We appreciate for your time on our manuscript . It is really a very high opinion of this article. In addition, we did not find the specific comments in  review system. If you have specific comments on our manuscript, please tell us, we will appreciate to revise the manuscript. Thank you again. Best wishes.

Small mammalian herbivores often create extensive disturbance on grasslands, and might affect the ecosystem services of grassland ecosystem. This study uses plateau pika as an example herbivore to investigate the effect of disturbance by small mammalian herbivores and its disturbance intensity on ecosystem services of alpine meadows on the Qinghai-Tibetan Plateau. This study finds that the presence of plateau pikas and its disturbance intensity have different effect on ecosystem services of alpine meadows when the indicators, which are used to estimate the ecosystem services of alpine meadows, are various. These findings can improve the understanding of small mammalian herbivores in relation to grassland ecosystem services. The experimental design is sound and results are striking. In my opinion, its topic is of interest to its audience of our journal. I suggest that the manuscript can be published after some minor corrections.

Response: Thank you for your positive comments, and we have revised the manuscript carefully according your advices.

General comments:

1. The plateau pika had been introduced in introduction sector and Field survey design sector. It had better move the description of plateau pikas in Field survey design sector to introduction.

Response: Good comment. We have moved the description of plateau pikas in Field survey design sector to introduction sector.

1. The confidence intervals are required to represent in the Figure 2.

Response: Thank you for your comment. We have added the confidence intervals in the Figure 2.

Some specific comments:

1. Line 41: Please add the Latin name of prairie dogs when it first appeared.

Response: Thanks for your advice. We have added the Latin name of prairie dogs when it first appeared (Line 41).

Line 44: What is the Latin name of European rabbit.

Response: Thank you for your suggestion. We have added the Latin name of prairie dogs when it first appeared (Line 44).

Line 141: The same alpine meadow is not good words. It may be alpine meadow with same dominant plant.

Response: Thanks for your comment. We have revised “the same alpine meadow” into “the alpine meadow with same dominant plant”.

Line 190-194: The quantity of samples in each plot with or without plateau pika is better to expressed consistently with Arabic numerals?

Response: Thank you for your advice. We have revised the quantity of samples in each plot with or without plateau pika with Arabic numerals.

Line 190-194: 5 soil samples in each plot were mixed into composite sample to measure carbon and nutrient concentrations, or each of 5 soil samples was used to measure carbon and nutrient concentrations. Please clarify it.

Response: Good question! In this study, we collected one soil sample from one subplot. 5 samples in each plot from 5 subplots were individually measured. The average value of five soil samples in one plot was considered as the representative data of that plot. We had clarified it in manuscript.

Line 255: Please move “Differences were considered significant at p < 0.05.” into the analysis of “LMM” in paragraph 2 of Data analysis sector.

Response: Thank you for your suggestion. We have moved “Differences were considered significant at p < 0.05.” into the analysis of “LMM” in paragraph 2 of Data analysis sector (Line 249).

Line 302: a quadrat scales?

Response: Thanks for your suggestion. We have revised “a quadrat scales” into “the quadrat scales”.

I suggest that the manuscript can be published after a little correction.

Response: We thanked you for your positive comments, and we have revised the manuscript carefully according your advices.

1. "this study" appeared in the abstract four times. It is necessary to further polish the language.

Response: Thank you for your advice. We have polished the language in the abstract.

1. For references in scientific hypothesis, please check whether there is the necessity of citation. So, it is suggested to summarize the existing research in the research progress. Then put forward scientific hypothesis.

Response: Exactly, we had summarized the existing research progress in front paragraphs of introduction. It is unnecessary to cite the references in hypothesis, and it seems redundant. To make the hypothesis more scientific, we have deleted the references in the hypothesis according to your opinion.

1. P96-97 “Plateau pikas can live in various habitats with different soil types, topographies, and microclimates.” This sentence lacks qualified region.

Response: Thanks for your suggestion. We have added the qualified region “on the Qinghai-Tibetan Plateau” in this sentence (Line 97).

1. What is the grazing situation of the experimental site and how to eliminate the impact of grazing intensity on the grassland.

Response: Thanks for your comment. This study used a stratified random and paired design to discover the general effect of plateau pika disturbance on ecosystem services of alpine meadows. In experimental design, each paired plot shared the same grazing intensity during the cold season, however, 50 paired plots consisted of different yak grazing intensity, and this can permit the general pattern relating to the effect of plateau pika disturbance on alpine meadow ecosystem services. We have supplemented this information into “Field survey design” sector.

