Ref: PID2021.126401OB.I00- Funding agency: Ministerio de Ciencia e Innovación. MICINN
- Realization: 01/09/2022 – 31/08/2026
- PIs: Sonia Raquel Gámiz Fortis, María Jesús Esteban Parra
- Researchers: Yolanda Castro Díez (University of Granada), Matilde García Valdecasas Ojeda (University of Granada), Juan José Rosa Cánovas (University of Granada), Patricio Yeste Donaire (University of Granada), Emilio Romero Jiménez (University of Granada), Joaquim Pinto (Karlsruhe Institute of Technology of Germany), Marcus Breil (Karlsruhe Institute of Technology of Germany)
Abstract
Reliable near-term climate predictions beyond the seasonal-to-interannual time scale, is one of the great current challenges for the scientific community dedicated to the study of climate. In this context, the objective of this project is to study the impact of near-term high-resolution climate change on extreme events in the Iberian Peninsula (IP) through dynamical downscaling of decadal predictions from the Coupled Model Intercomparison Project (CMIP) and the improvement by including initial soil conditions. Climate model initialization offers the potential to predict internal variability and externally forced climate change on decadal scale. The impact of the initialization of the oceanic state on the decadal predictability of the climate is beginning to be broadly investigated, and is taken into account in the decadal predictions of the CMIP5/6. Furthermore, the persistence of the climate induced by the memory of soil moisture and the coupling between the vegetation and the atmosphere can be amplified and extended through feedback mechanisms, being an important source of predictability on decadal scale.
Earth surface processes influence the boundary layer and the atmosphere and affect extreme events. Recent works have highlighted the importance of interactions between soil moisture and climate in Europe, particularly relevant in areas with transitional climate regimes, such as the IP. For regional applications, dynamical downscaling has shown an added value for climate assessment, for example, for the drought in the IP, as well as for climate projections. One of the specific objectives of this project is to analyze whether this added value can also be found in the regionalized decadal predictions of extreme indices and related variables, and if these can be improved by introducing initial soil conditions. For this end, a dynamical downscaling will be carried out using the WRF model, obtaining retrospective predictions for four decades initialized in the years 1988, 1990, 1992 and 1994, to evaluate the predictive skill of the regionalization.
Additionally, regionalized predictions will be made for the decades 2016-2025 and 2020-2029. The outputs of this downscaling will be used as input variables in the VIC hydrological model to obtain decadal predictions of streamflow and hydrological drought at high resolution in the main IP basins, which are under a strong overexploitation in the current climate change context.
The strong socio-economic impacts derived from the occurrence of extreme climatic events such as heat waves, torrential precipitation, droughts or wind storms, emphasize the need to make their predictions on decadal scale.