Atlantic hurricanes—Testing impacts of local SSTs, ENSO, stratospheric QBO—Implications for global warming

Abstract

Processes affecting hurricane development over the North Atlantic like the El Niño Southern Oscillation (ENSO), the stratospheric Quasi-Biennial Oscillation (QBO) and Sea Surface Temperatures (SSTs) are discussed. Global coupled climate model simulations cannot answer directly the question on enhancement of hurricane activities (or its absence) under increased greenhouse gas concentrations because of their too coarse resolution. Therefore large-scale quantities that affect hurricane formation are investigated in a future warmer climate. More frequent or more intense hurricanes are expected from an increase in the local SST, from more latent heat flux from the ocean to the atmosphere, from more westerly winds in the tropical stratosphere that reduces the occurrence of strong easterly phases of the QBO and from a more moist-unstable stratification of the atmosphere. However, a stronger vertical wind shear similar to the difference between El Niño and La Niña events suggests fewer hurricanes in the northern Atlantic. Also a more dry-stable atmosphere would lead to fewer hurricanes. Of the various forcing factors, the impact of wind shear appears to be more decisive, i.e. with a strong wind shear over the tropical Atlantic like during El Niño events strong hurricanes hardly happen while impacts from SSTs over the tropical Atlantic are less significant. As there are some factors favouring an increase of hurricane activity in a future climate and others favouring a decrease, it remains so far difficult to estimate their joint effect and to suggest any decisive trend. The area of hurricane development is limited among others by an increase of vertical wind shear towards the north and south from a minimum at 5–10°N. This wind shear pattern does not change in a future climate and has the potential of superseding impacts from ocean warming. A need of very long time-series for obtaining robust results becomes obvious. Here at least 50 years of data were used.