Translating Climate Variables into Everday Language: Sunshine Hour Projections from CORDEX Regional Climate Models

Sunshine hours are a widely used climate variable with relevance for ecosystems, agriculture, solar energy, and public health that are well understood by the average citizen. However, this variable is not routinely provided as an output variable by regional climate models (RCMs), including those in the CORDEX framework.

Here we develop and calibrate a practical two-stage method to estimate daily sunshine hours from daily surface downwelling shortwave radiation (rsds). The approach is calibrated using Danish observational data with month-specific empirical relationships based on the normalised radiation ratio x = R / Rₐ, where R is daily global radiation at the surface and Rₐ is daily extraterrestrial radiation from FAO-56. We describe how an empirical relationship has been derived from this approach using model selection techniques across a range of possible configurations. Model performance statistics are generally favourable and show a good fit to the observations.

We apply the relationship to outputs from the CORDEX5 and CORDEX6 family of regional climate models to generate projections of sunshine hours with daily resolution. The variable is then used to generate a new set of indicators for inclusion in Klimaatlas, the Danish Climate Atlas, including seasonally resolved mean number of sunshine hours: we find a reduction in winter sunshine hours of up to 13% projected for 2071–2100 under the RCP8.5 emission scenario. Other novel indicators, such as the frequency of “sunshine droughts” – prolonged periods with no or few sunshine hours – have also been developed. Finally, we perform a comparison of the results obtained between CORDEX5 and CORDEX6 models, with a particular focus on understanding differences between these generations of models, and how they relate to the aerosol scenarios used in each model.