Pacific salmon don’t feed in fresh water – everybody knows that, right?
Wrong. Brian Morrison, fisheries biologist and author of several recent posts on fish lifecycles in Western Alaska, is back today to tell us about some recent research findings that prove conventional wisdom wrong.
Salmon Feeding in Fresh Water?
Many anglers have long known that mature Pacific salmon (Oncorhynchus spp.) will strike at objects that resemble eggs while in fresh water. Concurrently, there has been a deeply entrenched belief held by both scientists and anglers that mature Pacific salmon cease feeding in fresh water, with numerous explanations having been developed to explain this behaviour, such as ‘instinct’, ‘aggression’, or a fish being ‘territorial’. It has even been stretched as a way for adult salmon to eliminate potential competitors that their offspring might encounter as juveniles. As a much simpler explanation, Pacific salmon continue to feed because it provides them with nourishment.
A group of researchers set out to test whether Pacific salmon do in fact feed, and try to unravel some of the myth around why they may feed. They looked at the feeding on eggs within adult Chinook Salmon, Chum Salmon, and Coho Salmon. Overall, 13% of stomachs examined contained eggs (up to 30% observed for Coho Salmon), and feeding rates were estimated at up to 14 eggs per day. Feeding experiments in a lab revealed that mature salmon could digest eggs, as fed salmon maintained significantly higher body mass than unfed salmon. The researchers in this study determined that the energy from consumed eggs could potentially allow salmon to migrate up to an additional 3.8 km per day of feeding, or extend the duration of spawning activity by up to 12%.
The researchers found that the energetic gains associated with egg consumption in Pacific salmon may thus be particularly important for precocial males (jacks). Alternatively, one of the most surprising findings was that large individuals were no less likely to feed than their smaller counterparts. The highest feeding prevalence was observed in adult female Coho Salmon from the Quinsam River, BC, in which 40% of the sampled females had consumed eggs. Importantly, female salmon may continue to feed in fresh water not only to offset the relatively higher costs of breeding that they incur, but to obtain physiologically important resources, such as carotenoids and dietary protein.
Although there is still a lot of uncertainty on why adult Pacific salmon will feed while in fresh water, the researchers conclude; given that Pacific salmon stocks show high homing to their stream of birth, it is possible that many populations have evolved local adaptations to their specific migration and spawning requirements that include nutrients from salmon eggs from their own species or other salmon species. Such a dependence on supplemental energy from freshwater feeding could explain why some populations fail to recover even when released from fishing pressure, as lower abundances of salmon (all species) may remove feeding opportunities that some populations and species depend on to successfully complete migration and spawning.
Want even more details? Here’s a link to the original research paper.
Also worth noting that a very small minority of scientific papers elicit direct rebuttals … and this paper elicited a direct rebuttal questioning a few key assumptions and conclusions of the study linked above
Armstrong, JB. 2010. Comment on ‘‘Egg consumption in mature Pacific salmon (Oncorhynchus spp.)’’ Can. J. Fish. Aquat. Sci. 67: 2052–2054
Abstract: In their recent article, ‘‘Egg consumption in mature Pacific salmon (Oncorhynchus spp.)’’ (Can. J. Fish. Aquat. Sci. 66(9): 1546–1553), Garner et al. observed salmon eggs in the stomach contents of mature Chinook, coho, and chum salmon. Through experimentation and simulation of salmon energetics, the authors conclude that the observed feeding represents
an important gain in energy, challenging the accepted paradigm that assumes salmon energy budgets contain no gains once adults return to freshwater. Here, I argue that Garner et al. have overestimated the energetic consequence of egg consumption and that the observed consumption rates do not represent biologically significant gains in energy.
In many freshwater systems with fishable returns of Pacific salmon, one can find anglers attempting to lure these fish with salmon eggs or imitations thereof. In the recent publication, “Egg consumption in mature Pacific salmon (Oncorhynchus spp.),” Garner et al. (2009) provide scientific documentation of what anglers have long known: adult salmon occasionally eat eggs. However, as many disappointed anglers can attest, Garner et al. found that egg consumption by adult salmon was generally rare. The authors explored the biological consequences of the observed levels of consumption through experimentation and simulation of salmon energetics, concluding that their results provide a “fundamental shift in our understanding of Pacific salmon ecology,” which previously assumed that energetic gains were negligible once adult salmon return to fresh water. Instead, Garner et al. claim that the energy from egg consumption helps salmon travel farther and survive longer during spawning migrations. While I find this work interesting and thought provoking, I challenge the authors’ claim that salmon feeding substantially alters the energy budgets of migrating salmon.
