Transcript Slide 1

Are
‘hotspots’
‘wetspots’?
By Clare de Villanueva, 4th yr F601
Are ‘hot spots’, ‘wet spots’?
What is a ‘hot spot’?
•enhanced rates of volcanism
•ascending plumes of (hot?) mantle material
• higher degrees of melting.
•‘plume’ assume that plumes or something of the sort exists
What is a ‘wet spot’?
We can constrain the concentration of H2O in Earth’s upper mantle by analyses of
quenched seafloor basalts and of H2O in minerals from mantle xenoliths.
However, the water concentration in deeper regions of the mantle is largely unknown.
By Clare de Villanueva, 4th yr F601
What is a ‘wet spot’?
•Schilling: hot spots could also be wet spots.
•elevated water contents in tholeiitic volcanic
glasses from the Azores platform may explain
the higher rate of volcanic activity, compared to
elsewhere along MAR.
•The presence of water lowers the solidus of
mantle material, increasing the degree of melting
at a given temperature.
P-T diagram (after Thompson, 1992). Near-solidus phase relations in peridotite and temperature distribution in the
upper mantle. The upwelling mantle plume adiabat shown is ~200 C hotter than average mantle. Shows the possible
minerals in which H2O could be stored.
The sublithospheric mantle has a heterogeneous distribution of water.
•N-MORB mantle source is anhydrous (100-200 ppm H2O)
•Water-rich mantle volumes exist in modern subduction zones where water may be
fed to the mantle wedge by dehydration of the subducting slab.
Hence, high water contents in regions of the upper mantle may be
related to ancient subduction zones.
The Azores hot spot
The upper mantle of the hot spot has undergone a high degree of melting
compared to mantle elsewhere.
Chemistry of primary minerals along the MAR show that <20% melting has been extracted
from the mantle rocks in the AHS region compared to elsewhere in the MAR, where the
degree of melting can be as low as 8%.
This could reflect higher upper mantle temperatures in the AHS than along the
rest of the MAR (hotspot theory).
If this were the only cause of increased melting, the estimated temperature
differences between the areas of highest and lowest extent of melting would be in
the region of 250-300C, at equivalent depths in the melting region.
BUT geothermometers suggest the temperature in the mantle was not higher
(indeed maybe lower!) in the hot spot than elsewhere along the MAR.
Al2O3 partitioning in opx and spinel indicate AHS rocks did not equilibrate at higher
temperatures than other MAR peridotites.
If it’s not hot, how else could all this melting occur..?
It has been suggested (by Schilling et al) that the AHS behaves also as a
wetspot and is the locus of metasomatism.
•CO2 or H2O, as free phases or within minerals (amphiboles, phlogopite, carbonates)
lowers the solidus temperature of peridotite. Even traces of either can have the effect of
lowering the melting temperature by hundreds of degrees. This would enhance partial
melting.
•Basalt and peridotite data both indicate the upper mantle in the AHS region is enriched in
H2O and other volatiles. MORB from the AHS region contain 2-3 times as much H2O as
normal MORB; are enriched in volatiles such as Cl, Br and F, and tend to have lower SiO2
contents, which is consistent with melting from a H2O enriched source.
Therefore some so-called hotspots may be melting anomalies unrelated to
abnormally high mantle temperature or thermal plumes. Instead, they may be
caused by a composition different from the underlying normal segments of
the Mid Atlantic Ridge.
What do we know about this wet spot?
Effects of the water
•H2O-rich (metasomatised) mantle domains would lower the melting temperature of
the hot spot mantle to give a high degree of melting without a mantle temperature
anomaly.
Simplified model of mantle plume by fluid release (I) and metasomatism (red) during mantle convection
(arrows), followed by partial melting (ii).
•Wet mantle that upwells adiabatically would cross the wet solidus at a higher
pressure than a similar but dry mantle body.
And where does the water come from?
It could be that a variably veined mantle may occur beneath the Atlantic,
so that the so-called hot spot mantle contains a high proportion of
metasomatic component, and the mantle ridge part a low proportion.
