This article is being shared in the media widely as a report that even common birds like blue tits and great tits risk extinction. But that can’t be what is meant. What did it really say?

It is notable enough to be a paper in Nature. But it seems to be more of a specialist paper for expert researchers in the topic area and has surely been misrepresented in the media reports. It is not challenging the IUCN results for these species which list them as being of least concern. The Great Tit for instance occurs in diverse habitatas all the way from the tropics to above the Arctic circle and its populatiom of 430 million to 710 million birds is listed as increasing.

This is another article I'm writing to support people we help in the Facebook Doomsday Debunked group, that find us because they get scared, sometimes to the point of feeling suicidal about it, by such stories.

So first, why am I so sure that great tits don’t risk extinction? Its range is from the mediterranean all the way to Scandinavia and East to China, south to India and Malaysia.

The number of mature individuals is listed as 430,000,000-709,999,999 and increasing.

Under conservation actions the IUCN red list says that it is one of the world’s best studied avian species and that currently no conservation measures are needed for this species.

Yet the paper on the face of it seems to say that Great tits are one of the species wth at least some risk of declining populations and eventual extinction. At least, this is how the media are reading it. You can understand why they read it that way. Here is the relevant passage:

Our comparisons of the actual vs. critical lags suggest that there is low but non-negligible probability that the degree of maladaptation is large enough for the majority of our study populations to be at risk. The actual risk of population extinction may in fact be larger because our estimations do not account for several sources of stochasticity. Moreover, our dataset predominantly includes common and abundant species (e.g. Parus major [Great tit], Cyanistes caeruleus [Blue tit], Ficedula hypoleuca [European pied flycatcher], Pica pica [Eurasian magpie]) for which collection of selection data is relatively easy. The generality of adaptive phenological responses among rare or endangered species, or those with different life histories, remains to be established. We fear that the forecasts of population persistence for such species will be more pessimistic.

Adaptive responses of animals to climate change are most likely insufficient

But how can that be? How could all the numerous studies of this very common bird, one of the best studied of all birds, not have not picked up on a risk of extinction due to climate change?

The others are also very common, as they say, and of least concern.

Eurasian Blue Tit - the IUCN Red List of Threatened Species

Similarly for

• Eurasian Pied Flycathcer - population trend unknown but around 25 million, of least concern
• Eurasian Magpie, stable, of least concern

It makes no sense. Perhaps some Great Tit populations could be at risk at the very north of its range in Scandinavia? Not globally.

This is just a meta study, not original research on any of the species. If it was true other researchers would be saying that the great tit risks extinction and it wouldn't have needed a meta study to uncover this.

They have to mean something different from what they seem on the face of it to be saying in their paper and press release, to those who are not expert on the topic.

• Climate changes faster than animals adapt (press release)

If you have a specialist understanding of all this please comment or send me an email or PM. I am writing this just because the media are running it with no comment at all as if these authors had suddenly discovered a new imminent risk of mass extinction which is clearly not true.

Meanwhile I’m doing my best to try to understand.

First of all, it's striking that the map only covers Europe and North America. For most of the world they found no studies at all.

This is clearly not meant as a world spanning review like the IUCN red list. Also the studies they focused on were of common birds like the Great Tit that are not listed as being at any kind of an extinction risk.

Let’s look at what the referees say - since unusually they share the referee comments on the paper.

First, one of the referees points out that there's likely an implicit bias against non significant results - if someone does a study and finds that there is no lag at all in adaptation - would they publish it? Especially for the type of papers that are found in the literature search they used (Web of knowledge).

There's likely implicit bias against publication of non-significant results, possibly for a couple of reasons. Authors may be disinclined to attempt publication of non-significant results, or such results may have a hard time finding their way into the literature, especially in journals tracked by Web of Knowledge (the source used in this paper). There's nothing the authors can do about this, but they should at least acknowledge this as a possible source of bias in their literature search

Referee 3 in the Reviewer’s comments.

The same referee also says that the variation in egg laying time and so on that they looked at may not be due to climate change, but other factors they do not consider, such as the result of varying abundance and population density leading them to lay at different times.

Category 2) Variation or trends in phenotypic and phenological traits may be due to variation or trends in environmental factors such as temperature (the focus here), variation or trends in abundance/population density (not considered here), or some interaction of these.

