Dear author, co-authors,

First of all, it has taken me longer than I hoped for to take my final decision on your ms submitted for publication in ESD. Sorry about that delay but it is also because of a decision that is not straightforward having now two contrasting reviews. One reviewer supports publication (with some minor comments) whereas one of the initial reviewers still has some major criticism on the revision. I have considered to invite another reviewer but that would again take substantial time and consequently decide at this point that I would like to invite you to check carefully the comments by both reviewers and provide then another revised version of the ms considering all suggested modifications on content but also on text issues.

Regards, Laurens Ganzeveld

Dear author, co-authors,

I have read your revised version as well as your extensive reply to the two reviews. One of the reviewers was still having some major comments and another reviewer, also called in for a review having to deal with two quite contrasting reviews of the first version of the paper, had only some minor comments. The comments range from discussions about the fundamental drivers of LAI changes to comments that mostly deal with different writing styles. There appears to be still some source of debate about the paper’s contents but also recognizing the serious efforts to properly address all the comments by the reviewers, I deem the paper now being acceptable for publication in ESD after you have addressed my last list of minor comments that were raised going again through the revised version. Hopefully, this publication will further trigger the debate about the most optimal way to appreciate future projections of LAI among other vegetation characteristics as a function of future climate change.
Laurens Ganzeveld

Minor comment (with page numbers/lines referring to the revised version pasted after your response to the reviewers comments):

Page 3, line 17: Obvious comment from my side: it is Ganzeveld et al. and not Ganzefeld

Page 11; lines 10-11: “with reanalysis derived data data (Qian et al., 2006; Harris et al., 2013) instead of model derived winds.”
There is something wrong with this line; data data.. and suddenly you are talking about derived winds where I would expect you to possibly refer to model derived precipitation. Check carefully and revise.

Page 12: Van Vuuren et al., 2011

Page 18: How would a higher LAI result in higher temperatures?? Yes, the LAI influences the surface energy fluxes but you would expect a higher evapotranspiration with the higher LAI and thus lower temperatures.
Then you might also need to reconsider to your statement on page 19 about LAI and surface fluxes and LAI and precipitation.

Page 20: “example, in the CLM, the..”

Page 21:
“Thus for high latitudes, especially, the projections of LAI appear to be dependent on the way the models’ simulate the carbon dioxide fertilization in the different models. This could also be, however, an artifact that the two models with the lowest carbon dioxide effect (CESM-­‐BGC and NOR-­‐ESM) use the same land carbon model (Thornton et al., 2009), which predicts low values of LAI in high latitudes for present day and does not tend to increase LAI much in the future. “
This appears to be a conflicting statement; there seems to be a relative strong correlation between the CO2 fertilization effect and LAI in high latitude regions but then your follow-up statement on the “artifact” of two of the models is actually implying that inclusion of the results for those simulations would tend to reduce the correlation between the CO2 fertilization effect and LAI for high latitude regions. These statements should be revised according to me. Also the use of the term artifact doesn’t seems appropriate here since it seems that there is no clear proof that inclusion of the nitrogen limitation in these two models results in incorrect simulations of the vegetation response to increases in CO2.

page 22: “...order to better understand and improve model projections on LAI” (or do you suggest on many other related parameters?

Page 22: “this correlation has been used in some studies to argue for a more correct answer, “. What is here the correct answer? Answer on what question, best estimates of LAI, best spatial distribution, best temporal evolution in LAI?

Page 23: “satellite data tend to be larger in tropical regions”; although it is not up to you to discuss all issues involved in LAI products from remote sensing observations, it might be worthwhile to also at least mention that also cloud cover is one of the main issues complicating LAI retrievals in especially tropical regions, besides the saturation issue for dense canopies.

Page 27: “as not statistically significant....,”; better say “not being statistically significant”

Page 30: “In the tropics, the top models tend to have lower future projections of changes in LAI than the average of all the models (0.07 m2/m2 instead of 0.16 m2/m2).” I was getting confused seeing those numbers and interpreting the statement on future projections in LAI as the absolute values in future LAI and consequently this suggestion.

Page 31; line 1: “projections using all the..”.

Page 31: The fraction of the time in drought in the future is increased in the tropics, if we only consider the top models” this expression of fraction of the time in drought reads weird; rephrase

Page 31: “project a more pessimistic future,....” pessimistic in what sense; less LAI, less CO2, less climate change; here giving this qualification it should be clear to what you deem being pessimistic.

Page 32: line 18, ...low LAI conditions..

Line 20: “or Low-­‐LAI incidence..”; I now wonder if I missed something but again, low-LAI and what you mean here with low-LAI incidence??

Dear author, co-authors,

I have read your response to my last round of editors comments and checked the revised version and accept now your paper as is for publication in ESD. Congratulations and let's see how the presented study will trigger more discussions/activities on the representation of LAI in ESMs,

regards, Laurens Ganzeveld