Ecosystem functioning is sensitive to climate extremes, but extremes of a similar magnitude not always elicit the same functional response. Thus, understanding vulnerability to climate extremes is utmost important. Rolinski et al. developed an innovative approach to this question by framing vulnerability, focusing on a given ecosystem function, into a probabilistic framework in relation to the relevant climate extreme (drought in this case, among other). While both reviewers have acknowledged progress from the previous version, and recognized the attractiveness of this approaches to the question of concern, which I agree, some major critical issues remain to be solved before this approach can be accepted.

A revised version will be considered once the following issues have been adequately dealt with: 1) greater explanations about mechanisms; 2) Calculations of the SPEI index; 3) the time window chosen and the likely consequences on observed results, including autocorrelation issues. The authors may wish to consider a greater focus on processes within the model that are more robust.
The text is much improved but homogenization of the concept of meterological extreme/climate extreme/weather extreme would be helpful to the readers.

Thank you very much for the revised version, which has appropriately addressed the main issues raised during the review process. I am happy to accept now the paper for publication. I have noted a few minor corrections (typos, etc.) that need your attention before submitting the final text.