Long-Term Stability of Harvest Mouse Population

Round 1

Reviewer 1 Report

The authors of the reviewed paper infer long-term stability in Lithuania's harvest mouse population. They analysed trapping data from 1975 to 2022 and owl pellet data from 1986 to 2009. Based on both datasets, the proportion of this species within the small mammal community remained consistent throughout all decades. Data regarding the long-term dynamics of Micromys minutus are scarce, and conservation assessment often relies more on the assumption that, as a rodent, the species is not subject to specific demographic pressures rather than concrete evidence of its status. Consequently, the work presented here can potentially be of broad interest. However, there are shortcomings in the paper that raise questions about its validity.
The trapping data used in the analyses came from different sources where traps were usually set in rows of 25 traps and usually left for 3 days (as mentioned in lines 122-127). This description indicates that the trapping procedures in various studies were not standardised. The reference studies must have used the same trapping protocols to compare capture data effectively. Using a different number of traps for varying days/nights introduces a source of error (greater trapping effort in some areas) that could influence the results.
Regarding the trapping sample size, there were 170 sampling sites and 669 samples. Thus, each site was sampled, on average, only 4 times. This confirms the heterogeneity of the data (sampling in different years from different areas and habitats), making comparisons quite challenging.
The paper does not provide a timetable for trapping in different environments. Trapping likely occurred in different habitats and years. Occasional trapping in different habitats cannot be used to infer long-term dynamics, as these may depend on both temporal and spatial variations. The relative abundance (RA) of M. minutus, expressed as the number of individuals per 1000 trap-nights, varied between 0.14 and 13.33 (Table 5) according to habitats and only 0.3-3.1 (Table 2) or 0.3-2.5 (Table 3) according to the year (long-term data from the same sites). Thus, inter-habitat variations were much more significant than inter-annual variations, and spatial variations can potentially obscure temporal trends.
Catches of M. minutus were very low (e.g., 1 to 14 long-term in Ignalina). With such small numbers, it is crucial to demonstrate through a power analysis that the comparisons can detect temporal variations.
The revised paper does not specify the raptor species from which the analysed owl pellets originated. Based on the reference paper, it is understood that they were from Strix aluco. Since this raptor is a forest species, there may be better options for sampling an open environment species. It would have been valuable to have owl pellets from Tyto alba, a species that hunts in open habitats, has a more varied diet, and provides a better sample of micro-mammal communities.
Overall, the data used in this paper was derived from catches in different habitats in different years and also with non-standardised protocols. Catches of M. minutus have always been very limited (low densities), making it difficult to have robust data to detect even large temporal variations. With these limitations, the work presented here cannot be published.
The authors also present four long-term small mammal trapping events. Since these areas have been sampled hopefully with standardised protocols, I suggest the authors analyse this data, including conducting a power analysis. They could prepare a paper with only this homogeneous dataset if they can demonstrate that demographic variation can be detected with the available data.

I am not a native English speaker, but English sounds good to me

Author Response

Rev#1 comments and answers

Comment: The authors of the reviewed paper infer long-term stability in Lithuania's harvest mouse population. They analysed trapping data from 1975 to 2022 and owl pellet data from 1986 to 2009. Based on both datasets, the proportion of this species within the small mammal community remained consistent throughout all decades. Data regarding the long-term dynamics of Micromys minutus are scarce, and conservation assessment often relies more on the assumption that, as a rodent, the species is not subject to specific demographic pressures rather than concrete evidence of its status. Consequently, the work presented here can potentially be of broad interest. However, there are shortcomings in the paper that raise questions about its validity.

Answer: thank you, we are going to clarify some of the shortcomings in the answers below.

Comment: The trapping data used in the analyses came from different sources where traps were usually set in rows of 25 traps and usually left for 3 days (as mentioned in lines 122-127). This description indicates that the trapping procedures in various studies were not standardised. The reference studies must have used the same trapping protocols to compare capture data effectively. Using a different number of traps for varying days/nights introduces a source of error (greater trapping effort in some areas) that could influence the results.

Answer: in only 1.3% of trappings there were less than 75 trap/days (one 25 trap line and 3 days). Rest was according standard. We changed word “usually” to explain real situation.