The metasomatised mantle domains inferred for the AHS area might be
remnants of proto-Atlantic rift mantle that were left behind during the
opening of the North Atlantic.
A metasomatised H2O- and CO2-rich domain
It could therefore be argued that the suggestion
of an Azores hot and wet spot is modified in to
the idea of wet but not particularly hot spot.
Hawaii: primary Kilauean magmas
Kilauea View north-northeast
across Kilauea's summit
caldera and Halema`uma`u
crater (left of centre)
•The mantle plume source for Kilauea is estimated to contain 450190 ppm H2O.
•This is ~3 times greater than that estimated for the mantle source for depleted
MORB.
•Water undersaturated melting of the upwelling Hawaiian plume begins at a depth
of ~250 km compared to ~120 km beneath spreading ridges.
•This is primarily because of the greater water content of the plume source.
Where does this water comes from?
•There may be an enrichment of water in the transition zone relative to the upper
mantle due to recycling of water stored in subducted ocean crust.
•If subducted slabs penetrate into the lower mantle and suitable OH-bearing
phases are stable, then the lower mantle could also be enriched in water.
•Because mantle plumes are mainly believed to originate from the 660 km
discontinuity or deeper, water contents of basaltic magmas formed in upwelling
plumes can provided information on water in the deep mantle.
Arguments for a dry plume component
•In 2001, Dixon et al studied basaltic glasses from the Loihi seamount of Hawaii, and found
evidence for a relatively dry plume component.
•They used H2O/Ce as an indication of enrichment or depletion of H2O relative to other
incompatible trace elements.
•They found mantle water concentrations of ~400 ppm. This is greater than that for
MORB mantle (100 ppm) but slightly lower than that for other estimates from Hawaii
source regions.
•The differences in mantle water concentrations are small, but do support a wet
rim/dry core theory, with most Loihi glasses having lower H2O/Ce than other
Hawaiian lavas.
•Their model is of a zoned plume, in which the mantle components within the plume
are drier than the exterior.
I have not found any research which supports compositional zoning within a
plume model, relating to other ‘hotspots’
Summary of Dixon and Clague’s work
•The most volatile enriched lavas occur ‘in-front’ or updrift of the plume where small
volumes of plume derived hydrous melts have metasomatised the over-lying
asthenosphere-lithosphere.
•The core of the Hawaiian plume is wetter than the MORB source, but the amount of
water present is equal to (not anomalously wet) or less than (relatively dry) that expected
based on concentrations of other incompatible elements.
•The Hawaiian plume therefore does not represent primitive ‘undegassed’ lower mantle.
So, Hawaiian magmas have higher water concentrations than MORB but
only in advance of the plume.
Small scale hydrous regions in the upwelling plume melt, segregate and
metasomatise the overlying mantle.
The Reykjanes Ridge
•Iceland lies on the mid Atlantic ridge, above
a ‘hot spot’
•Compare geochemistry of N-MORB and
Icelandic samples.
The Reykjanes peninsula (right) is where the
Mid-Atlantic ridge appears above sea level
•Water contents are higher in samples from Iceland than Along the Reykjanes Ridge.
•The H2O values increase sharply northwards, towards Iceland, about 650 km from the centre of the
Iceland plume.
•Estimated source water contents using these melt fractions increase from ~165 ppm at the S end
of the Reykjanes ridge furthest from the influence of the mantle plume to 620-920 ppm beneath
Iceland.
•Further studies have determined water concentrations in olivine melt inclusions in tholeiites erupted in
the Lakagigar eruption on Iceland in 1783 that are 1.5-4 times higher than typical for MORB.
Nichols et al 2002
This figure shows the variation of H2O values with distance from the Iceland mantle plume
(corrected to account for crystallisation). Data from Iceland are plotted as open symbols
and from the ridge as filled symbols.
What does this mean?
Such a rise in the mantle water content, will increase the degree of melting by up to
10%, from about 10% to between 20 and 30% beneath Iceland.
An increase in mantle temperature of almost 100C would be required here to
generate a similar rise in the melt fraction.
