Re: Another Radiometric Challenge

Joel Duff (crinoid@midwest.net)
Sat, 05 Sep 1998 09:45:12 -0700

David,

I appreciate the response and given the information I provided these are valid concerns.  Let me try to address a few of these concerns and expand the example some more.

At 05:07 PM 9/4/98 David Tyler wrote
>David Tyler responding to Joel Duff (Fri, 28 Aug 1998).
>
>> The Hawaiian Islands are
>> a linear chain of Islands that  is youngest at the southeastn most point
>> (i.e. where the active volcanoes are) and get progressively older as one
>> travels toward the northwest.  Many rounds of radiometric dating have been
>> done and a very consistent relationship has been found between the distance
>> of a volcano from the active island and its age. 
>
>DT:  This is an interesting set of data - which must mean something!  The
>problem for earth scientists is: how should this trend be
>interpreted?
>
>As an alternative to the Plate Tectonic model, consider that
>developed by Warren Carey (Professor Emeritus of geology in the=20
>University of Tasmania).  I refer to his book "Theories of the Earth
>and Universe", Stanford University Press, 1988.
>
>His analysis is one which promotes the concept of tectonic control.
>He reconstructs the Pacific Floor in terms of age provinces and shear
>zones.  The Hawaiian ridge runs parallel to the Melanesian Plateau,
>and the Emperor Ridge runs parallel to the Mariana Ridge.  He calls
>the gap between the the Hawaiian Ridge and the MP the "Central-Pacific
>Sphenochasm" and the gap between the Emperor Ridge and the Mariana
>Ridge the "Shatsky Spenochasm".  By applying a dextral global torsion
>(by 34 degrees) he closes these chasms.  These changes are all linked
>to Tethyan widening.  In this model, hot spots are not required, and
>the volcanism is mainly contemporaneous.

A lot to discuss here.  I have not read Carey's book yet but will attempt to obtain it since I am very interested in the topic.   Does Carey suggest that hot spots are not required to explain any linear volcanic chain or just the Hawaiian Islands?  If not at least we can see that the Hawaiian Islands appear similar to other Pacific island chains that have active islands at the very tip of the most southeastern point .  More on other hot spots below.

Secondly, let me add one more bit of Radiometric correlation.   Drowned coral reefs (or whatever you believe them to be) are found in a successive layer of rings around the Hawaiian Islands.  On the Islands of Hawaii these reefs have been dated and a strong correlation is found between depth and age such that a rate of subsidence (we can discuss the evidence that is subsidence later if you wish) can be calculated very simply.  On the active end of the island this rate of subsidence is nearly the same as the measured rate of subsidence using tide gauge data over the past 30 years and dating of drowned Hawaiian rock paintings.   Interestingly, drowned reefs on the far end of the island demonstrate a slightly lower rate of subsidence.  This makes sense given the  loading of new material at the active end of the island and thus a faster rate of "settling."

The  linear relationship between coral age and depth suggests that the rate of subsidence has been the same over a long period of time (100kyr).  To avoid such a conclusion either the radiometric dates must be suspect but then why would a rate determined from radiometric dating be the same or even remotely similar to that determined from modern tide-gauge data?
>
>And what of the radiometric dates?  Carey writes:
>"Plate tectonicists place much emphasis on the decrease in the ages of
>the volcanic rocks along the Emperor and Hawaiian ridges, and
>interpret them in terms of motion of the Pacific "plate" over a
>"hotspot" fixed in the mantle.  But these ages record the times at
>which volcanism _terminated_.  The lavas at Hawaii are 12 km thick.  I
>suggest that the lava at the base is very much older than those
>measured at the surface, perhaps as old as any in the Emperor Ridge."
>(page 321).

It would appear that the speculation that the base of the islands are much older is meant for us to think that the "older" Emperor Ridge seamounts are simply eroded stumps of much larger mountains and so they have simply eroded down into much older rock and the same could occur to the active islands eventually.  But the older islands are not just eroded stumps but rather the depth at which they may be found is due to subsidence.  This is attested to by the fact that these seamounts have the remains of ancient coral reefs on their tops.  If they had been eroded to the point they are the reefs would have been the first to go. 

