On The Significance Of Tectonic Underplating Beneath Continental Margins

1
On The Significance Of Tectonic Underplating
Beneath Continental Margins

• Mexico and Southern California allochton

Regional Tectonics, 2009, Sept 21, Lecture 14, 39
slides, 1 hour
2
Outline

• Setting up the problem - why is mass transfer of
  upper crust into the lower crust and mantle
  important?
• Subduction zones - sediment subduction and
  tectonic underplating. The case will be made that
  underplating is significant among subduction
  erosion processes examples from Mexico, central
  California and Pacific NW
• Implications for (1) continental evolution
  (throughout the talk), (2) mechanics of plate
  boundaries and (3) magmatism (the 2nd and 3rd
  briefly addressed at the end).

3
IDEALIZED ISLAND ARC
4
How much sediment in a trench?

• Use the thickness of pelagic sediment on an
  oceanic plate (gt100m)
• Convergence rate (1-10 cm/yr)
• Assume a simplified geometry of an accretionary
  wedge
• Calculate the width of a wedge for a given
  geologic time scale
• WEDGES should be on the order of hundreds of km
  wide for long lived (gt50 My) subduction zones.
  Many are not, not even close.

5
Accretionary vs. erosional trenches
6
Sediment subduction, erosion and underplating
Most geochemical and physical models assume
sediment is being subducted.
7
Destinations for eroding trench sediments

• 1. Recycled into the mantle - documented with
  isotopes e.g. EM1
• 2. Tectonically underplated - sediments were
  first transported onto the lower plate and then
  back to the upper plate- exposed at the surface
  in places and seen on seismic profiles
• 3. Recycled via the magmatic arc- documented with
  isotopes in arc volcanoes e.g. 10Be.
• We know that magmatic recycling has to be minor,
  but we dont know the relative significance of
  the first two mechanisms.

8
Subduction erosion

• WHAT WE KNOW
• 1. Instantaneous pelagic input and rates of
  erosion
• 3. Many trenches are sediment starved.
• 4. Subduction erosion is particularly important
  at flat subduction settings.

• WHAT WE KNOW LESS
• 1. Long term subduction erosion rates
• 2. The input of terrigenous sediment, especially
  when the forearc is uplifted

Long-term processes require investigation of the
geologic record
9
Fingerprinting underplating via forearc
exhumation- a local example

• Why Little is known about forearc exhumation
  rates, especially in areas of flat slab
  subduction and in places where turbidites
  dominate over pelagic sediments
• Forearc unroofing erosional or tectonic?
• What is the flux of sedimentary mass delivered to
  the trench?
• Example the Sierra Madre del Sur, southern
  Mexico and the Acapulco trench.

10
(No Transcript)
11
Acapulco trench and the Sierra Madre del Sur

• Site of modern flat slab subduction of Cocos
  plate beneath mainland Mexico
• The modern arc (Trans Mexican Volcanic Belt) is
  300 km inland due to the shallow angle of slab
• Subduction and arc magmatism operated at this
  margin for much of the late Mesozoic and
  Cenozoic.
• The margin is truncated today the Eocene
  Oligocene arc is very close to the modern trench
  no Eocene-Oligocene forearc or accretionary
  wedges exist.

12
SMS geology - the Xolapa terrane or complex is a
high grade basement complex (Ducea et al, 2004)
of mainly magmatic arc rocks and no sedimentary
cover. Typical exposure levels are 25-35 km. Much
of the basement is deformed and migmatitic.
13
(No Transcript)
14
Offshore geology

• 25 Ma arc rocks extend all the way into the
  trench (Watkins, 1988)
• Sedimentary cover thin, 25- 0 Ma (DSDP)
• Mostly greywackes with an average composition of
  an andesite (Plank and Langmuir, 1998)
• Prism only some 30 km wide

Sediment accumulation rates are known from DSDP
Leg 66
15
Ducea et al., 2004, JGR
16
Interpretations

• Cooling rates are due to exhumation, not heat
  advection
• Exhumation rates of 0.1-0.3 km/m.y. are
  erosional, and not tectonic (i.e. crustal
  thinning) Corollary- the forearc did not
  collapse, which is consistent with surface
  geology
• Miocene erosion rates accumulation rates in the
  trench are an order of magnitude larger than
  sediment preserved in the trench.

17
Implications for subduction erosion

• At least 90 of the terrigenous trench sediment
  has been subducted, and either transferred to the
  upper plate or into the mantle (dont know)
• The Acapulco trench is a prime example of an
  erosive margin, most of the sediment accumulated
  is being recycled
• The long-term, Miocene rate of subduction erosion
  is 23 km3/m.y. km length of trench, consistent
  with classic von Huene and Scholl estimates.

