Hiding Data

JOE MEERT

I am just curious as to whether or not we will ever be presented with the
evidence for discarding of radiometric ages that was made on this board in
November of last year. This is a serious accusation aimed at modern
science. In my experience, this is not true at all but I am willing to be
convinced by a preponderance of data. Remember, published works don't
count. You must supply evidence that most radiometric ages are tossed out
before publication. I read lots of articles and the good are presented
alongside the bad so I would like to know who these people are that are
hiding the data. Anyone?

 

HELEN

The following quotes and notes are from a number of sources and should back up the point that there are quite a few dates which are discarded where radiometric dating is concerned – they never get mentioned in any publication.

<BLOCKQUOTE>quote:</font><HR>One of the most obvious problems is that several samples from the same location often give widely-divergent ages. Apollo moon samples, for example, were dated by both uranium-thorium-lead and potassium-argon methods, giving results which varied from 2 million to 28 billion years. Lava flows from volcanoes on the north rim of the Grand Canyon (which erupted after its formation) show potassium-argon dates a billion years "older" than the most ancient basement rocks at the bottom of the canyon. Lava from underwater volcanoes near Hawaii (that are known to have erupted in 1801 AD) have been "dated" by the potassium-argon method with results varying from 160 million to nearly 3 billion years. No wonder the laboratories that "date" rocks insist on knowing in advance the "evolutionary age" of the strata from which the samples were taken -- this way, they know which dates to accept as "reasonable" and which to ignore. Of one thing you may be sure: whenever "absolute" radiometric dates are in substantial disagreement with evolutionary assumptions about the age of associated fossils, the fossils always prevail. ]http://www.gennet.org/facts/metro14.html<HR></BLOCKQUOTE>

A collection of radiometric data sets were reanalyzed to determine their relevance to defining previous sea level positions. Criteria for analysis included core
location, material dated, and position with the core sequence. Unsuitable dates were removed from the data set based upon five possible parameters: reworked
shell material, peat compaction (invalid or uncertain vertical placement), root contamination, insufficient sample or positioning data, or self-conflicting data. The
reduced data set is shown in Figure 13.
www.geology.iupui.edu/research/sedlab/data/Lenore/sea%20level.doc

(in checking links, the article this was part of appears to have been pulled. Interesting… )

<BLOCKQUOTE>quote:</font><HR>Another disconcerting fact: the "correct" ages which
appear are the result of a selection (Holmes, 1965), the
author having only accepted the "best values" (those
confirming Lyell's stratigraphical scale), rejecting the
others as "anomalous."
http://www.rael.org/int/english/evidence/evidence/body_evolution2.html
<HR></BLOCKQUOTE>

Dave Tyler sent me the following in a private email:

In June 2000, Paul Garner wrote:
"Last night I spoke at Histon Baptist Church, Cambridge on
"Creation or evolution?" .. .. Another gentleman was a geochemist
at the British Antarctic Survey, working in the field of
geochronology. He is a Christian but convinced of an old earth. We
had a good conversation afterwards, and he said that although he
disagreed with my position, he felt that I'd "made a good case for
it". I was encouraged by some of what he had to say about
radiometric dating. This is his field of work and so he was very
knowledgeable on the subject. In his opinion, rubidium-strontium
(Rb-Sr) isotopic data are very difficult to interpret because false
isochrons are so common, and potassium-argon (K-Ar) dating is so
fraught with problems that "nobody believes K-Ar dates" (his words)
anymore. However, he pinned his confidence in radiometric dating
on other methods, such as uranium-lead (U-Pb) dating of zircons -
which has a whole set of problems all of its own."

The point I would make here is that most people would never know
from the literature that false Rb-Sr isochrons are so common, and
that "nobody believes K-Ar dates". Such knowledge comes from
insiders who see the results before they are published.


In a private email from Dr. Tim Standish at GRI:

<BLOCKQUOTE>quote:</font><HR>In Arial Roth's book he cites Runge et al 1973 “Radiocarbon chronology and problems in its interpretation for Quarternary loess deposits-South
Canterbury, New Zealand”, Soil Science Society of America Proceedings 37:742-746. in which they give a sequence of C14 dates of soil layers going from top to bottom of 9,900, 12,000, 27,200, 17,300 and 15,650 years. Obviously the last two dates can't be right and they are dropped out of a subsequent paper, Tonkin et al. 1974. “A study of late Pleistocene loess deposits, South Canterbury, New Zealand. Part 2: Paleosols and their statigraphic implications.” Quarternary Research 4:217-231.<HR></BLOCKQUOTE>


The following came in from a friend who is a graduate student now:

Hi Helen,

This shouldn't even be a debated issue, because it's a fact. I used
radiometric dates in my masters thesis, and if they seemed to be "off" a bit, they were simply said to be wrong. Of course I did what I was told to do, but my feeling was that the commonly accepted wisdom is that the methods are fallible, and so if the date is wrong, no worries! My question then becomes, how do we know when it is right? It's not talked about much. God bless!

--Casey


A personal friend of ours is Dr. Graham Mortimer:
http://wwwrses.anu.edu.au/egg/Pages/eggMortimer.html


Barry and I both took the time individually to talk with him on the phone about the discarding of dates. Graham works with zircon crystals. He stated that about 15-20% of the dates were considered anomalous and thus not used. Keeping in mind that zircon dating is probably the most reliable and yet, as Andrew Snelling has mentioned, still frought with problems
http://www.icr.org/pubs/imp/imp-319.htm

it is quite reasonable to assume that more than 15-20% of the dates obtained by other methods would be considered anomalous/unreliable and thus not used.
I am willing to take back “most” as me believing something I was told without checking it myself, but the more I ask and read, the more aware I am of two things:
1. There are quite a number of dates that are discarded by lab or researcher or both as being wrong from the start
2. and almost no one is willing to talk about it.

Helen

[ February 20, 2002: Message edited by: Administrator ]

 

JOE MEERT

Helen, once again your answer is laced with 'what-if' scenarios and a few
unverifiable personal communications. The evidence you do seem to cite is
from published sources. Since I not only conduct geochronological studies
but also review geochronologic studies, read the literature and work in a
University with two geochronologic labs, I think that I am quite familiar
with the mode of operation. However, let's assume, for the sake of argument
that 10-15% of ages produced are anomalous. That leaves 85% of age
determinations that are not. How do you explain the 85% of ages that are
not anomalous? You must also remember that we are dealing with natural
systems where alteration can and does take place. Geochronologists have
devised a number of ways to deal with alteration of systems and do so on a
continual basis. I went back and looked at the percentage of age
determinations that I personally have worked on and less than 5% are
troubling. Each one of those 5% have a reasonable explanation for why they
didn't work (from excess argon clearly evident in the plateaus to severe
hydrothermal alteration of the samples). In any case, your argument that
most dates are toosed out before publication fails yet again. Think about
it for a minute. If geochronology was so utterly unsuccessful, the geologic
community would stop using it. They do not. The ages produced in labs have
verified field relationships time and time again. This fact argues my point
much more strongly than your accusation of subversion on the part of the
geochronologists. Finally, I might also add that none of these anomalous
ages points to a young earth. You would think that somewhere in that 10-15%
of anomalous ages that you claim exist, there would be a bunch of 6000-10000
year old ages. Why don't you produce some of this evidence? In fact, let's
assume that your 'friend' Mortimer thinks that dating is unreliable. Why
does he continue to use the method and publish articles about the method?
Seems like if something is so horribly awry with radiometric dating, Graham
would stop using it and call for its abolition. Yet, there he is publishing
away on the subject and continuing to use it on a daily basis. Seems like
that would be such a waste of time for such an awful technique! I must
conclude on the basis of your arguments that you really don't have any hard
data available to substantiate your claims. Your insistence is simply that
geochronologists are dishonest. I find that absurd.

 

DAVID PLAISTED

My recollection is that the great majority of isotopic dates are K-Ar
dates, so that if 85 percent of dates agree it could be largely because
K-Ar dating agrees with itself much of the time. A better statistic
would be how often different dating methods (different parent and
daughter substances) agree on the same sample or formation.

I recall that there is a technique called "mineral isochrons" in which
different minerals from the same rock are dated by the same method, say
Rb-Sr dating or U-Pb dating -- I'm not sure which would be appropriate.
This can be treated as an isochron since if all minerals cooled at
about the same time, any changes in parent to daughter substance would
have arisen since then (if the system were not disturbed etc.). One
would expect different minerals to absorb different amounts of parent
and daughter substance on cooling so one should get a good isochron.
At least I can't think of anything that would throw off this method.
So my question is how often are such "mineral isochrons" or should I
say "multiple mineral isochrons" done? I read that they are not done
much because of the difficulty of isolating the different mineral
fractions, but if they give more reliable dates it might be worth the
effort. What would be especially interesting is if two mineral
isochrons from the same rock, using two different parent-daughter
substances, gave nearly the same date -- has this been found? If such
a method gives "good" dates then either the sample is old or decay
rates were faster in the past.

