Dam Construction Impacts Fish Biodiversity in a Subtropical River Network, China

Round 1

Reviewer 1 Report

Water impediment due to dam construction is a major threat to riverine diversity, especially the freshwater fishes. As a result, studies on how the dams can affect freshwater fish ecology can contribution to conservation biology and help in environmental impact assessments projects. The current manuscript is therefore an interesting study and important for informing conservation management activities. There are, however, some issues that authors will have to address in a revision.

1. Lines 15-16: author follow the words “taxonomic homogenization” with “increased similarity” in parenthesis. However, it is more than just increased similarity. It is loss of beta diversity due to loss of endemic rare taxa and increase in the invasive introduced taxa leading to increased similarity. Just mentioning “increases similarity” does not convey the negative connotation associated with “taxonomic homogenization” used in conservation ecology. Author should rephrase this to highlight it better.

2. Line 18: Replace “has” with “have” before “changed”.

3. Line 21: It should read”…29 native species are lost, while there is a gain of 6 alien species.”

4. Line 22: 1% increase during what span of years?

5. Line 27: “current trajectories” of what? Biodiversity change/loss? 

6. Lines 36-39: Another major ecological impact of dams is the hindrance to local migration, especially for fish that show altitudinal migration for breeding. 

7. In the methods, author should make it clear how they have treated recent taxonomic descriptions. If the same species was known with a different name earlier and now it is known by a different name, have they updated the historical list accordingly? 

8. Figure 2B: Since authors are providing family names for every second bar, the figure makes little sense. The highest alpha is likely to be for Cyprinidae (as per line 147) but that family name is not even mentioned on the x axis. Provide vertical family names and for all bars. 

9. Figure 2C: Comparison of IUCN redlist statuses for before and after dam construction hardly make any sense. The authors must have used the current IUCN statuses so for both the historical and current bars. So it makes no sense to make a comparison. For instance, the only reason there are slightly more NE for current species composition is because these might be recently described species that were not known in historical times. The bar graph gives wrong impression between historical and current times. For example, there appears to be less DD species earlier and now there are more, but this could be just because there are some taxonomic changes in the recent times. 

10. Figure 3 is not a very effective way of looking at the familywise alpha diversity in river sites and basins. I will suggest that instead of using the percentage values authors can plot the actual number of species in various families and across the rivers/basins using correspondence analysis biplot. Authors can do this in the free software PAST (https://www.nhm.uio.no/english/research/infrastructure/past/)

11. Line 167: Provide F values with degrees of freedom and actual value of P and not just P < 0.05. 

12. Figure 5 is not much informative. Another way authors can show this data is by showing the map (similar to figure 1) and then drawing an arrow pointing to the river sites and then showing the pie chart with the IUCN statuses at the end of the arrow. This way we know which part of the rivers, what is the threat status distribution. For the basins, authors can show larger basins and pie charts for them in the same manner.

13. Table 2: There are too many statistical tests that authors have conducted. So they need to account for familywise error rate (some tests will be significant just by chance alone). Alternatively, authors can perform a multivariate analysis. In this case authors can take all the geographical and environmental variables as multiple independent variables and the diversity related variables (provided in various columns) as multiple dependet variables and perform Partial Least Squares (PLS) or redundancy analysis (RA). An example of PLS can be found in – Riyas et al. 2021, Marine Biology Research 17, 185-199.

Author Response

Manuscript ID diversity-1739215

Dam construction causes fish biodiversity change in a subtropical river network, China

Dear Reviewer 1,

Thank you for your thoughtful and constructive of our manuscript. We have made revisions to the manuscript as suggested by you. Below are the responses to the comments. The responses are in red text.

Water impediment due to dam construction is a major threat to riverine diversity, especially the freshwater fishes. As a result, studies on how the dams can affect freshwater fish ecology can contribution to conservation biology and help in environmental impact assessments projects. The current manuscript is therefore an interesting study and important for informing conservation management activities. There are, however, some issues that authors will have to address in a revision.

Lines 15-16: author follow the words “taxonomic homogenization” with “increased similarity” in parenthesis. However, it is more than just increased similarity. It is loss of beta diversity due to loss of endemic rare taxa and increase in the invasive introduced taxa leading to increased similarity. Just mentioning “increases similarity” does not convey the negative connotation associated with “taxonomic homogenization” used in conservation ecology. Author should rephrase this to highlight it better.

