[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Paleontology and Signal-Processing

I have just read an interesting article titled "Notes from the Underground" 
by Alan Cutler of the NMNH at the Smithsonian, published in the Jan/Feb 1995 
issue of The Sciences. The article is an explanation as to why there are no 
dinosaur remains found at the KT boundary. The following are excerpts from 
the article. I recommend it for an interesting approach to the 
paleontological record.

"Even the famed iridium layer, which many investigators consider 
near-irrefutable evidence of a collision with an iridium-rich 
extraterrestrial body, seems less impressive when viewed up close. In the 
deep oceans, where sediments accumulate at the rate of millionths of a meter 
a year, the iridium peak is smeared vertically through several centimeters 
of sediment - hardly what one might expect of the debris from an 
instantaneous impact."

"... But in this case the event of record is a cosmic catastrophe that 
killed all the dinosaurs in the world. Shouldn't the so-called iridium spike 
be a little more spike-like? Shouldn't the concentration of bones in the 
fossil record be, at very least, above average?"

"The ups and downs of life on earth constitute a signal, as real as the 
voltage in a telephone wire or the rays of light striking your retina. In 
that sence, a paleontologist studying the KT boundary resembles an 
astronomer peering at a distant star. Just as the light reaching the 
astronomer from the star is distorted by imperfect lenses and roiling 
 atmospheric gases, so the signal of the KT event reaching the 
paleontologist is distorted by the imperfect record of life preserved in 
sedimentary rock. And just as atronomers have devised a host of digital 
signal processing techniques for sharpening blurred images and enhancing 
contrast, paleontologists are starting to develop ways of squeezing as much 
information as possible out of the fossil record. "

"... the mixing (of sediments) shifts upward and the particles (or bones) 
are left behind as part of the historical record. In effect, mixing acts as 
a convolution operation, and the mixing layer serves as the "kernel" ... In 
the mixed ... record spikes in the abundance of species are attenuated and 
tail off exponentially ... Thus the sudden extinction of a species shows up 
not as an abrupt disappearance of fossils but as a gradual petering out." 
(at least in marine sediments).

"... Anna Behrensmeyer and I developed just such a convolution model of 
fossil preservation ... Starting with a hypothetical population of 
dinosaurs, we estimated their normal annual mortality from ecological data 
collected for large mammals in African wildlife preserves. Next, to estimate 
what fraction of the dinosaur's bones would end up safely buried, we drew on 
data from Behrensmeyer's 15 year study of the decay of mammal carcases in 
Amboseli National Park. For our kernel we assumed that the layers were time 
averaged over 100 years or more. Finally, we ran the model to see what sort 
of bone spike would result if the entire population of dinosaurs suddenly 

"The answer, we discovered, was no bone spike at all. For all reasonable 
estimates of our parameters - annual mortality between 5 and 25 %, bone 
burial at about 3% - time averaging completely swamped any such effect. Only 
by assuming circumstances unusually favourable to preservation ... could we 
raise the accumulation of bones above the normal background value."

"If normal mortality was 10% (to choose an intermediate value), 100% 
mortality merely compresses 10 years of mortality into 1 year. In a record 
that "convolves" several hundred years, that would hardly generate a blip. 
Put another way, consider that the impact of an asteroid 10 km in diameter 
would have increased the input of iridium to the earth almost 15,000,000 
times above background level. Yet in most deep-ocean sediments that iridium 
spike has been so smeared that it now measures less than 100 times the 
background - a reduction by a factor of 150,000. Small wonder that the bones 
of the dinosaurs left such a feeble signal."

I hope that this is not too long for the list. If so, please excuse me. It 
was an insightful article about an area of paleontology that I had not 

Miles Constable
Toxic Chemicals Biologist by necessity,
By-stander and Observer of Paleontology by interest