Climate models and scenarios

Climate models are extensive computer programmes that rely on certain assumptions when calculating the future development of the climate. These assumptions are combined into greenhouse gas scenarios, resulting in climate projections. Projections are not forecasts or predictions ("this will happen"), but rather "if-then" statements: if this scenario occurs then this could happen... They form the basis for assessing the risks and opportunities of future climate change and for developing adaptation measures.

Greenhouse gas scenarios play an important role for the calculation of potential climate changes. They are based on a number of assumptions that concern different global trends such as population growth, economic and social developments, technological changes, consumption of resources and environmental management. On the basis of these assumptions it is possible to comment on how the emission of greenhouse gases (emissions scenarios) and, as a result, the concentration of greenhouse gases in the atmosphere (concentration scenarios) will develop. The climate projections for the fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) (2007) were based on the SRES emission scenarios.

New scenarios have been developed for the IPCC's 2020 Sixth Assessment Report. They consist of two complementary components: the Shared Socioeconomic Pathways (SSPs), which describe possible future socioeconomic developments, and the Representative Concentration Pathways (RCPs), which depict possible concentration pathways of atmospheric greenhouse gases and thus possible future climate developments. In principle, more than one SSP can lead to a given RCP. They are calculated by coupled consistent models (Earth System Models and Integrated Assessment Models, or IAMs) that represent the climate system, ecosystems, economy, and land use. Scenarios derived from such integrated models (SSPx-y scenarios) thus encompass nearly all factors influencing radiative forcing, i.e., the "additional/increased" energy supplied to the Earth by human activities such as greenhouse gas emissions and land cover change, as well as albedo (e.g., desertification) and aerosol changes, and allow plausible projections of potential climate change. The characteristics of five currently used scenarios with selected socioeconomic development pathways and the resulting concentration pathways are described in “recommendations for characterizing selected climate scenarios” (only in German).

Climate models on a global scale often have a very coarse resolution (approx. 100*100 km) and are therefore spatially quite imprecise. In order to better align the climate projections with the specific circumstances in Germany, the models are specified using regional climate models up to a resolution of approx. 25* 25 km and less locally. There are two different methods for this regionalisation of global climate models: dynamic and statistical methods – these also include so-called downscaling approaches, which can be used, for example, to further downscale the results of statistical models. An example of dynamic models is REMO and for statistical models WETTREG.

Dynamic regional models simulate different climate parameters by solving non-linear equations for regional sections of the global model. These simulations are based on input data provided by the global model. By focusing on a smaller section, a greater spatial and temporal resolution of climate projection can be achieved with the same computer capacity.

Statistical regional models use knowledge from climate observations. For example, WETTREG draws on the observed statistical relationships between large-scale circulation patterns in the atmosphere and local and regional weather events. Based on the large-scale circulation patterns of the global models, possible regional developments can be calculated with the regional models.

The results of regional model runs are called (regional) climate projections, similar to the global climate models.

In order to analyse and assess the future impacts of climate change in Germany within the framework of climate impact and vulnerability studies, socio-economic scenarios are needed in addition to climate projections. The socio-economic scenarios are combined with climate projections to depict the socio-economic structures that could be affected by climate change in the future. Socio-economic structures influence where people and systems, for example infrastructures, forests, industrial areas, are exposed to climate change and how strongly they can be influenced by climate change. For example, the socio-economic scenarios provide indications of where and how older people who suffer particularly from heat will live in the future. With this information, preventive measures, for example in the health sector, can be planned more specifically. For this reason, three socio-economic scenarios at national and regional level were developed in 2019 on behalf of the Federal Environment Agency, which were also used to calculate local land use scenarios up to 2045.

Future projections and scenarios always contain uncertainties. Even if climate models were able to depict the physical and chemical relationships in the atmosphere very accurately, uncertainties would still remain. Uncertainties arise from limited system knowledge, e.g. about the formation of thunderclouds or land-ocean-atmosphere coupling, limited data and limited computer capacities. Moreover, socio-economic future scenarios on future population growth, economic and social development, technological changes, resource consumption and environmental management are always associated with unforeseeable developments, such as technological leaps, economic crises or social disasters such as wars. In addition, climate projections become more and more uncertain the further into the future the point in time under consideration and the smaller the region under consideration is.

In order to be able to deal with this uncertainty, it is important to know how wide the range of possibilities is. This range or bandwidth can be estimated with the help of ensembles formed from a variety of climate model results.