The Blackboard

I am deeply impressed that Australian bird size data for the last century is such that a 2 to 4% change in size in warbler and wren-sized birds can be confidently stated, much less attributed to a particular cause.

The birds studied are in the range of 8 to 10 grams and most well under six inches long. Measurements would need to be made at almost the exact same time of year since their body weight may fluctuate 50% as they fatten up for migration. So we have to marvel at the precision of the data collection.

And of course that fact that Australia has had more severe droughts this decade than average probably has an effect on the size and growth rates species who feed entirely on leaf-eating bugs. So presumably, a model was built to adjust the data accordingly.

Sure.

For a comparison, the Northern Cardinal started moving into the Northern US before AGW (pre-1950). Cardinals do not appear to be larger in Minnesota than Alabama. Given that they do not migrate and are very territorial and even have regional accents, you would think that the Connecticut variants would be comparative giants by now.

If all it takes is 1.3 deg F to create a 2-4% change over 100 years, then the 13 deg difference between Michigan and Georgia ought to make for 20 to 40% bigger cardinals up north over the same time period. As a birdwatcher, I can say with confidence that one of us would have noticed that by now.

I would like to say as politely as I can, that I find this study unpersuasive. I have the distinct feeling that If the PC villain du jour were were not AGW but air pollution or PCBs or EMFs, etc, these authors would just as confidently assert that ___(fill in au current cause here)___ was the cause.

I blame AGW for the decrease in capacity for embarrassment among tenured populations. There is clearly a link.

JohnFPittman,

I did not know the state bird of Illinois has regional accents. I wonder if the well known “northern/southern Illinois accent” line running through Urbana applies to cardinals?

For the benefit of Europeans who may be unfamiliar with this bird, and wonder why it is called a cardinal, here is an image of a male cardinal from
http://www.birdsandblooms.com/.....te-Birds-/

Male and female:

My apologies to everyone but after reading “Waxman Markey Hotdogs” and “Global Warming and Shrinkage” all I can think about is that commercial from about 10 years ago where a guy walks out of the ocean without his swimsuit and 2 girls look at him and just say, ‘That water must be really cold!’ Again, my apologies, I should have not posted this but with these two topics so close together I could not help myself.

Lucia, though I have a southern acccent; I can’t claim the point. It was G Tobin. Though “region accents” are sometimes pointed out by other biologists, it is not an expertise of mine. My efforts were in ecology and waste treatment areas.

Sean… Now that you mention it, those two titles next to each other do have a certain “what’s it?”

JohnF–Opps. I must have been reading a response in my mail box and swapped names. (Or I’ll pretend that’s the cause of confusing the name.)

Americans have been getting considerably more obese over the same period of time, so I think it would be fair to say that climate change has been a leading cause of the this expansion in our society. A plausible explanation is that warmer weather is linked to lessened desire to exercise and more consumption of ice cream.

George Tobin didn’t read the paper very carefully:

The studied species are 7-79 grams (not 8-10) grams.

They do not fatten for a long distance migration – they are mostly sedentary.

Historical measurements were not used, they measured museum specimens.

All are insectivores but not just consuming foliovorous insects.

The authors found no trend in nutritional condition as estimated from feather growth bars.

Latitudinal clines in avian body size are well known, and were present in several of the species, but are certainly not universal. 0 versus

They compared specimens 1860-1950 versus 1950-2001 so
very few specimens will have come from this decade.

Atttribution is of course challenging, the effect seems too large to be explained by temperature change only.

Oliver: Great observation. No different than other cause-effect results. Further, the obesity will lead to a tipping point, whereby more greenhouse-gas flatulance from the obese will cause more warming. A dangerous positive feedback!

Cardinals have regional accents that are kinda the reverse of people–Southern birds sing in more rapid bursts, higher pitched, northern birds have a longer, slower song.

I first heard about regional accents in non-migrating birds in a recorded lecture from Donald Borror, the now deceased Ohio St prof who collected thousands of bird song and insect recordings way back before technology was very conducive to that sort of thing. They have since named a bioaccoustics lab after him.

The shorter buzzy songs may work better in warm moist swampy areas whereas dense forest echoes are better suited for slower more distinct notes. Who knows.

Birds that do migrate practice the company song together in the tropics in the off season and thus maintain audio uniformity (“consensus”?). Cardinals don’t migrate and thus local differences can emerge fairly quickly.

