Ensemble sub-selection

Studies on the impacts of climate change require robust climate change information on a local scale in order to simulate the potential environmental and socio-economic sectoral impacts of climate change. The CH2018 climate scenarios provide this data. For each emission scenario considered, the CH2018 data is based on a more or less large ensemble of bias-adjusted climate model simulations downscaled to the local scale. For each climate model considered, the two daily CH2018 datasets DAILY-LOCAL and DAILY-GRIDDED contain daily time series or daily fields for several climate variables in the period 1981-2099. Since impact models can often only deal with a limited number of climate simulations, a practice-oriented sub-selection of the climate ensemble members is desired by the users of the climate scenarios and was recently developed at MeteoSwiss. It is presented below.

Graph_EnsembleSubselektion_EN
Overview of the impact model chain and the climate model ensemble sub-selection based on the climate change signal.

Sub-selection

Future projections from climate models show how the climate could develop under the influence of certain scenarios of human greenhouse gas emissions. As each model is only a simplified representation of reality in different ways, different models deliver different results even when identical framework conditions are specified (model uncertainty). This uncertainty can be limited and quantified by using larger climate model ensembles. In practice, however, it is often not possible for climate impact modellers to integrate every single climate model ensemble member into the downstream impact models. A sub-selection of case-specific representative ensemble members can help here.

To meet this user need, MeteoSwiss and ETH Zurich have now developed a new method for sub-selecting members from a climate model ensemble and applied it to the DAILY-LOCAL and DAILY-GRIDDED products of the CH2018 climate scenarios (Sikorska-Senoner et al. 2024). The method selects ensemble members based on the ranking of the climate change signal (CCS) calculated for a range of climate indices (figure above). Based on the strength of the CCS, three ensemble members are selected for each application and each emission scenario (Representative Concentration Pathways, RCPs), representing the strongest, the weakest and the medium CCS and recommended to the users of the climate scenarios as ensemble sub-selection if - and only if - a consideration of the complete ensemble is not possible. It is generally recommended that the complete ensemble be analysed.

The sub-selection of the CH2018 climate scenarios is available to users for seven predefined use cases and one expert case. This sub-selection is available for the whole of Switzerland and for the five major regions of Switzerland. The pre-calculated tables and graphics are available in the MeteoSwiss Product Browser as a new product of the CH2018 web atlas, i.e., CH2018: Ensemble sub-selection.

Application cases

Anwendungsfaelle

There are currently seven application cases available to users of the Swiss climate scenarios. There are initially four applications that cover the key messages of CH2018: drought, heatwave, heavy precipitation and snow-scarce winter. These are supplemented by three further applications that illustrate climate change using the most important basic climate variables: temperature, precipitation, and temperature and trecipitation combined. In case of uncertainty regarding the selection of an application case, it is recommended to select the overarching case ‘temperature’, which is based on five temperature indices calculated on an annual and seasonal basis.

Expert case

Expertfall

The expert case is currently only available on request via the CH2018 contact form or e-mail (klimaszenarien@meteoswiss.ch). The personal exchange between the users and MeteoSwiss is necessary in order to optimally define the respective application case and to enable the best possible sub-selection of ensemble members. Feedback from users on the current or possible future application cases for sub-selection is very welcome and is systematically collected in order to expand the available selection of application cases.

User feedback

More information

Sikorska-Senoner, A. E., Rajczak, J., Zappa, M., Kotlarski, S. (2024) User-tailored sub-selection of climate model ensemble members for impact studies, Science of The Total Environment, 952, 175769, https://doi.org/10.1016/j.scitotenv.2024.175769.

CH2018_ensemble_subselection_guide (PDF, 2 MB, 07.10.2024)CH2018 Ensemble Subselection guide (English)

Understanding Climate Change Scenarios

The Climate Scenarios are based on the latest computer simulations. Here you will find more information on how the Climate Scenarios were generated and how to interpret them.

Last modification 28.10.2024

Top of page

Contact

Federal office of meteorology and climatology MeteoSwiss
Operation Center 1
P.O. Box
CH-8058 Zurich-Airport

Contact

Print contact

https://www.nccs.admin.ch/content/nccs/en/home/climate-change-and-impacts/swiss-climate-change-scenarios/ensemble-subselektion.html