Future change of North Atlantic tropical cyclone tracks: projection by a 20-km-mesh global climate model

Possible future change in tropical cyclone (TC) activity over the North Atlantic (NA) was investigated by comparison of 25-year simulations of the present-day climate and future change scenario (A1B) using a 20-km-mesh version of the Meteorological Research Institute and Japan Meteorological Agency atmospheric general circulation model. The present-day simulation reproduces many essential features of observed climatology and interannual variability in TC frequency of occurrence and tracks over the North Atlantic (NA). For the future projection, the sea surface temperature (SST) used to drive the model includes a trend projected by IPCC AR4-multi-model ensemble and year-to-year variations derived from the present-day climate. A major finding is that the model suggests total TC counts in NA will not change significantly, but the frequency of TC occurrence will decrease in the tropical western NA (WNA) and increases in the tropical eastern NA (ENA) and northwestern NA (NWNA). The projected change in TC tracks suggest a reduced probability of TC landfall over southeast US and and increased influences of TC on northeast US. The projected track changes are not due to changes in large-scale steering flows, but due to alteration in TC genesis locations. An analysis of genesis potential index (GPI) suggests that the increase in TC genesis in the ENA arises from increasing upward motion and convective available potential energy due to local ocean surface warming. In contrast, the reduction of TC genesis in the WNA is attributed to decreases in midtropospheric relative humidity and upward motion caused by remotely forced anomalous descent. This indicates that impact of remote dynamical forcing seems to be larger than local thermo-dynamical forcing in the WNA. The increase in the NWNA is attributed to reduced vertical shear of horizontal wind and most significant local ocean surface warming in the NA. These TC changes appear to be most sensitive to the future change of spatial distribution of rising SST. Since majority of IPCC AR4 models project larger increase in SST in the ENA than in the WNA, the eastward shift in TC genesis is probably robust.