Assessment of a full-field initialised decadal climate prediction system with the CMIP6 version of EC-Earth
Roberto Bilbao,Simon Wild,Pablo Ortega,Juan Acosta-Navarro,Thomas Arsouze,Pierre-Antoine Bretonnière,Louis-Philippe Caron,Miguel Castrillo,Rubén Cruz-García,Ivana Cvijanovic,Francisco Javier Doblas-Reyes,Markus Donat,Emanuel Dutra,Pablo Echevarría,An-Chi Ho,Saskia Loosveldt-Tomas,Eduardo Moreno-Chamarro,Núria Pérez-Zanon,Arthur Ramos,Yohan Ruprich-Robert,Valentina Sicardi,Etienne Tourigny,Javier Vegas-Regidor
DOI: https://doi.org/10.5194/esd-2020-66
2020-09-10
Abstract:Abstract. In this paper we present and evaluate the skill of the EC-Earth3.3 decadal prediction system contributing to the Decadal Climate Prediction Project - Component A (DCPP-A). This prediction system is capable of skilfully simulating past global mean surface temperature variations at interannual and decadal forecast times as well as the local surface temperature in regions such as the Tropical Atlantic, the Indian Ocean and most of the continental areas, although most of the skill comes from the representation of the externally forced trends. A benefit of initialisation in the predictive skill is evident in some areas of the Tropical Pacific and North Atlantic Oceans in the first forecast years, an added value that gets mostly confined to the south-east Tropical Pacific and the eastern Subpolar North Atlantic at the longest forecast times (6–10 years). The central Subpolar North Atlantic shows poor predictive skill and a detrimental effect of the initialisation due to the occurrence of an initialisation shock, itself related to a collapse in Labrador Sea convection by the third forecast year that leads to a rapid weakening of the Atlantic Meridional Overturning Circulation (AMOC) and excessive local sea ice growth. The shutdown in Labrador Sea convection responds to a gradual increase in the local density stratification in the first years of the forecast, ultimately related to the different paces at which surface and subsurface temperature and salinity drift towards their preferred mean state. This transition happens rapidly in the surface and more slowly in the subsurface, where, by the tenth forecast year, the model is still far from the typical mean states in the corresponding ensemble of historical simulations with EC-Earth3. Our study thus highlights the importance of the Labrador Sea for initialisation, the relevance of reducing model bias by model tuning or, preferably, model improvement when using full-field initialisation, and the need to identify optimal initialisation strategies.