1. ABED (active burrow entrance densities), when is the investigation period?

Response: Thanks for your comment. The ABED (active burrow entrance densities) was measured in the process of field sampling. As described in field sampling, this study firstly measured the density of active burrow entrance, and then measured the area of bare soil patches, collected vegetation and soil samples.

1. In Fig.1, the presence of plateau pika, What pika density are based on?

Response: Thanks for your comment. Fig.1 shows that plateau pika density is a qualitative description, which is to present the difference in ecosystem services of alpine meadows between the presence of plateau pikas and the absence of plateau pikas. Fig.2 shows that the changeable trends of each ecosystem service of alpine meadows as the disturbance intensity of plateau pikas increased. In Fig.1, 50 disturbed plots with different plateau pika densities (it ranged from 83 to 1384 entrances ha^-1) was considered as a whole to compare with the 50 undisturbed plots. We supplement some sentences to clarify this research approach in data analysis sector.

Dear Editor

Thank you for your patience with our manuscript numbered “BG-2021-313”. Based on the further comments from the reviewer and editor, we have revised the manuscript carefully again. We found that these questions and comments are valuable to improving the quality of this manuscript. Here, we submit both a clean and a track changed versions of manuscript to “Biogeosciences”.

The responses to the reviewer’ comments are following:

Chen and others explore the role of pika presence at different densities to ecosystem services. The manuscript makes a few interesting points but could be strengthened by focusing more on fundamental ecological theory.

1. I wasn't entirely sure how it was identified that pikas caused some bare patches and not others, for example if the bare patches were caused by past pika activity. Therefore it was unclear to me how controls were used to test the objectives.

Answer: this is a common comment we met. The soil bare patches caused by plateau pikas is easily to identify because one soil bare patch caused by pika is paired with a visible burrow entrance (Pang et al. Geoderma, 2021, 115098). Other soil bare patches are not paired with visible burrow entrance.

The soil bare patches caused by past plateau pika activities can be divided into two types. Some can be gradually restored, and others are still be bare.

In this field survey, we firstly identified the soil bare patches caused by plateau pikas though the paired visible burrow entrance. And then, we continued to identify whether these soil patches are bare. If bare, we divided them into soil bare patches caused by pika; if not bare, we did not divide them into soil bare patches caused by plateau pikas. We supplemented this information into 2.3 Field sampling.

1. 'plateau pika presence' in the title is a bit smoother (and section 3.1).

Answer: Thanks for your comment. We have revised “plateau pika presence' into “the presence of plateau pikas” in the title and section 3.1.

1. 17: 'ecological service of' can be deleted here because it's mentioned below. Honestly the word 'ecological service' needs to only be stated once or maybe twice in the abstract; it's continued usage is redundant. Also in the main text. A previous reviewer may have recommended this but it needn’t be repeated so much.

Answer: Great comment. According to you this comment, we have only stated once in the abstract, and the other have been deleted in the revision manuscript.

1. The abstract doesn't contain any qualitative values and it would be of greater use to the reader if it did. 'Possible pattern' is also speculative, you can say what the pikas did in your study.

Answer: Thank you for your comment. We have revised the abstract and result sectors, in which some qualitative values are added into manuscript. This study showed that the forage available to livestock and water conservation were 19.74% and 15.86% lower in the presence of plateau pikas than in their absence, while biodiversity conservation, carbon sequestration, soil nitrogen, and phosphorus maintenance were 14.58%, 29.15%, 9.97% and 8.89% higher in the presence of plateau pikas than in their absence.

In addition, we have deleted possible, which confirmed that the findings of this study present a pattern of plateau pikas influencing the ecosystem services of meadow ecosystems in alpine regions.

1. 62: 'with averaging' isn't correct usage. A few minor usage changes throughout the manuscript would make for an improvement, using an automated grammar checker will probably catch almost all instances. (see also 66 'with average' and numerous other instances, e.g. line 137 'can turn green until July'. "don't turn green until July"?).

Answer: Thank you for your suggestion. Based on your comment, we have revised “with averaging” into “with an average weight of”; We have revised “with average” into “with an average length and depth of”; we have revised “can turn green until July” into “don't turn green until July”. In addition, we have used an automated grammar checker to make for an improvement throughout the manuscript.

