Mar 05 2010
Reaching Scientific Consensus – On Dinosaurs
Controversies in science are fun, and the spectacle of such controversies being worked out by competing groups of scientists is a wonderful way to learn about the process of science. But as science progresses, we hope to occasionally resolve controversies and come to a reasonable consensus.
One such controversy that I have been following is the question over what killed the dinosaurs, and much of other life on earth, 65 million years ago. The two leading contenders were an asteroid impact at the Chicxulub crater near Mexico, and volcanic eruptions at the Deccan trap in India. Now a new review of the literature has resulted in a solid consensus supporting the asteroid theory.
Actually, the consensus is somewhat of an anti-climax in that this consensus has been slowly building for years. In recent years the vast majority of scientists already agreed on the impact theory, with just a handful of holdouts rooting for the volcano alternative. So this latest report is no surprise.
In fact, this is often how such things will play out. It would be premature to try to impose consensus on a controversy when the controversy is still raging and the data is not yet definitive. On the other hand – why bother formalizing a consensus that already exists de facto?
In this case, at least, while there was a solid consensus, there was still a robust minority opinion and new data had been coming in fairly quickly. So it seemed like a good time to step back, take a thorough look at all the evidence, and see where the science really is. In this case, a thorough review of the evidence strongly supports the impact theory.
The biggest piece of evidence is that there is a layer of debris spread around the earth. This layer contains iridium, which is more common in asteroids than the earth’s crust. It also contains things like shocked quartz, which is created only during a sudden violent impact. There are also spherules – melted rock that hardens into spheres as it rains back down onto the earth.
Moreover, this layer varies in thickness around the world, pointing like a bull’s eye to Chicxulub. But most importantly, when you look at those layers that are about half-way around the world from Chixculub, you find nicely organized debris layers with a sharp demarcation of dinosaur fossils. Below the line – dinosaurs. Above the line – no dinosaurs. Get too close to the impact and the layers are jumbled by the impact itself. Get too far away and they are too thin to separate out the fossils nicely. But at the perfect distance, the layering becomes clear.
The consensus also dispenses with some of the objections. The timing of the mass extinction does line up with the impact. And the global effects of the massive volcanoes at the Deccan traps were small and short-lived – not enough to explain the mass extinction.
I wonder if this consensus will have any effect on those few remaining holdouts. I also wonder if it will affect future media reporting (probably not). One of my peeves of science news reporting is that the media will report a science news story by a scientist in the small minority as if it represents the new or consensus opinion. So whenever the volcano advocate published a study, the media declared that the impact theory was being rejected – not quite.
The same is happening with the dino-to-birds theory – this is by far the consensus opinion. But a few holdouts grab headlines occasionally and get the media to declare that the birds from dinosaurs evolution story is in doubt. It’s the journalists’ job to put the minority opinion in context, and they rarely do, which is irritating. The public is left confused by each new headline that seems to contradict the last.
And that may be the biggest benefit of such a consensus – if we can get the word out, it may help to make sense of the science to the public.
![[del.icio.us]](http://images.del.icio.us/static/img/delicious.small.gif){kind=small}
Print This Post
11 Responses to “Reaching Scientific Consensus – On Dinosaurs”
There seems to be this idea in US journalism that being impartial means giving equal time to all sides of an issue, even if one of those sides is represented by a tiny minority of people. Pointing out that one side is a miniscule minority may lead to accusations of bias, as for whatever reason, that appears to terrify our news organizations. I imagine that they treat science news the same way because, these days, science news is often also political news, and perhaps simply out of habit.
The age of the bird linage still amazes me. Their class has been around for so much longer than ours and it shows in every aspect of their anatomy.
