UC Davis

January 10, 2003

Review of Final Report "Documenting Biodiversity of Coastal Salmon (Oncorhynchus spp.) in Northern California” by Hedgecock et al. Dec. 2002

By BERNIE MAY

General Comments

Above all this report describes very valuable research in salmonid genetics, with special relevance to those of Northern California. SCWA clearly received their money's worth over the past five years from the efforts of the Hedgecock laboratory. The comments detailed below are queries and suggestions for the overall improvement of this final report.

In general the report is very clear and well organized, although the text features (capital letters, boldface, underlining, italics, etc.) used to indicate hierarchy of headings and subheadings can be confusing. Hierarchical numbering would improve the readability of the text and use of the TOC.

Good explanations of basic genetic principles like H- W equilibrium and FST are given and used effectively throughout the report. If any of this material is still unclear to SCW A, I encourage them to ask for further clarification. This is a great opportunity to get genetic concepts explained.

The use of their computer program Kinship to remove the bias of related individuals from juvenile samples is well explained. Removing this bias is a significant contribution by this group of investigators to population genetic studies of any organisms.

The report does an excellent job of addressing the objectives outlined in the SOW with a few exceptions. The section on "Alternative male-types" was underdeveloped. The attachment of three vague abstracts provides limited useful descriptions of the experiments and results. In the objectives for the coho study it is stated that proteinencoding gene and mtDNA markers will be developed as well as micro satellite loci; why this was not done should be mentioned more extensively. Also in the coho objectives, it was stated that estimates of migration rates among and effective population sizes of spawning runs would be given, although only the effective population size of the green Valley Creek samples are presented. Some of these issues were resolved in the 2001 report, but probably should be repeated in this Final report since they are pertinent to the SOW.

A final overall section on "Future Needs" or "Recommendations" should probably be added that are directed towar5ds SCW A and not simply scientific interest.

Specific Comments

TITLE:

While "biodiversity" was used in the SOW, the results described are primarily about salmonid "genetic diversity".

SUMMARY:

Steelhead work is mentioned in the introductory sentence but not discussed further in the summary. The discontinuance of the several portions of the SOW (e.g., steelhead work, historical samples, mtDNA, phylogeny) should be mentioned in the summary.

Possible admixtures are offered as an explanation for the deviations from random mating equilibrium. How do you distinguish the effects of admixtures from the effects of stock transfers?

"Diversifying effects of genetic drift" (bottom of 1st full paragraph on page 3)awkward wording since you associate genetic drift with a loss of heterozygosity. Perhaps 3 and 4 could be addressed in the discussion of genetic principles.

POPULATION GENETICS OF COASTAL CALIFORNIA COHO SALMON POPULATIONS:

Introduction

In the list of objectives, number 4 would be better split into two objectives.

One of the conclusions of the report is that the genetic data corroborates the previously established ESUs for Coho the U.S. Pacific coast. While it is certainly beyond the scope of the report to describe how the ESUs were established in the first place, it would be nice to see a reference to the document (possibly a Federal Register) that describes the ESUs for Coho. Additionally a map showing the geographical distribution of the ESUs would be helpful.

Materials and methods

Page 5 fix wording so as not to use "(p+qi=Np2+2Npq+Nq2" which is not mathematically correct.

8. Pg. 6- last paragraph; the following sentence describing F ST could be more clear. "The genetic correlation between gametes drawn from different demes or subpopulations, with respect to the allelic frequencies in the total population, is given by FST, the ratio of the variance of allelic frequencies among subpopulations to the maximum." Perhaps the above sentence would be clearer if the term "maximum" were replaced wi the phrase "variance in allelic frequencies among all subpoulations".

9. Fig. 1 legend should indicate that sample sizes are found in Table 2.

10.Fig. 1 and Table 2 should be included in the text after page 12 (the page where they are first mentioned).

11.Page 8 under "Micro satellite DNA markers": The text states that there were 67 microsatellites for testing, while the table refers to 69.

12.Top of page 13: remove the parentheses around the decimal values ... otherwise, it appears that you're multiplying. Try "1/10 or 0.1"

13.In determining significance of F ST values multiple tests were done iteratively by testing a group of populations, removing the most divergent one, and then retesting the group until none of the remaining populations were significantly different. Some discussion of adjusting P values (i.e. Bonferroni corrections) for this kind of testing should be included, since adjustments would greatly effect which populations are excluded and included from the final set.

14.Define LOD abbreviation on page 14.

15.Page 15, KlGHA description: N=15, not 13.

16.Page 16, ESPRS99: Should the populations be 68-92 and 96-11O?

17.Page 17 , MATS: What biological data supports the initial division of the samples collected 517-5/11 and the samples collected 5/l2-5/l6?

18.Page 18, RRGV98b: It might be of concern that some of the same individuals from 98a may be represented in the 98b population. Was a test for homogeneity of these two populations performed? It's interesting that both 98a and 98b have the same proportion of loci pairs showing significant associations.

Results

Color coding (or alternate acronyms) of populations by ESU/river in Figs. 6-8 would help the reader see the geographic relationships presented.

Table 3: Does NA signify no amplification?

An estimate of the number of breeders was done for the Green Valley samples. Why wasn't this done for any other populations to see if genetic drift can explain the random mating deviations? Alternatively, why were the temporal variation data not used to estimate Ne?

Page 30, last paragraph: It is stated that the phylogenetic trees correspond to the designated ESUs. However, the CC and SSF clusters is not supported by the bootstrap analysis. Therefore, I am unclear how the distinction between those 2 ESU s can be confirmed.

Pg. 30- I believe they mean Fig. 8 and not Fig. 9 in the last paragraph on this page.

Discussion

Under Departures from random mating equilibrium in CA coho salmon populations, third paragraph, it is suggested that the Wahlund effect probably doesn't explain the observed deviations from random mating equilibria due to small spatial and temporal scales involved. Give examples and references that show temporal and spatial homogeneity in other salmonids populations (some researchers have shown significant heterogeneity on small spatial scales (i.e. within tributaries) for east coast U.S. Atlantic salmon).

Under Departures from random mating equilibrium in CA coho salmon populations, fifth paragraph, some examples with references regarding temporal genetic heterogeneity would be helpful.

STOCK ORIGIN ESTIMATES FOR CHINOOK JUVENILES CAPTURED IN THE RUSSIAN RIVER

Results

Results are written in the present tense, while results of the coho work were presented in past tense.

Page 40, Results: Table 10 should be referenced, not Table 2.

Since there are no historical samples of Russian River fish, derivation of the current Russian river fish seems unknown. The possibility for inclusion of some hatchery component would still seem feasible. It would seem that more coastal chinook populations should be included in an analysis before any conclusions are drawn.

DEVELOPMENT OF GEOGRAPHIC INFORMATION SYSTEMS

Testing a GIS map server was interesting. However, I was unable to use the web site at all. Even trying to retrieve a document gave the response "The document that you have requested is currently being processed or updated. Please check back later." Use of these technologies to map genetic data with biogeography will certainly extend our understanding of the effects of environmental variation on the numbers of salmonids and consequently genetic variation. The discussion is clear and extensive about the value of these tools and ways they can be developed and extended. However, this is a general scientific discussion. Specifically, what should SCW A do now? Should they be involved? How should they be involved? Who will maintain the existing mapserver?

Final Note: This is an excellent example of a report from a university investigator to a funding agency. It is unusual to have these reports reviewed. That being said, such reviews should be common occurrences.