Skilful predictions of the Summer North Atlantic Oscillation

Abstract

Abstract The Summer North Atlantic Oscillation is the primary mode of atmospheric variability in the North Atlantic region and has a significant influence on regional European, North American and Asian summer climate. However, current dynamical seasonal prediction systems show no significant Summer North Atlantic Oscillation prediction skill, leaving society ill-prepared for extreme summers. Here we show an unexpected role for the stratosphere in driving the Summer North Atlantic Oscillation in both observations and climate prediction systems. The anomalous strength of the lower stratosphere polar vortex in late spring is found to propagate downwards and influence the Summer North Atlantic Oscillation. Windows of opportunity are identified for useful levels of Summer North Atlantic Oscillation prediction skill, both in the 50% of years when the late spring polar vortex is anomalously strong/weak and possibly earlier if a sudden stratospheric warming event occurs in late winter. However, we show that model dynamical signals are spuriously weak, requiring large ensembles to obtain robust signals and we identify a summer ‘signal-to-noise paradox’ as found in winter atmospheric circulation Our results open possibilities for a range of new summer climate services, including for agriculture, water management and health sectors.