Physically, they have had evolved into their niche so well that it boggles the mind. From their specialized bones to their system of air sacks, it seems like every physical feature of the birds has not just been adapted for flying, but perfected for it. When you compare their adaptations to the crude modifications that occurred in the tetrapods, it’s almost embarrassing. Four legged mammals still bear the hallmarks of the reptile-like creature we evolved from (especially in our pelvic girdle). Don’t even get me started on humans! We’ve got more design flaws than a Toyota (too soon?). But birds…if any group of animals can be said to have evolved elegance, it’s gotta be them.
Not only are they physically amazing, they’re brains are extraordinarily efficient. Birds tout some of the smallest and most intelligent brains in the animal kingdom!
Dr. Novella, I’m not sure how much is known about avian neurobiology, but I’d be interested to hear some theories about how animals that are so smart can have brains that are so small!
PS: The fact that we can solve some of these great mysteries, which at first must have seemed so intractable (“What are these crazy giant bones in the ground!?”), makes me proud to be a part of the scientific community.
In some ways, it’s sad that so few scientists get to discover something so amazing that their name lives on for generations. But seen in another light, this is a great thing because science IS truly a collaborative endeavor and every one of us who is part of the process (even the armchair skeptics) can claim to have advanced the progress of our species in some small way.
Pah, another bird proponent…
*beakflip*
An impact extinction can also have some extended events. The most likely impact is not a “bolt from the blue”, with a single large object coming in from beyond Pluto and on the first pass smacking into Earth. A more likely scenario is multiple passes with orbits being perturbed multiple times until there is an impact. A scenario like that could put lots of debris into the near Earth system that could have impacts on weather and climate over thousands of years before and after the catastrophic event.
Better mapping of craters on the moon, and knowing their exact timing from in situ measurements might be able to sort that out. What might be very interesting is if something did get splashed off the surface of Earth by that very large object and landed someplace “safe” on the moon, like one of those craters in permanent shadow. There might be organic material preserved by the cold. The cumulative radiation dose would probably completely destroy any DNA, but there might be short bits left.
Some of the problems with false balance in non-troversies have their origins in ignorance, and some are due to a misunderstanding of what bias is in journalism.
Often writers are writing about topics beyond their scope of understanding. This is because most science writers in the press are not physcians or scientists. This makes it difficult for the writer to have proper perspective on certain subjects, and can result in poor quality of information.
The bigger problem is the perception that journalist need to have balance in order to minimize bias. Of course in reality balance should be weighted, so that it is understood when certain perspectives are minority positions. It should also be clear why a certain position is a minority one, and what it lacks in comparison to the majority position. Journalists understand this with some topics (Holocaust deniers and 9/11 deniers), but they struggle with scientific ones (vaccine nontroversy).
Again some of this is ignorance, but some of it is intentional to create a good story. This is why the retraction of the Wakefield paper was important. Even though the study has been ridiculed and disproven for many years, it wasnt until the reatraction took place that the media really did the topic any justice. It made for a better story than the reality that “the weight of the evidence indicates…”
After all, we are trying to reach the large segments of the population who arent paying attention in life, but are still making important decisions for themselves and others
I’d actually quibble with CivilUnrest. If reptiles are going to be considered a single clade, to include turtles, snakes, crocodiles, dinosaurs, etc, then birds are also reptiles. Mammals did not evolve from reptiles. The mammalian line split off through the synapsids before the ‘reptilian’ line. So mammals as a clade have actually been around longer than reptiles as a clade. But every species that is around today has been evolving for exactly the same length of time as any other. It’s just that some seem to have evolved more quickly than others. When the non-avian dinosaurs went extinct, it just opened opportunities for surviving mammals and avian dinosaurs to expand into available niches. It’s arguable whether birds or mammals did better.
Hey Steve,
I’ve been a long time reader and appreciate all the effort you put forth in promoting science. However, this is a topic that all too easily will get distorted if individuals don’t have time to follow the literature (which is very unfortunate, as this is usually a topic that generates a lot of excitement). Please bear with me, as I seem to have gone into some depth.
