Monday, April 29, 2013

Yes, species are adaptable, but there are limits

For a species, habitats are successful in the long term when they sustain seasonal balancing between food sources and prey.  Flowers that insects feed on have to be in bloom when the insects that feed on and pollinate them have hatched and are looking for food, insects that a given species of bird feed on have to have hatched when that bird has arrived back from its winter or summer migrations, and so on.  So, climate change has the potential to disrupt complex habitats if migration times change, flowering times change, the timing of thaws and hibernations change, food species can't survive the changes, etc., all of which are occurring at the present time. It is likely that some of this has always been taking place, but generally at not nearly the rapid pace we see today.

So a paper in last week's Science is of interest.  In "Population Growth in a Wild Bird Is Buffered Against Phenological Mismatch," Reed et al. describe the effect of recent climate change on the cycling of a European songbird, the great tit, and its food sources.  In theory, gradual environmental change is less of a threat than rapid change because directional selection would favor organisms with extreme trait values that allow them to best adapt to the change, and because that change would be gradual enough that useful variation would be present and the selective intensity would not simply make the species extinct.

When this is the case, the average organism would do less well in the new environment than a subset of organisms with more extreme but adaptive trait values.  If environmental change is rapid, however, there may not be enough genetic variation in a population to allow it to adapt and keep pace with the changes.  And so the population is expected to experience reduced fitness, or a 'demographic cost' as the speed of change increases. If too severe, extinction is the ultimate price.

The Reed et al. study was a test of whether disruptions in the phenology, or seasonal timing, of predator and prey interactions affected the demography of the predator population.  If food is no longer available in abundance at the right time due to climate changes, is there a measurable effect of natural selection on traits that affect phenology in, in this instance, the tit?
We studied great tits (Parus major) in the Netherlands in relation to the phenology of their caterpillar food supply. This part of Western Europe has experienced substantial spring warming in recent decades related to global climate change. Great tits rely on caterpillars to feed their chicks and strive to match their breeding time with the pronounced seasonal peak in caterpillar biomass, which enhances offspring survival. Previous studies illustrated how climate change has produced a steadily increasing mismatch between great tit and caterpillar phenology in our study area, because the caterpillar food peak has advanced in response to rising spring temperatures at more than twice the rate of great tit laying dates. When temperatures during the period after great tits have laid their eggs (late spring) are high, the mismatch is larger (by 2.96 ± 0.43 days per 1°C increase, F1,36 = 47.40, P < 0.001), because caterpillars develop faster under warmer conditions and hence the food peak is early relative to the great tit nestling phase. The greater this mismatch, the stronger is directional selection for earlier laying dates (linear regression slope = –0.007 ± 0.003, F1,36 = 5.066, P = 0.031).
That is, the hypothesis is that great tits that lay their eggs early enough to take advantage of the peak supply of caterpillars to feed this chicks will produce more surviving offspring than those who continue to lay eggs at the later date.  And, this would be an 'extreme' trait, so presumably the great tit population would decline in years when the peak caterpillar supply was later than usual.  And, keep in mind that if we are considering this as an evolutionary effect, the variation must have a genetic basis.

But, the researchers don't find this.  Nor do they find an association between population growth and directional selection; phenological mismatch in this instance doesn't affect fitness (the genetically based reproductive success).  They controlled for the effects of variation in winter food supply (beech nuts), but found that it had no effect on fitness either.

Population growth as a function of (A) annual population mismatch
and (B) annual standardized selection gradient.  Source: Thomas et al.,
Science 26 April 2013 Vol. 340 no. 6131 pp. 488-491
Why is fitness not affected by this predator/prey timing mismatch?  The authors suggest two reasons.  One, the timing of egg laying is never optimal for all females relative to the peak supply of caterpillars because the peak supply window is always narrower than the variation in timing of egg laying.  So, fitness is never optimal for the entire population.

And second, reduced survival of hatchlings due to scare food supply is offset by improved survival of fledglings because there's less competition due to lower population size.  The survival of young birds is highly correlated with population density.

Again, we would note that even if there were  a difference in fitness, it would have no evolutionary relevance unless it were due to specific genetic variation.  This is easy to forget, even if an effect were found. 

In our book, The Mermaid's Tale, after which we named this blog, we suggested a set of general fundamental principles of life that 150 years of observation since Darwin have made apparent.  (We've blogged about these principles before, including here.)  We referred to one such principle as 'facultativeness', or the ability to adapt, which we have suggested must have been one of the earliest traits to evolve given that it is so ubiquitous.  Another word we've used to describe this is 'slop', or imprecision -- as opposed to the exquisitely finely-tuned adaptation that many people think of when they think of the effects of natural selection.

Timing of egg laying in the great tit, however it is determined, is clearly imprecise, and there's still variation in the trait.  And, importantly, that imprecision is tolerated.  It has not been drummed out of the population by natural selection, balancing or not.  And that's a good thing for great tits, given that their food supply varies.

It might be tempting to interpret these results as meaning that climate change isn't going to be a problem because organisms will just adapt.  And yes, organisms are adaptable -- within limits. Including the great tit.  But many species have already gone extinct, and many more will do so.  The Reed paper documents the response to a 3 degree temperature change.  As with the proverbial frog in a pot of water set to boil, there's no reason to think these results are generalizable to the greater changes to come.

If evolution has led to, or tolerated, facultativeness, there is still the question as to how it is maintained.  We don't suggest that there is a gene 'for' such a trait.  Instead, genetic mechanisms have evolved the ability to react to environmental conditions, rather than requiring very precise environments.  The result is that organisms can survive and reproduce in a range of circumstances.  This should not be a surprise to anyone, but may seem so if your view of evolution is that natural selection fine-tunes every species for  highly specific, restricted circumstances.  There's a word for species that were so programmed:  extinct.

No comments: