Unveiling the Mysterious American Mastodon

[Andre]

With some regular intervals we have a look at the Late quartenary megafauna extinction. Currently the American Mastodon (Mammut americanum) has captured our attention.

When diving in the literature we observe many dating "problems" like this for instance:

http://www.calvin.edu/academic/geology/mastodon/poster/proto4.htm

"The date returned was 3400 + 130 years B.P. The sample is believed to be contaminated with organic carbon from the surrounding sediment, based on a 13C of -28."

or this real gem:

https://kb.osu.edu/dspace/bitstream/1811/5422/1/V68N06_257.pdf

"For various reasons, particularly contamination and sampling of material which was not contemporaneous with the bones, several of the youngest dates have been rejected as nonreliable (Hester, 1960; Skeels, 1962; Dreimanis, 1967a; and Martin, 1968)."

Well, there were many problems with carbon dating in those days but on the other hand, can you reject dates that do not fit your idea / hypothesis / theory? In that timeframe, Martin really had forced his will upon the community, the megafauna was over-killed when the first humans allegedly arrived in North America somewhere between 12-10ky crossing the Beringia land bridge. No doubt about that (Not!). So what is your choice when you find datings of 8000 years or 6000 years or 3400 years? It does not fit the theory so it must be contaminated.

An excellent example of circular reasoning, the first half that is.

Nowadays, we know that extinction took place at a much greater intervals. The horses and woolly rhinos disappeared shortly after the onset of the Bolling event some 14,000 Calendar BP, the last woolly mammoths on Wrangel island at 4150 years Cal BP (3700 14C years). The Irish Elk was also one of the last to go around 6000 years ago. So, why not the American Mastodon? After all, the animal was a browser, specializing on consuming trees, which were abundantly available in the early wet phase of the Holocene, the Preboreal. So, if he survived the Younger Dryas there was no more reason to get extinct in the Preboreal when his biotope restored nicely.

Worth ducking into. So we need all the datings of the mastodon and for comparison his buddy, the woolly mammoth (Mammuthus primigenius) who lived in a completely different biotope (steppe) though.

 

[Mk]

From what I've read there are a lot of problems with especially radiocarbon dating techniques. Why is it still used so much?

 

[Andre]

No, carbon dating is rather thrustworthy provided that you know what you're doing, because it is pretty complicated. Most of those problems are created by creationists. Actually, the many calibrations between dendrochronologic proxies and other sorts with countable annual layers show the limits of the reliability as well as the error margins. he reliability shows when these experiments proof to be reproduceable. Much is known about fractination when also monitoring d13C, which helps determining initial delta14C.

The problems come when you blame carbon dating for your own errors.

For instance. The hypothesis is that Mastodons died out between 12-9 Kya (carbon dated) due to the humans invading North America and overkilling the species.

Now we find a mastodon carbon dated to 3400 years. So we don't accept the date due to contamination of the sample. That must have been "in situ" (since we don't contaminate in labs anymore). So perhaps biologic contaminated water penetrating the bone, which changes the delta14C contents.

But there are many more mastodon dates rejected because those did not fit the hypothesis.

Now I've made a little spreadsheet which simulates that process. So if I assume that the basic age of the bone is 9000 carbon years, I need a contamination of 50% carbon of 100 years age to attain a mixed result of 3400 yeas.

So which is in error, the carbon dating or the hypothesis?

edited, adding some info

 

[Mk]

How do daters find what the background carbon levels were? How many years back is it reliable?

 

[Andre]

Point is to first calibrate carbon date readings against countable dating. Dendrochronology is the simplest example. You can carbon date that wood but also use series of tree rings to count every year. I believe that this goes back to some 13000 years now.

Independent series of tree rings give persistently the same reading within narrow error margins, which is very important for thrustworthiness.

Another very important countable proxies are stratified annual sediment layers in lakes (known as http://www.evolvingearth.org/mcabee/lakemcabee.htm , - Swedish, this is where the technique was develloped. Yearly there is blooming of diatoms which gives a white layer against the accumulation of dust. Those diatoms can also be carbon dated, thus calibrating the carbon dates against the counted varve- dates.

