Comparison of Otolith Readability and Reproducibility of Counts of Translucent Zones Using Different Otolith Preparation Methods for Four Endemic Labeobarbus Species in Lake Tana, Ethiopia

Round 1

Reviewer 1 Report

I have reviewed the manuscript “Comparison of age determination methods for the 2 endemic Labeobarbus species in Lake Tana, Ethiopia” and recommend it be accepted for publication in Water after moderate revision. The paper is well written and uses standard techniques and analyses. Below you will find specific comments the authors should consider.

Line 16: “Nevertheless” is incorrectly used here.

Line 19: Not clear why studying methods gives “good insight into the age of fish worldwide.”  Use the plural form “fishes” here.

Line 21: Best methods to age what?  Be specific with species and the most applicable methods are species-specific.

Line 21. What kind of samples? There is an issue with logical sequencing in the abstract.

Line 23: I think it would be informative to list the methods examined. At the very least, state what method one was.

Line 33: Is flock the correct term?

Line 35. What is the most important species? Perhaps edit sentence to “… are the most economically and ecologically important fishes in Lake Tana.”

Line 36: Change to “behaviours”. Change “species” to populations”.

Line 38: Capitlaize “Red List”.

Line 41. What conservation measures have been taken?

Line 43. Be specific about whose life history traits are being referred to. As written it implies the life history traits of the authorities, which is clearly not the intention.

Line 47. Which parameters?

Line 48: Change “profile” to “structure.” I would suggest making this change throughout the manuscript.

Line 49: Insert “age of” before “maturity.”

General comment: In order to determine which method is best it will be necessary to validate.

Line 61. Should this state “fishes within the genus”?

Line 74. How were the sampling sites systematically designated?

Line 76: Inset “all” before “four sites”.

Line 78. Delete from and they.

Line 78 and 78. These are confusing. Were all nets the same?  Why is the word “monofilaments” used rather than gill nets? Why were there two different setting times? Was “set” in line 79 meant to be retrieved?

Line 79. Need to add “The” at the beginning of the sentence.

Line 80. Change to captured. Need to add “The” at the beginning of the sentence.

Line 84. Need to indicate what other pertinent data were collected (e.g. length, sex, weight, etc.)

Line 85. As written it states that an ice box was used for otolith extraction.

Lines 89-90. Edit for clarity. Suggest deleting “each” and inserting “from each species” after “collected”.

Line 100. What were those time intervals?  Was this controlled?

Line 101. Following which “same procedures”.

Line 103-105. How was the effect of glycerol and water examined by placing otoliths in one or the other? This is not clear and should be revised.

Line 108. How were they cleaned?  Sodium hypochlorite?

Line 109. Was the resin sprayed or poured?

Line 115. What was the solution?

Line 133. A figure including images of otoliths prepared using the three methods would be informative.

Line 135-136. This would be better suited earlier in the methods.

Line 140. Would translucent zones be light and opaque zones be dark?

Line 141. Use plural form annuli.

Line 145. Change “to” to “in’.

Line 155. What are those limitations? Should the plural form of agreement be used here?

Lines 157-159. Place after formulas.

Line 165. Without validation you are actually examining degree of error in growth increment counts rather than age. I would state earlier in the methods that you assumed one pair of growth bands were deposited annually.

Line 172. State test ANOVA was used to determine if there were there was a significant difference among …

Line 175. What symmetry test was used?

Line 176. Add “among” before “aging”.

Line 181. Edit to “…. R1 as the independent variable.”

Line 186. Used plural form annuli throughout manuscript where appropriate. Paired “annulus counts” implies that two translucent zones were used meaning two years. I am not sure this sentence says what was intended and needs to be revised.

Line 190. Dunn’s test.

Line 192. Needs revision for clarity. How were there reading to discard if an otolith was unreadable or missing. This is confusing.

Line 199. What is meant by “hard to follow around most of the sections”?

Lines 203-210. This section needs to be revised for clarity. As written, there seem to be a number of contradictions and redundancy. For example, the section begins by stating there were no differences for two of the species. A couple sentences later it is stated that there were differences for all four species.