The range of values may reflect variable source compositions, degrees of partial
melting or the effects of degassing of water that have not been distinguished.
So, water plays a significant role in the generation of melts
beneath Iceland and it does appear to be a wet spot compared to
the southern end of the Reykjanes Ridge.
The North Atlantic Province (produced by the mantle plume now located under Iceland)
It shows two different initial water
contents. If the basalts and picrites
are generated by 10-20% batch
melting they will have 5-10 times
higher H2O contents than the
corresponding mantle source.
For this range of melt fractions, a
depleted mantle source with 100
ppm H2O would give rise to melts
with 0.2-0.4 wt% H2O.
Depleted source (100ppm)
Enriched source (400ppm)
Alkaline rocks
Melt water content (wt%)
This figure shows the expected
water contents of mantle-derived
melts as a function of degree of
mantle melting and the initial
mantle H2O contents.
PICRITES
N-MORB
Degree of mantle melting (%)
Jamtveit et 2001
•Basaltic melts with H2O contents >0.2 wt% cannot be obtained by
melting a mantle containing less than about 300 ppm H2O at any
reasonable degree of mantle melting.
•It may be that here we have a source region within a ‘wet’
asthenospheric mantle (i.e. ‘plume’ material), though contributions
from a ‘fertile’ lithosphere cannot be ignored.
•The temperature anomaly associated with the Iceland hotspot is hard
to constrain by geochemical data alone.
Summary
•Hawaiian lavas contain more water than MORB, about 3 times as much.
•Water contents determined for submarine basalts along the northern mid-Atlantic ridge
show that the water contents increase towards the Iceland hot spot.
•On the basis of the water contents of olivines from across the North Atlantic Volcanic
Province, the magmas are derived from sources significantly enriched in water relative to
those of MORB.
•The water may be derived from subducted material recycled to the surface by a mantle
plume. However, whether water can reach the deep mantle through subduction is a matter
for debate.
•Alternatively, metasomatism of shallow mantle may have occurred by water derived from
both hydrous and nominally anhydrous minerals residing in the upper mantle.
•Overall, the data support for the wet hypothesis of Schilling.
References
Enrico Bonatti. Not So Hot So Hot “Hot Spots” in the Oceanic Mantle. Science 250 (1990) no 4977
pages 107-111
Paul J Wallace. Water and partial melting in mantle plumes: inferences from the dissolved H2O
concentrations of Hawaiian basaltic magmas. Geophysical Research Letters 25 (1998), pages 36393642
A. R. L. Nichols, M. R. Carroll and A. Hoskuldsson. Is the Iceland hot spot also wet? Evidence from
the water contents of undegassed submarine and subglacial pillow basalts. Earth and Planetary
Science Letters, 202, No. 1, pages 77-87
J. G. Schilling, M.B. Bergeron et al. Halogens in the mantle beneath the North Atlantic. Philosophical
Transactions of the Royal Society of London A 297 (1980), pages 147-178
J. G. Schilling, M. Zajac et al. Petrologic and geochemical variations along the Mid-Atlantic Ridge
from 29N to 73N. American Journal of Science 283 (1983) pages 510-586
R. Poreda et al. Helium and hydrogen isotopes in ocean ridge basalts north and south of Iceland.
Earth and Planetary Science Letters 78 (1986) pages 1-17
N. Metrich et al. The 1983 Lakagigar eruption in Iceland: geochemistry, Co2 and sulphur degassing.
Contribution Mineralogy and Petrology 107 (1991) pages 435-447
J.E. Dixon and D.A. Clague. Volatiles in basaltic glasses from Loihi seamount, Hawaii: evidence for a
relatively dry plume component. Journal of Petrology 42 (2001) pages 627-654
G.A.Gaetani and T.L. Grove. The influence of water on melting of mantle peridotite. Contributions to
Mineralogy and Petrology3 (1998) pages 323-346
B. Jamtveit et al. the water content of olivines from the North Atlantic Volcanic Province. Earth and
Planetary Science Letters 186 (2001) pages 401-415