Schulte et al. 2010 seems, to me, to be an attempt to generate a false consensus. This paper is nothing more than a restatement of Alvarez et al., 1980 with the added information provided by Hildebrand et al., 1991.
I concede this paper does a brilliant job laying out that the evidence of a bolide impact at approximately the same time as the KPg boundary, however this was already established by Alvarez et al., 1980. I further concede that this paper does a good job laying out Chicxulub as a likely crater, though this was already established by Hildebrand et al., 1991. However, these points were never in contention. (Keller et al. 2009 simply noted an unreworked layer of shocked quartz which would require further research. Keller’s paper never claims a bolide didn’t impact the Earth at the end of the Cretaceous, she merely questioned the specific timing).
However, Schulte et al., 2010 fails, like practically all extinction studies, to demonstrate causation. Ultimately it is not enough to demonstrate a correlation between the end-Cretaceous extinction event and a bolide impact layer, causation must be demonstrated. The jump from correlation to causation is the weak point.
The problems with demonstrating causation, under these circumstances, are exacerbated by the limitations of the stratigraphic record. Terrestrial records, from the time of the end-Cretaceous, offer chronostratigraphic resolution of ± 40,000 years (Swisher et al., 1993). Considering that the time scale of models of the impact event only suggest that the effects will last days- ~100 years (Schulte et al., 2010), this is a significant source of uncertainty.
Furthermore, there have been several documented phenomena that hamper establishing a causal link between an event and extinction.
The most significant is the Signor-Lipps effect (Signor and Lipps, 1982). The Signor-Lipps effect was initially a thought experiment that argued that a gradual extinction and a catastrophic extinction would be preserved in the stratigraphic record in the same fashion.
A brief synopsis of the Signor-Lipps effect goes a little something like this. When an organism dies, varying factors influence whether or not the organism will be preserved as a fossil. The potential that an organism will be, in whole or in part, fossilized can be referred to as ‘preservation potential’. The preservation potential varies between species. For example, a clam lives its life buried in sediment. If it dies while still buried, it has a very good chance of becoming a fossil. A bird that lives in trees and flies around in the sky, will need to die and get transported to a location that is likely to be preserved. So the bird in this example has a lower preservation potential than a clam. Now expand this concept to discuss populations, rather than an individual. We see that rare species with relatively low preservation potentials aren’t going to be fossilized as often as abundant species with high preservation potentials.
Now, if we apply this concept to the study of extinction events (this is where the Signor-Lipps effect comes into play), we would expect to see certain species disappearing from the fossil record before other species, regardless of the cause of the extinction. Rare species with low preservation potentials should disappear lower in the stratigraphic record, than abundant species with high preservation potentials. This will give the appearance of a gradual extinction (where species diversity is declining over a period of time) even if the extinction event is catastrophic (where all the species disappear at once). The implication of this is that gradual extinctions are indistinguishable from catastrophic extinctions in the stratigraphic record. This has subsequently been demonstrated to be an actual artifact, not just a thought experiment, by MacLeod (1996).
Additionally, depositional hiatuses can generate a false extinction signal (MacLeod and Keller, 1991). This study showed that a sharp extinction horizon in one portion of an oceanic sedimentary basin might be the result of minimal sediment deposition, and if you correlated to an area that underwent more consistent deposition, you would observe a broader transitional zone rather than a sharp horizon.
Additionally, a study by Smith (1995) raised the possible implications of mistaking an extirpation event as an extinction event. He was examining the Permian-Triassic extinction event in the Karoo basin in South Africa. This boundary is marked by a change in depositional environment, from a river dominated environment (fluvial) to an arid environment (aeolian). This shift in depositional environment is coincident with a faunal turnover from dicynodonts to lystrosaurs. However, Smith found it impossible to rule out extirpation. Since dicynodonts were associated with wetter environments, it is possible that when the environment became drier, the dicynodonts simply migrated to wetter areas. While lystrosaurs, associated with drier settings, might have migrated into a newly available area.