I don’t think you notice it unless you are listening for it and can already notice difference in styles among birds in your own neighborhood. I once awakened my wife well after midnight to have her listen to a serenade from an old mockingbird in our yard who did more than a dozen species really well. Much to my surprise she was not anywhere near as impressed by his repertoire as was I and remembered only the loss of sleep.

This audio skill will also ruin some movies — sound track inclusion of eastern US deciduous forest birds in a Western or in a Amazon jungle setting can be jarring.

Sean and Lucia,

I was reminded of George Costanza: “I was in the pool!”

George–
I don’t think I’ve ever heard a mockingbird!

I can report that the local bird populations have changed recently. It’s been attributed to west nile. Crows died out; we now see lots of hawks. I’m also seeing more humingbirds and some very, very beautiful bright yellow birds.

andrewt:

Didn’t read it at all—just the linked article about the paper and one other.

Three fourths of the species studied are in the exact small size range I cited. The notion that one can get size measurement precision in the 2-4% range from pulling old dried birds of that size pulled out of museum drawers is a stretch.

My point about migration (yes some Australian birds do migrate) and drought is that there is a great variance in individual bird weight, especially from seasonal factors. The mass or volume of an old dried dead bird is probably not particularly reliable, but even if it were, the date of collection would be relevant. A seasonal bias in specimen collection would be enough to overwhelm the reported findings.

I think you were being charitable to say “attribution is of course challenging.” The new climate in which no scientific finding is significant unless it can be made ideologically relevant is scary.

George–
There is another issue. Back in 1860, even the most objective scientists would not have known the potential future uses. So, there is the possibility that they birds preserved were not biased toward slightly larger birds that might be easier to handle.

Of course, there also is no guarantee they weren’t biased towards smaller sizes either. But the fact is, when you pull out specimens from 1860, you can’t interview the scientists to learn whether they took special pains to sample randomly.

So, at best you can know is that birds you randomly sample in 2010 are large or small compared to bird collected in 1860, which may, or may not, have been randomly sampled.

Perhaps bird size vs. global temperature can be a new proxy for Mann to write a paper on. I just read an article that says “Using proper terminology, birds are avian dinosaurs”. Perhaps I should quit reading and take 2 aspirin with a scotch and water.

lucia:
Bird populations are pretty dynamic. Lots of demographic boom and bust. I visit the same places at the same time of year and time of day to look for old friends and sometimes they seem to disappear only to rebound years later. One path near a lake in PA was once filled with Northern Juncos who simply disappeared for over 5 years. Now they are back but in fewer numbers. Disease? Competitor? Hurricane victims on migration? I Dunno.

Your bright yellow birds are probably goldfinches. I think the housing bust was a boon for them — they love weed seeds and new brush so unattended properties and overgrown lots are a real treat for them.

George, I assumed when you said you found the study unpersuasive, you’d actually read it.

If you were to look at Table 1 in the paper you’ll see 2 of 8 species (not 3/4) overlap the size range you nominated.

Yes Australia has long-distance migrants, but these 8 species are largely sedentary.

They measured wing length not mass or volume, which would be problematic with a skin.

Some of your claims about bioacoustics are incorrect too. Dialects are common in migrants. Generally migrants are thought not to sing in wintering areas. Migration is thought to affect song,e.g. perhaps increase repetoire, and presumably the scale of geographic variation. I’d recommend Kroodsma’s popular work on bird song: “The Singing Life of Birds”.

I recommend “The Beak of the Finch” for a really rigorous, in-depth, and well-written analysis of the impact of weather on finches in the Galapagos from year to year.

Yes it makes perfect sense. With all these birds dying through excessive heat they MUST be getting smaller because they can shed heat faster. I mean, just look at the mounds of heat afflicted dead bird carcasses; don’t you see they are all just a few percent larger than the ones chirping over there ?

PMH (Comment#18030):

“Perhaps bird size vs. global temperature can be a new proxy for Mann to write a paper on. I just read an article that says “Using proper terminology, birds are avian dinosaurs”. Perhaps I should quit reading and take 2 aspirin with a scotch and water.”

I just put to and two together, and the Cretaceous must have been damn cold! So much for the faint young sun paradox if the Earth was a snowball all along!

