Miracles In, Creationism Out
On June 16, 1997, a major weekly magazine, US News & World Report published "The Geophysics of God" reporting claims that a sophisticated computer program written at the Los Alamos National Laboratory (LANL) and "used by geophysicists around the world... proves the Bible is correct" (Chandler 1997). By mid-July NCSE had received copies of letters submitted to a large state textbook commission making similar claims. One letter said that textbooks should include "Flood geology as a theory for mass extinction... [since] this idea is currently being studied in some of the national laboratories in New Mexico"; another letter cited a sophisticated-sounding argument used by geophysicist John Baumgardner, the scientist profiled in the article, claiming that different methods for dating ancient rocks yield different results.
NCSE members need to know the flaws in these arguments because this "evidence against evolution" may appear in an opinion piece in your local newspaper, or, worse, a proposal to your local board of education. NCSE has consulted several geologists for detailed comments on the US News article. Excerpts from the article are presented in italics followed by responses from scientists working in the relevant fields.
Terra [a computer program for modeling the flowing movement of the earth's mantle]...exists because its creator, John Baumgardner, is a fundamentalist Christian who believes, in accordance with the Bible, that the Earth was created by God less than 10 000 years ago...[and] created Terra expressly to prove that the story of Noah and the flood...happened as the Bible tells it".
There are good scientific reasons for developing a computer model of mantle flow, independent of Baumgardner's motivation. His views do not represent those of LANL. According to Dr. Jerry M Boak, another geologist at LANL, "Creationist modeling was distinctly not part of the work scope. Scientific publications of research using the Terra program have nothing to do with ...flood geology..." (For the history and the results of this research, readers should examine Baumgardner 1985, Bunge and Baumgardner 1995, or Bunge, Richards, and Baumgardner 1996).
Boak requested further information from Dr. Chick Keller, Director of the Institute for Geophysics and Planetary Physics at LANL, who explained that Baumgardner had originally written Terra as a part of his doctoral dissertation. According to Keller "...[Terra] fell into disfavor when it could not adequately reproduce results [derived] from other 3-D spherical mantle convection codes." Keller told NCSE that Peter Bunge and other workers later "improved it considerably, to the point that the community now accepts it as [reasonable].... [However], [Baumgardner's] version is somewhat different from [Bunge's] and still needs benchmarking — that is, testing against other codes to assure that it gives reasonable answers to well-defined problems."
Not only did he come up with a tool used by geophysicists around the world but his "numerical code" actually proves the Bible is correct. Or at least in Baumgardner's view it does.... Terra proves that this is true — or, more precisely, that it could be true, provided one accepts certain assumptions.
Computer models do not prove anything; they can help work out the consequences of certain assumptions. One can feed a computer program data about cheese and rats and find out how large a rat must be in order to take crater-sized bites out of a moon made of green cheese, then illustrate the process with realistic graphics. This would not prove that the moon is made of cheese or that there are giant, space-going rats. The article's author admits that Baumgardner's explanation of the flood "contradicts almost every existing piece of evidence" (emphasis added). For example, accepting Baumgardner's assumptions would include changing the values of known physical constants such as rates of radioactive decay.
Keller commented that it was unfortunate that "The article gave the impression that one need only change a few numbers like viscosity and you can get anything from mantle convection to continents' zooming around." He added that there are important questions about other variables that Baumgardner's model does not consider. As for the computer program itself, Baumgardner's "Terra" is not the only mantle modeling program in use and, Keller commented, while Bunge and possibly some other scientists have gained "considerable respect and support" for their use of some versions, "I don't think it's very widely used."
Baumgardner believes that around 6000 years ago,...[God] caused an enormous blob of hot mantle material to come rushing up at incredible velocity through the underwater midocean ridges. The material ballooned, displacing a tidal wave of sea water over the continents.... Then, after 150 days (Genesis7:24), the bubble retreated with equal speed into the Earth.
According to University of New Mexico geologist John W Geissman, this series of events would create "an enormous volcanic province in a single region. So, where is it??" Keller added, "Blobs don't just emerge and retreat...; one needs a source mechanism.... Current blobs which emerge, for example, out of a volcano or midocean ridge, are rather local and short lived...."
[A]nd the continents began re-emerging above the water, sending the runoff back to the oceans at around 100 miles an hour....Baumgardner says that this runoff would have been sufficient to create the Grand Canyon and other massive geologic features and to deposit the various sedimentary layers in about one week.
The phrase "various sedimentary layers" glosses over a great variety of geological formations, many of which could not possibly have been deposited in one week, least of all by currents traveling one hundred miles per hour. Examples of these include sediments that settled in still water, those transported by wind, and those formed by evaporation or precipitation of dissolved chemicals (Strahler, p 61, 170). Not only does one week provide too little time for sediments to be deposited, Baumgardner does not explain inter bedding of sediments clearly formed by different processes; for example, a layer formed of wind-borne particles sandwiched between layers of sediment that had settled from still water.