Reply: This was revised. Changes in pairwise compositional dissimilarity among regions, calculated as Δβsor = βsor-historical– βsor-current, indicates either taxonomic homogenization (i.e. positive Δβsor) or taxonomic differentiation (i.e. negative Δβsor) (Olden & Rooney, 2008). Therefore, We found evidence for fish faunal homogenization in the Gan River Basin, with a slight (1%) increase in taxonomic similarity among river basins from the historical period to the current period.

Line 18: Replace “has” with “have” before “changed”.

Reply: Done.

Line 21: It should read”…29 native species are lost, while there is a gain of 6 alien species.”

Reply: Done.

Line 22: 1% increase during what span of years?

Reply: Done.

Line 27: “current trajectories” of what? Biodiversity change/loss?

Reply: Done.

Lines 36-39: Another major ecological impact of dams is the hindrance to local migration, especially for fish that show altitudinal migration for breeding.

Reply: Done.

In the methods, author should make it clear how they have treated recent taxonomic descriptions. If the same species was known with a different name earlier and now it is known by a different name, have they updated the historical list accordingly?

Reply: We used FishBase (Froese and Pauly, 2016) to update all scientific names of fish species to accepted nomenclature in the historical and current list.

Figure 2B: Since authors are providing family names for every second bar, the figure makes little sense. The highest alpha is likely to be for Cyprinidae (as per line 147) but that family name is not even mentioned on the x axis. Provide vertical family names and for all bars.

Reply: Done (See Figure 2B).

Figure 2C: Comparison of IUCN redlist statuses for before and after dam construction hardly make any sense. The authors must have used the current IUCN statuses so for both the historical and current bars. So it makes no sense to make a comparison. For instance, the only reason there are slightly more NE for current species composition is because these might be recently described species that were not known in historical times. The bar graph gives wrong impression between historical and current times. For example, there appears to be less DD species earlier and now there are more, but this could be just because there are some taxonomic changes in the recent times.

Reply: Deleted Figure 2C.

Figure 3 is not a very effective way of looking at the familywise alpha diversity in river sites and basins. I will suggest that instead of using the percentage values authors can plot the actual number of species in various families and across the rivers/basins using correspondence analysis biplot. Authors can do this in the free software PAST (https://www.nhm.uio.no/english/research/infrastructure/past/)

Reply: The assemblage structure of fish was analyzed using the non-metric multidimensional scaling (NMDS) and hierarchical cluster analysis of species composition (See Figure 3).

Line 167: Provide F values with degrees of freedom and actual value of P and not just P < 0.05.

Reply: Done.

Figure 5 is not much informative. Another way authors can show this data is by showing the map (similar to figure 1) and then drawing an arrow pointing to the river sites and then showing the pie chart with the IUCN statuses at the end of the arrow. This way we know which part of the rivers, what is the threat status distribution. For the basins, authors can show larger basins and pie charts for them in the same manner.

Reply: Done (See Figure 5).

Table 2: There are too many statistical tests that authors have conducted. So they need to account for familywise error rate (some tests will be significant just by chance alone). Alternatively, authors can perform a multivariate analysis. In this case authors can take all the geographical and environmental variables as multiple independent variables and the diversity related variables (provided in various columns) as multiple dependet variables and perform Partial Least Squares (PLS) or redundancy analysis (RA). An example of PLS can be found in – Riyas et al. 2021, Marine Biology Research 17, 185-199.

Reply: We added redundancy analysis (RDA) in the manuscript (See Figure 6).

Author Response File: Author Response.docx

Reviewer 2 Report

GENERAL COMMENTS

The manuscript focuses on a topic that is great interest for the readers of Diversity. The manuscript is overall well-written though some grammatical improvements are needed. If possible, please let you manuscript be revised by a native English speaker. Regarding the scientific component, several details are lacking on the Material and Methods (see details below on Specific comments) that need clarifications.

SPECIFIC COMMENTS

L32 – Dams are by definition, artificial structures, I suggest removing “artificial”.

L96-99 – You repeat “Gan River” in this sentence, 5 times”. Please write improve readability.

L104 – Refer in the caption what the red symbols represent. I don´t see the river sites on the map, only the dams (I suppose the red symbols represent the dams).

L107 – What criteria did you use for the search? What database(s)? Web of science? What were the key-words or topics used for the search? Note that methods description need to be reproducible.

L109 – Please remove this sentence or provide tools to support it (see my comment above).