... For instance, significant trends toward earlier timing may in some instances be an artifact of population trends if the distribution of the phenological trait is approximately normal and the variance about the mean is sensitive to population size. In addition to such effects, trends in abundance or density might drive or contribute to trend in phenotypic and phenological traits through density dependent competition. This isn't acknowledged in the methods description, but should be.

However the main thing likely to surprise you about these comments, after reading the media stories, is that the word “extinct” does not occur anywhere in the Reviewer’s comments. It is all stuff like

The manuscript represents an impressive and rigorous attempt to assess adaptive responses to climate change.

Surely if the remarks about extinction were new and important, if they had really uncovered a new and surprising risk of extinction for great tits, the referees would pick up on them as either saying that this result is important or saying there is something wrong.

However this part of the paper is not even mentioned, it is just like "meh". It is not possible that the referees understood this paper as contradicting the conclusions of the IUCN red list.

It is clearly a paper that looks different to expert eyes than it does to the ordinary person like you or me reading it. So what did they read it as saying?

WHAT IS THE PAPER REALLY ABOUT?

It seems to be about how fast birds can respond if they stay in the same place rather than moving, e.g. to cooler conditions.

It is an important topic especially for populations that have limited range, and can’t move in response to climate change.

An earlier article from 2013 found that the rate of change of temperature and precipitation expected for many vertebrate species is far more than they have had to adapt to in the past.

• Rates of projected climate change dramatically exceed past rates of climatic niche evolution among vertebrate species

But this doesn’t mean they go extinct. They may be able to step up to a higher rate of evolution when needed. Or they may just move. As the authors of that earlier paper from 2013 say in their conclusion:

We show that projected rates of climate change exceed typical rates of climatic niche evolution among vertebrate species by 10 000-fold or more, based on results from 17 clades and 540 species. These results do not necessarily mean that there will be widespread extinction of vertebrate species from climate change. For example, many species may be able to simply follow their climatic niche as it moves poleward in latitude or upward in elevation.

However, these results do suggest that in situ adaptation of populations to changing climatic conditions would require rates of climatic niche evolution that are largely unprecedented among species in these clades.

The same would apply to this new study. The species could just move.

It also doesn’t have any timescale for extinction - they do not look at details that could distinguish between a possible extinction after ten thousand years of constant lag behind climate change, and a possible extinction within a century.

Their methodology is based on an earlier paper that said that the adaptations of a species if there is a constant change will lag behind the change and that by the amount of the lag you can estimate whether they are adapting fast enough to keep up, or are lagging behind enough to eventually risk extinction.

Their conclusion was that for many species they are not adapting fast enough.

So to understand what they mean by this, we need to understand the earlier paper

WHAT DOES THE EARLIER PAPER SAY?

This is rather techy - I will indent it. But short summary, it is a simplified model over a wide range of possibilities where depending on the rate of change, and the amount of lag, it could be that extinction would occur after climate warming and evolution lagging for thousands of generations (thousands of years in the case of birds which have generation times typically at least one year).

Amongst other assumptions they assume a constant birth rate - that no matter what happens the parent always have the same number of children - and a population that produces offspring whose fitness to survive depends directly on the amount it lags behind the optimum by a fix percentage for each increment (e.g. it might be, say, that for every week the breeding time lags behind the optimum, the ability of the offspring to survive is halved).

Techy summary was based on a very simple computer model with a single phenotype (e.g. that might be the time of breeding of the population, though they do not give any examples). They assume that the parents are randomly mated and have always the exact same number of offspring, 2B. Selection only operates through the viability of the offspring. There is a carrying capacity K and if the population exceeds K then K individuals are selected at random to make the next generation.

The viability in this simple model depends on how far it lags behind the change in the climate, a parameter they call Wz,t governed by this equation

There the Wₜ depends on how far the phenotype value z (say time of breeding) varies from the optimal value θₜ and the ω² is a measure of how stable the selection process is (inversely proportional)

Then they model the θₜ or optimal pheonotype as varying linearly due to time as

θₜ = kt + (random stochastic element)

If the pheontype z lags too much behind the optimal value θₜ for the changed conditions, then Wₜ will decrease. Once this gets to less than 1/B then the population is in decline and will eventually go extinct (since they assume that there is always a fixed number B of births - (most populations actually have an increased birth rate when the population decreases because there is less competition and more resources for individual birds but in their simple model then the birth rate is constant up to the carrying capacity K).

