Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Natural isotope markers in salmon


To optimize conservation and restoration strategies for fish stocks it is important to identify the rearing habitats that produce the most successful individuals. However, tracking millions of migratory fish through multiple life stages seems to be impossible using conventional techniques. Using differences in the ratio of stable isotopes of strontium (Sr), found naturally in stream water, we have been able to distinguish juvenile Atlantic salmon (Salmo salar) from eight of ten rearing sites studied in Vermont streams. This demonstrates the potential for using environmental signals to determine the rearing stream from which salmon originate, and should help to guide restoration activities.


Seven million Atlantic salmon fry (about 30 mm long) are stocked annually into tributaries of the Connecticut River in the United States. Virtually all adults in this restoration effort are captured before returning to their rearing streams and are used in breeding programmes. Therefore, there is currently no way to determine which rearing stream produced a returning adult salmon. Because the fry are too small to be physically tagged before stocking, we investigated the use of stable isotopes of Sr, which seem to provide an environmental signature1, to distinguish salmon from different rearing streams.

The isotopic composition of Sr (87Sr/86Sr, expressed as δ87Sr; Table 1) dissolved in stream water reflects that of the watershed geology2,3,4. Given the geological diversity across our study region, we expected Sr isotope ratios to vary significantly among tributaries. Within a site, stream-water δ87Sr remains relatively constant across seasons and years5,6. Fish take up Sr isotopes in direct proportion to their availability in water, and substitute Sr for calcium in calcified tissues during growth7. Because the Sr isotope ratio does not change during uptake and assimilation, δ87Sr of fish bones and otoliths (ear-stones) should be the same as that of the stream water8.

Table 1 Strontium isotope compositions in West and White River drainage basins

Earlier attempts to identify natural fish markers have focused on differences in elemental concentrations, for example, distinguishing migratory from non-migratory fish populations using the large difference in Sr concentrations between marine and fresh water9,10. Isotope ratios provide a much more sensitive method than elemental concentrations for distinguishing multiple populations11.

We collected fish and water from 10 salmon stocking sites in two major watersheds of the Connecticut River. Three months after stocking, the vertebral δ87Sr signature in the salmon was the same as the signature of the stream where they were collected. During this time the salmon undergo explosive growth, and the Sr signal of hatchery fry (~4‰) is lost because of exchange and accumulation of Sr from the rearing tributary. We found only minor seasonal fluctuations in stream-water δ87Sr values (Table 1).

The δ87Sr values in salmon aged 0 and 1 years from the same sites did not differ, so we pooled all fish from each site and found a strong correlation between the δ87Sr value in the stream water and that in the tissues of 90% of fish. We believe that the most likely explanation for the two ‘mismatched’ fish is that they migrated from an isotopically distinct site. Thus, our results suggest that few juvenile salmon (87Sr signal.

On the basis of δ87Sr values in salmon backbones and water, we separated fish from 10 rearing sites into 8 distinct groups (Table 1 ). There was significant overlap of δ87Sr between only two pairs of sites. For one pair (two sites separated by less than 3 km), the upper site drains into the lower site. The second pair of sites lie further apart but seem to drain geologically similar catchments.

Finally, the δ87Sr values of sagittal otoliths closely matched the vertebral signature (Table 1). Otoliths grow in discrete annuli which can be isolated and used to trace the age-specific chemical history of a fish12. In returning adult salmon, ocean-derived Sr will be in the outermost otolith bands. Because the world's oceans have no measurable spatial or temporal variability in δ87Sr, ocean-derived Sr in returning adult salmon can be reliably and easily separated from the freshwater signal. Thus we should be able to distinguish between stocks on the basis of their freshwater signature and to estimate tributary-specific contributions to smolt and adult salmon production.


  1. Koch, P. L. et al. Earth Planet Sci. Lett. 108, 277–287 (1992).

    Google Scholar 

  2. Fisher, R. S. & Stueber, A. M. Wat. Resour. Res. 12, 1061–1068 (1976).

    Google Scholar 

  3. Graustein, W. C. & Armstrong, R. L. Science 219, 289–292 (1983).

    Google Scholar 

  4. Åberg, G., Jacks, G. & Hamilton, P. J. J. Hydrol. 109, 65– 78 (1989).

    ADS  Article  Google Scholar 

  5. Miller, E. K., Blum, J. D. & Friedland, A. J. Nature 362, 438– 441 (1993).

    ADS  CAS  Article  Google Scholar 

  6. Bailey, S. W., Hornbeck, J. W., Driscoll, C. T. & Gaudette, H. E. Wat. Resour. Res. 32, 707–719 (1996).

    Google Scholar 

  7. Simkiss, K. in The Ageing of Fish (ed. Bagenal, T. B.) 1-12 (Unwin, Old Woking, 1974).

    Google Scholar 

  8. Graustein, W. C. in Stable Isotopes in Ecological Research (eds Rundel, P. W., Ehleringer, J. R. & Nagy, K. A.) 491-512 (Springer, New York, 1989 ).

    Google Scholar 

  9. Kalish, J. M. Fish. Bull. US 88, 657–666 (1989).

    Google Scholar 

  10. Limburg, K. E. Mar. Ecol. Prog. Ser. 119, 25–35 (1995).

    Google Scholar 

  11. Chamberlain, C. P. et al. Oecologia 109, 132– 141 (1997).

    Google Scholar 

  12. Campana, S. E. & Neilson, J. D. Can. J. Fish. Aquat. Sci. 42, 1014–1032 (1985).

    Google Scholar 

  13. Hodell, D. A., Mueller, P. A., McKenzie, J. A. & Mead, G. A. Earth Planet. Sci. Lett. 92, 165–178 (1989).

    Google Scholar 

Download references

Author information

Authors and Affiliations


Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kennedy, B., Folt, C., Blum, J. et al. Natural isotope markers in salmon. Nature 387, 766–767 (1997).

Download citation

  • Issue Date:

  • DOI:

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing