Abstract
The response of tropical cyclone activity in the predicted warmer future climate is still a topic of scientific research. However, by the nonlocality hypothesis, tropical cyclone activity depends rather on relative than absolute sea surface temperature (SST). This hypothesis is investigated through idealized experiments performed with the global atmospheric climate model PlaSim. The model includes a prescribed SST and adopts a spectral resolution of T170 for resolving tropical cyclones. An idealized land-sea configuration with two oceans and two continents has been used to study the nonlocality mechanism. The sensitivity experiments WARMBASIN and COLDBASIN include a positive and negative SST anomaly of 2.5 K in the northeastern basin. The tropical cyclone activity in the control run is similar in all four ocean basins while experiment WARMBASIN simulates a striking local increase of tropical cyclones. However, they nearly vanish in the other three ocean basins. The response is weaker and in the opposite way in COLDBASIN with a local reduction and a nonlocal increase. Analysis of well-known cyclogenesis indices shows that nonlocality could be explained by the vorticity, relative humidity, and upper tropospheric temperature changes in experiment WARMBASIN while only vorticity is in agreement with nonlocality in experiment COLDBASIN. The vorticity anomalies determine the presence or absence of a steady state large-scale low at the location where tropical cyclones favorably form. This low is modified by an SST-induced change of a Walker-like planetary circulation.