Initial decision bg-2020-428

Development of global temperature and pH calibrations based on bacterial 3-hydroxy fatty acids in soils. Pierre Véquaud, Sylvie Derenne, Alexandre Thibault, Christelle Anquetil, Giuliano Bonanomi, Sylvie Collin, Sergio Contreras, Andrew Nottingham, Pierre Sabatier, Norma Salinas, Wesley Phillip Scott8, Josef P. Werne, Arnaud Huguet.

Dear Arnaud,

First of all, I would like to thank the two anonymous reviewers for there comments and you for your reply. After reading the reviewer comments I thought they were quite severe, especially the comment about the old versus new brGDGTs, losing the brGDGTs completely and the model/regression issue. After reading your reply I noticed that one of the old brGDGT datasets is now available as state-of-the-art brGDGT dataset, which already helps. I also have the impression that a lot of the comments are based on the writing not being as clear as it could have been, the models not being described thoroughly enough, the link between brGDGTs and OH fatty acid data and quite a few more. I am willing to accept this, but you will have to make it extremely clear why you made the choices you have for instance for the models and the terrestrial paleo archive used. Based on your reply I think this should be possible.

I am looking forward to the next version of the manuscript and rebuttal.

Best regards,

Marcel

Decision on BG-2020-428

Dear Arnaud Huguet and co-authors,

First of all, let me thank both reviewers again for their review and you for your reply. I don’t want to be too difficult but a slightly more cleaned up version of the annotated manuscript would have been nice. This multi-coloured mash up, I assume with different colours coming from different co-authors, with French comments in the side-line is not great to read. I also have some other comments. I think new proxies always come with their issues, nothing to be ashamed off. I also think old or more established proxies also have their issues, that is one of the reasons new proxies are always welcome. Multiple proxies for the same parameter and based on different compounds, organisms and analytical techniques are always great to have and use together. Therefore, I do not think you need to sell this proxy based on hydroxy fatty acids by making it very clear the more established proxy has a relatively large RMSE. It is not like your proposed proxy has no issues itself. Personally, I would have framed this, line 88 to 95, a bit different. My guess is that would have also helped with the response. The multi proxy approach is great and wonderful, but for that you need multiple proxies. Something along these lines would have been great. The “despite improvements it is still not great” statement is not necessary and completely ignore the fact that all proxies have their good and bad sides. As long as your proxy is not approaching perfection, universally applicable with extremely small errors just be careful.
I like the multiple models used. I also like the explanation of why one works better under certain conditions etc. It also clearly indicates that this is in the trail and error phase. As the authors mention, more soils need to be analysed both for the calibrations as well as applications. The application here is nice and it helps explain some of the pros and cons of the different models, but it is also fairly limited and definitely not the dataset that ends all discussion on the applicability of these models. Be careful on how much emphasis you want to put on the example.

I also have some technical issues:

Line 114-115: suggesting that regional relations may be more adapted to apply RAN15 as a temperature proxy in soils. I assume you mean something like “suggesting that regional or local RAN15 calibrations maybe more appropriate to apply”.
Line 146-150: The sentence that starts with “In addition” and ends with “settings”. This has become a very long and difficult to follow sentence.
Line 155-157: again without any indication on how great the 3-OH FA proxy might be the authors are very negative about a more established proxy. This is not necessary and only puts readers off.
Line 386: never start a sentence with But, ok, only very rarely start a sentence with But. Why not however? Think about rewriting up until line 431. Quite difficult to follow.
Line 530: order of magnitude, from 0.1 to 1? Sounds very dramatic order of magnitude, varied between 0 and 1 is much less dramatic.
Line 662: low Ph ranges, I think you mean narrow or small.
Line 794: are pooled very narrowly?
Line 844: “and sources of these compounds” I can guess what you try to say, but I think it is best if it is written more clearly.
Line 845-847: Who is using bacterial relative abundances. No one measured bacterial relative abundances for this manuscript, right?
Line 937: All or the C10 to C18? If I understood it correctly there were more.
Line 953: whole suite limited to C10 to C18?
Line 1181-1182: along some, but not all, of the altitudinal transects … At least I guess that’s what you want to say.
Line 1345-1346: by working on the microbial level, please explain?
Line 1813: is it really much smaller than the 6.5 to 19.7 of the k-NN method?
Line 217-2019; I don’t this is needed in this way. The proxies can strengthen each other, use that. I mean the information is correct, I would just frame it a less “confrontational”.
In your reference database, could you please change J.S.S. Damsté into J.S. Sinninghe Damsté.

I would like to ask the authors to go through the manuscript again and clean up some of the language, the different multi coloured text additions and the French on the side with removed or added things, very distracting. It is a relatively novel proxy with the different models it looks promising, the pros and cons are being discussed, I would really like to see it again after it has been cleaned up a bit more.

Best regards,

Marcel

Final decision on BG-2020-428

Dear Arnaud Huguet and co-authors,

Thank you for putting in the effort to go through the manuscript again and work on the suggested changes. I am happy to accept the manuscript as is.

Best Regards,

Marcel

Dr. Marcel van der Meer