On Thursday 20th February, Glenn Morton wrote:
"Those who might think that water flow through an overlying rock
is rapid should realize that water flows about half as rapidly
through rock as does oil ( Chris Clayton, Differential Flow
Rates of Petroleum and Water in fine-grained sediments, AAPG,
sept. 1993, p. 1613) The authors attribute this to the lack of
unbound water in rocks."
The neocatastrophist scenario opens many avenues for tectonic
disturbance of rocks, creating channels where none existed
before. Even without catastrophism, I do not consider that a
batholith can be emplaced without considerable disruption to the
country rocks.
GM: "While I can not find a reference, I would be willing to bet
that groundwater flow in a hydrothermal region, cooling a
batholith, would still take longer than a global flood could
accommodate, because water does not travel rapidly through
shale. Shales have permeabilities of the order of a few
millidarcies."
Tectonically disturbed shales present a rather different picture:
the permeabilities will be larger. Furthermore, when people
start to look for evidences of convection, they are likely to
find them. Parmentier and Schedl (1981) have considered the
thermal aureoles of the Mull intrusive complex, the Skye Cuillin
gabbro, and the El Salvador porphyry copper deposits. The shapes
of the metamorphic aureoles were inconsistent with purely
convective heat loss, and yet could be explained by invoking
convective activity.
>GM: "Large batholiths can take 100,000 to 1,000,000 years to
> cool down...."
DT:
>The first sentence is again model-dependent. If you rely only
>on conduction to cool the hot body - it takes ages!
>What is the link with reality?
GM: "To state this objection is easy. To produce a realistic
model that cools all batholiths within a year is hard. What
other mechanisms are you proposing?"
First, I am not defending the idea that batholiths are all cooled
within one year: I am promoting neocatastrophism. Second, I
have already indicated that a variety of mechanisms may be
relevant: convective cooling and alternative models of granite
emplacement. In addition, I am sympathetic with the expanding
earth concept - which has thermodynamic implications which favour
rapid cooling.
GM: "Not all batholiths were cooled by being eroded to the
surface. That also would create problems fitting all this into
a single year."
Of course. I was a bit surprised you brought these uncovered
batholiths into the discussion, as they are much more easily
cooled.
GM: "Secondly, I am aware that large crystals of the various
minerals in a cooled batholith are indicative of slow cooling
rates. Lots of batholiths have large minerals. Rapidly cooled
batholiths have small crystals. To fit all the cooling into a
single year would need a miracle."
Again, it is not my intention to fit all the cooling into "a
single year". But are your points about crystal sizes
substantive? I accept that this is an area where quantitative
measurements are needed. Sure there are time implications - but
how much time? Dowty (1980) cites maximum rates of crystal
growth in a granite melt with 3.5% water as:
quartz 1.0 x 10^-8 cm/s
alkali feldspar 1.0 x 10^-7 cm/s
plagioclase 1.0 x 10^-6 cm/s
(Note: dry granites have higher growth rates).
The number of seconds in 1 year is 3.15 x 10^7. When compared
with the maximum growth rate parameters for wet granitic melts,
there are implications for time: large crystals need not imply
time intervals much greater than a single year. Luth (1976)
comments:
"It is frequently assumed that the presence of large
crystals in these phases implies slow growth over long periods
of time. Although this may be the case, the intent here is to
demonstrate that it does not necessarily hold" (p.405).
Regarding the Deccan Traps:
GM: "If each flow required a month to cool to this point and each
was 100 feet thick, then the Deccan Traps would require 100
months or 8 years to be deposited. The Deccan Traps have a volume
of 1.7 x 10^6 cubic kilometers of lava.(W.S. Holbrook and P.B.
Kelemen, Nature July 29, 1993, p. 433-436) They cover 200,000
square miles. This does not fit a global flood well."
My objective is not to force data to fit a specific model of
earth history. If the Deccan Traps are better interpreted as
being formed in 8 years, that would be my preferred option.
