Climate change and human activities impact living creatures in complex ways. In response, organisms may acclimate to those challenges during their lifetime or adapt over several generations, through genetic changes. Organisms with characteristics matching best their new environment, such as size at reproduction, will spread more their genes to the next generation. With the development of new technology, we can now associate genes with organisms’ characteristics (size, diseases…) and track how the frequency of these genes changed over time in populations. Those methods can be used, for instance, for tracking the spread of new COVID-19 variants, which can evade immunity and/or be more contagious. This exercise is more difficult for organisms with long generation times, such as vertebrates, as it requires historical samples from which DNA can be extracted and genetic variants determined. 

In River Teno, today Atlantic salmon spend less time at sea, feeding and growing, before returning to their natal river to spawn. We have previously reported that it was the consequence of changes, over the last 40 years, in the frequency of the gene variant associated with late age at maturity (vgll3 L). Thus, it was the result of adaptive evolution (https://doi.org/10.25250/thescbr.brk191). However, the environmental drivers of age at maturity evolution were unknown. We hypothesized that salmon sea environment changed and lowered the survival of salmon maturing late, influencing the frequency of the associated gene variant. We also investigated the consequence of sea and riverine fisheries which can target salmon differently according to their size (and age).  

We linked changes in the maturation gene frequency over a 40-year period with changes in sea temperature, the abundance of prey at sea, and fishing efforts at sea and in the river. We also used a sonar to gather data about the number and size of salmon ascending River Teno and compared them with the size of salmon caught with different fishing methods. It allowed us to determine how size-selective were the different fishing methods and how it impacted salmon carrying the different vgll3 gene variants. 

First, we found that evolution of age at maturity in River Teno was driven by the abundance of capelin, a small fish eaten by salmon. When prey at sea is scarce, salmon are likely to have a lower survival rate, and it particularly impacts those that need to stay longer at sea before being mature. Capelin abundance collapsed three times over the last 40 years because of predations by other species and overfishing. We showed that capelin harvesting indirectly impacted salmon and made them evolve to mature earlier, at a smaller size. The small omega-3 rich forage fish has been harvested mainly for feeding farmed animals, including farmed salmon. Our discovery illustrates a rare example of an indirect effect of human activity on a wild animal, strong enough to change their genetic material over several generations. 

Second, we found that riverine net fishery was, overall, targeting early maturing salmon. This was due to a traditional net fishing method (weir) with small mesh sizes and used in shallow water, where small salmon swim. Other net fishing methods were targeting preferentially large salmon but captured lesser numbers. As fishing pressure has decreased in the last decades, early maturing salmon survival to fishing likely improved.  It has been known for a long time that fishing can have evolutionary consequences on the targeted fish, but we lacked clear evidence of the impact at the genetic level before our findings. 

Salmon are exposed to many threats by humans, direct and indirect (fishing, aquaculture, dams…). Yet, we observed that they have an incredible ability to adapt, which should lower the negative impact of those threats, and may help them to persist in their new environment. Still, we are keen to understand why the abundance of Atlantic salmon, and many other wild fish, has been declining worldwide over the last decades. It remains critical to determine the drivers of those changes, quantify their impacts, and try to find solutions. Wild fish are important sources of proteins and lipids in aquafeed formulations. The relative amount of wild fish required to produce farmed fish has decreased over the last decades due to progress by aquaculture operations. However, aquaculture production has considerably increased and will continue to do so in the future. About 18 million tons of wild fish are currently used to feed farmed animals, with consequences going beyond the targeted species. Progress by aquaculture operations must continue to achieve sustainable aquaculture.