More generally, I was SURE that you were going to be talking about a, uh, “different” kind of “shrinkage”….

Are you really telling me that someone actually PAID money to draw some dried out century old bird carcasses from a cellar , measure them , weigh them and then waste energy and matter to report about his delirium tremens induced “findings” ?
Because if the answer is yes then those people are definitely much too rich , lazy and on drugs and deserve the Apocalypse that will be visited upon them in form of very small man eating bird mummies .
.
Besides our bird mummy amateur made a trivial mistake anyway as everybody familiar with Peacock,Biture & al (1935) would know .
These scientists , authority in the mummology and authors of the seminal textbook in 15 volumes “Non euclidian metrics of undead avians in a divergence free thaumaturgical field” have proven beyond any reasonable doubt that topology and mass of bird mummies varies in quite interesting albeit surprising ways .
And this guy didn’t know that , ha ha !
I laugh .

TomVonk–
What I want to know is whether they had to file an FOI to gain access to the birds. After all, the birds were originally mummified by someone else. That person might still need a grace period in which to analyze the carcasses and write more papers. Does even death end that periods? This is a matter that should be debated along with the CRU data.

andrewt:

I am pleased that you’ve read Kroodsma’s book which means that I am communicating with another aficionado. I had been afraid this was going to devolve into one of those tiresome blog comment opposition-for-its-own-sake things grounded in Googlized instant expertise.

However, you do have to be careful with the word “accent” since Kroodsma mostly uses that term to indicate a different kind of song rather than regional difference. That migration chatter is important for transmitting certain types of songs or calls to the next generation is one of his points and exactly why migrating birds are much less likely to have regional accents.

Good catch on the size of the Australian birds. Of the 6 birds I had estimated to be in the 9 gram range, several could indeed be as much as 11-12 grams.

Of course, I still don’t get why that matters.

Because with those little birds we are talking about rather variable wingspans of around 9 inches which means the study we are discussing is making claims of significance in a range of as little as 1 or 2 mm per wing on old, dead, dried birds from a museum drawer when the natural variability within each species is probably in the range of at least 10 mm.

As lucia opined, a preference for larger, better-looking specimens in an earlier age would be more than enough to explain the difference. And given the natural variability in size (perhaps as much as 25% for the larger species in the set), any one of a number of possible collection biases could completely determine the outcome.

Of course, all of this is beside the main thrust of my original argument which was:

How is it that Australian birds could be so sensitive to climate that a mere one-degree change over a relatively short period of time (100 years) could be said to cause significant size differences when that does not happen in a species such as the Northern Cardinal.

As you pointed out, latitudinal size differences among bird populations are known to exist, so why wouldn’t cardinals, who have (via territorial expansion) subjected themselves to a temp difference ten times larger over the exact same time period (100 years) exhibit even larger size changes? Or any size changes, for that matter?

It is utterly depressing to see that this story is now in every paper in the world thanks to the credulous AP wire. For millions of people, it is now received wisdom that little birds in Australia are shrinking into oblivion in a desperate struggle to cope with a trivial temperature change so God help them if the increase becomes, dare I say it, 2 degrees F? What a farce.

There is the old research into the human ear during the last century that reveled that ears were getting smaller in successive generations.

Then the other researchers pioneered research that would counter this, by showing that ears got bigger throughout the century because they never stop growing. They gave a linear trend with confidence intervals. Ears only appeared to get smaller across generation samples if you assumed that they do not grow after the commencement of adulthood.

(I am not making this up!)

Further research showed that the trend was not linear. Human ear growth is cyclic with a disernable circaseptennial rhythm.

With that the science was settled, we knew that we were in for relentless auricular extent rises for decades to come.

“Act before it is too late.” “We must move Auricle Change up the agenda and get a UN body to look into ears.”

“Be on your watch for the ear growth deniers and prepare to rubbish any hint of auricular stasis.”

The rumour that ears are only getting bigger as a result of a rebound to “normal” size as a result of previously warm and uncapped medievalist suffering serious frostbite in later centuries is just that; “A Rumour”;

put about by anarchists and the infamous clandestine order of the Knights of van Gogh.

They msut be stopped!

We cried out for capping legisation:

“Give us “muff laws”, Halt auricle growth by 2050.” We chanted.

Things have moved on since then.

