Fidelity to Natal Tributary Streams by Kokanee Following Introduction to a Large Oligotrophic Reservoir
Round 1
Reviewer 1 Report
This is a nice, well-written study on an interesting topic. I have made a number of comments using sticky notes on the pdf and do not need to repeat them here. There are some places (marked, and similar ones elsewhere) where a bit of "word economy" is possible but this MS is in good shape. However, I must state for the record that I am expert in neither the details of otolith chemical analyses, nor in the statistical methods. So, my review is at a somewhat higher level. Mostly, I think the study could explain the life history of this species/form better and earlier in the paper. I think some of the many papers indicating stream-specific homing by sockeye salmon or kokanee as inferred from genetic tools would be useful, to establish the principle that they home to the stream where fry emerge, not the river where smolts leave, as one might believe from studies on other species. I am happy to discuss this with the authors if they wish but this is a suggestion rather than a requirement.
Comments for author File: 
Comments.pdf
Author Response
General Comments
Response: We appreciate the comments and outline below how we have modified the manuscript to incorporate the suggestions. We have also moved information into the introduction to describe the life history of Kokanee and movement to spawning locations and fry emergence downstream to lakes.
SPECIFIC COMMENTS
Line 14: insert ", non-anadromous sockeye salmon" here for the benefit of those not knowing this term
Response: Done
Line 24: please clarify where and how the adults were caught, as "fish" here is a bit vague
Response: We have changed “fish” to “spawers”, but not indicated “where and how” as this would add too much detail to the sentence.
Line 24: how many streams were samples in this manner for chemistry and kokanee?
Response: We have indicated that homing was assessed in four tributaries to the reservoir.
Line 48: again, I'd add "non-anadromous sockeye salmon" here to help readers not familiar with this term.
Response: Done
Line 55: from where? The natural colonization is not clear to me.
Response: Kokanee occur in two known locations in the watershed; Arctic Lake in the headwaters of the Parsnip River in the south and Thutade Lake in the headwaters of the Finlay River in the north. Our genetic analysis has revealed that the native Williston Kokanee likely colonized from Thutade Lake. We had added “headwaters” to this section as it is not clear how long the fish were actually in the lower Finlay before formation of the reservoir.
Line 56: can delete "showed an"
Response: Done
Line 107: Might you indicate when spawning occurs and when fry are likely to migrate to the lake? Some readers may not know that fry of this species do not tend to feed in streams.
Response: As suggested below, we have moved a paragraph from the discussion into the introduction to clarify these points.
Line 230: I think "emergence from the gravel" is better, as the gravel itself is not emerging!
Response: Excellent – done
Line 231: not clear what is meant by "novel" - perhaps delete the word?
Response: Novel has been removed
Line 314: I think "estimated" would be better than "determined" here.
Response: Done
Line 367: can just say "occurs" rather than "has been documented to occur"
Response: Done
Line 384: This paragraph is clearly needed but I think it belongs much earlier in the paper, either the Introduction or Methods, as it helps non-salmon experts get the sequence of the life history which is key to this study (e.g., no extended rearing in the stream by fry).
Here, it would be good to indicate the months when these events occur in these populations, as "fall" is as early as July for Alaska sockeye and as late as January farther south.
Response: We have followed the suggestions.
Line 486: This is a lot of detail but can the authors perhaps step back and indicate roughly what distance from the likely natal stream the strays came from? More likely from more proximate streams or not? Similar in physical features or not, etc.
Response: The reviewers comment is well taken and we really like the suggestion, however, the streams chosen for the analysis were based on their differences from the other streams in the watershed and unfortunately there was considerable overlap in elemental signatures among many of the putative locations where the fish may have strayed from. It is therefore, not possible to estimate distances that the fish might have strayed from. This would make a really cool study, but there does not appear to be enough distinct stream elemental signatures to use this approach to estimate straying distance.