18
MASE
No intervening mantle wedge beneath the upper
plate crust, at least for gt150 km inland from the
trench.
Unusual geometry!
19
(No Transcript)
20
A geologic example - Southern California and
Salinia
21
Ducea et al., 2009
22
Upper plate
Lower plate
Schist of the Sierra de Salinas
Salinian block
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
A regional perspective

• Much of southern California is underlain by an
  equivalent schist (locally known as Rand, Pelona,
  Orocopia, etc.)
• This schist represents tectonically accreted
  material during the Laramide orogeny
• It contains wedge sediments, but probably also
  forearc and even arc rocks.

Barth et al., Grove et al., 2003
Saleeby, 2003
28
Swakane gneiss, WA, another example

• Accretionary sediments underplated after 73 Ma,
  and metamorphosed before 68 Ma (Matzel et al.,
  2004)

29
Global significance

• Underplated wedge sediments and forearcs have
  been recently described in several young (e.g. NA
  Cordillera in Canada) and old orogens (e.g.
  Variscan Europe)
• Probably a very common mechanism of shuffling the
  crust in a subduction zone.

Lowest crustal exposure of the Coast Mountain
Batholith, Kwinitza, BC, Ducea et al., in
progress.
30
Criteria for identifying underplated sediments

• a. The existence of distinct lower and upper
  plates, in which the lower plate is composed of
  metagreywackes and mafic/ultramafic oceanic rocks
  and the upper plate is a part of the continental
  interior of a subduction system (e.g. the arc
  region)
• b. The existence of a shallowly dipping structure
  (a ductile structure, most likely), that puts the
  two plates in contact
• c. The lower plate is cooler than the upper plate
  and may have an inverted thermal gradient
• d. The lower plate was thrusted under the upper
  plate at plate tectonic rates the time
  difference between depositional age (constrained
  by the youngest detrital zircon ages) and
  metamorphic age (garnet Sm-Nd or Lu-Hf ages, U-Pb
  ages of dikes cross-cutting foliation, etc.) is
  on the order of only a few million years

31
Processes of sediment removal
One mechanism
A different mechanism
32
Consequences

• If tectonic underplating is taking place at plate
  kinematic rates, what does that mean for the
  mechanics of faults?
• Are underthrusted sediments prone to melting?

33
Ducea et al., 2007
34
The original ductile thrust is preserved- The
Salinas shear Zone
This shear zone represents the fossil subduction
megathrust during the early Laramide- between 71
and 76 Ma
The Salinas shear zone developed locally
granulite facies metamorphism - higher grade than
both the upper and lower plates.
Ducea et al., 2007
35

• Implications
• This is a subduction megathrust, which puts in
  contact rocks that were originally gt150 km apart
• The window into this fault is 30-35 km deep,
  within the seismogenic zone
• The fault is dry and does not seal instead it
  releases water
• One can calculate 50-80 MPa stress drops (
  large individual earthquakes from cpx
  microstrctures
• The development of granulite facies metamorphism
  suggests water loss along megathrust faults and
  is an analogue to modern slow/silent
  earthquakes in places like Southern Mexico,
  Cascadia, etc

36
Significance

• If faults propagate through the continental crust
  at plate convergence speeds - the thermal regime
  and mechanical behavior of the plate boundary may
  be significantly different than standard
  subduction zone models -
• Lower plate is hotter, and approaches granulite
  facies metamorphism
• Lower plate can melt (see next slide)
• The plate boundary is entirely intracrustal.

37
Magmatism?
Wilderness Granite, Catalina Mts., AZ
38
Magmatism

• Wet metagreywackes subducted-accreted under arc
  sections can melt
• PT record of at least the structurally higher
  levels of units like the Kwinitza Gneiss, Sierra
  de Salinas schist, or Swakane Gneiss is fully
  consistent with this hypothesis
• Predicted magmatic compositions - rocks very
  similar to the crustal peraluminous granitoid
  rocks of the SW USA Cordilleran interior, which
  have previously been temporally linked to
  shallowing of the Farallon plate subduction. Can
  look like adakites if melting took place at
  depths in excess of 40-50 km.

39
Take home message(s)

• The geologic record strongly suggests that
  various forms of underthrusting/ tectonic
  underplating are important in all convergent
  settings not all of the missing volume of trench
  deposits is subducted into the mantle much can
  be underplated, during punctuated shallow
  subduction events
• Transferring rocks back to the upper plate
  results in higher thermal gradients than classic
  subduction complexes melting and the mechanics
  of fault zones are different and need to further
  be explored.