My recollection is that different methods often agree on certain
meteorites, and maybe even on certain very old (appearing) rocks on
earth, but much less often in the Phanerozoic (Cambrian and later).

Dave Plaisted

(Let me also add that it is not so easy to know whether the dates are
biased -- if geologists believe zircons to be more accurate then it would
appear that more and more of the dates in the literature would be from
zircons. Thus the proportion of the dates in the literature could be
skewed towards methods (or sub-methods, or formations) that seem to give
more accurate results, and thus the system could be in some sense
self-fulfilling. Of course this is just speculation but it's hard to see
how this would not bias the results.)

 

HELEN
What I find absurd, Joe, is that you don’t seem to actually read my
posts.

1. I did not say geochronologists are dishonest. I think they are being
quite as straight as they can be in line with the faith in long ages and
the fact that almost all radiometric data points to long ages at the
very least.

2. Dr. Mortimer does not think the dating is unreliable. He simply said
that there was a certain percentage of the dates which were considered
unreliable. There is a big difference.

3. I would not expect short ages out of radiometric dates for the very
reason of the change of decay rates in the past – this change from much
faster than now to now would guarantee that few or no dates would
indicate a recent age.

4. I think the references pointed out quite clearly that when field
assumptions and radiometric dates differ the field assumptions win.

5. I retracted, with apologies, the statement that MOST dates were
thrown out and admitted I had understood or been told wrongly about
this.

 

JOE MEERT

To Dave Plaisted
Your intuition may be correct that most geochronologic data
available are K-Ar determinations. I would not bet against that
conclusion, since it was one of the first and cheapest methods for
determining ages. It also has a wide range of applications so the
cheapness and the range made it extremely popular. It still is used
quite a bit today and it does quite a good job.

Today, our
understanding of closure temperatures and the behavior of the K-Ar
system is much better constrained by using the 40Ar/39Ar method. In
general, most of the K-Ar ages that I have personally examined are quite
robust. In the cases where the ages seemed a bit askew based on more
recent datings, the 40Ar/39Ar system usually reveals the reason why the
K-Ar ages were skewed. In fact, the 40Ar/39Ar has been a nice
addition and way to double check for isotopic disturbances in the K-Ar
system.

In terms of your experiment using different isochrons for different
minerals in the same rock, it has been done a number of times (some of
these can be gleaned from the database in Meert, 2002). You must
remember a few things about such an experiment. The first has to do
with the idea of isotopic closure. Different isotopes close under
different conditions (depending on the isotope, temperature and grain
size). Let me give you two examples that I am personally familiar with.
There are many others out there, but I am at home and these are both in
my brain. The first is from the Carion granite in Madagascar. The
granite has been dated using a variety of methods. U-Pb SHRIMP ages and
U-Pb evaporation ages overlap with small errors at ~535 Ma (Kroner et
al., 2000; Meert et al., 2001). The U-Pb system closes at temperatures
well above 850 C. 40Ar/39Ar studies on hornblende and biotite and
K-feldspar yield younger ages of ~513 Ma and ~480 Ma consistent with
their experimental closure temperatures of 500 C and 350 C respectively.
Multidomain K-feldspar studies yield younger ages and cooler
temperatures that helped us develop a composite cooling curve (see

http://www.indstate.edu/gga/pmag/radiomet.htm).

