How do escaped farmed salmon affect their wild cousins? Four major research institutions are now applying their respective expertise to building a common, fact-based knowledge platform on this vital topic.
Four key research institutions in Norway have established the knowledge platform "Quantifying genetic effects of escaped farmed salmon on wild salmon (QuantEscape)" to find out what actually happens when farmed salmon escape. The institutions are the Norwegian Institute for Nature Research (NINA), the Institute of Marine Research (IMR), the Norwegian Institute of Food, Fisheries and Aquaculture Research (Nofima) and the Centre for Integrative Genetics (CIGENE).
"We are working to increase the knowledge needed to ensure that farmed and wild salmon can coexist," says Kjetil Hindar, Research Director at NINA, who is in charge of research collaboration on the QuantEscape platform.
Heredity and environment
Using the best genetic tools available, the QuantEscape researchers are seeking answers about how much genetic influence the escapees have on wild salmon and the extent to which the offspring of farmed salmon adapt in the wild as a result of natural selection.
"Even though farmed salmon are selectively bred for growth traits, they do not always grow faster than wild salmon when they are out in nature," explains Dr Hindar. "Their ability to compete for survival is affected by the pressures of the natural environment. The offspring of escaped farmed salmon may possibly exploit their growth potential when food is plentiful – but when it is scarcer, their bred capacity for rapid growth does not appear to be an advantage."
Quantifying gene flow
Dr Hindar emphasises that research shows that escaped salmon have a negative impact on wild salmon, but that much is still unknown about the ways in which the escaped salmon affect wild stocks.
"We don't know the current true scope of hybridisation within wild stocks, or the long-term biological impacts of escapees on wild stocks. For the QuantEscape platform we will be quantifying the gene flow from farmed to wild salmon in a large number of rivers, and then calculating the relative reproductive success of escapees versus wild salmon. From there we will conduct experiments to measure the extent to which natural selection slows the gene flow from escaped salmon."
Differing tolerances to environment
The research collaboration also seeks to identify the river conditions and wild salmon traits that are significant for gene flow from escaped salmon to wild salmon, as well as how wild traits are influenced by the addition of escapee genes.
"We know that there are genetic and ecological differences between Norwegian wild salmon stocks, and significant differences in how strong the populations are. Escapees are larger than the local wild salmon in some rivers but smaller than wild salmon in others. We also believe that environmental factors in the rivers play a significant role in escapees' ability to reproduce naturally. In areas where we know there are escaped salmon, we find farmed genes in certain rivers but less introgression (gene flow between species) in other rivers nearby."
Better resource management
The research director is convinced that this broad approach to gene flow challenges will culminate in new insights that will be valuable for not only resource management but also the aquaculture and sports fishing industries.
"The knowledge we produce through QuantEscape will enable us to develop good models for wild salmon stocks that absorb escaped farmed salmon. These models would give us greater certainty in predicting both the long-term impacts of escaped salmon and the success of various management measures on mitigating those impacts. The models will be useful tools for sustainably managing production salmon and wild salmon," concludes Dr Hindar.
Detecting gene flow from escapees
Researchers know more than ever about the genetic profile of Norwegian salmon, both wild and farmed. This enables them to identify gene flow from farmed salmon to the wild salmon of Norway's rivers.
In 2012 alone, researchers generated genotype data from over 5 000 individuals from more than 60 stocks of wild and farmed salmon. This data enables researchers to determine, with relatively high certainty, the degree of introgression (gene flow between species) from farmed genotypes in many of Norway's major salmon rivers.
Researchers study genotype data of salmon stocks by employing a set of genetic markers with "snips" (SNP = Single Nucleotide Polymorphism), which are small genetic mutations found throughout the salmon genome. In a project under the now-concluded National Programme for Research in Functional Genomics in Norway (FUGE), researchers pored through more than 4 500 snips to find just the right ones that distinguish farmed salmon from their wild cousins, regardless of an individual's production lineage or wild stock. They narrowed down this diagnostic set of snips by comparing four year-classes of farmed salmon from the three largest selective breeding companies, using 13 wild stocks from throughout Norway. To ensure that the genomes of the wild salmon being compared for this purpose were in all likelihood not altered by introgression from escaped salmon, the researchers used scale samples collected in the 1970s and 1980s.
Samples from Norway, north to south
The researchers have now analysed scale samples from wild salmon specimens caught in a wide variety of salmon rivers, from Finnmark county in the north to Vestfold county in the south. In addition they have analysed a number of production stocks, using the same genetic methods on samples from farmed individuals today and back to 20 years ago. There is also a wealth of scale samples available from wild salmon from the early 1900s and from production fish from the late 1980s through today.
Using the genotype data from the "original" wild salmon stocks as a baseline, the researchers will now investigate changes in wild salmon genes through both natural selection and introgression from farmed fish.
"Based on the identifications we have done, we will be able to assist the wild salmon gene bank by excluding any stored material that has been influenced by farmed genes," adds Dr Hindar.
Mapping the threat of farmed genes
The researchers at NINA have now developed a map based on preliminary stock modelling for a large number of rivers in Norway. The model incorporates records of escaped farmed salmon found in rivers during summer and autumn from 1989 to 2009 as well as figures for the reproductive success of escapees and survival rates of their offspring in the wild. The figures are derived from field experiments in Norway and Ireland. The model estimates stock changes over a period of four years (equal to one salmon generation in the model); the researchers can then let a computer run the model's calculations through the desired number of salmon generations.
The map provides a preliminary risk overview of the cumulative effects of escaped farmed salmon over the 20-year period in selected rivers along the Norwegian coast. At this point, rivers in Western Norway look to be most affected.
"In QuantEscape," continues Dr Hindar, "we are now comparing the model's depiction of the threat with analyses of genetic material from salmon in the selected rivers. Preliminary results indicate that the model generally provides useful descriptions of conditions."
"But at the same time we see that the model projects different outcomes for genetically based measurements of farmed salmon hybridisation. While the model's projections for some rivers reflect the genetic changes caused by escapees, they are misleading for other populations. Certain wild stocks seem to be less affected than the escape figures would suggest." He stresses that the model is still general and will be continually fine-tuned as new knowledge is accumulated under the knowledge platform. The consortium of research institutions is also working with several models that can describe the impacts of escaped salmon on genetic variation and traits of wild salmon and the development of wild stocks.
Low success in the wild
In rivers, escapees' reproductive success rates are moderate at best. Their offspring tend to have lower survival rates than wild salmon offspring.
"Problems arise for the wild salmon primarily where there is a high proportion of escapees among the spawning stock, and where the local wild stock is under stress from other pressures," explains Dr Hindar.
It is substantially more difficult to predict the effects of low proportions of escapees in the long run, both because escapees do change genetically over time and because different production lineages can have different effects. Moreover, the field experiments conducted have extended no further than the second generation of offspring.
In the longer term, the researchers will be developing more powerful genetic models, and will search for snips that distinguish better between escapee offspring groups and generations in the wild. The researchers also hope to know more about genetic changes for traits that are essential for wild salmon.
"But first we will publish the results of a large-scale study using the exciting genetic tool we now have," says Dr Hindar.
Explore further: New model helps explain how provisions promote or reduce wildlife disease