I think the authors' responses to the reviewer comments are mostly fine. I don't plan to send any further resubmission out for review, I'll handle it myself because I appreciate that there have been some conflicting requests. Having said that, the referees *have* made some constructive and valid points and it would be good to try and address these if feasible.

For myself, the section on pyroclastic density currents is still out of line with the rest of the paper: this doesn't articulate the sources of uncertainty at all clearly and, indeed, looks like a standard "science" contribution that focuses more on "doing better modelling" than on handling uncertainty. It would be good if this could be brought to the level of the rest of the paper.

More broadly: I spoke to Thorsten about the paper a couple of months ago, and we agreed to take the discussion offline so as to avoid the very cumbersome Copernicus system. He is aware of my remaining concerns therefore, so I'm not going to rehearse them in detail here. Just for the public record though: my *main* concern with the revised paper is the continued criticism of probabilistic approaches on the basis that they implicitly consider everything as being aleatory in nature. This is the authors' view, and they're entitled to it. However, not everyone views probabilities in this way: in a review paper, the authors need at least to acknowledge the existence of alternative interpretations, and to make clear that such alternative interpretations can potentially alleviate (some of) their criticisms.

All the best,
Richard

I think we've got as far as we're going to get with this paper now, so I won't hold it up any further. I am disappointed that the revision still does not engage adequately with my request to acknowledge alternative interpretations of probabilistic reasoning (indeed, there is now a very contentious statement in the abstract, about probabilities and aleatory uncertainties, that wasn't there before). It's clear that we're not going to get anywhere with this however, and therefore I'm giving up.

Since these comments will be public when the paper is published, I take the opportunity to alert readers to the fact that (a) some of the discussion is predicated on a notion that to treat uncertainties probabilistically is implicitly to treat them as though they are aleatoric in nature (b) this viewpoint is not universally accepted. In particular, an alternative viewpoint is held (strongly) by many mathematical scientists who are familiar with, and who understand, mathematical decision theory. Decision theory does not have all the answers to the kinds of complex problems faced when dealing with natural hazards; but readers should be aware that the present paper represents the authors' personal perspective. It would be inappropriate, therefore, to cite it as justification for an assertion that probabilities should only be used to represent aleatory uncertainties.