The Seasonal Effects of ENSO on Atmospheric Conditions Associated with European Precipitation: Model Simulations of Seasonal Teleconnections

Abstract

Abstract The seasonal upper-tropospheric teleconnection between ENSO and the North Atlantic/European sector is explored through a series of model experiments. A barotropic vorticity equation model is linearized about climatological conditions for each season of the year, and divergence forcing is applied over the equatorial Pacific to mimic El Niño–related convective activity. During boreal fall, winter, and spring, this forcing similarly excites a northeastward-propagating stationary barotropic Rossby wave train that extends across the North Atlantic to the European coast. Strong anomalies develop over the British Isles in the vicinity of the North Atlantic jet exit. Solutions during boreal summer produce no clear wave train; however, evidence exists for a North Atlantic response because of both eastward- and westward-propagating signals. These direct responses over the Atlantic and Europe are qualitatively similar to observed ENSO-associated anomalies during boreal spring and fall, but differ structurally during summer and winter. Further experiments with the vorticity equation model using full Rossby wave source forcing, which included vorticity advection, increase the amplitude of the response over Europe during some seasons; however, structural differences persist. Finally, experiments with the Community Atmosphere Model (CAM), version 4, reveal that the basic northeastward-propagating response is modulated by downstream feedbacks. These changes are most profound during boreal winter and engender an arching wave train pattern that, matching observations, reflects off the jet over North America, propagates southeastward over the North Atlantic, and fails to reach the European coast. Overall, the simulations with CAM correctly depict observed seasonal changes in the magnitude of the ENSO–North Atlantic/European teleconnection by producing a strong fall and winter response but a weaker spring and summer response. The CAM experiments also indicate that the seasonal response is not dependent on antecedent conditions; however, CAM simulations fail to project the upper-tropospheric anomalies appropriately to the lower troposphere.