Multidecadal variability of the ENSO teleconnection to Europe in early-winter and implications for seasonal forecasting

El Niño-Southern Oscillation (ENSO) is the leading mode of global climate variability. Through its associated teleconnections, ENSO can impact the climate of numerous regions worldwide at seasonal timescales, highlighting its role as the main source of seasonal predictability. Numerous studies have demonstrated a significant influence of ENSO on the climate of the Euro-Atlantic sector, but the impacts and mechanisms of the teleconnection in early-winter (November-December) remain unclear. Besides, in early-winter, ENSO teleconnections involve tropospheric pathways, which may change in response to different background states of the ocean. Thus, a crucial research question to address is whether the early-winter teleconnection to the Euro-Atlantic sector has changed under the different background states of sea surface temperature (SST) over the Pacific Ocean. 

 

This work aims to analyse the ENSO early-winter teleconnection to the Euro-Atlantic sector from a nonstationary perspective. Specifically, the teleconnection is analysed under different background states of SST over the Pacific Ocean, related to changes in the phase of the Pacific Decadal Oscillation (PDO). Using observational and reanalysis datasets for the period 1950-2022, results reveal that the tropospheric pathways of the teleconnection change under the different Pacific SST background states, leading to distinct responses of the North Atlantic atmospheric circulation to ENSO. We also confirm that these distinct responses in the North Atlantic entail significantly different impacts of ENSO on the surface climate across Europe, particularly on surface air temperature. Furthermore, the teleconnection is analysed in the SEAS5 state-of-the-art dynamical seasonal prediction model. The analysis within the model is also conducted from a nonstationary perspective, and aims to determine whether the model successfully reproduces a shift in the teleconnection in the late 1990s identified in reanalysis and observations. Results show that the model accurately captures the spatial pattern of the teleconnection impacts across Europe after the late 1990s, but not before. In turn, significant changes in the skill of seasonal forecasts are observed between before and after the late 1990s. However, skill after the late 1990s is just moderate due to a significant underestimation of the teleconnection impacts. 

 

The results of this study shed light on the nonstationary behaviour of the early-winter teleconnection to the Euro-Atlantic sector and have important implications on seasonal predictability in Europe. Particularly, the nonstationarity of the teleconnection gives rise to the emergence of windows of opportunity for seasonal forecasting, in which forecast skill may be greater than initially expected from a stationary analysis.