>The least that can be said about this is that there is here an
>alternative model - and I am a great believer in the evaluation of
>multiple models!  I wonder how many geoscientists have considered
>Carey's explanation?

I may be way off but the description of Carey's explanation sounds like it would fit within the alternative plate tectonic models promoted by Meyerhoff and Meyerhoff in the 1970s (American Association of Petroleum Geologists Bulletin 56(2):269-336, 1972).

>>  What is truly amazing
>> though is that this relationship was established 30 years ago and from that
>> data an estimate of how fast the Pacific plate was moving could be made
>> with a gradeschool knowledge of math.  Now in the last 10 years satellites
>> have measured the rate of motion of the Pacific plate and the numbers
>> 8.6-9.0 cm/yr compared to 9.0 cm/yr are the statistically the same.   
>
>It is an impressive correlation.  But is it soundly based?  In recent
>years, it has become apparent that the alleged hot spot under Hawaii
>is NOT stationary.  The data suggests it is moving.  This throws a
>spanner in the works - the satellite measurements give a movement
>rate higher than would be expected.  My source for this=20 is:
> Geology Today, 14(3), May/June 1998, 101-103.

How far off?  Sorry don't have the article on hand and won't be able to get if for a few weeks.   I wouldn't be shocked at some deviation.  Also, it is acknowledged that there was a different rate of at least plate movement if not hot spot movement (it is a tough problem to separate the two variables)  for the Emperor Seamounts and Hawaiian Ridge.   Even a degree of magnitude would still be amazing given a young earth model would suggest the islands were created over a short period of time.  Why would radiometric dates of points covering some horizontal distance (vs. depth related which is a separate argument) give an estimate that was anywhere even close to a satellite measurement? 

>Misinterpretation in the Pacific?
>
>The Hawaiian-Emperor chain of volcanic islands and seamounts in the
>Pacific Ocean (Fig. 7) is famous, because it was in that connection
>that many of the ideas on mantle plumes and hotspots were worked out.
>The conventional wisdom, reached after much early debate, is that the
>islands/seamounts are the product of a mantle plume, fixed in
>(lateral) position relative to the mantle, that gave rise to a series
>of volcanic hotspots at the surface as the Pacific lithospheric plate
>moved above it. The chain is also famous for its bend, interpreted as
>a sudden change in the direction of motion of the Pacific Plate about
>43 million years ago.
>
>This story is now so widely accepted that it almost has the status of
>`fact'. But what if it's not true? What if it's merely wishful
>thinking? The alternative is perfectly obvious, of course, if not much
>thought about. Could the bend in the chain be due not to a change in
>the motion of the Pacific Plate but to a sudden movement of the mantle
>plume itself? Possible movements of mantle plumes have been=20 much
>considered in the past and not always ruled out, but envisaged motions
>have generally been small, barely invalidating the general concept of
>a `fixed hotspot frame of reference'. Needless to say, any proved
>deviation from that would be devastating, requiring a major rethink on
>a number of fronts.

True, it is difficult to establish the absolute fixity of the spots and a major deviation would require some new thinking.  This still does not get us out of the correlation between age and distance even in the Emperor Seamounts.   If we have dramatic movement of the hotspot then how are the radiometric dates to be understood?  Paleomagnetic positions are notoriously difficult to determine and the error bars on most of the dates determined for the seamounts are huge (see Clague and Dalrymple, 1987).   With the new technology for determining paleomagnetism coming on line this fall it will be interesting to see what happens when new and much more extensive studies can be made.     Still outlying data points are not unexpected in the current data and are much easier to explain than the radiometric dates.    But I am willing to allow that there appears to be contradictory data regarding the exact origin of the Emperor Seamounts.  Tarduno and Cottrells data are very interesting and I hope further data is gathered to test their proposal.  I would note that the error bar around the paleomagnetic  measurement is huge (Fig 4) and the traditional view of the origin of the seamounts appears to encompassed within that value though is on the fringe.  Further the studies of the other island chains in the Pacific that also have an apparent bend have either not been studied enough to draw definitive conclusions or do not have bends that only loosely correlate with the Emperor-Hawaiian Ridge bends.   I willing to leave the challenge solely on the data of the Hawaiian Ridge.  