Like everyone else I bought into this big time, having originally been an inocent and unwitting follower of the ear shrinkage church. I flipped in the face of overwhelming evidence. The aural index could be seen to have risen steeply during the latter of the 20th Century. Massive El Nino ‘ears followed in repeat cycles for what seemed like decades. Until we had had enough and cried no Spock. We saw the light and all became Spockites or joined our local Noddy and friends society.

I can not recall for certain whe I first had doubts.

News of the ear shrinkage research finally reached me in the 1990s. I had bought into the standard model but I was shaken by the audacity of this old seldom read paper. Despite the risk of ridicule I prepared to come out as a auricle change skeptic.

Since then I have been keeping close eye on my ears and can announce that I have conclusive evidence that none of this is my fault!

I regularly score less than half a planet on on-line earprint calculators.

My ears are not growing and are definitely much smaller than my farther’s were at my age. In public I normally only confess to an adherence to auricular stasis but I have deep auricle shrinkage leanings.

Alexander

Considering the amount of habitat changes that have affected SE Australia in the last 150 years and the changes in selection pressures, I think it would be strange if bird morphology had not changed during this time period.
However since the changes are only on the order of 1 mm, there is a lot of factors that could affect the result.

Shrinkage (usually 1-3% over the first decade or so, can’t remember any longer-period studies).

Abrasion: Primaries are significantly abraded during the year from molt to molt, this must be corrected for if collection dates are not constant.

Degree of migration/nomadism: Birds which migrate longer distances have longer wings than birds of the same species which migrate shorter distances. Most Australian passerines are not regular migrants, but many (most?) are nomadic. The degree of nomadism might have changed over time and this might well affect wing-length.

Was the measurements for each species done done by a single person? It is a well known fact from bird observatories that different persons get systematically different results when doing biometry.

Were the measurements done “blind”? I. e. did the person making them know the date of the skin he was measuring? You can get significantly different results depending on the amount of pressure applied on the wing to straighten it when measuring. There is a significant risk for “confirmation bias” when dealing with such small differences.

And finally, was there collection bias? Probably impossible to determine, but more recent skins are more likely to be a random sample than old skins according to my experience.

George, I’ve no idea why Cardinals don’t show latitudinal size variation but some species don’t, including several in this study.

George, if you were to look at table 1 in the paper you still have the sizes wrong. I’m not sure what you mean you mean by the *range* of natural variability but the standard errors for their measured wing lengths (same table) are all less than 1mm.

tty, yes quite a few Australian passerines are nomadic to some degree but these 8 passerines are sedentary with one exception (M. fascinans) which might in some areas make seasonal or other short-range movements.

A collection bias towards larger specimens doesn’t sound particularly likely but its hard to exclude. I gather retrospective study of museum specimens are not uncommon so it should have look at generally.

It was no doubt a FOAA filing.

What?

You haven’t heard of the Freedom Of Avians Act?

Philistine!

“I’m not sure what you mean you mean by the *range* of natural variability but the standard errors for their measured wing lengths (same table) are all less than 1mm.”

It’s not rocket science: Lisa Leslie vs Mary Lou Retton. Ed “Too Tall” Jones vs Billy Shoemaker. Shaquille O’Niel vs Muggsy Bogues. John Wayne vs Tom Cruise. Variation within a species goes far beyond “2 to 4 percent”.

Were the birds collected over the last century “average” sized? Or did they perhaps tend to collect larger specimens? Was the collection larger at some point in the past? It’s not outside the realm of possibility that someone decided to “make room” by getting rid of a number of superfluous bird carcasses. Hmm, which ones might they keep, and which ones might they destroy?

Finally, the author’s decision to state her guess as to the reason behind these dubious findings as “likely” … no wonder she is being mocked.

But averages are different things than ranges.

It is quite clear, for example, that the average height of humans has increased, even though the natural variation easily covers that increase in average.

If you have sufficient data points, you can find changes that are statistically significant even when the changes are very small.

Given that the data showed birds getting smaller, given that Australian temperatures have increased and given that there is already documented size differences in birds of the same species that correlate with latitude (and therefore average temperature), I do not think that the conclusions that the size changes are real and are happening because of the increase in temperature are unreasonable.

There may well be sampling problems. But, given that those sampling problems might just as well be in the other direction – for example, space considerations might have meant that smaller birds were kept and larger ones thrown out – I do not think that there are any concrete reasons to doubt this study as yet.

