The manuscript “Invasion-mediated mutualism disruption is evident across heterogeneous environmental conditions and varying invasion intensities” assesses the role of invasive garlic mustard abundance over space and time to assess the impact of invasives density on mycorrhizal mutualistic interactions and on native species richness.

The introduction does an excellent job in reviewing the background literature thoroughly, and driving the reader towards the relevant hypothesis. The methodology is well described and clearly it has been well-tested over time in a network of long-term study sites. The dataset is highly valuable, and it has been well analysed and contextualised, while acknowledging potential limitations (e.g. the potential for spurious significant results on line 294) which they reasonably deal with and justify. The discussion reads well and adequately contextualises the findings.

I do not have any major comments on the manuscript. It is obvious that it has been thoroughly reviewed and polished. It is well designed, presented, and discussed, and provides with highly valuable long-term observations. The conclusions are very interesting and, although they might not be conceptually ground-breaking, they do provide with hard evidence under realistic field conditions that confirm theories that, to date, have only been partially supported by hard data.

Minor comments:
-I suggest the authors to have a look to the publication below. I am not involved at all on this citation, but I feel that the authors might find useful information there, perhaps relevant for their discussion on possible interactions with climate change on pages 17-18. Feel free to disregard if not relevant.
Pugnaire, F. I., Morillo, J. A., Peñuelas, J., Reich, P. B., Bardgett, R. D., Gaxiola, A., Wardle, D. A., & van der Putten, W. H. (2019). Climate change effects on plant-soil feedbacks and consequences for biodiversity and functioning of terrestrial ecosystems. Science Advances, 5(11), eaaz1834. https://doi.org/10.1126/sciadv.aaz1834

The manuscript “Invasion-mediated mutualism disruption is evident across heterogeneous environmental conditions and varying invasion intensities” uses a long term 20 year dataset to assess garlic mustard invasion effects on native vegetation while accounting for the known ability of natives to establish below-ground mutualisms, and assesses this at different spatial scales. The manuscript deserves praise for the combination of long-term quality observations of both above and below ground biodiversity data. However, I found that the text did not make the most of this setup, and somewhat undersells the importance and context of some of the results.

For example, the term “quadrat” is used in the abstract, figures, and thorough the text, but it is only defined in the methodology. I could only learn what quadrats were sub-units within each site when I got deep into the methods. Most readers will read the abstract, skim through figures (and their legends), perhaps even the discussion, and will likely be unable to understand and interpret your results adequately. I myself guessed that quadrat was some kind of larger geographic unit, instead of a sub-unit of a site, and was quite confused until I got to the methods. I suggest that you replace the term/dichotomy “quadrat/site” by “small/large spatial scale” or “within site/among site variability” or something on those lines that is more intuitive to the casual reader, and that also links better with the concepts in the literature. I felt that by not using macro/micro scale (or the like) as your anchor words the text tends to focus too much in your specific results and misses the larger context at times.

I found that the introduction was sometimes unnecessarily long and meandering. Of course, I might be missing something and then you might disregard this comment but, for instance, consider paragraph on L66-80. I feel that it could be fully deleted without any real loss of context for your work. The introduction section is adequate overall, but I would have wished that it were focused in a more straightforward way on your two main areas, which as far as I understand are: the importance of the ability to establish below ground mutualisms, and the importance of spatial scales to detect this.

A similar situation can be found in the discussion, where the last three paragraphs (including the last one) seem to be mostly devoted to discussing caveats. Caveat discussion is important, but when it takes nearly half of the discussion it can become distracting (and even somewhat concerning). I suggest that they are reduced in length, consider also moving some parts to the methods, possibly.

Finally, and perhaps more importantly, I do largely miss some clear presentation and discussion of the magnitude of the patterns presented. Figure 2 shows significant differences, but the magnitude of those differences seems so small that one wonders if they are biologically significant or somehow a side effect of very large sample sizes (which, see below, we can’t assess as it is unclear how many sites were included). Figure 3 shows a heatmap from where magnitudes can be inferred, but since it is a relative measure one wonders whether the entire range of variation in which the heatmap is based is very narrow too. It would be very illustrative to have a figure, or at least a clear quantitative description in the text, of how large are the differences detected, for at least the main variables (e.g. richness of mutualistic vs. not mutualistic plants).

Minot comments

L173- How many site replicates were used for this study? And how many for each year, exactly? This should also be described in the results section as appropriate.
L323- The first half of the first paragraph in the discussion does more for summarising the manuscript’s hypothesis than for presenting the main take home messages to discuss. Consider whether you would prefer to start on L329, for instance, and lead the discussion from there. Again, remember to define quadrat here, or to replace it by a more standard term.

The authors have carefully reviewed and answered to my queries. They have chosen to follow my suggestions only partially, but their decisions are well justified and do not detract from the quality of the paper.

I still find difficult to infer the magnitude of the differences found, and although the authors have made improvements in that regard, I feel that it is not enough to clearly and intuitively understand, e.g. what is the % of increase of garlic mustard abundance and mychorrhizal status. This might not detract from the scientific quality of the manuscript either, but I do feel that there is a missed opportunity here to make the manuscript more accessible. I will leave to the Editor to decision to pursue this further.