Deterministic and Stochastic Aspects of El Niño’s Impact on Atlantic Tropical Cyclones

Due to a short data record that is complicated by Atlantic sea surface temperature (SST) variability, observationally based studies produce conflicting claims regarding how the location of tropical Pacific SST warming during El Niño, which varies from the East Pacific (EP) to Central Pacific (CP), impacts Atlantic tropical cyclone (TC) activity. We attempt to resolve this issue using large ensembles of simulations with a tropical-extratropical channel model forced by different boundary conditions to isolate the El Niño influence from the confounding effects of Atlantic SST and stochastic atmospheric variability. We find statistically significant reductions in seasonal Atlantic TC activity in response to both CP and EP El Niño. Despite large differences in the location and magnitude of tropical Pacific warming during the two El Niño types, moderate warming east of the Pacific warm pool is common to both and satisfies the SST threshold for deep convection, leading to tropical Atlantic vertical wind shear enhancements that suppress Atlantic TCs. CP El Niño, which has occurred more frequently and with increasing intensity in recent decades, is 50% and 150% as effective as East Pacific El Niño at suppressing Atlantic TCs for similar percentile-based (top 90^th) and absolute (2.25 K) intensities of warming, respectively. We also note the importance of stochastic atmospheric variability in contributing to variability in TC activity, as there is a considerable range between the maximum and minimum ACE in the ensemble of experiments for each El Niño type, despite having fixed surface and lateral boundary conditions. This large stochastic variability may explain some of the inconsistencies in observational analyses, as it is clear that analysis on an insufficient sample size could easily produce a misleading result that CP Niño drives an increase or no significant change in Atlantic TC activity.