The complexities in this model are mainly to do with modeling how the phenotype z varies assuming a constant rate of genetic modification. They assume five gene variations (alleles) at each of 50 locations contributing to the phenotype. I am not going to attempt to understand this part of the paper which is very techy.

Anyway - that’s the basic idea. This earlier paper finds that in a broad range of conditions the results of this simple model are in good accord with what actually happens. However it remains a very simple model.

The graphs are all labeled in generations. For a small bird like the Great Tit typically live for only a year or two, a generation would typically be one year. If they live longer of course that means each individual has many more offspring, which would influence the results, so that’s another factor, with fewer numbers they would be likely to be more successful in rearing young and individuals would likely live longer too. But they do not take account of factors like this in their simple model. For great tits, the oldest one found in the wild in Germany so far as of writing this is over 15 years old according to this constantly updated online longevity list:

EURING | Longevity list - Fransson, T., Kolehmainen, T., Kroon, C., Jansson, L. & Wenninger, T. (2010) EURING list of longevity records for European birds.

With all these assumptions, for populations that lag a long way behind the change they found extinction was rapid, but for ones that only lag a short way behind, it could take a thousand generations like this run

And then depending on birth rate it can take up to tens of thousands of generations for anything to happen. Here the various values of k are to do with how rapidly the environment is changing, with other factors kept constant.

The axes are logarithmic here. So 10¹ is 10 generations, 10² is 100 generations, 10³ is 1000 generations, and so on.

WHAT DOES IT MEAN FOR GREAT TITS?

Great tits can have more than 10 young in a single clutch, and with less competition could live for many years, increasing the birth rate even more. Unless the change was very rapid and deleterious to their ability to survive in the wild, then they would survive for thousands of generations even with this simple fixed birth rate model before the effect of the lag is so much that the population starts to decline.

I.e. if the climate kept warming, and depending on the rapidity of change, with the response lagging for thousands of years, maybe tits would not be able to keep up and it would have some effect on their populations.

For great tits however, given the high birth rate and constant fluctuations of population anyway I would have thought that was extremely unlikely even over thousands of years.

They say that they were not able to extract many of the parameters they need to do the modeling. They instead used numbers from other studies, and then did a probabilistic analysis to see what the chance was that none of the species they studied were at risk. They come to the conclusion that the chance of this is very small.

The paper doesn't discuss clutch size or birth rate. I have read it through and also done a page search - it has no occurrences of the word "clutch" or "birth".

Nor are either of those mentioned in the supplementary data.

This suggests that they are not trying to figure out timescales for extinction, or whether any particular species are at risk. By the earlier paper they would have to discuss birth rate to do this even with this very simple model.

So, it is more like a preliminary meta study to see if this is something that should be considered in future more detailed work.

They did not try to come up with as much as tentative assessments of extinction risk. It just wasn't that type of a study where they could do such a thing. It is just preliminary research to see if adaptation lags behind change by enough to be a concern. They found that it does.

I think we can understand from this why the Nature reviewers didn’t comment on the matter of extinction risk. That was not the most important part of the paper for them, even though this is what the media have picked up on as its “new result” as if it overrides the IUCN red list.

So now if it is going to be applied rather than remain in the realm of pure theory and hypothesis, presumably the scientists need to look into whether this is a factor for the more important at risk species with limited ability to move in response to climate change.

The co-authors have conflicting views of what the long term implications of their work are. Steven Beissinger takes a positive view that it shows that the species are adapting to climate change. Alexandre Courtiol takes the view that the research shows they are not adapting fast enough. But it is all speculation, and not hard prediction for any species.

Co-author Steven Beissinger (Professor at the University of California in Berkeley) says “This suggests that species could stay in their warming habitat, as long as they change fast enough to cope with climate change.” However, senior author Alexandre Courtiol (Leibniz-IZW) adds “This is unlikely to be the case because even populations undergoing adaptive change do so at a pace that does not guarantee their persistence”. Co-author Thomas Reed (senior lecturer at University College Cork, Ireland) explains “These results were obtained by comparing the observed response to climate change with the one expected if a population would be able to adjust their traits so to track the climate change perfectly”.

It is certainly not saying that Great Tits are at risk of extinction any time soon. There is no need to adjust their IUCN red list classification as being of "least concern".

In their press release they say:

The scientists hope that their analysis and the assembled datasets will stimulate research on the resilience of animal populations in the face of global change and contribute to a better predictive framework to assist future conservation management actions.