1. 192: use 'W' if it is the shape of a W. I wasn't sure if it meant "West".

Answer: Thank you for your comment. We have revised “a W pattern” into “ the shape of a W pattern”.

1. Figure 1: why does it go 'a' then 'b' for some bars and 'b' then 'a' for others?

Answer: This is a statistical expression. Lower case represents a significant difference between the absence and presence of pika in Figure 1. Generally, bars with “a” represent the bigger value, and bars with “b” represent the smaller value.

1. 2: this looks like the famous "Intermediate disturbance hypothesis" in some cases. (There’s even a wikipedia page for it: https://en.wikipedia.org/wiki/Intermediate_disturbance_hypothesis). A little bit of pika activity is good but too much and they take over. De-emphasizing the comparisons against rabbits and prairie dogs and thinking more about ecological principles could structure the Discussion more and add references, which are a bit lacking. In other words, if you expand the interesting discussion beginning line 457 a bit and focus on fundamental theory in addition to these nice examples of intermediate disturbances, the Discussion could be further improved. (This paragraph is also too long; breaking it up would help the reader.)

Answer: Thanks for your comment. At the beginning line 457 in former revision (line 455 in present revision), we firstly focus on the Intermediate disturbance hypothesis. We looked up some literatures (Dial and Roughgarden, Ecology, 1998, 79(4), 1412-1424; Gao and Carmel, Oikos, 2020) to refer the methods that introduce the fundamental theory. The revision discussion is as follows:

This study also shows that the disturbance intensity of plateau pikas also affects the forage available to livestock, biodiversity conservation, water conservation, carbon sequestration, and soil total nitrogen and phosphorus maintenance. As found in plant-species richness and aboveground plant productivity (Dial and Roughgarden, 1998; Gao and Carmel, 2020), the response of plant-species richness and palatable plant biomass to the disturbance intensity of plateau pikas follow the pattern for the intermediate disturbance hypothesis in this study. In addition, the soil organic carbon stock, soil total nitrogen and phosphorus stocks at home-range scale also support the intermediate disturbance hypothesis. However, the top soil water storage does not conform the intermediate disturbance hypothesis.

At lower disturbance intensity, stronger competition of dominant sedges often restrains the grass to grow well (Pang and Guo, 2018) and the rare plants to coexist (Wang et al., 2012), which leads the forage available to livestock and biodiversity conservation of alpine meadows to be maintained at a low level. Although the presence of plateau pikas can increase the input of soil organic matter, this increase is low (Pang and Guo, 2017; Pang et al., 2020b), which enables the soil organic carbon sequestration and soil nitrogen and phosphorus maintenance of alpine meadows to maintain a relatively low level.

At intermediate disturbance intensity, the activities of plateau pikas improve the growth potential of grass plants (Wang et al., 2012), and increase the input of organic matter, soil total nitrogen (Li et al., 2014), organic carbon accumulation (Yu et al., 2017b), which contributes to higher the biodiversity conservation, forage available to livestock, carbon sequestration, soil total nitrogen and phosphorus maintenance services.

At higher disturbance intensity of plateau pikas, frequent bioturbation can enable all species to be at risk of going extinct (Dial and Roughgarden, 1998). Low soil water content in alpine meadows (Liu et al., 2013) only sustains the xerophytes and mesophytes, most of which are unpalatable (Pang and Guo, 2018). This contributes to relatively lower forage available to livestock and biodiversity conservation. Low vegetation biomass decreases the input resources of soil organic matter (Sun et al., 2015; Pang and Guo, 2017), contributing to a decrease in the soil organic carbon sequestration and soil nitrogen and phosphorus maintenance of alpine meadows.

Additionally, the linearly negative relationship between the water conservation of alpine meadow and disturbance intensity of plateau pikas is ascribed to evaporation and more water infiltration on bare soil patches, as the amount of water evaporation and infiltration tends to increase as the area of bare soil increases (Liu et al., 2013).

The manuscript has been revised carefully and strictly according to your comments. We hope our modification and explanation is clear enough, however, if there is still any question, please do not hesitate to contact us.

Yours sincerely

Ying Ying Chen, Huan Yang, Gen Sheng Bao, Xiao Pan Pang, Zheng Gang Guo^*

First author: Ying Ying Chen, Email: chenyy2019@lzu.edu.cn

Corresponding authors: Zheng Gang Guo, E-mail: guozhg@lzu.edu.cn