The US News article also does not say what kind of land surface was being eroded. Arthur Strahler's comments on erosion by forty days and forty nights of rain apply to Baumgardner's model of streams rushing back to the ocean: "Fully lithified, hard, dense rock — such as ... [various] kinds of igneous and metamorphic rock... — could withstand forty days and nights of torrential rainfall with scarcely a measurable quantity of erosional removal.... Even on the assumption that a thick (100-meter) layer of decayed rock (saprolite) was available... it would be woefully inadequate to supply the quantity needed to form all existing Proterozoic and younger sedimentary and metasedimentary rocks" (Strahler, p 201). Simply put, sudden erosion cannot explain the sheer volume of material in the earth's "sedimentary layers".
Almost all physicists calculate the age of the planet at 4.6 billion years because they assume that mantle viscosity...has been consistent throughout time and so use the value that applies today. They add other ingredients like the speed of the tectonic plates... and arrive at the conclusion that one full deformation cycle of the mantle occurs about every hundred million years, giving the 4.6 billion figure. But Baumgardner says scientists wrongly assume that geology happens consistently,...."If you look at the geological record," he insists, "there are fingerprints of catastrophe everywhere one looks."
G Brent Dalrymple, author of The Age of the Earth, told NCSE, "The viscosity of the planet does not enter into calculations of the age of the earth. There is no way to calculate the age of Earth from plate tectonics or from mantle convection." Boak added, "Certainly there are catastrophes in the geologic record, but they aren't the catastrophes Baumgardner describes, and they don't represent the fingerprints of God spinning up a planet last week."
Baumgardner...notes, first, that different radiometric dating methods give vastly different ages. To date rock, geologists commonly use three types of unstable (radioactive) 'parent' isotopes — samarium, rubidium, and potassium — which decay into stable 'daughter' elements.... But different isotopes yield different dates for the same rocks. As an example, Baumgardner points to the Cardenes [sic]basalt, a Precambrian volcanic rock found in the Grand Canyon's inner gorge.
In fact not three, but at least seven isotopes are used for dating rocks (Dalrymple 1991, p 80). More significantly, Baumgardner simply misinterprets what is to be learned by comparing different isotopic dates.
Boak pointed out that in the Cardenas rock "The sequence of these dates [found by the different dating methods] is exactly what would be expected for such old rock." Rocks can be altered by heating or melting, and Dalrymple commented, "Alteration is readily visible in thin sections of the [Cardenas] rock." The different dating methods date different events in the history of the rock and Dalrymple added, "The results tell us that this particular rock is not a reliable chronometer, but say nothing about the dating methods themselves.... There are many more instances of agreement than disagreement; if Baumgardner finds a wristwatch that doesn't work would he conclude that all watches and clocks are unreliable?" When rocks are altered by heating, the geological clock is reset so that, as Boak put it, "The Cardenas rocks ... may be more like modern stopwatches, which may record several lap times. Because the degree of heating required to reset the various geochemical clocks is different, we may be able to identify several different events."
The science Baumgardner uses to account for these extraordinary happenings is a sort of niche physics called runaway subduction ...[which] posits that the potential energy in the cold, heavy crust of the Earth is like the potential energy in a rock held above the ground. Drop the rock, and its potential energy is turned by gravity into kinetic energy, and into heat when it hits the ground. As gravity pulls the rock, so it pulls the gigantic, heavy plates of ocean floor under the continents into the hotter, lighter mantle, which is silicate rock.
As the plates deform the surrounding rock, the mechanical energy of deformation is converted into heat, creating a superheated 'envelope' of silicate around the sinking ocean floor. Silicate is very sensitive to heat, so it becomes weaker, allowing the plates to sink faster and heating the envelope still further, and so on, faster and faster. As the plates pull apart, the gap between themgrows into a broadening seam in the planet. This sends a gigantic bubble of mantle shooting up through these ridges; [w]hich displaces the oceans; [w]hich creates a huge flood.
"Fringe" physics might be a better description. The article does go on to quote another geophysicist who pointed out that one would have "to believe that by some miracle the diffusivity of the Earth was different before we learned to measure it;" in fact, the miracle would have to increase thermal diffusivity by a factor of 10,000. Boak commented, "The problem with his work on 'runaway subduction' is that the frictional heating he requires to accelerate subduction may not continue, given the reduced viscosity of the subducting materials as they are heated. To my knowledge, this aspect of his model has never survived a peer review by others in ... mantle modeling."
[Baumgardner] cites the common geological feature of erosional channels, like the sunken rivers running through Zion National Park. The walls of these channels, created by rainwater eroding uplifted terrain, show the cross sections of sedimentary layers laid down over millenniums [sic]. But while the evidence of erosion and sedimentation is all around (the Mississippi Delta, the Ganges taking soil from the Himalayas), surprisingly few erosional channels can be seen in the sedimentary layers themselves.
Just as it is erroneous for Baumgardner to state that sediments clearly formed by different processes could have been deposited in a single flood, it is erroneous to lump together many types of channels. Besides, as Dalrymple points out, the article does not state that there were no such channels, but that there were "too few". Yet we are not told how many would be "enough", or why.
