Linking local climate scenarios to global warming levels: applicability, prospects and uncertainties

Abstract

Global warming levels (GWLs) are increasingly becoming a central concept in climate change studies. In recent years, their integrative quality for climate change impact analysis has been demonstrated, and methodological advancements have helped to compensate for some inherent shortfalls of the concept. However, their applicability at the regional and local level is debatable, and no study to date has examined the possibility of linking local climate scenarios to GWLs. For the case of Austria, we evaluate the relation between global and regional warming patterns, and whether version changes of global climate models could be incorporated into local climate scenarios by means of the GWLs concept, without updating the actual data. We apply the time sampling approach, where GWLs are determined as periods when global mean temperature anomalies cross a certain threshold. GWL periods are sampled both from the global models in the background of the local climate scenarios (CMIP5), and from an equivalent ensemble of newer-generation climate models (CMIP6). Uncertainties resulting from sampling GWLs from different global climate model ensembles are examined, and prospects for local climate change impact assessments are discussed. Accounting for updated global climate model versions might be useful when the changes at certain GWLs are related to fixed reference periods, but temperature increments between GWLs remain relatively constant across model versions, even on the regional level. The study bridges a significant gap to link local climate projections to GWLs. Climate change impacts assessments that build on those datasets can benefit from the integrative character of GWLs, making studies comparable across multiple disciplines and model versions, and thus fostering a way to communicate local climate change impacts more comprehensible.