A bias-corrected CMIP6 climate projection dataset for the Danube Basin

As the climate keeps changing, it is essential to develop effective solutions to support adaptation and mitigation to the observed and projected climate change. It is imperative that our long term strategies rely on in-depth understanding of the prospective range for expected climate scenarios.

The Coupled Model Intercomparison Project (CMIP) framework was developed in response to this need, integrating an extensive collection of global climatic models developed by different modelling groups, each of them with a focus on specific physical and atmospherical processes. These models generate valuable future climate projections that could be integrated in decision-support solutions to foster the delivery of key ecosystem services in water-dependent habitats and improve their resilience to climate change.

This study focuses on the Danube Basin, a representative climate change hotspot at European level, which integrates the joint impacts of a wide range of socio-economic factors (i.e., urbanization, land use change). This region is subject to the ecosystemic and biodiversity threats arising from increased temperatures, decreased precipitation and reduced river flow in both present and future climate, challenging the management of restoration actions. This study is aimed to provide an improved understanding of the expected future climate change signals for the timeframe 2015 - 2100 in the top priority CMIP6 scenarios (i.e., SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) at the Danube Basin scale, using bias-corrected data. The model outputs are calibrated on the historical period 1961–2014 with E-OBS gridded dataset at 0.1° spatial resolution.

Our approach includes a performance comparison of several bias-correction methods (e.g., univariate - quantile delta mapping, multivariate – R2D2, MBCn) used within a spatial disaggregation algorithm, to minimize the uncertainty induced by systematic errors in the projection data. The bias-corrected climate projections of air temperature and precipitation will feed the Restore4Life decision-support system for wetland restauration.

Acknowledgements

This research received funds from the project “Restoration of wetland complexes as life supporting systems in the Danube Basin (Restore4Life)” funded by the European Union Horizon Europe programme, under Grant agreement n° 101112736.