Tuesday, July 26, 2016

Fishing-induced evolution


How fish have changed my understanding of evolution, and how they brought me to the field of eco-evolutionary processes, particularly to study how species-specific factors can shape fishing-induced selection


When I started studying evolution, about 20 years ago, things were relatively simple; individuals differed in some phenotypic trait and environmental conditions would fuel natural selection and select for the most successful individuals in such an environment. It had been well described about a hundred years ago, by a famous English bearded naturalist named Charles Darwin, and it was sufficiently simple that it could be applied to everything I knew in biology.
I later discovered that it was a little more complex, as it is in fact genes that respond to selection, and not directly individuals, thanks to another English evolutionist, a hairy one, Richard Dawkins [1]. Nothing really perturbing when you preach to the choir, just a slight scale adjustment.

But after that, I started to work on fish and everything changed…

First, I discovered phenotypic plasticity; which means that your environment could completely modify your appearance and that one genetic combination could result on a vast array of traits. I particularly appreciate how the owner of this blog illustrates that phenomenon by regularly modifying his facial hair template (sorry Andrew, but you knew somebody would eventually collect all these pictures…).

Beside this Homo sapiens case, one of the most fascinating examples of phenotypic plasticity is how trout (brown trout or rainbow trout) can stay in their native river and become resident, or can move downstream (in lakes, sea, or ocean) and become enormous in a few years. Even more dramatic, also debatable if this is truly plasticity, some clownfish will change their sex when the dominant female dies. For the young evolutionary biologist that I was, it implied an immense adjustment to my system of thoughts, and it had big consequences on how selection and environmental processes interacted. Sometimes, it’s not the phenotype that is selected for, but how well the phenotype can be adapted to be more fit.  Studies on bichir fish (Polypterus sp.) have even suggested that plasticity could be a source of evolution [2,3], i.e., that when individuals experience a dramatic environmental change (such as migrating from sea to land), their body could change without mutation involved (in this case they learned to walk), and natural selection will only then favor mutations to fix this transition.

Image source
Another consequences of plasticity, sympatric speciation, brought me to the next puzzling conundrum implying evolving fish. Sympatric speciation means that from one founding, relatively homogeneous, population, several new species can flourish in the same environment. Meaning that in a particular environment, several paths could be taken by evolution to reach a stable state, and from one initial population, different genes combinations could be best adapted to such particular environment. This is very puzzling, as individuals sharing the same space shouldn’t become sexually separated, but still there is plenty of work, and particularly on cichlids, that demonstrate its reality [4]. These cichlids evolved into many different species, and this process could even be replicated in different lakes!


Much closer to me, both in terms of location and species of interest, is the adaptive radiation of alpine whitefish [5]. Almost every lake in Switzerland has several species of whitefish, usually one fast growing species specializing on larger preys (benthic fauna), and one slow growing species specializing on smaller preys (zooplankton). These “species” are still very close genetically, and can interbreed under some conditions, so they should be considered an intermediate step between morphs and species. Interestingly, the genetic distance between these populations is smaller between two morphs in the same lake than between the same morphs in two different lakes. Most stunningly, this evolution towards different morphs happened so recently, that it shouldn’t be even conceivable. Lakes in Switzerland were frozen during the last glaciation, some 10’000 years ago, so this specialization was extremely rapid (evolutionarily speaking). This question of rapid evolution was one of the biggest revelations of my early career, evolutionary processes did not have to wait millions of years to be meaningful, but instead could be as rapid as ecological processes! I guess that you have a good idea of where I am going… Are there any interactions between both? And if yes how does it work? I spent the last ten years trying to understand these questions by studying how environmental factors could shape the response to fishing-induced selection.

During my PhD work in Switzerland, I measured change in growth through time in several Alpine whitefish populations, and contrasted it with estimations of selection differentials on growth. I found out that both ecological (phosphorus change) and evolutionary processes (fishing selection) where involved in the growth decreases observed. However, and very interestingly, the contribution of both processes was very variable. Fishing-induced selection could explain between 20 and 70 percent of the change observed in growth and this contribution was mediated by ecological processes, such as the trophic state of the lake, and species-specific processes, the type of life-history (slow or fast growers).

This interaction between harvesting, species-specific parameters, and fishing-induced evolution has been modeled by Erin Dunlop, Anna Eikeset and Nils Stenseth [6], and we found it so interesting that we spotlighted their research in the latest TREE edition [7]. Their model explores how the interaction between species life-history speed and selection for earlier maturation can impact the population growth rate and therefore the stock productivity. They contrasted the outcomes of fishing on three harvested fish species, characterized by different life histories: Atlantic cod (Gadus morhua) with a “slow” life history, lake whitefish (Coregonus clupeaformis) with intermediate” life history, and yellow perch (Perca flavescens) with “fast” life-history. They used individually-based simulations that modeled the fisheries-induced evolutionary change of the probabilistic maturation reaction norm (PMRN), which describes the maturation probability given a certain age and size. In particular, they measured how the evolvability of the maturation parameters, the fishing intensity, and the strength of density dependence interacted to influence population growth rate.

Their study describe that, under the influence of fishing, the population growth rate follows what we called a transient dynamic, i.e., fishing had a reducing effect on population growth rate when fishing starts (with the removal of the largest individuals), and that reduction is followed by a recovery as density-dependent processes lead to increase growth (overcompensation). When fishing stops, after the introduction of a moratorium for example, they observed a reversed transient dynamic, i.e., an increase in growth rate due to the suppressed harvesting of the largest fish, followed by a compensatory density-dependent response when the population reaches carrying capacity. This transient dynamic after fishing cessation was highly influenced by fishing-induced evolution, but in a species-specific way. Species with slow life-histories can “adapt” to fishing by maturing at an earlier age, which reduces the risk of collapse during fishing but also induces a genetic legacy that impairs recovery during moratoria. On the other hand, for species with fast life histories, they observed a reduced impact of FIE, as fast growing species were already reproducing as early as possible. The consequences for these species was a faster recovery during moratoria, due to the reduced genetic legacy, but at the cost of an increased risk of collapse at high harvest rates as such species could not adapt to fishing.

Such impact of species-specific factors might explain the discrepancy observed in studies investigating the response to size-selective fishing; or why some species with slow life histories that have experienced high harvest rates have shown slow recoveries even after several years of fishing moratoria. Understanding how and when evolutionary change matters are key management questions but are problematic as each population can respond differently. I believe that disentangling how environmental and species-specific factors influence the response to size-selective fishing is the next step necessary to develop robust management policies.




  1. Dawkins, R. (1989) The selfish gene (rev. ed.), Oxford: Oxford University Press.
  2. Standen, E.M. et al. (2014) Developmental plasticity and the origin of tetrapods. Nature 513
  3. Hutchinson, J. (2014) Evolutionary developmental biology: Dynasty of the plastic fish. Nature 513
  4. Seehausen, O. (2004) Hybridization and adaptive radiation. Trends Ecol Evol 19
  5. Vonlanthen, P. et al. (2009) Divergence along a steep ecological gradient in lake whitefish (Coregonus sp.). J Evolution Biol 22
  6. Dunlop, E.S. et al. (2015) From genes to populations: how fisheries-induced evolution alters stock productivity. Ecol. Appl. 25
  7. Nusslé, S. et al. (2016) When Should Harvest Evolution Matter to Population Dynamics? Trends Ecol Evol 31(7)





A 25-year quest for the Holy Grail of evolutionary biology

When I started my postdoc in 1998, I think it is safe to say that the Holy Grail (or maybe Rosetta Stone) for many evolutionary biologists w...