Another piece of evidence [Baumgardner] points to is the fact that coal — fossilized plant matter — is found in concentrated seams rather than spread out, as forests generally are. This indicates to Baumgardner that a huge mass of water — a flood — swept floating trees together, depositing them in thick layers.
Baumgardner's argument is a red herring. There are many types of forest environment, not all of them conducive to the initial steps of coal formation. Coal seams originated in swamps in deltas, and their boundaries are those of the deltas. Besides, layers of marine and non-marine sediments can be found between layers of coal, and Baumgardner neither asks how this alternation of layers could occur nor offers an explanation. While the standard geological explanation accounts for these alternating layers, Baumgardner's flood does not. The flood currents bearing trees would have had to alternate quickly with currents moving in a different direction, carrying different materials. Currents moving as swiftly as 200 km/hr would have to alternate approximately every half hour, not allowing enough time for rafts of water-borne trees to settle between the other strata (Strahler, p 218-221). Furthermore, according to Dr. Steven M Getty of the University of New Mexico, fast-moving water currents pick up large quantities of sand and gravel, and mix them with organic matter, while many large coal seams are almost purely organic, often preserving details of plant structure too well to be consistent with Baumgardner's model.
Baumgardner himself says, "The only way to square the radiometric data with a flood that caused all these changes is to conclude that one aspect of the catastrophe was rapid radioactive decay." ...This is not impossible.
As the article itself comments, this is only possible in "a Through-the-Looking-Glass world where nothing is as it seems and no scientific principle — from gravity and electromagnetism on down — exists as it exists today." Moreover, this objection can't be met by calling on miracles, not if one considers other events that would have occurred in Baumgardner's scenario — something the article does not do. For example, Dalrymple points out that "If one 'speeded up' 4.5 billion years of radioactive decay into just a few years, then the heat released over such a short time would melt and probably vaporize the Earth." In effect, readers are being told that it is "not impossible" for Noah's Ark and its inhabitants to have withstood the melting of a planet. Was the "ocean" on which the ark floated one of molten lava (which should have burned the Ark to a cinder), or is the earth supposed to have recondensed and generated new plant life while the Ark's passengers waited?
[T]here is universal agreement that Terra, created to prove the Bible literally true, is one of the most useful and powerful geological tools in existence. "Baumgardner is seen as one of the world leaders in numerical models of mantle convection," says Hager. Agrees Gerald Schubert of the University of California - Los Angeles Department of Earth and Space Sciences: "As far as the code goes, Baumgardner is a world-class scientist."
NCSE wrote to Hager and Schubert to ask about their remarks attributed to them. Schubert replied, who explained, "I did not agree with what Hager [reportedly] said. I simply stated that John Baumgardner had written a state-of-the-art numerical code to study mantle convection.... In no way can John's numerical code say anything about the possible validity of a literal biblical view of creation." Hager, too, told NCSE that he had only been commenting on the computer code, and had not in any way endorsed Baumgardner's opinions.
It is Baumgardner's "scientific" claims that are most likely to be repeated by advocates of "creation science", especially now that coverage in a national weekly has given Baumgardner a much larger audience than the Institute for Creation Research Graduate School where he is an adjunct faculty member. However, the article also underplays the manner in which Baumgardner "takes issue with the teaching of evolution in public schools". He is not just a local "rabble-rouser"; he has been appointed to the committee that is working to develop "performance standards" measuring what New Mexico's school-children have learned about science. These standards are supposed to redress the omission of evolution from New Mexico's content standards (RNCSE, 17 p 4, 17 p 8-9), but Baumgardner's participation in the committee might prevent that from happening.
The US News article concludes, "Belief does not need the blessing of science. But to John Baumgardner ... apparently it does." That, of course, is a personal choice. But as a matter of public policy, public school science education should not be required to receive the blessing of unscientific beliefs.
[With thanks to (in alphabetical order) Jerry M. Boak, G. Brent Dalrymple, John W. Geissman, Stephen R. Getty, Chick Keller, and Brad Hager for their comments and, in some cases, detailed explanations.]
Baumgardner J. Three-dimensional treatment of convective flow in Earth's mantle. Journal of Statistics in Physics 1985; 39:501-11.
Bunge P, Baumgardner J. Mantle convection modeling on parallel virtual machines. Computers in Physics1995; 9: 207-15.
Bunge H-P, Richards MA, Baumgardner JR. Effect of depth-dependent viscosity on the planform of mantle convection. Nature 1996; 379: 436-8.
Chandler B. The geophysics of God. US News & World Report 1997 Jun 16; 122:55-8.
Dalrymple GB. The age of the earth. Stanford (CA): Stanford University Press; 1991.
Strahler AN. Science and earth history : The evolution/creation controversy. Buffalo (NY): Prometheus Books; 1987.
Author(s): Molleen Matsumura Volume: 17 Issue: 3 Year: 1997 Date: May–June Page(s): 29–32