Now don’t get me wrong. I’m not opposed to the bolide extinction hypothesis. I just don’t think there is currently enough data to demonstrate causation. Certain issues need to be overcome before there can be a scientific consensus.
1) We need to be able to surmount the Signor-Lipps effect. If all extinction events generate the same stratigraphic signal, it is impossible to falsify whether the extinction was “Gradual”, “Stepwise”, or “Catastrophic”.
2) Speaking of “Gradual” and “Catastrophic”, these terms need to be properly defined in this context. If you think it is straight forward, answer this question: Does “Catastrophic” refer to a rate of extinction or a magnitude? If you answered rate, where is the cutoff? What is the cutoff based on, or is it an arbitrary cutoff? Can this cutoff be resolved chronostratigraphically, i.e. is it ~ an order of magnitude greater than 40,000 years? The order of magnitude difference is necessary to assure a “complete section” (Dingus and Sadler, 1982; Dingus, 1984). Similar questions apply to magnitude.
3) We need to be able to delineate an extinction event from an extirpation event. If we can’t distinguish these events, there is little hope of truly establishing a causative association for the disappearance of any taxa from the stratigraphic record.
Once these issues have been properly addressed, it may be time to reassess extinction mechanisms in the rock record. Until that time this “controversy” will continue.
Again, I apologize for the length of my post, and I thank anyone who has read to this point. It’s a crash course in the current workings of extinction studies. I run a small geoblog if anybody has any questions or wants to read further (www.in-terra-veritas.blogspot.com) or would like to find out how to get a copy of the cited papers. Though this comment is similar to a post I did on the same publication, I’ve included links to prior posts covering certain topics in more depth.
Works Cited:
Alvarez, L. W., Alvarez, W., Asaro, F., and Michel, H. V., 1980, Extraterrestrial cause for the Cretaceous Tertiary extinction: Science, v. 208, no. 4448, p. 1095-1108.
Dingus, L. W., 1984, Effects of stratigraphic completeness on interpretations of extinction rates across the Cretaceous-Tertiary boundary: Paleobiology, v. 10, no. 4, p. 420-438.
Dingus, L. W., and Sadler, P. M., 1982, The Effects of Stratigraphic Completeness on Estimates of Evolutionary Rates: Systematic Zoology, v. 31, no. 4, p. 400-412.
Hildebrand, A. R., Penfield, G. T., Kring, D. A., Pilkington, M., Camargo Zanoguera, A., Jacobsen, S. B., and Boynton, W. V., 1991, Chicxulub Crater; a possible Cretaceous/Tertiary boundary impact crater on the Yucatan Peninsula, Mexico: Geology, v. 19, no. 9, p. 867-871.
Keller, G., Adatte, T., Juez, A. P., and Lopez-Oliva, J. G., 2009, New evidence concerning the age and biotic effects of the Chicxulub impact in NE Mexico: Journal of the Geological Society, v. 166, no. 3, p. 393-411.
MacLeod, N., 1996, Nature of the Cretaceous-Tertiary Planktonic Foraminiferal Record: Stratigraphic Confidence Intervals, Signor-Lipps Effect, and Patterns of Survivorship, in MacLeod, N., and Keller, G., eds., Cretaceous-Tertiary Mass Extinctions: Biotic and Environmental Changes: New York NY, Norton, p. 92.
MacLeod, N., and Keller, G., 1991, Hiatus distributions and mass extinctions at the Cretaceous/Tertiary boundary: Geology, v. 19, no. 5, p. 497-501.