The energetic consequences of egg consumption are overstated by the authors’ analysis. The energy density assumed for salmon eggs, 18.7 kJ·g–1 wet mass, is roughly two times larger than reasonable estimates of this parameter. Garner et al. cite Meka and Margraf (2007), who did not actually measure this parameter, but reference an unpublished Master’s thesis (Eastman 1996), which also did not measure the parameter. While it is not clear how this estimate was derived, it is clearly inaccurate. Hendry and Berg (1999) researched salmon energetics in the same populations studied by Eastman and found a mean energy density of 7.8 kJ·g–1 wet mass for ripe eggs (standard deviation = 0.5). A survey of the literature reveals that this parameter varies from approximately 7 to 12 kJ·g–1 across different species of Oncorhynchus and Salmo (Table 1). Accordingly, Garner et al.’s estimate of the energy density of salmon eggs, and thus their conclusions regarding energetics consequences for egg predators, are exaggerated by about two times.
Regardless of the energy density of salmon eggs, it is not clear that Garner et al.’s observed consumption rates represent meaningful feeding or energetic gains. The average feeding rate Garner et al. observed in adult salmon was less than 1 egg·day–1. This value is decidedly low when placed into context with other fishes that exploit Pacific salmon eggs (Scheuerell et al. 2007; Denton et al. 2009; Armstrong et al. 2010). For instance, adult rainbow trout (Oncorhynchus mykiss) and Dolly Varden (Salvelinus malma are regularly found with hundreds of Pacific salmon eggs in their stomachs (Fig. 1) (Denton 2008; T.A. Jaecks, Alaska Department of Fish and Game, Sport fish Division, 546 Kenny Wren Road, Dillingham, Alaska, unpublished data). In contrast, the highest ration reported by Garner et al. was 14 eggs·day–1 and appears to occur infrequently, as 87% of the salmon surveyed did not have eggs in their stomach. The authors suggest that these small rations are biologically meaningful because the fish in their experiments consumed a similar number of eggs per day and were able to maintain their body mass (losses < 0.03% body mass·day–1). However, the fish in the authors’ feeding experiment weighed 1/7 as much as the fish they sampled in the wild (mean mass = 565 and 4200 g, respectively), so the mass-specific rations of observed and experimental fish were not similar. In fact, bioenergetics analysis (Hanson et al. 1997) suggests the authors’ maximum observed feeding rate (14 eggs·day–1) was less than 1/4 of the ration required for a 4200 g coho salmon (Oncorhynchus kisutch) to maintain body mass (68 eggs·day–1 in 7.5 °C water). While the authors’ experimental results convincingly suggest that adult salmon can digest and metabolize eggs, they do not suggest that the low rations observed in wild adult salmon were biologically significant.
Using the average feeding rate observed by Garner et al. (1 egg·day–1) and an energy density for salmon eggs of 9 kJ·g–1, the average adult salmon consumed only 2.4 kJ·day–1, producing 1.6 kJ·day–1 of available energy (accounting for indigestible portions and egestion according to Garner et al.). Following the energetics approximations used by Garner et al., a 5 kg salmon feeding at the average observed rate would gain an additional 3 min of survival per day on the spawning grounds. Feeding activity itself likely increases energetic costs, increases predation risk, and decreases the ability to find and defend mating sites. The egg consumption observed by Garner et al. is unlikely to increase energy or fitness in biologically significant ways and actually reinforces the assumption that salmon, in effect, stop feeding upon reaching freshwater.
Thanks for sharing that Steve.
The response does not refute that Pacific Salmon indeed consume food during their spawning migration, just the authors estimates about the energetic benefits and extent to which it occurs. The jury is out on these two points, but ultimately, the fundamental paradigm in which we’ve viewed these species has been shaken. The authors also submitted a response to the rebuttal:
Good and interesting article. However, how many of the fish referenced in this study of hatchery origin or otherwise artificially propagated? H fish are, of course, aberrant on any and every level studied.
Any inclusion of H fish in the study will minimally require their exclusion and a reevaluation of your percentages.
Thanks for the provocative article.
Arthur Frederick says
I have heard that a small percent of salmon never leave freshwater. These salmon live their lives in freshwater and do not mature in the manner of their saltwater siblings. Is this true?