Ludwig-Maximilians-Universität München (Germany)
Local Project ID:
HPC Platform used:
SuperMUC of LRZ
Predicting weather and climate and its impacts on the environment, including hazards such as floods, droughts and landslides, continues to be one of the main challenges of the 21st century – in particular for the European region as it is exposed to intense Atlantic synoptic perturbations. Scientists performed for the first time long climate simulations over the European domain at a very fine cloud-permitting resolution of about 4 km with explicitly resolved convection and a sharp representation of orography, thanks to the possibility of running very computationally and data storage demanding simulations on SuperMUC.
Predicting weather and climate and its impacts on the environment, including hazards such as floods, droughts and landslides, continues to be one of the main challenges of the 21st century with significant societal and economic implications. At the heart of this challenge, lies the ability to have easy access to hydrometeorological data, to share predictive models, and to facilitate the access to High Performance Computing facilities supporting leading edge hydro-meteorological simulations. Advances in the science and observation of climate change are providing a clearer understanding of the inherent variability of Earth’s climate system and its likely response to human and natural influences.
Numerical simulations at the global scale of future climate in different emission scenarios, produced in the framework of the fifth climate model intercomparison project (Coupled Model Intercomparison Project Phase 5, CMIP5), have been typically obtained using hydrostatic global climate models (GCM) and are available only at quite coarse spatial resolutions which do not allow an accurate representation of intense precipitation events over complex topography areas such as Europe. An improved generation of regional climate models (RCM) is being developed and applied in the framework of the CORDEX (COordinated Regional climate Downscaling Experiment) initiative, with the aim to produce regional climate change projections worldwide for input into impact and adaptation studies, considering multiple forcing GCMs from the CMIP5 archive. Recent results for the European branch of the CORDEX initiative (Kotlarski et al. 2014) confirm, with simulations on grid-resolutions up to about 12 km (0.11°), the ability of RCMs to capture the basic features of the European climate for the period 1989-2008, but also show non-negligible deficiencies of the simulations concerning selected metrics, certain regions and seasons: for example seasonally and regionally averaged temperature biases are mostly smaller than 1.5 °C, while precipitation biases are in the ±40% range.
The EXPRESS-Hydro approach and supercomputing power role
This reality, together with the fact that the European region is exposed to intense Atlantic synoptic perturbations and potentially their north shift due to climate change, make it critical to carefully understand all model sensitivities before drawing conclusions and assessing climate change impacts from the numerical simulations outputs.
Starting from these ideas, the EXtreme PREcipitation and Hydrological climate Scenario Simulations (EXPRESS-Hydro) project moves one step further, performing very high-resolution regional dynamical downscaling of historical climate scenarios produced by the ERA-Interim reanalysis using the state-of-the-art non-hydrostatic Weather Research and Forecasting (WRF) regional climate model.
To the best knowledge of the current literature, EXPRESS-Hydro performed for the first time long climate simulations (1979-2008, Pieri et al. 2015) over the European domain (Inner European Region, IER, Figure 1) at a very fine cloud-permitting resolution of about 4 km (0.037°) with explicitly resolved convection and a sharp representation of orography, thanks to the possibility of running very computationally and data storage demanding simulations (about 700 Tb of outputs produced) on the state-of-the-art SuperMUC Petascale System, one of the fastest supercomputers in the world.
The EXPRESS-Hydro results
EXPRESS-Hydro explored the WRF regional climate model capability in reproducing observed precipitation extremes, over Europe and in detail over the Greater Alpine Region, comparing simulations results with available high-resolution gridded observational data sets. Overall our results indicate that increased resolution with explicitly resolved convection helps to obtain a closer representation of the precipitation field, reducing the overestimation of precipitation (about 25% on average over the European domain) and allowing to better reproduce the distribution and the statistics of the rainfall rate, particularly over the Alps (Figure 2).
EXPRESS-Hydro complemented the present climate with a future scenario projection at 0.11°, in which the WRF model is forced with CMIP5 output from the EC-Earth global climate model, with RCP4.5 greenhouse gas and aerosol concentrations. Furthermore part of the EXPRESS-Hydro team undertook also high-resolution simulations, using again SuperMUC Petascale System, down to the cloud-resolving range (200 m), for high impact weather events (HIWE) occurring in the Mediterranean region, such as the recent Genoa 2011 and Genoa 2014 flash-flood events (Fiori et al. 2014, Fiori et al. 2015, Hally et al. 2015).
The post-processed datasets produced in this project will be used as inputs for regional impact studies, in particular forcing hydrological models and ecosystem dynamics models over selected study areas. The data will be distributed in the framework of major international projects such as DRIHM, and the Italian Project of Interest NextData.
Fiori, E., Comellas, A., Molini, L., Rebora, N., Siccardi, F., Gochis, D. J., ... & Parodi, A. (2014). Analysis and hindcast simulations of an extreme rainfall event in the Mediterranean area: The Genoa 2011 case. Atmospheric Research, 138, 13-29.
Fiori, E., Ferraris, L., Molini, L., Siccardi, F., Kranzlmueller, D, Parodi, A. (2015). Morphology of the triggering and evolution of a micro-α convective system in the Mediterranean Sea. Quarterly Journal of the Royal Meteorological Society, submitted.
Hally, A., Caumont, O., Garrote, L., Richard, E., Weerts, A., Delogu, F., ... & Clematis, A. (2015). Hydrometeorological multi-model ensemble simulations of the 4 November 2011 flash flood event in Genoa, Italy, in the framework of the DRIHM project. Natural Hazards and Earth System Science, 15(3), 537-555.
Pieri, A. B., Hardenberg, J. V., Parodi, A., & Provenzale, A. (2015). Sensitivity of precipitation statistics to resolution, microphysics and convective parameterization: a case study with the high-resolution WRF climate model over Europe. Journal of Hydrometeorology, (2015).
Research team and scientific contact:
Jost von Hardenberg1, Antonio Parodi2, Alexandre Pieri1, Antonello Provenzale1, and Dieter Kranzlmueller
1 Institute of Atmospheric Sciences and Climate - National Research Council (ISAC-CNR) and
2 CIMA Research Foundation, Italy
Prof. Dr. Dieter Kranzlmüller
Institut für Informatik, Ludwig-Maximilians-Universität München
Oettingenstraße 67, D-80538 München/Germany
e-mail: Kranzlmueller (@) ifi.lmu.de