Abstract:
|
This talk analyzes the North Atlantic Basin's tropical cyclone record since 1850 for changes in storm counts and strengths. Based mainly on climate model simulations, many climate scientists believe that our current era of warming should give rise to slightly stronger storms, but not significantly more storms. This talk assesses this hypothesis with a changepoint analysis of the North Atlantic Basin's record. The methods employ a penalized likelihood with a minimum description length penalty. In the end, we find the exact opposite of what the climate scientists claim: that we entered into an era of increased storm frequencies circa 1995 that we are still in, but that the storm strengths are not increasing. Analysis of annual accumulated cyclone energies also support the circa 1995 changepoint to increased tropical cyclone activity.
|