Figure 2. Edit first line. Fish percentage is a confusing term.

Table 1. Be specific what significant difference is being referred to.

Line 221. I don’t understand the intent of this sentence.

Line 222. More detail is needed here. Are the authors referring to the samples or population?  If the latter, then will need to show that the population was representatively sampled. It would be informative to include length frequency histograms of the fish sampled. Also a scatterplot of each species showing length and age should be included so a reader can assess if the assigned ages are reasonable.

Line 223. How were potential differences in variability tested?

Figure 3. Need labels on axes.

Line 245. I would not consider PA of 69% to be relatively high. This should be thoroughly discussed in the discussion section. Based on this section and Table 2, I do not think the authors can state they have found a reliable ageing method, especially in the absence of validation.

Line 259. This is counterintuitive in light of the low PA.

Line 288. Change from “between” to “among.”

General. Did the readers look at samples beforehand for calibration purposes? The age bias plots bring the reliability of all methods used into question. Just because one was a little better does not mean it is good.

Lines 310- 311. How is this so in light of the low PA?

Line 312. Figures 4 and 5 indicate differently.

General. Authors need to test to determine if assumptions of parametric statistics were met. This should be included in the methods and results.

Line 347. Insert “experience level” before “age readers”.

Line 350. Also need validation and verification studies. If the structure does not accurately record age of a fish then all else is essentially meaningless for these purposes.

Line 352. Validating ages of fishes has been done for many species and is not always practically difficult. Also need to mention validation is needed across all age classes.

Line 360. Problems similar to what?  Be specific.

Lines 360-364. These sentences do not logically flow and seem like unconnected statements. This section should be revised.

Line 368. This statement is a bit strong as the PA was what many would consider low.

Line 377. I do not think age reading can be underestimated as they are what they are. Perhaps the authors meant to say age reading underestimated actual age, which goes back to the validation issue. The authors should consider rearing specimens in captivity to further test their hypotheses. At the least, this should be recommended as a direction for future research.

Line 396. How was it determined that rings were “false”?

Line 400. It needs to be clearly stated that the periodicity of growth band formation needs to be examined before a structure or method can be conclusively considered the most accurate.  

Appendix A. Figure is too large and last panel cannot be viewed.

Author Response

Thank you very much for the editorial corrections comments and valuable comments and suggestion. We have incorporated all the comments accordingly.  

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Alexandra,

Thank you for the opportunity to review the article by Gebremedhin et al. entitled “Comparison of age determination methods for the endemic Labeobarbus species in Lake Tana, Ethiopia”. Conserving endemic fish species, particularly those in freshwater systems that are under stress from our warming, drying climate, is an important issue. This manuscript has the potential to provide the much-needed ageing data for Labeobarbus species in Lake Tana that is required to understand growth and mortality rates enabling better management of the fisheries that exploit these species. However, as it stands, this manuscript has one large, glaring problem, the lack of age validation.

The title of this paper indicates that there is some sort of age determination carried out, but there has not. What has been done is a comparison of the suitability of different ageing structures/methods and comparisons of the counts of the number growth zones (unvalidated) between multiple readers in those structures/methods. I do not have any issues with what the authors have done in this regard, it is the fact that they have then gone on to say that the counts of growth zones equate to age estimates of the fish (lines 221, 306, 368 and many more), when there is no validation to show the periodicity of the formation of those zones, which is even acknowledged by the authors (line 163). The authors are also aware of the errors in age determination can result from issues surrounding periodicity of growth zone formation (lines 345-347). The term ‘age estimation’ that is used throughout the manuscript in relation to the results should be changed to ‘number of growth zones’. The last paragraph of the Discussion (lines 396-402) is further indication that validation needs to be carried out. It is concerning that false rings, incomplete annuli and general difficulty in reading some of the structures was encountered, yet no validation was carried out. In the methods, the authors state that samples were collected monthly over a 12-month period (lines 75-76), so I don’t not understand why validation was not carried out?