Schulte, P., Alegret, L., Arenillas, I., Arz, J. A., Barton, P. J., Bown, P. R., Bralower, T. J., Christeson, G. L., Claeys, P., Cockell, C. S., Collins, G. S., Deutsch, A., Goldin, T. J., Goto, K., Grajales-Nishimura, J. M., Grieve, R. A. F., Gulick, S. P. S., Johnson, K. R., Kiessling, W., Koeberl, C., Kring, D. A., MacLeod, K. G., Matsui, T., Melosh, J., Montanari, A., Morgan, J. V., Neal, C. R., Nichols, D. J., Norris, R. D., Pierazzo, E., Ravizza, G., Rebolledo-Vieyra, M., Reimold, W. U., Robin, E., Salge, T., Speijer, R. P., Sweet, A. R., Urrutia-Fucugauchi, J., Vajda, V., Whalen, M. T., and Willumsen, P. S., 2010, The Chicxulub Asteroid Impact and Mass Extinction at the Cretaceous-Paleogene Boundary: Science, v. 327, no. 5970, p. 1214-1218.
Signor, P. W., and Lipps, J. H., 1982, Sampling bias, gradual extinction patterns, and catastrophes in the fossil record, Geological Implications of Impacts of Large Asteroids and Comets on the Earth: Special Paper – Geological Society of America, p. 291-296.
Smith, R. M. H., 1995, Changing Fluvial Environments Across the Permain – Triassic Boundary in the Karoo Basin, South Africa and Possible Causes of Tetrapod Extinctions: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 117, p. 81-204.
Swisher, C. C., Dingus, L. W., and Butler, R. F., 1993, 40 Ar/39 Ar dating and magnetostratigraphic correlation of the terrestrial Cretaceous – Paleogene boundary and Puercan Mammal Age, Hell Creek – Tullock formations, eastern Montana: Canadian Journal of Earth Science, v. 30, p. 1981-1996.
I’d also like to see science journalism improved. For instance, in this case the reporter might have asked how many of the reviewers were neutral or held a different position from the one put forth by the review. It wouldn’t be particularly impressive if a group of partisans got together to review the evidence and found that it supports them. That would just seem like an attempt to marginalize other positions even if the reviewers are part of the majority. A “robust minority opinion” isn’t extinguished by majority declaration. This whole we’re the majority so shut up and go away approach is a disservice to science.
Four legged mammals still bear the hallmarks of the reptile-like creature we evolved from (especially in our pelvic girdle). Don’t even get me started on humans! We’ve got more design flaws than a Toyota.The mammalian line split off through the synapsids before the ‘reptilian’ line. So mammals as a clade have actually been around longer than reptiles as a clade.An impact extinction can also have some extended events. The most likely impact is not a “bolt from the blue”, with a single large object coming in from beyond Pluto and on the first pass smacking into Earth.
New York liposuctions
Bryan, Thanks for the detailed analysis. I am aware of the factors you raised, and have discussed them with paleontologists in the past. In fact I interviewed Macleod aboutt this very topic.
I disagree with your conclusion, however. While many definitions are fuzzy around the edges and have demarcation problems, that does not mean they are not useful. We can still ask:
Is the existing evidence consistent with the impact theory – yes.
There is solid evidence for the impact, which was large enough to be a plausible cause for the extinction. The extinction event occurs at the impact to within the resolution of the evidence (which is not perfect, but the best of it is better than you suggest, in my opinion.)
There is always uncertainty in science, but I do not think this is a false consensus. My conversations with paleontologists suggests this is, in fact, a very strong consensus.
bachfiend, I’m not trying to say that the birds are doing better than the mammals — that sort of declaration would require a very precise definition of “better”. I guess what I’m getting at is that birds SEEM much more elegant and well adapted to their niches than mammals.
Of course, now that I read that I see it is a very unscientific thing to say and likely has a lot do to with my admiration for birds from an engineering standpoint.
I’d like to restate my unscientific claim in more refined (but equally unscientific) language:
Mammals seem like a more adaptable group overall as opposed to birds, which seem better adapted. Hopefully that makes sense outside of my head.
Either way, I think we can all agree that humans bodies have some serious design flaws (like the vertebral arteries — those things really freak me out).