Relationship of the Atlantic Meridional Mode with Atlantic hurricane tracks

How the Atlantic Meridional Mode (AMM) influences Atlantic hurricane track types is explored here. A clustering technique is applied to tropical cyclones in Atlantic best-track dataset in the period 1950-2005, with at least tropical storm intensity. The clustering technique consists of building a mixture of polynomial regression models (i.e. curves), which are used to fit the geographical ``shape'' of the trajectories. This technique was applied previously to the western and eastern North Pacific tropical cyclones.

Two well-separated track types (clusters) are identified. Correlations of the number of tropical cyclones and ACE in each cluster type show significant relationship of the clusters with AMM and ENSO (El Nino-Southern Oscillation). The two track types are very closely aligned with the track types defined by the phase of the AMM, and the mean values of the AMM are significantly separated within each cluster. Sea surface temperature (SST) anomaly composites show the influence of various modes in the clusters. Besides AMM and ENSO, the Atlantic Multi-decadal Oscillation (AMO) pattern is present in the SST composites. Composites around the clusters suggest that Atlantic tropical cyclone tracks are shifted according to the phase of the AMM. On interannual time-scales the correlations with AMM are stronger than with ENSO indices.