Validation is a key component to any ageing study (read any Campana paper to see this) and can not be skipped. The current manuscript could be repackaged as methods paper on the preparation of ageing structures and reader comparisons, but it will lose a lot of impact as this has already been done for Labeobarbus species (Winker et al. 2010; Ellender et al. 2012; Gerber et al. 2012). A comparison of the age, growth and mortality rates of the four congeners would be far more interesting and valuable.

As it stands, I don’t think that this paper is worthy of publication unless revised substantially.

Below are some other comments regarding the text.

Abstract.

Line 24. It would be useful to say what “Method one (MO1)’ is in the Abstract as we don’t find out until the methods section. A lot of people will only read the Abstract so this is really important.

Introduction.

Lines 43-45. The statement that biological information on Labeobarbus species are ‘virtually non-existent’ is completely incorrect, unless the authors are saying that biological information for such species in Lake Tana is non-existent, but this is not clear. I found a few with a quick Google Scholar search (i.e. Winker et al. 2010; Ellender et al. 2012; Gerber et al. 2012). What is even more strange is the fact that the authors have actually referred to some these papers later in the text??

Materials and Methods.

Line 64. What is the importance of the ‘source of the Blue Nile’?

Figure 1 Labelling needs to be changed so that ‘a’ is first.

Line 112. The otolith section thickness is far too thick, and thus not surprising that sections were not very good for reading. The authors should consider thinner sections, ~0.3 mm.

It would be nice to include a figure that includes images of the same asteriscus for each of the reading methods.

Results.

Figure 2 is showing “The percentage of otoliths at each confidence ranking……”

Table 1. heading. Not sure what the last line is referring to (lines 218-219). There are no subscript letters.

Line 221. Age determination was not applied to the four fish species. This line needs to be reworded.

Figure 3. No y-axis or x-axis labels!! Figure heading should not include the words “age readings”.

Lines 257-259 are not necessary.

 

References

Ellender, B. R., Weyl, O. L. F., & Winker, H. (2012). Age and growth and maturity of southern Africa's largest cyprinid fish, the largemouth yellowfish Labeobarbus kimberleyensis. Journal of Fish Biology, 81(4), 1271-1284.

Gerber, R., Smit, N. J., & Wagenaar, G. M. (2012). Age, growth rate and size at sexual maturity of Labeobarbus aeneus (Teleostei: Cyprinidae) in the middle Vaal River, South Africa. African journal of aquatic science, 37(1), 49-58.

Winker, H., Ellender, B. R., Weyl, O. L., & Booth, A. J. (2010). Validation of growth zone deposition in otoliths of two large endemic cyprinids in Lake Gariep, South Africa. African Zoology, 45(1), 133-138.

Author Response

Thank you very much for your valuable comments and suggestions. We have thoroughly incorporated them.

Author Response File: Author Response.docx

Reviewer 3 Report

This manuscript deals with the age determination methods for endemic Labeobarbus species of the Lake Tana in Ethiopia. The approach was to collect asteriscus otoliths from large size samples of four species and read the age using six preparation techniques in order to determine the best aging method. In the lack of knowledge of the age of the populations of these species due to the unavailability of an appropriate assessment method, the study covers a particularly interesting aspect that can help to fill this gap. But with some regret to be too methodological study that should associate the aging methods with growth parameters (e.g. length, sex). This should help the lake managers and the fisheries biologists to better understand the population dynamics of these fish species stocks and therefore to implement the appropriate management measures to ensure the sustainability of Labeobarbus species. Overall, I feel that the manuscript may suitable for publication in Water. However, before it is considered for publication, I would encourage the authors to consider specific comments and suggestions listed below aiming the manuscript improvement.

Line 24. Here should be described in a few words the MO1 as the best aging method identified.

Line 46-56. The question here is how to make your method of age reading using asteriscus otoliths less invasive and more life-saving for fish and to interest managers in the effective management of fish stocks.

Line 58-61. This approach is hard to follow because your aging methods are all invasive and require the sacrifice of fish. Similarly, there is no mention of the reasons for the choice of this approach, including the limitations and benefits of using non-invasive methods for fish in the Lake Tana.