It only looks at some European and American species and they only look at some common species with lots of data. It should be applied more to vulnerable species with a limited distribution but their methods of doing a literature survey didn't permit this.

MORE ABOUT GREAT TITS AND CLIMATE CHANGE

Indeed, the Great tit has many features that suggests it will be highly adaptive and climate resilient.

Its clutch size varies a lot e.g. in Wytham Wood Oxford which has been intensively studied for a long time, year on year the numbers of great and blue tits vary hugely.

The average clutch size varies between 12.5 and 7.8. There are four sorts of variation in clutch size. Smaller later in the season, smaller at higher densities, smaller in habitats with fewer large trees, and smaller for first time breeders.

These are all conditions where the chances for raising the young are not so good. Obviously if the population drops then the ones left can find better habitats, there is less density and the population will increase. But most of the young tits die in the first winter. The numbers of surviving tits depend hugely on the amount of the beech mast crop - many more survive if there are than if there are not (probably because the other crops vary synchronously with the beech mast).

• Population Fluctuations and Clutch-Size in the Great Tit, Parus major L.

EXAMPLE OF LEMMINGS - WHERE BIRTH RATE IS HARDLY RELEVANT AT ALL

For a more striking example, the Arctic lemmings. These are affected by climate change, but not by adaptation lags. Every few years the numbers increase a thousand fold, then suddenly crash. Not through swimming into the sea - that's a myth.

Just through predators and starvation.

See the Amazing Lemming. For some reason there have been no population peaks in recent years, possibly a side effect of global warming and less snow cover.
 
 There are many ideas about why these population explosions happen, not well understood. E.g. that it is due to abundance or scarcity of moss, their main food source. Or that it is due to reductions in populations of snowy owls and other predators. The predators are inversely correlated with the lemmings - that's well established - but it's hard to disentangle what is cause and what is effect.

A dead lemming on a stone in the river Revåa in Norway. After a lemming boom and bust, so many die, that drinking the water becomes a health hazard for hikers. Photo by Bjørn Christian Tørrissen

This boom and bust cycle is true of many small rodents in the Arctic region. This shows fluctuations for the lemming, tundra vole and grey sided vole in the tundra ecosystem at Varangar, Svalbard.

Small rodent module

Climate change will impact on this. But not because the small rodents are unable to adapt, but rather because it will lead to warmer and wetter winters. With more rain, the conditions are less optimal for the lemmings, who like layers of snow to run around in hidden from arctic predators. This then impacts on their predators, the Arctic fox and snowy owl. Instead there would be more generalist predators and a more steady cycle instead of the boom and bust - and that also affects the vegetation which will be different due to the different ways the rodents nibbling at it affects it.

You can’t expect the lemmings to start behaving like grey voles, when the climate around them shifts from lemming favourable to grey vole favourable conditions.

Instead the grey vole numbers will increase and lemming numbers reduce.

Here is how they summarize it:

Climate change is expected to lead to warmer and wetter winters. This will lead to increased proportion of precipitation coming as rain instead of snow, leading to melting-freezing events and ice crust formation. Icy snow limits the rodents, especially lemmings, access to plant foods. Dampened, irregular or lost rodent population cycles are a likely result. Sensitivity of lemmings may render the rodent guild dominated by voles.

These changes are predicted to have cascading impacts in the ecosystem. Boreal generalist predators will replace the lemming dependent arctic predators. Reduced rodent herbivory will contribute to vegetation state changes.

Meanwhile further north, rain falling on snow leads to hard layers of ice below the snow, which make it harder for the lemmings to build their homes and burrows at a crucial time of the year.

• Snow physical properties may be a significant determinant of lemming population dynamics in the high Arctic

So, lemmings are going to have a hard time under climate change. This will have a knock on effect on the entire habitat and lead to changes. In those changes there will be some winners and some losers.

The lemmings lose, the grey voles win.

But the Lemming is not at risk of extinction. It is stable over its range in Norway and some northern parts of Sweden, Finland and Russia.

The IUCN Red List of Threatened Species

SEE ALSO

This contrasts the hugely positive central message of the news conference with the dismal reporting of it by journalists.

• Let’s save a million species, and make biodiversity great again - UN report says we know how do it

Also my Positive side of climate change facts

Positive side of climate change facts, after two years of climate change action, heading for 3°C with 1.5°C well within reach

Also

• Yes our generation’s children are headed for a world with nature and wonder in it - and their children too
• 24 ways the world is getting better - good news journalists rarely share