Based on 79 and 76 paired observations for soil and plant δ¹⁵N from 20 published studies, the authors applied a global meta-analysis method to evaluate the influence of the global warming on soil and plant δ¹⁵N. This work is helpful for readers and can be useful in the study of ecosystem nitrogen cycling. The topic of this manuscript fits well under the scope of “SOIL”. Moreover, it is well written and figures are of high quality. In my view the paper should definitely be published in “SOIL” with a moderate revision. However, the following comments should be taken into account before acceptance.

Major comments

1. In this paper, few literatures (only 20 papers) were selected for meta-analysis. Fortunately, there are 79 and 76 paired observations for soil and plant δ¹⁵N, and I believe that these samples are enough for meta-analysis. However, the manuscript lacks of important information on soil, such as organic matter content.

2. The Introduction section still needs to be improved to clarify the key scientific issues. Such as, one of purpose of this study is to identify the main factors influencing the warming effect on the soil and plant δ¹⁵N. In fact, since a relatively large number of investigators from around the world were involved in the 20 studies that were selected, there is no indication that all these studies used similar protocols, and coordinated their experimental conditions. For example, we know that soil water content has an impact on soil and plant δ¹⁵N, and also impacts the analysis results. The authors should check what is the cause of the differences between soil and plant δ¹⁵N?

3. The M & M section is a little rough and please reorganize this section to make it more clearly. Please provide detailed soil information which appeared in this section.

4. The Discussion section, which is the most important part, is relatively weak because the contents and explanations are not well organized and are still not convincing. Less descriptive of results and more mechanism discussion are encouraged in this section. More references should be added to the discussion section.

 Specific comments

Line 12: The nitrogen-15 (¹⁵N)...

Line 14: global warming -> experimental warming

Line 15: for -> of

Line 29: ...soil δ¹⁵N was more effective than plant δ¹⁵N in indicating....

Line 40: cycle -> cycling

Line 48: becomes a useful tool -> is often used

Line 68: In addition to soil warming, air warming was also conducted.

Line 93: more than 50? Please specify

Line 99: can be obtained -> were provided

Line 109: Did you do resampling with bootstrap? What are the resampling times?

Line 122: Mean effect sizes: This concept was not specified earlier. I think that it is technique, soil type, warming treatment... . Is it so? Please, clarify

Line 154-159: The authors should add more references to support this statement.

Line 160-164: Is that true? There were two kinds of ¹⁵N, three soil acidity-alkalinity types, three vegetation types, and three warming treatments groups. Were there enough number of observations for these groups?  For example, for plant δ¹⁵N, there were only 9, 3 and 2 observations for acid, neutral and alkali, respectively.

Line 167: add a reference here

Line 182-184: awkward sentence

Line 185: If you want to use “significantly”, you have to give the P value.

Line 188: Add a reference.

Line 211: when investigating -> in order to better investigate

In the manuscript, ‘Soil and plant δ¹⁵N have a different response to experimental warming: A global meta-analysis’, the authors assess 20 experimental warming field studies and conclude that soil and plant δ15N had negative and positive responses to warming at the global scale, respectively.

Overall, the study is a nice contribution because it looks at both plants and soils. But I think the title oversells and misleads. Also, I realize that this is a short communication, but more detail is needed to support the hypothesis, to rationalize why the specific environmental variables were chosen over others, and to relate this study to other results in the literature.

The word ‘significantly’ is overused in the abstract, and the presentation of results that are not significant as effects is not appropriate. It would be better to not use the word significantly and to only present the significant results (after defining p-value cut-off in methods). In other words, remove the inference from the title and abstract that plant δ15N had a positive response to warming – this was not significant. I think the finding that soil δ15N is a better indicator than plants of environmental cues is a more appropriate conclusion or story lead. Because really, the pattern of response of plants and soils to environmental drivers tested here was not different, it was just weaker in plants than soil.

Comments by line number:

l. 44. Please explicitly define ‘openness’ in the introduction (and if possible, in the abstract). Although some readers will understand, those unfamiliar with the δ15N literature will read this as jargon.

l. 50-51. This should be reversed: The isotopic fractionation effect results in gradual 15N enrichment.

l. 74-75. The hypothesis is not supported by any rationale in the introduction, please provide some preamble that supports why they should be different.

l. 85-86. It is not clear why temperature gradient studies are being excluded, as they will also include a treatment and control. Perhaps the authors could clarify - do they mean climate gradients (space-time substitution), or lab incubations?

l. 114. I am quite surprised that length of warming was not considered, as multiple studies illustrate different responses in plant and soil CNP for short and long-term warming experiments. Also, see Craine et al (2015), which suggests that soil δ15N is directly controlled by soil C and texture, and only indirectly controlled by temperature. Did you consider length of warming, SOC, or soil texture as sub-groups? I see l. 47-67 in the intro provides justification for the subgroups, but I remain unconvinced that the chosen subgroups are more important than the ones not assessed. Perhaps this needs better support from the literature.

l. 176. Actively layer > active layer

l. 177. This is not true for all air warming treatments.

Table 1 should have a column for soil types – at the very least, organic or mineral soil, but especially pH, since this was a main factor in the analysis.

Many grammatical typos, please correct: l. 40, 42, 43, 49, 68, 83, 85, 171