Here is the most recent calibration site for carbon date to calendar date which is a combination of dozens of counted - carbon dated proxies. You'll find that the this also enables to calculate the original delta 14C levels with remarkable fluctuations. Difference between carbon date and calendar date can accumutale to several thousand years.

The INTCAL04 table goes back to 26000 calendar years, or 21341 +- 108 carbon years.

 

[Mk]

Point is to first calibrate carbon date readings against countable dating. Dendrochronology is the simplest example. You can carbon date that wood but also use series of tree rings to count every year. I believe that this goes back to some 13000 years now.

You have to have really old trees in the same area right? Around the world won't do, even if you took an average.

Independent series of tree rings give persistently the same reading within narrow error margins, which is very important for thrustworthiness.

That makes me think that you can take a world average.

Another very important countable proxies are stratified annual sediment layers in lakes (known as http://www.evolvingearth.org/mcabee/lakemcabee.htm , - Swedish, this is where the technique was develloped. Yearly there is blooming of diatoms which gives a white layer against the accumulation of dust. Those diatoms can also be carbon dated, thus calibrating the carbon dates against the counted varve- dates.

This one sounds like circular reasoning. Based on the carbon isotope levels, you find how old the varves are, which tell you the carbon isotope levels?

 

[Andre]

You have to have really old trees in the same area right? Around the world won't do, even if you took an average.

To clarify,

You don't need a single tree for counting, just a lot of dead wood from the same area of different age. The typical patterns in tree width variation are shared by all trees and allow for matching up to construct a continuous chronology

http://www.ltrr.arizona.edu/dendrochronology.html

This one sounds like circular reasoning. Based on the carbon isotope levels, you find how old the varves are, which tell you the carbon isotope levels?

No you don't need the carbon dating to establish age. In some places in the world it is possible to count varves from zero to 45,000 years like http://www.cio.phys.rug.nl/HTML-docs/Verslag/97/PE-04.htm . So you got the age of the samples you're counting without having to do any processing or making assumptions, then those samples are carbon dated giving the difference between carbon date and real age.

And again, the strenght of the method is reproduction word wide. The calibration table for instance ends much sooner than this 45,000 years of Lake Suigetsu because there are not enough proxies old enough to validate the results.

Especially the period around the end of the last ice age has been studied intensely like http://www.sciencemag.org/cgi/content/full/290/5498/1951 because of it's strong variation of delta-14C.

Another way of validating varve counts for discontinuing proxies and carbon dating is using widespread volcanic ash layers or "tephra" Those can be dated using several radio-active uranium series. Those can be used as starting or end calibration date, this game is known as Tephrochronology.

It's this kind of validation that should restore the confidence in the posibilities and limitations of carbon dating and it should be clear that refusal of dates, which don't fit the hypothesis, should not be done anymore.

 

[Andre]

So what would this do to you if you'd just decided that the Mastodon did survive the end Pleistocene extinction and lived on well into the Holocene:

http://www.etheric.com/Superwave/Ch10.html

Galactic Explosions, Cosmic Dust Invasions, and Climatic Change"

From Chapter 10 "The Terminal Pleistocene Extinction Episode"
©1983 Paul LaViolette


. . .In the remainder of this section it will be argued that the majority of the extinctions actually took place in a short space of time about 12,000 C-14 years BP (~14,000 calendar years B.P.),* and that the apparent temporal dispersion of the dates may be due to a number of causes. For example, excessively old dates may be due to contamination of the remains with old carbon or to the alluvial reworking of older fossils from preexisting deposits. Excessively young dates may be due to contamination with young carbon or may reflect an unusually high C-14 content in the remains associated with a brief episode of intense cosmic ray bombardment. Martin (1967), for example, argues that many radiocarbon dates assigned to the remains of Pleistocene megafauna should not be trusted. He cites the problematically young date of 2,040±90 C-14 years BP found for the terminal Pleistocene deposits in St. Petersburg, Florida and also questions the youth of several mastodon remains found in Michigan and dated at around 6000 C-14 years BP. Because of these few dates, which are in obvious error, he warns that all C-14 dates of mastodon remains may be in error to some degree. He states (p. 98):

Is it possible that all postglacial dates on mastodons are overshots? No skeptical archaeologist would consider accepting a radiocarbon date of 6,000 to 8,000 years on an alleged Clovis site before subjecting it to the most minute excavation and examination, without demanding an effort at replication of the date on the critical beds, without considering carefully all the possibilities of intrusion, and without a field demonstration of the evidence to equally critical colleagues.

