Extreme precipitation in the Greater Alpine Region: a CMIP6 and EURO-CORDEX model view
Assigned Session: FS 3.135: Elevational stratification of climate change: impacts and driving mechanisms in global mountain ecosystems
Abstract ID: 3.13096 | Accepted as Poster | Requested as: Poster | TBA | TBA
Mirsada Cravero (1)
Olivia, Ferguglia (1); Elisa, Palazzi (1); Enrico, Arnone (1)
(1) University of Turin, via Giuria 1, 10125 Torino, IT
Abstract
Under recent climate change, the Greater Alpine Region (GAR) is experiencing an increase in the intensity and frequency of meteoclimatic extreme events. In particular, precipitation extremes are of primary importance, given their role in defining mountain hydrological resources and triggering geo-hydrological hazards, including downstream impacts. In areas with complex orography, it is crucial to assess how such extremes and their temporal changes are stratified with elevation.
In this study, daily precipitation data from ERA5 reanalyses, 29 CMIP6 global climate models and 18 EURO-CORDEX regional climate models were employed to evaluate extreme precipitation through 10 indices, following the ETCCDI definitions. The analysis was performed over the entire GAR, with a specific focus on the Piedmont region for regional models, taking advantage of their higher spatial resolution.
Firstly, the capability of the models to represent precipitation and its extremes in the area was evaluated, particularly considering the effects of spatial resolution. Both the probability distribution of precipitation and its extremes, and the spatial distribution of the indices in the recent past were analysed. A clustering technique was adopted to group models based on the probability distributions of the indices, and a score indicating the similarity between the ERA5 and modeled spatial distribution computed. Eventually, a bias correction method was applied to specific CMIP6 models in order to evaluate a possible improvement in their performance.
Our results highlight a group of models that clearly clustered similarly to ERA5, showing the best scores for both the precipitation distributions and their spatial patterns. The cluster could be further enlarged by adopting the bias corrected models. Since such a cluster improved the model analysis as compared to the ensemble mean, the selected models were used to evaluate projections of precipitation extremes to the end of the century, with particular attention to their elevational dependency.
N/A | ||||||||
|