• In 3 cases (1995, 2014, 2020), there were one line with 25 traps kept two days (third was heavy rain, therefore, standard trapping not completed).
• Three other cases were trappings in commensal habitats, where we had no chance for 25 trap line, as habitat is too small.
• The last 4 cases, all in mowed meadow, 2018 and 2019, there was one line with 25 traps kept two days, due to the rain.
• Finally, in none of these 10 trappings, minutus was not trapped.

Comment: Regarding the trapping sample size, there were 170 sampling sites and 669 samples. Thus, each site was sampled, on average, only 4 times. This confirms the heterogeneity of the data (sampling in different years from different areas and habitats), making comparisons quite challenging.

Answer: to lessen heterogeneity of data, we pooled trappings into decade-long periods. Main habitats for M. minutus, namely wetlands and meadows, were samples in all decades. While we agree with your warning on heterogeneity, this is an unavoidable issue in perhaps all countries where small mammals were snap-trapped.

Comment: The paper does not provide a timetable for trapping in different environments. Trapping likely occurred in different habitats and years. Occasional trapping in different habitats cannot be used to infer long-term dynamics, as these may depend on both temporal and spatial variations. The relative abundance (RA) of M. minutus, expressed as the number of individuals per 1000 trap-nights, varied between 0.14 and 13.33 (Table 5) according to habitats and only 0.3-3.1 (Table 2) or 0.3-2.5 (Table 3) according to the year (long-term data from the same sites). Thus, inter-habitat variations were much more significant than inter-annual variations, and spatial variations can potentially obscure temporal trends.

Answer: A timetable of trappings, yes, paper do not show this, but it is too long to provide – it is over 600 lines in the table. It might result separate data paper later.

With each decade is represented by the average of 87–278 trappings, we believe that inter-habitat variations are smoothed. 11 trappings from 1981 to 1990 have averaged quite intensive investigation – three times per year, three habitats (meadow, forest and wetland) in several sites covering region with 20 km in diameter. The 70’s period is the only less represented in trapping effort, no more data present.

In Lines 245–250 we write “The cumulative effect of two factors, decade and habitat, on the relative abundance of M. minutus was found to be statistically significant (F = 2.78, p < 0.001). However, the model's explanatory power is relatively weak, accounting for only 9.1% of the variation in RA. Specifically, the influence of the time factor appears to be insufficient (F = 1.96, p = 0.08), while that of habitat is significant (F = 3.16, p < 0.001).” So yes, we agree, that habitat is more important!

Comment: Catches of M. minutus were very low (e.g., 1 to 14 long-term in Ignalina). With such small numbers, it is crucial to demonstrate through a power analysis that the comparisons can detect temporal variations.

Answer: power analysis confirm total sample in Ignalina was sufficient to say, that in 1985 proportion of M. minutus was higher than on average. The other comparisons would lack of required power (0.80).

Comment: The revised paper does not specify the raptor species from which the analysed owl pellets originated. Based on the reference paper, it is understood that they were from Strix aluco. Since this raptor is a forest species, there may be better options for sampling an open environment species. It would have been valuable to have owl pellets from Tyto alba, a species that hunts in open habitats, has a more varied diet, and provides a better sample of micro-mammal communities.

Answer: we add owl species to the text. Prey of Asio otus represented only 4.3% of all preyed small mammals, and no M. minutus were recovered in it. We also add explanation of the pellet collection sites, to show that Strix aluco preyed not only in forest, due to specific location of the nestboxes in fragmented open landscapes. Unfortunately, there are no data on the prey of Tyto alba prey from Lithuania, therefore this cannot be used.

Please note, that this is retrospective investigation, using all available data

Comment: Overall, the data used in this paper was derived from catches in different habitats in different years and also with non-standardised protocols. Catches of M. minutus have always been very limited (low densities), making it difficult to have robust data to detect even large temporal variations. With these limitations, the work presented here cannot be published.

The authors also present four long-term small mammal trapping events. Since these areas have been sampled hopefully with standardised protocols, I suggest the authors analyse this data, including conducting a power analysis. They could prepare a paper with only this homogeneous dataset if they can demonstrate that demographic variation can be detected with the available data.

