LARGE IGNEOUS PROVINCES: Results of Delamination?

1
LARGE IGNEOUS PROVINCES Results of
Delamination?

• Don L. Anderson
• Caltech

2
A new GSA book
3
DelaminationThe Eclogite Engine

• Kay, R.W. Kay, S.M., Delamination and
  delamination magmatism, Tectonophysics, 219,
  177-189, 1993.
• mechanism can explain some long-standing
  geophysical problems, e.g.
• subsidence prior to LIP emplacement
• short duration
• bottoming of seismic tomography anomalies beneath
  hot spots
• but what happens to this lower crust?

4
Summary of model

• When crust thickens to gt 50 km
• converts to dense eclogite
• delaminates
• sinks
• heats up
• rises
• eclogites have low Vs for their density - may be
  confused with high T

5
Rocks and minerals arranged by density crust
upper mantle
eclogite here used as a general term for garnet
pyroxene-rich rock
6
Rocks and minerals arranged by density crust
upper mantle

• delaminates when crust gt 50 km thick

• warmer than MORB

7
Rocks and minerals arranged by density upper
mantle
Where does delaminate reach neutral buoyancy?
Vs
r
8
Delaminated roots warm quickly

• will start to melt before reaching same T as
  surrounding mantle
• already in TBL, so starts off warm
• when 30 melt, garnet mostly gone will start to
  rise

9

• pink eclogite is only temporarily stable at these
  depths
• arclogites less SiO2 than MORB eclogite do
  not sink so far
• Vs of eclogite low at depth
• low melting point
• as it warms, it rises

10
Mantle stratification

• irregular chemical discontinuities expected
• difficult to see in tomography
• can be seen in receiver functions

11
Underside reflections 0 1,000 km depth

• 410 660-km discontinuities clear
• 10 others
• may be chemical

12
Delamination cycle

• dense roots
• fall off
• warm up in ambient mantle
• rise
• possible mechanism for Atlantic Indian ocean
  plateaus DUPAL anomaly

13
Many ways for eclogite to get into the mantle

• collision belts, arcs
• can fuel melting anomalies at normal T

14
LIPs are associated with continental breakup

• reconstruction at 30 Ma
• dual volcanism
• on breakup
• 30-40 Myr later
• oceanic plateaus form 1,000 km offshore
• rising of delaminated root?

15
Eclogite 70 molten before peridotite starts to
melt

• eclogite sinkers warmed by conduction
• rise before T has risen to that of ambient mantle
• eclogite 70 molten at peridotite solidus

16

• delamination controls crustal thickness
• very sharp cut-off at 50 km
• interpreted as eclogite phase change

from Mooney et al., 1998
17
Example 1 Rio Grande riftAre LVZs delaminated
roots?
hot?
eclogite?
from Gao et al., 2004
18
Example 2 Sierra Nevada
attenuation
P-wave slowness
anisotropy
Vp/Vs
from Boyd et al., 2004
19
Example 3 Iceland

• Restricted LVZ
• possibly Caledonian arc roots delaminated on
  breakup
• Cold, dense, sinking eclogite can be LVZ
• warmed, melted, rising eclogite can also be
  buoyant if 1/2 garnet eliminated

Ritsema et al., 1999
20
Summary

• Dense, mafic cumulates may be twice the thickness
  of arc crust

• Delamination accompanied by upwelling adiabatic
  decompression of the asthenosphere a whole cycle
  may take 30-40 Myr

• The global recycling flux of arcologite is 10
  that of oceanic crust, i.e. hotspot volume rate

• It starts out hotter by-passes normal
  subduction zone processing

• Delaminated arclogites preferentially melt form
  a unique component of hotspot ridge magmas
  (e.g. suggested DUPAL Gondwana crust).

21
Resources

• Please visit
• www.mantleplumes.org/Eclogite.html
• www.mantleplumes.org/LowerCrust.html

End