>Yet that is what Norton proposed a few years ago (Tectonics, v.14,
>p.1080, 1995) and is what Tarduno and Cottrell consider in more detail
>now (Earth & Planetary Science Letters, v.153, p.171,=20 1997).
>Specifically, Norton suggested that the hotspot moved southwards up to
>43 million years ago, creating the Emperor seamount chain, and then
>became fixed in the mantle, the Hawaiian part of the chain having been
>generated by the method outlined in the first paragraph above. As
>Tarduno and Cottrell now point out, this is testable by
>palaeomagnetism, for if the hotspot has always remained fixed, the
>original palaeolatitudes of the now-extinct seamounts of the Emperor
>chain should be the same as the present latitude of Hawaii, under
>which the supposed mantle plume now rests. Moreover, thanks to samples
>collected from Detroit seamount (see Fig. 7) during the Ocean Drilling
>Program, the palaeomagnetic test can now actually be carried out.
>The results are troubling. For one thing, there is an age
>inconsistency. Hotspot-plate motion models suggest an age of 75-65
>million years for Detroit seamont, but new 40Ar/39Ar data indicate an
>age of about 80 million years. More seriously, the palaeolatitude
>given by the Detroit rocks is about 36=B0N, whereas the present latitude
>of Hawaii is only about 19=B0N. Taken at face value, that means that the
>plume could not possibly have remained stationary before 43 million
>years ago - i.e. the Emperor chain cannot be explained entirely, if at
>all, by plate motion above a stationary plume. However, the
>palaeomagnetic data do not have the resolution to allow decisions on
>whether the Emperor chain is entirely due to plume movement or partly
>to plume movement and partly to plate motion.
>
>Either way, the palaeomagnetic data, if confirmed (and confirmation is
>necessary because of the limited sampling of the Emperor chain thus
>far and the possibility that movement of the Earth's pole may have
>biased the data to some, unknown, extent), would appear to demolish a
>cherished hypothesis and thus require `a major change in how we view
>mantle dynamics and the history of plate tectonics'.
>
>[end of quote]
>----------------------
>
>Hope this feedback is of interest.
Yes, it is - thanks.  The Geology Today issue is at the binders (it seems everything I ever want is!).  I am surprised I missed that one in my literature search was researching the subject recently.

Cheers,

Joel Duff

<http://scribers.midwest.net/crinoid/origins.htm>
<http://scribers.midwest.net/crinoid/hawaii.htm>

Relevant refs:


Johnson, H. P. and D. Van Patten. 1996.  Age-dependent variation in the magnetization of seamounts.  Journal Geophysical Research 101(B6):13701-13714.

Larson, K. M., J. T. Freymueller, and S. Philipsen. 1997.  Global plate velocities from the global-positioning system.  Journal Geophysical Research 102(B5):9961-9981.

Ludwig, K. R., B. J. Szabo, J. G. Moore, and K. R. Simmons. 1991.  Crustal subsidence rate of Hawaii determined from 234U/238U ages of drowned coral reefs.  Geology 19: 171-174.

Moore, J. G., B. L. Ingram, K. R. Ludwig, and D. A. Clague. 1996.  Coral ages and island subsidence, Hilo drill hole.  Journal Geophysical Research 101(B5):11599-11605.

Sharp, W. D., B. D. Turrin, and P. R. Renne. 1996.  the 40Ar/39Ar and K/Ar dating of lavas from the Hilo 1-km core hole, Hawaii Scientific Drilling Project.  Journal Geophysical Research 101(B5):11607-11616.

Szabo, B. J. and J. G. Moore. 1986.  Age of -360-m reef terrace, Hawaii, and the rate of late Pleistocene subsidence of the island.  Geology 14: 967-968

Wessel, P. and L. W. Kroenke. 1998.  The geometric relationship between hot spots and seamounts:  implications for Pacific hot spots.  Earth and Planetary Science Letters 158:1-18.