Answer: as we already answered, most of the trapping were according standardized protocol, with only 1.3% of non-standard trapping effort.

We also performed a power analysis to calculate the required sample size to detect differences in the proportions of M. minutus between decades. Calculations were done for de-sired significance α = 0.05 and power = 0.80, using online calculator (https://www.stat.ubc.ca/~rollin/stats/ssize/b2.html, accessed on 15 October 2023). Results were added as Table 6 in Discussion. From the Table we conclude that sample size was sufficient to analyze differences of proportions between decades, but ANOVA analysis confirmed proportion remained stable.

Monitoring of small mammals across several decades using the same sites, habitats and trapping effort is not possible in the absences of dedicated financing. We have no examples of such monitoring in Europe, therefore, thank you for your comment – after the power analysis we see our main conclusion on the stability of M. minutus in Lithuania is correct.

Reviewer 2 Report

All notes are marked directly in the text

Comments for author File: Comments.pdf

Author Response

Rev#2 comments and answers

Comment Line 16: one "dot" is extra, it needs to be deleted

Answer: done

Comment Line 193: in my opinion it is unnecessary= to repeat "M. minutus" 3 times in the table, it is already clear from the header. Maybe just once for the total number (n) is enough. However, I leave it up to the authors to decide.

Answer: we decided to leave headed in one line. Not repeating the word “M. minutus” would require to make two-line header.

Comment Line 213: I recommend to change the colours for more contrasting ones, especially for "M. oeconomus" and "Other species"

Abswer: colors changed as per your comment. Idea, however, was to make only M. minutus in high contract. But, perhaps you are right.

Reviewer 3 Report

This manuscript is well organized and written. The figures and tables are appropriate. The article uses long-term data to describe a species that is widespread but always occurs in low abundance, uses a variety of habitats and a variety of plant species for food and nesting. There are no particular management suggestions, nor evidence that the species needs more monitoring or research. Hence, I'm not sure of much value to the effort put into this manuscript. 

Author Response

Rev#3 comment and answer

Comment: This manuscript is well organized and written. The figures and tables are appropriate. The article uses long-term data to describe a species that is widespread but always occurs in low abundance, uses a variety of habitats and a variety of plant species for food and nesting. There are no particular management suggestions, nor evidence that the species needs more monitoring or research. Hence, I'm not sure of much value to the effort put into this manuscript.

Answer: paper value is to show, that in Lithuania in particular, the species is not declining and does not require protection, but this is not the case in all countries. There are practically no such long-term data reviews published on M. minutus, so our article (hopefully) will be useful as a reference for other countries. By the way, there are no references for this species in Latvia or Estonia, so this manuscript represent three Baltic States.

Reviewer 4 Report

A well written and interesting paper and a great contribution to the knowledge of the relatively elusive harvest mouse in Europe. I have only very minor suggestions for improvement in the PDF file.

Please check the References section for consistent style, as annotated.

Comments for author File: Comments.pdf

The English is mostly excellent.  I made some suggestions for improved wording in the PDF file

Author Response

Rev#4 comments and answers

A well written and interesting paper and a great contribution to the knowledge of the relatively elusive harvest mouse in Europe. I have only very minor suggestions for improvement in the PDF file.

Please check the References section for consistent style, as annotated.

Comment Line 16: Delete extra period.

Answer: deleted

Comment Line 57: Delete:  )

Answer: deleted

Comment Line 61: Change r to "s"

Answer: changed

Comment Line 79: better: "mice of the genus Apodemus"

Answer: changed

Comment Line 102: Change to: "Land use map of Lithuania..."

Answer: changed

Comment Line 117: Please try not to split Table across two pages.

Answer: Table is placed according requirements, after the first mention. We are sure in the layout version it will be in one piece.

Comments Lines 123–124:

Answer: All changes done as suggested, sentence excluded. Now paragraph sounds “We analysed small mammal trapping material from the author's personal data and from data published by various authors in Lithuania. All the data we have used on owl diet have been published. Review of the sources is given in [50].”