Line 74-75. The four sampling sites are well presented in Figure 1 but why did you select them? The reason given is not convincing.

Line 76-77. What do you mean by “We used 10 multi-mesh gill nets consisting of 6, 8, 10, 12 and 14 stretched mesh sizes”.

Line 76-77. What is the selectivity of the size of fish catch from your fishing nets?

Line 79-83. Identification of Labeobarbus spp. to only species level for age determination limits the implication or interpretation of your results in stock management at the lake manager level. The morphometry (e.g., total length or fork length) of the individuals caught, combined with aging methods, may be more relevant.

Line 144. Why these important data, especially those related to the fish size, are missing throughout the text.

Line 82-84. A brief morphological description of these four selected species should be provided in the material and methods section.

Line 144-146; line 191; line 201-203. The reader’s confidence scale in the readability of otoliths can be simplified in the text and legend of figure 2 as well as in table 1 and legend (see in the figure 2). Please try the following scale: easily readable (Q1), moderately readable (Q2) and unreadable (Q3). The letter A in front of Q can be removed.

Line 220-227. These results of one year of data collection should be presented by site while specifying the selectivity of the capture gear. An opportunity was missed because they should provide information on the biological conservation status of the species at each sampling site given the large surface area of the lake.

Line 69. Figure 1. Please change the order of the figures. And adapt this change in the legend of figure. Please put (a) (Ethiopia in Africa) and (b) (Lake Tana in Ethiopia) on the left and (c) (the Lake Tana watershed) on the right.

Line 225-227. In Discussion: It is also necessary to discuss the significance of your results in the context of the viability of the population of the selected species. For example, What do you mean the predominance of age readings and maximum age in age distribution plots from different between species.

Line 289 Figure 4, line 298 Figure 5, line 432 Appendix A and B. Please improve the resolution of the figures. They are not easily readable.

Line 353-420. It is a little regrettable to reduce this study of great scientific value to the simple dimension of aging methods. Once the best method of reading aging has been determined, a relationship between this method and the reliable growth descriptor (e.g., fork length) of the fish caught for this discovery should be tested. This should help the lake managers and the fisheries biologists to ageing without killing fish and to better understand the population dynamics of these Labeobarbus species (e.g., maximum age, age structure, age or size at maturity) and thus to take the necessary measures to promote their sustainability.

Author Response

Your comments and suggestion were very helpful to improve the quality of the manuscript. Thank you very much for that and we have incorporated them. 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear Alexandra,

In my initial review of this paper, I suggested that the authors needed to consider what it was that they were trying to achieve, i.e. providing counts of growth zones or providing actual ages, as there were lots of inconsistencies between what was done (counts of growth zones) and what was being said in the text (age estimates). As such, I was reluctant to highlight all the changes I thought were necessary as this major issue required fixing first.

In this re-review of the paper, for which I have given considerable time, I find that little has been done in addressing this issue. The authors still do not shown an understanding of the difference between counts of growth zones and ageing. The former is simply a count of growth zones in a hard structure, while ageing requires knowledge of 1) spawning time, 2) when growth zones are formed,3)  capture date and 4) validation for growth zone formation. Throughout the manuscript the authors use the terms “aging” in reference to counting growth zones and “aging methods” for the different methods they are using. “aging” should be changed to “counts of opaque zones” and “aging methods“ should be changed to “methods”. I have not highlighted all those instances where this needs to be addressed, as they are to numerous, and it is something that the authors should be doing, but haven’t.

At lines 225-226, the authors have included a new statement that the growth zone deposition has been validated, citing their own work in another paper that is currently under review. If the growth zones have been validated as annual, why would you not then use the calculated decimal ages for each species in the analyses present in this paper, instead of just growth zone counts?

In the attached PDF, I have provided many comments and suggested changes that should be considered before this manuscript is accepted. Currently, I don’t feel that the manuscript is of a quality worthy of publishing.

Kind regards PC

Comments for author File: Comments.pdf

Author Response

Thank you for your in-depth comments and suggestions. We have incorporated all your comments and suggestions and they have been really helpful to improve the quality of the manuscript.

Author Response File: Author Response.docx