The increased radioactivity is an interesting point and revived recently by http://www.lbl.gov/Science-Articles/Archive/NSD-mammoth-extinction.html.

There are however two problems with that, there is no close supernova known from that time and the anomaly appears to pertain only to Mastodons.

So the simple solution is that the Mastodon survived and the carbon dates are just real within their error margins.

 

[Mk]

I was looking at different hypothesized causes for climate change in the past, and the title involving cosmic dust and galactic explosions... it occurred to me that it would be very interesting for the general megafaunal extinction to have been caused by something like say, a large influx of cosmic rays. Maybe they were the most suseptable animals to cancer back then? Or maybe the only ones that could get it? Or cosmic dust, tons and tons of some kind of nanoparticle hit the Earth and the animals breathed it in, in addition to eating it in. The largest got the highest concentration of the nanoparticle, which interfered with their brain. I just read a study last night in a nanotech news report talking about how subjects breathed in some sort of nanoparticle bromine compound, which they found through MRI scans to have reached the inside of the brain within hours.

Shaviv and Veizer (2003, http://www.envirotruth.org/docs/Veizer-Shaviv.pdf ) proposed that the biggest long-term influence on temperature is actually the solar system's motion around the galaxy, and the ways in which this influences the atmosphere by altering the flux of cosmic rays received by the Earth. Afterwards, they argued that over geologic times a change in carbon dioxide concentrations comparable to doubling pre-industrial levels, only results in about 0.75 °C (1.3 °F) warming rather than the usual 1.5-4.5 °C (2.7–8.1 °F) reported by climate models (http://www.grida.no/climate/ipcc_tar/wg1/122.htm) . They acknowledge (Shaviv and Veizer, 2004) however that this conclusion may only be valid on multi-million year time scales when glacial and geological feedback have had a chance to establish themselves. Rahmstorf et al. 2004 (http://www.pik-potsdam.de/~stefan/Publications/Journals/rahmstorf_etal_eos_2004.html) argue that Shaviv and Veizer have highly and arbitrarily tuned their data, and that their conclusions are unreliable.

 

[Andre]

Well attempting to make a 545 million year hockey stick is liaible to a lot of critique. I would think that the changing geography, with forming supercontinents twice, floating around and passing poles would have much more effect on global climate than any other fluctuation. Anyway, that's off thread.

About a galactic explosion scenario and the extinction, I mentioned the interesting hypothesis of Firesone:

http://abob.libs.uga.edu/bobk/nuclear.pdf

in which he argues that a supernova may have caused the megafauna extinction especially in the Great Lakes area. The intense neutron bombarment may have reset the caron time clock by changing 14N in the animal tissues into 14C.

I think that there are several problems with that. First of all I don't think that supernova's light years away do cause neutron bursts given the half time of neutron being some 19 minutes I seem to recall. Second, the neutrons would have encountered much more 14N in the atmosphere to change into 14C when looking at the delta in the dead animal tissues that would be enough to cause an unprecented spike in many proxies. Although there are many spikes in many proxies but none stands out as produced by a supernova.

Lastly, why a supernova only affect a relatively small area, why not a complete hemisphere albeit it with a more gradual diminishing towards the horizons.

His new idea about a super radioactive comet requires some thought

http://www-library.lbl.gov/docs/LBNL/582/08/PDF/LBNL-58208_Ext_Abs.pdf

 

[Mk]

the half time of neutron being some 19 minutes I seem to recall.

The half-life is about 10.3 minutes, and the mean lifetime about (±2)885 seconds.