New research, published in Climate Dynamics and involving members of the MRI Elevation-Dependent Climate Change Working Group, reveals that climate model biases, particularly in temperature and precipitation, increase with elevation in the European Alps. The study also stresses the importance of dataset selection for accurate climate evaluations.
A comprehensive evaluation was conducted on climate model data over the European Alps using the EURO-CORDEX ensemble of regional climate models and the CORDEX-Adjust dataset. The study examined biases—discrepancies between climate models and observed data—across different elevations and climate indices, including both average and extreme weather conditions.
Key Insights
- Biases in seasonal temperature, seasonal precipitation, and wet-day frequency were found to increase with elevation.
- Temporal trends in biases were noted, driven by trend differences between the regional climate models (RCMs) and observational data, but could be addressed by detrending both sets.
- The selection of reference observational datasets was found to have a greater impact on bias adjustment outcomes than the choice of bias adjustment methods.
“The strong elevation dependency we found highlights the need to look beyond spatial averages in regions with complex terrain, and cautions against generalising results across elevation.” – Michael Matiu – main author, Postdoctoral Researcher at the University of Trento
“Addressing elevation-dependent model biases is crucial for enhancing the reliability of future climate projections in mountainous areas.” – Sven Kotlarski, co-author and Head Climate Evolution at the Federal Office of Meteorology and Climatology, MeteoSwiss
In summary, the study reveals the challenges encountered when using climate model data in mountainous environments. The latter are highly sensitive to climatic changes, but describing the complexity of mountain weather and climate, including their past and future changes, requires well-designed and high-quality observational networks and climate models capable of resolving the governing processes. To guarantee both, continuous investments in observational and atmospheric modelling capabilities and a better integration of observations and models are required.
Citation
Matiu, M., Napoli, A., Kotlarski, S. et al. Elevation-dependent biases of raw and bias-adjusted EURO-CORDEX regional climate models in the European Alps. Clim Dyn (2024). https://doi.org/10.1007/s00382-024-07376-y
Cover image by: Sven Kotlarski, MeteoSwiss
