Second round

The authors have revised this protocol manuscript in a very satisfying way. Well done!

Just two very minor comments:

L246: Jakobsson et al. 2018 do not categorise in to the three categories high, medium and low risk of bias, do they?

L346: Package => package

First round

Title

You totally mislead the reader by not specifying that you will focus on seed addition restoration methods, not other methods which are way more common in northern temperate zone of Europe.

In addition, are you targeting restoration or re-creation? In other words, will you include studies looking at restoration of e.g. abandoned or over-grazed grasslands, re-creation of former arable fields, or both types? I strongly suggest you specify this more clearly (see also my comments below). The title suggests you will investigate restoration for species-rich grasslands, but if you are targeting/including former arable fields, you are restoring for re-creating/creating species-rich grasslands.

Abstract

L15: Why "have for long… …among the biodiversity richest…"? First of all, "for long" and "among the… …richest" statements do not tell the reader anything of value. Secondly, we do not know for how long this has been valid, such that this type of introductory sentence does not make sense. L20: Here it is very clear that your focus is on seed addition. Please clarify that also in the title.

Background

L52-53: have been introduced => were introduced L57: As far as I know, "lowland extensively managed grasslands" are not the AES, but among the targeted habitats. L58-60: Unnecessary complicated sentence. "Got depleted" is in the wrong tempus, but please rephrase and simplify the sentence to more direct writing. L61-62: This does not make sense. The sentence refers to one article published in 1997, which I would not regard as recent, and the other article does not even deal with seed addition. Please find and add proper references to this statement, which is fundamental for the whole review proposed in this protocol manuscript. L63-65: In my opinion, you fail to capture the proper references for your statement again. There is one German-only reference which I cannot understand, and the other reference is one single study. You cannot refer to one or two individual experimental studies with this statement, referring to others introduction is not a proper way to refer to (as I understand your reference to Kiehl et al.). In addition, I would recommend the authors to carefully clarify this statement. Seed addition is NOT the most common restoration method in northern parts of temperate Europe. L68-78: To be honest, I do not understand what you aim to target with these sentences. First, there already seem to be six reviews on the topic. Although not systematically conducted, I would suggest the authors to carefully go through the motivation for this review and clarify that in the protocol. In addition, when reading one gets the impression that you might not target restoration of arable fields to extensive grasslands ("…rather on the recreation of grasslands on arable fields"), but it seems like you will include this type of restoration, which I would rather call re-creation. The second part of that sentence ("…also looked at…") sounds like this is what you would focus on? First of all, that seems not to be the case, and secondly, what do you mean with species-poor grasslands (also L69)? Abandoned? Overgrazed? Please clarify these kinds of statements throughout the manuscript. As it is now it is not clear what types of grasslands you are targeting (does not need to be exclusively one type) and whether you focus on restoration, re-creation or both? See also my comment on the title (concerning "restoration" and "species-rich grasslands"). L80-83: This is far from a satisfying description of the stakeholder engagement in the review project. Please elaborate this section for transparency. If stakeholder engagement is too poor, I would like to see further efforts to include this essential aspect of the systematic review.

Methods General: There is no "Methods" heading. Please add. There is no specification of the PICO components. Please add, and please consider my other comments concerning terminology when adding this to the manuscript. L122: Why "enhance*"? There is a high risk of bias in including a positive outcome here, and not the negative counterpart. I suggest you remove this. L123: I see your motivation behind including "NOT tropic" (L137-142), but I do not agree. Just because the search term does not fail to find any of the articles in your test list it does not mean it would not fail finding other papers. This seems to be done only to save the review team time. Screening literate to exclude those studies does not take that much time, and I would like to see this negative Boolean operator and the term removed. L142-143: Why no studies before 1990? Please motivate and clarify in the text. L176-177: Please use "grassland" and not "meadow" or "pasture" when you are referring to grasslands in general. Potentially "managed grassland" to clarify the management part. L199-200: What do you mean? If outcomes are not properly reported? Later on, you state you will contact other for additional information (e.g. regarding measures of variation: L219-220). Why will you not contact the authors for outcomes not properly reported if method descriptions suggest that proper data are available? I find this a major drawback, and strongly suggest the authors include contact with study authors to receive outcome data, at least try to, for studies published the last decade or so. L246-247: What do you mean with "restoration projects"? Seems like you just refer to publications and the duration of or time since restoration for grasslands that those publications use data from. Please clarify.