If it can be shown that it is likely that there sampling problems *and* that those sampling problems distort the data in a specific direction, that would of course throw doubt on the study and its conclusions.

David Gould

But averages are different things than ranges.

It is quite clear, for example, that the average height of humans has increased, even though the natural variation easily covers that increase in average.

Believe it or not, the bird example shows that you grasp the difference between the t-test I use to compare the multi-model mean trend to an observation, and the test Gavin discussed at RC.

His test insists that we can’t say the models are off unless the earth’s temperature falls outside the full range and I test whether the two means are identical.

FWIW: With the birds, I agree we could detect a change in the mean if we used a sufficient number of birds. However the issue of possible sample bias in the historical museum collections, human bias that could lead someone to stretch the wing span more or less still need to be considered.

In principle, the ornithologist peer reviewers who happened to review the paper would have thought about these things and made their judgment. In practice, all sorts of stuff passes peer review, and it may be that the birds haven’t changed size.

The connection to climate change does seem speculative. Maybe that’s the cause. Maybe the cause is pesticide, hunting, change in territory etc. I have no idea. I do think there is a tendency for people to want to report reasons even when the main value in the paper should have been simply reporting the change in size of birds.

“It is quite clear, for example, that the average height of humans has increased,”

Not in North Korea, http://english.donga.com/srv/s.....6112256018 , or here, http://www.dprkstudies.org/200.....h-koreans/

Of course, it might be hard to pin the blame on “Global Warming Shrinkage”, but give it time, I’m sure the studies are in the works now.

“There may well be sampling problems. But, given that those sampling problems might just as well be in the other direction – for example, space considerations might have meant that smaller birds were kept and larger ones thrown out – I do not think that there are any concrete reasons to doubt this study as yet.”

Yes, there may be (and probably are) sampling problems. Yes, these might go in either direction. This lack of knowledge means that the error bars MUST be larger than would otherwise be the case – and this should easily swamp the 2 to 4 percent difference reported in the study IMHO. Therefore there is no legitimate statistical difference to report and the conclusions are complete garbage.

I just wanted to add a little follow-up from MIT Physics 101:

“Now, all-important in making measurements which is always ignored in every college book is the uncertainty in your measurement.

Any measurement that you make without any knowledge of the uncertainty is meaningless.

I will repeat this.

I want you to hear it tonight at 3:00 when you wake up.

Any measurement that you make without the knowledge of its uncertainty is completely meaningless.”

Pity I can’t get the original article – our library doesn’t have a subscription. In the 70’s, there were rather common studies about size and shape differences in animal population. Several of people studying with me did their Theses on such subject. It encompassed catching, measuring, weighing and killing the animals, then bone cleaning and again measurements of skulls, teeth, long bones etc. The bones then were stored, so if the compared old bird specimens were from such study, it could be reasonable.
And about what can pass today through peer review – just last week I compared my fungal sequence with those deposited in the GenBank database. Among the related fungi, five sequences were annotated as originating from small monkeys!
I found the publication, where the alignment of fungal monkey sequences was published with database-downloaded chimp one and some evolutionary speculations were presented. Even the wildly differing chimp sequence did not persuade the authors to think twice about their results… and apparently neither the referees could be bothered to check the sequences themselves. Oh well.
I wrote to the journal’s editors, but no reply yet…

Lucia,

Re reporting reasons, the scientist(s) in this particular study looked at a number of possible reasons, including changes in diet, but ruled them out. They did not simply say, ‘Warming.’ When an observation or series of observations are consistent with other data – as in this particular case with bird sizes at different latitudes – it does not seem unreasonable to conclude that the cause is the same.

I also think that simply reporting data is not the main purpose of scientific papers. At the very least, some possible causes for the data you are seeing should be *raised* as paths for possible future investigation. Science is not simply measurement.

David– Of course simply reporting observations does not encompass all of Science. But that doesn’t mean many experimental papers would not be improved by dumping the speculations as to cause. For many years, I’ve thought many experimental papers would benefit from reporting what was observed while being cautious about suggesting tenuous reasons. Still, scientists like to speculate as much as anyone.

I agree the authors believe they ruled out some possibilities they thought of.

Lucia,

On averages, yes, I realised that as I was thinking about the issue.

My first inference assignment is due Monday, so I hope I am starting to understand this more.