Thank you for your revised version. We now have one additional review that requires your attention. In addition to the reviewers comments, I think it would be good to add a bit more insight as to why you think the trends differ between Stroeve et al., 2014 and this study. The time-periods are slightly different and could be made consistent when discussing how they differ. It would be interesting to understand why the melt onset trends differ between the two studies in regions such as the Beaufort/Chukchi and East Siberian seas. Adding a discussion on the mean melt onset dates between the two studies may also be insightful, i.e. how much earlier/later do your dates start relative to Stroeve et al? What are the implications for the ice-albedo feedback? Note the additional review still rates the significance (impact) of the study as "fair". Thus, more discussion on the importance and the significance of the study is warranted.
"
Additional reviewers comments:
Arctic sea ice melt onset from passive microwave satellite data: 1979–2012 by Bliss and Anderson

General Comments
This revised version of the manuscript is much improved. The addition of Section 3 certainly adds more value to the manuscript and better justifies using V3 of the data. I do have a one minor concern that should be addressed prior to publication.

1. The justification for why the authors have selected 1992 and 2004 is not explicit until the third paragraph and this needs to come first. Frequent users of PM data understand the differences between F11 and F13 but casual users not so much. It would be better to use all the years for a more robust V2 and V3 comparison illustrated in the scatterplots. This will improve the utility of Figures 4 and 5. I think you can still keep Figure 3 to show the differences spatially.

I am pleased to accept tc-2014-84 "Arctic sea ice melt onset from passive microwave satellite data: 1979-2012" for publication in The Cryosphere

Julienne Stroeve