Comment Line 126: Do you mean 7 x 14 cm? Please change. AND What kind of traps are these, live traps or killing (snap) traps? If killing traps were used, where are the reference specimens stored in Lithuania?

Answer: yes, we used snap traps, necessary changes done. There is was and is no requirement to collect reference specimens in Nature Research Center. We have only a collection of skulls of small mammals.  

Comment Line 135: change to: "trap sites" ?

Answer: We intended to meant not only trapping, but also pellet collection and case study sites. Thank you for your observation, word “sites” was missing. Now changed to “Schematic maps showing the locations of small mammal trapping, pellet collection and case study sites”

Comment Line 153: Please italicize

Answer: done

Comment Line 169: change to:  "this species"

Answer: done

Comment Line 170: change to: "preyed on"

Answer:  changed

Comments Line 178: use either n, or N not both in figure AND please italicize

Answer: italicized, but, based on https://languagetips.wordpress.com/category/nn-and-sample-size/

See what the American Medical Association Manual of Style had to say: N: total number of units (eg, patients, households) in the sample under study;  n: number of units in a subgroup of the sample under study.

So we refer total number of trapped small mammals by N, and that of M. minutus by n

Comment Line 184: delete “and”

Answer: we consulted about the “and” and leave it as is. Word “seemingly” deleted after the same consultation.

Comment Line 193: please italicize

Answer: done

Comment Line 195: replace with: "capture"

Answer: replaced

Comment Line 201: delete "the"

Answer: deleted

Comment Line 209: replace with "captures" ("trappings" has a completely different meaning in English)

Answer: changed with “records”. We rely on https://wikidiff.com/trap/capture?utm_content=cmp-true, saying “As verbs the difference between trap and capture is that trap is to physically capture, to catch in a trap or traps, or something like a trap while capture is to take control of; to seize by force or stratagem.”

We believe, that “capturing” is a good word for livetraps, but not killing traps.

Comment Line 214: Can the Latin species names be set in italics in this figure?

Answer: done

Comment Line 219: change to: "eastern"

Answer: changed

Comment Line 320: change to: "a lack"

Answer: “a” added

Comment Line 342: do you mean: "modern harvesting techiques"?

Answer: yes, thank you

Comments Line 381: change to: "the long term" AND change to: "the Nature Research Centre"

Answer: changes done

Comment Line 425: Either spell out all journal titles or abbreviate all. Consistency!

Answer: according https://images.webofknowledge.com/images/help/WOS/A_abrvjt.html , not all journal names have abbreviations. We rely on this source and the technical check of the publisher

Comment Line 440: Correct to: "Zeitschrift"

Answer: done, thank you

Comment Line 507: Correct to: "Westfälischen"

Answer: done, thank you

Comment Line 513: Use lower case consistently in English titles.

Answer: original source was exactly as presented. We rely on the Publisher here.

Thank you very much for your detailed reading and language suggestions. With the exception of two, all have been acknowledged with explanations of the rebuttals.

Round 2

Reviewer 1 Report

I read the authors' responses. I am not convinced of everything reported, but the role of the referee is to highlight pitfalls in the methods and analyses. If they think their analyses are robust enough, that's fine. In any case, these are data on a little-studied species; therefore, useful, and readers will be able to judge what is worth it.

However, regarding this author's comment, Monitoring small mammals across several decades using the same sites, habitats, and trapping efforts is only possible with dedicated financing. We have no examples of such monitoring in Europe; therefore, thank you for your comment.

I have to say that the authors are not very informed on what is going on in Europe because there are small mammal projects across several decades using the same sites, habitats and trapping effort without dedicated financing using volunteer work, for instance, in Spain and the UK.

In Spain SEMICE small mammal monitoring program (www.semice.org; (Torre et al., 2021). The program monitors common species with high detectability (Torre et al., 2018), ensuring no sampling biases of population and occupancy estimates (Mackenzie et al., 2018).

See also Flowerdew et al. 2003 for a more extended project in the UK. Live trapping to monitor small mammals in Britain. Mammal Review. Volume 34, Issue 1-2 p. 31-50

Reviewer 3 Report

The authors have improved the text in several places. They have also explained the uniqueness of the data and its value to the region.