Regarding the Oceanic Ridge systems:
DT:
>... I cited the oceanic ridge systems:
>buried several kilometres down, but with enormous water
>convection cells cooling them. The amount of water moving
>through the oceanic crust is enormous. Conductive heat loss is
>negligible compared with convective loss.
GM: "Can you give me a source for how much water flows through
these systems? I just looked and couldn't find one. What does
enormous mean?"
Edmond et al (1982) suggest that "a volume of water equivalent
to the whole ocean must circulate through the high temperature
intrusion zone in the ridge axis ... every 8-10 Myr".
Macdonald et al (1980) estimate that a single vent provides a
hydrothermal heat loss of (6+-2) x 10^7 cal/s. This is compared
with the conductive heat loss over a 60 square kilometre area of
0.23 x 10^7 cal/s. The heat flows are so high that the authors
estimate the vents are active for less than 10 years. This study
"emphasises the importance of hydrothermal activity in the global
heat budget".
Cann and Stiens (1982) argue that the heat source for black
smokers must be magma chambers - nothing else will satisfy the
modelling constraints. Large sulphide deposits require the
crystallisation of large volumes of magma. If larger deposits
have to be formed, more black smokers can be invoked and there
can also be replenishment of the magma chamber. Thus, three
black smokers could produce a 1 million ton ore deposit in 320
years, solidifying 7 cubic kilometres of magma in this time. One
black smoker is said to have a mass flow rate of 160 kg/s.
[Note: even though these timescales are geologically short, I am
of the opinion that they are unrealistically long. I say this
after visiting some of the Cyprus sulphide deposits and comparing
them with what we know of the present day deposition of
sulphides. The latter are quite minuscule in comparison: there
are no adequate modern analogues for producing 1 million ton
deposits. We need catastrophist mechanisms and models.]
Anderson et al (1979) say "By carefully measuring the
nonlinearity of temperature profiles, we have calculated both the
conductive and convective components of the total heat flow at
the sea floor. Significant convective heat transfer is occurring
through the crustal and sedimentary layers even at the oldest (55
x 10^6 years) experimental site".
For a popular overview, see Macdonald and Luyendyk (1981).
Regarding models of batholith formation
DT:
> ... Large magma
>bodies moving upwards through the crust of the earth have the
>problem of "What creates the space into which they move?"
GM: "melting and incorporation of the rock into the magma. as
well as uplift. Mt. St. Helen underwent an uplift and even an
expansion of the mountain prior to its eruption."
Are you serious? How much country rock do you think a magma body
can ingest before it becomes too cool to melt it? Remember that
granitic magmas are far cooler than basic magmas. Stoping will
not get you very far! And where is the physical evidence for
extensive stoping? We do see xenoliths - but they are hardly a
major component of batholiths. Regarding Mt St Helens: how much
uplift are you prepared to defend? A few metres? If you want
much more, you are moving into the domain of tectonic emplacement
- which will disrupt the surrounding rocks and create channels
for water flow.
GM: "... I haven't seen any quantitative estimates of how you
would alter the present presuppositions. Everything has
presuppositions so it seems to me that you should propose new
values and show that a batholith can cool in the timeframe you
prefer."
Fair comment. However, I'm not ready to do this. I'm not
satisfied with the conventional models of granitic magma
formation nor of batholith emplacement. When a more adequate
model emerges, then it will be worthwhile putting forward
quantitative estimates of timescales.
It may be apparent by now that my references all date from 15
years ago (the last time I set out to research this area). I've
retained an interest in the subject - and nothing I've seen in
the intervening years has suggested to me that I'm on the wrong
track. Perhaps it is now appropriate to look at the topic area
again - but I would want to build on the kind of thinking I've
expressed in this post and not on the old orthodoxies.
As I've run out of time, I'll send the references in a separate
post, early next week.
Best wishes,