. The vulnerability of traffic infrastructure has many different factors. One of these factors is the weather and climate impact. For the assessment of the vulnerability and possible adaptation strategies for transport infrastructure on a local scale, climate information is needed on the same scale.

Thus, we perform convection-permitting climate projections with COSMO-CLM on 2.8 km grid width for time periods of more than 30 years using the RCP 8.5 scenario, dynamically downscaled two-fold from MIROC5 global model data. In order to estimate the robustness of our climate projections we aim at developing a high resolution climate multi-model ensemble using a combination of statistical and dynamical downscaling. Thus, we apply a statistical downscaling technique for EURO-CORDEX ensemble members by linking high-resolution dynamical simulation at 2.8 km grid width to the simulations at 12 km grid width. For this procedure we use the principal component analysis (PCA) method. It has been tested for a 30 year period of climate projections driven with MIROC5.

Comparisons of dynamically and statistically downscaled simulations show overall good agreement, except that the latter produce lower temperatures and more rain. Compared to gridded observations, it turns out that the COSMO-CLM simulations produce too much precipitation over the Alps and too less over the northern parts of Germany. Comparing the historical run and the scenario, we might expect more frequent strong wind events, especially over Northern Germany. This will be important for the adaption of railway infrastructure, as it is especially vulnerable to storm events.