The manuscript has been greatly improved since the last revision. Based on the reports from the two referees, the authors are expected to make minor revisions to address their comments before the manuscript can be accepted for publication. In particular, Referee #2 pointed out a previous study by Robbins et al. (2011; Changepoints in the North Atlantic Tropical Cyclone Record, Journal of the American Statistical Association) that appears to be highly relevant. The authors are encouraged to compare their results to that work.

The manuscript will be ready for publication after the following technical corrections.

* Please give the source (IBTrACS? HURDAT?) and relevant reference and/or website of the Atlantic tropical cyclone dataset used, probably in the first paragraph of Section 5 (around line 585).
* The manuscript has used “hurricanes” and “tropical storms” loosely, although the two terms are not interchangeable and not all tropical cyclones reach hurricane intensity. The data presented in the manuscript are apparently of tropical cyclones. Therefore I suggest replacing hurricane with tropical cyclone throughout the manuscript.
* Scale Fig. 1 properly so that it does not appear stretched.
* Change the x-axis tick label in Fig. 3 from 0-60 to 1950-2008.