Reviewer 2 Report
- ICPMS can measure many trace elements of the otolith that is very useful to discriminate the natal origin, but you only analyze a few major and minor elements, Ca, Sr, Ba, Mn and Mg, in this study, why? Sr and Ba are ok for discriminating the diadromous fishes migrating between freshwater and seawater. To discriminate the microhabitat of the natal origin, it is suggest to use more kinds of elements particularly the trace elements.
- Ca, Sr, Ba, Mn and Mg were measured as described in the material and method, but Mg was not shown in the result, and Zn was shown in the result but its measurement was not mentioned in M & M, why?
- Mn and Mg are non-conservative elements, easily affected by metablic rate and fish physiology. It is not suggested to be used as environmental indicator.
- Anyway try to add the trace elements that would be possible to increase the assignment % of the discrimination function because Sr and Ba only are not enough to discriminate the natal origin of the fish from different streams and tributaries of the same river because the difference in elemental composition of the water is very small among stream and tributaries of the river.
- It is suggested to use the element:Ca ratio rather than absolute value for analysis in this study because relative value is more acurate and precise than absolute value for the measurement of ICPMS.
Author Response
COMMENTS
- ICPMS can measure many trace elements of the otolith that is very useful to discriminate the natal origin, but you only analyze a few major and minor elements, Ca, Sr, Ba, Mn and Mg, in this study, why? Sr and Ba are ok for discriminating the diadromous fishes migrating between freshwater and seawater. To discriminate the microhabitat of the natal origin, it is suggest to use more kinds of elements particularly the trace elements.
Response: The reviewer is correct in that many studies have used more elemental signatures and we actually have data for more elements. This certainly adds power to the analysis when discriminating among watersheds for water elemental signatures. The approach has been extended to otolith microchemistry, but we were cautious in using elements in the analysis that do not show a direct or strong relationship between water elemental composition and otolith microchemistry such as Mn and Mg. The relationships for Ba and Sr between water and otoliths are highly significant, whereas Mn was below detection in two of the streams where we had Kokanee spawners, Russel and Stevenson. We conducted the discriminant function analysis with Sr, Ba and Mn and did find a modest improvement; 70% using Sr and Ba and 73% using Sr, Ba and Mn. The improvement was due to otolith signatures with higher Mn being assigned to the Osilinka that had low Ba; a characteristic for some of the streams in the larger watershed, but not the Osilinka River specifically. As we saw an improvement in regional assignment, we have included the 3 elements in the otolith analysis, but not the water analysis as Mn was below detection for some of the streams examined.
- Ca, Sr, Ba, Mn and Mg were measured as described in the material and method, but Mg was not shown in the result, and Zn was shown in the result but its measurement was not mentioned in M & M, why?
Response: We have removed the reference to Mg since it was not incorporated into the analysis, however at line 158 in the original manuscript we indicate that we measured Zn.
- Mn and Mg are non-conservative elements, easily affected by metablic rate and fish physiology. It is not suggested to be used as environmental indicator.
Response: We agree with the reviewer’s comments and this is the why they were not included in the analysis. Mn, however, is useful because the core has such high values.
- Anyway try to add the trace elements that would be possible to increase the assignment % of the discrimination function because Sr and Ba only are not enough to discriminate the natal origin of the fish from different streams and tributaries of the same river because the difference in elemental composition of the water is very small among stream and tributaries of the river.
Response: The large differences throughout the Williston Reservoir for Sr and Ba and the strong relationship between water elemental signature and otolith microchemistry drive the spatial patterns. Inclusion of Mn in the otolith analysis slightly improved our ability to discriminate natal origin and consequently was included in our analysis for the revised submission.
- It is suggested to use the element:Ca ratio rather than absolute value for analysis in this study because relative value is more acurate and precise than absolute value for the measurement of ICPMS.
Response: We used the elemental ratio to Ca for all of our analyses. This is stated at lines 114 and 162 of the original manuscript.
