Non-Hydrostatic Climate Modelling

Those characteristics of climate which have direct impact on human society and ecosystems are often related to small scale phenomena. In this context, particularly hydrological features and extreme climate events are of highest relevance and considerable demand for highly resolved climate analyses and projections exists on the part of climate impact research. Current regional climate models are generally operated at 20-50 km horizontal grid resolution, a scale where many regional features can be well resolved but too coarse for proper representation of important features such as severe precipitation events and their spatial distribution or diurnal cycle of precipitation, particularly over complex terrain. Climate simulations on finer, local scales (grid spacing less than 10 km) are conceptually very promising to resolve or at least mitigate these problems, but many methodological challenges hamper the reliable application of local scale climate models (LCMs), such as shortcomings in physical parameterisations also in climate models principally capable to resolve local scales (non-hydrostatic climate models, NHCMs). Significant obstacles exist also to adequate spatial validation of high-resolution model results due to poor availability of suitably dense observations. Up to date, the uncertainty range of LCM simulations has thus not been systematically investigated. In the framework of the proposed project Non-Hydrostatic Climate Modelling (NHCM-1), systematic sensitivity studies will be conducted with focus on two test areas in Austria (an alpine region "Hohe Tauern" and a hilly area "Oststeiermark") at 10 km, 3 km, and 1 km horizontal grid resolution with two LCMs (German CLM and U.S. MM5 model) aiming at elucidating the resolution-dependant performance of various model components, particularly physical parameterisation. In its investigations in one of the test areas, the project will be backed by a novel, high-resolution dataset from 150 climate stations ("WegenerNet") available for model evaluation purposes. As part of the project, an international intercomparison campaign for LCMs will be conducted. This will enable for the first time to quantitatively investigate the uncertainty range associated with LCMs using probabilistic techniques. The overall goal of the project is to examine yet unclear potentials and shortcomings of current LCMs and to provide estimates on uncertainties related to these highly relevant tools. The broader and longer-term background aim is to build a better basis for LCM development and of providing climate impact research with more suitable, more reliable, more comprehensive, and more consistent data than available today.

The FWF project "Non-Hydrostatic Climate Modelling" (NHCM-1) did the first steps towards the next but one generation of high resolution climate scenarios, which will be based on convection resolving dynamical models that are capable to explicitly simulate thunderstorms and many local climate features. Such small scale processes are the key to understand the impacts of climate change on society and ecosystems, but they can so far only be predicted with weather prediction models a few days into the future, or projected into longer future with indirect empirical methods. Convection-resolving climate models are expected to give a boost to the quality of high resolution climate scenarios. In NHCM-1, convection-resolving climate models passed their first test towards climate application. It could be demonstrated that they improve the quality of climate simulations on the small scale and that they are of comparable quality with their established coarser scale predecessors at larger scales. While the technical development of high performance computing systems promises to enable century-long convection-resolving climate simulations within the next five years, the results of NHCM-1 motivate for the next phase of preparation, which will transfer the lessons learned to decadal timescales and set up century-long climate simulations, that are needed for climate impact research.