L117-125 – Please include the description/method to collect each for these environmental variables? Once again these details are needed to allow reproduction of methods. GIS? Multiparametric probes (provide brand and model), measured on fields? With which devices (brand and model)…

L153-155 – Does this refers to the historical or actual period?

L190-194 – This information is not clear on figure 5. What do (A) and (B) represent? It is not said on the caption of Figure 5. What is the meaning of the acronyms on the x-axis of both 5A and 5B? It is not said on the caption, either.

L191 – “these imperiled species”. Which ones? Those not-evaluated as you refer before on L190? Not clear.

L204 – were instead of was.

L213 – Discussion instead of Disscussion.

L254-255 – This sentence is not clear, please re-write clearer.

L272-274 – What stressors could be? Based on your results, what future lines of research would you recommend?

L282-284 – This is not a recommendation (it confounds that reader in that construction of small and large hydropower plants should be promoted), and should be removed (you already said this earlier on the Introduction).

Author Response

Manuscript ID diversity-1739215

Dam construction causes fish biodiversity change in a subtropical river network, China

Dear Reviewer 2,

Thank you for your thoughtful and constructive of our manuscript. We have made revisions to the manuscript as suggested by you. Below are the responses to the comments. The responses are in red text.

GENERAL COMMENTS

The manuscript focuses on a topic that is great interest for the readers of Diversity. The manuscript is overall well-written though some grammatical improvements are needed. If possible, please let you manuscript be revised by a native English speaker. Regarding the scientific component, several details are lacking on the Material and Methods (see details below on Specific comments) that need clarifications.

SPECIFIC COMMENTS

L32 – Dams are by definition, artificial structures, I suggest removing “artificial”.

Reply: Deleted.

L96-99 – You repeat “Gan River” in this sentence, 5 times”. Please write improve readability.

Reply: Done.

L104 – Refer in the caption what the red symbols represent. I don´t see the river sites on the map, only the dams (I suppose the red symbols represent the dams).

Reply: This was revised.

L107 – What criteria did you use for the search? What database(s)? Web of science? What were the key-words or topics used for the search? Note that methods description need to be reproducible.

Reply: This was added.

L109 – Please remove this sentence or provide tools to support it (see my comment above).

Reply: This was added.

L117-125 – Please include the description/method to collect each for these environmental variables? Once again these details are needed to allow reproduction of methods. GIS? Multiparametric probes (provide brand and model), measured on fields? With which devices (brand and model)…

Reply: We calculated a number of basin descriptors for each river and a suite of water chemistry parameters in the current period from published online databases, books and literature (Zou, 2011; Su et al., 2012; Liu et al., 2017; Guo et al., 2018; Liu et al., 2019; Wang et al., 2019; Tong et al., 2019; Shi et al., 2020; World Clim-Global Climate Data (http:// www.worldclim.org/bioclim); Bureau of hydrology in Jiangxi Province (http://www.jxssw.gov.cn/)).

L153-155 – Does this refers to the historical or actual period?

Reply: This was revised.

L190-194 – This information is not clear on figure 5. What do (A) and (B) represent? It is not said on the caption of Figure 5. What is the meaning of the acronyms on the x-axis of both 5A and 5B? It is not said on the caption, either.

Reply: This was revised.

L191 – “these imperiled species”. Which ones? Those not-evaluated as you refer before on L190? Not clear.

Reply: This was revised.

L204 – were instead of was.

Reply: This was revised.

L213 – Discussion instead of Disscussion.

Reply: This was revised.

L254-255 – This sentence is not clear, please re-write clearer.

Reply: This was revised.

L272-274 – What stressors could be? Based on your results, what future lines of research would you recommend?

Reply: This was revised. In light of the mounting threats to China’s freshwater fishes (Xing et al., 2015; Liu et al. 2017), we offer the following two recommendations. First, we call for systematic conservation planning efforts that seek to efficiently select a comprehensive and representative set of areas for conservation management to ensure the long-term persistence of fish biodiversity. Second, changing socio-economic activities and climate change is only further intensifying the risk of species invasion risk across China (Liu et al., 2019; Li et al., 2021). We call for enhanced biosecurity frameworks that incorporate the risks from species invasions on fish biodiversity, including improved monitoring programs for invasive alien species.

L282-284 – This is not a recommendation (it confounds that reader in that construction of small and large hydropower plants should be promoted), and should be removed (you already said this earlier on the Introduction).

Reply: This was deleted.

Author Response File: Author Response.docx