This study was done in
conjunction with a paleomagnetic study on the same rocks. A
pre-existing apparent polar wander path had been established for the
continent of Gondwana (Meert and Van der Voo, 1997). The pole for the
Carion granite fell on this APWP at 510 Ma. Temperature studies on
magnetic minerals within the granite combined with the cooling curve for
the granite yielded a magnetic blocking age of 508 +/- 11 Ma entirely
consistent with its location on the APWP (details are described in Meert
et al., 2001a, 2001b).
A second study that has been done using multiple isotopic
systems, in multiple labs over a number of years resulted in wholly
consistent and overlapping ages for the Fen carbonatite complex in
southern Norway. This study is also described in the above link. Such
agreement between systems and labs is pretty tough to refute without
calling on conspiracy or some miraculous (and yet undescribed physics).
It may happen, but a much simpler explanation is that science has it
right and radiometric dating does, indeed, work. As for your assertion
that radiometric dating is better for older rocks, I’d love to see data
confirming that. As far as I can tell from my knowledge of the
literature, it works well across the board. In addition, given that the
longer rocks are around, the more likely the possibility for younger
disturbance, I would have anticipated the opposite assertion. That is
younger rocks should yield the most consistent ages. The best part
about the whole process is that we can now tell, with some ease and
care, when isotopic systems have been disturbed. In fact, some of these
thermal disturbances have helped us decipher tectonic histories with
greater precision than ever (see Meert, 2002 available at

http://www.clas.ufl.edu/users/jmeert).

I would indeed argue that
radiometric dating is healthy, happy and wonderfully consistent with the
rock record.

Finally, let me state that while it is time consuming to separate
minerals, it is not all that big of a deal and there are many more Rb-Sr
mineral isochrons generated now than previously. This is in response to
valid criticisms regarding the co-genetic assumptions of whole rock
dating. I don't think there is anything inherently difficult about
separating minerals at all.


FOR HELEN

Thanks for withdrawing the blanket claim against radiometric
dating and for acknowledging that science, for the most part, is a noble
pursuit. As I mentioned, even if we assume your 15% bad age estimate is
correct, that still leaves 85% supporting the method. As I also
mentioned above, I would guess that we can explain the anomalous
behavior in 90% of those 15% of bad ages. This is part and parcel of
the method. We constantly check and double check the results for signs
of isotopic disturbance. What constantly amazes me is how robust the
data really are. Radiometric dating is attacked by creationists quite
often because it is such a powerful tool for showing the age of the
earth. Young earth creationism, in particular, falls apart in the face
of radiometric data and they must maintain that it doesn’t work. At
the same time, there is very little evidence available that supports the
notion of a young earth from the radiometric side of things. I don’t
fault you for trying to disprove the methods because you have to. I
just hope that the arguments amount to more than innuendo and slander
against geochronologists. Most of the ‘case-studies’ conducted by
creation scientists suffer the same problem. None of the scientists are
trained in the field and laboratory methods and it causes
misinterpretation of the data (see

http://www.indstate.edu/gga/pmag/crefaqs.htm).

I notice that the RATE
group at ICR has hopes of overturning radiometric dating, but their
methods are also somewhat suspect at the moment (see

http://www.indstate.edu/gga/pmag/rate.htm).

Still, I know that
radiometric dating is a big threat to the young earth creationist
community and they must continue to attack it at every opportunity. So
far, not much has come of the attacks, but at least there is no lack of
trying.
As for your earlier innuendos based on e-mailed accounts of personal
recollections, I can tell you that K-Ar ages are viewed within their
full context and the statement that 'nobody trusts K-Ar ages' would mean
that they would not get published, yet they are. As with all isotopic
studies, the authors must document the proper tests so that the data can
be fully evaluated.

Kroner, A. et al. Age and magmatic history of the Antananarivo block,
central Madagascar as derived from zircon geochronology and Nd-isotope
systematics, Am. J. Sci., 300, 251-258 (2000).

Meert, J.G. A Synopsis of Events Related to the Assembly of Eastern
Gondwana, Tectonophys., in press. Tectonophys., (2002, in press).

Meert, J.G., Nédélec, A., Hall, C.M., Wingate, M.T.D. and
Rakotondrazafy, Paleomagnetism, geochronology and tectonic implications
of the Carion granite, central Madagascar, Tectonophys. v. 340, p 1-21
(2001a).

Meert, J.G., Hall, C., Nédélec, A. and Razanatseheno, M.M, Cooling of a
late-syn orogenic pluton: evidence from laser K-feldspar 40Ar/39Ar
modeling of the Carion granite, Madagascar, Gondwana Research, 4:3,
541-550 (2001b).

Meert, J.G., Torsvik, T.H., Eide, E.A. and Dahlgren, S., Tectonic
significance of the Fen Province, S. Norway: Constraints from
geochronology and paleomagnetism, J. Geology, 106, 553-564, (1998).

Meert, J.G. and Van der Voo, R., The assembly of Gondwana (800-550 Ma),
J. Geodyn., 23, 3-4, 223-235, 1997.