Download Javier Quinteros and Stephan Sobolev

Why does the convergence rate between Nazca and South
Poster T43F-2738 America decrease since the Neogene? Javier Quinteros and Stephan Sobolev
Introduction
Introduction
Facts
Facts
The
Theclassic
classicexample
exampleofofthe
thenot-well-understood
not-well-understoodrapid
rapidchange
changeofoftectonic
tectonicplate
platemotion
motionisisthe
theincrease
increaseand
andthen
then
decrease
decreaseofofthe
theconvergence
convergencerate
ratebetween
betweenthe
theNazca
Nazcaand
andSouth
SouthAmerica
Americaplates
platesduring
duringthe
thepast
past25–20
25–20m.y.
m.y.
that
that coincided
coincided with
with the
the growth
growth ofof the
the Andes
Andes Mountains.
Mountains. Currently,
Currently, the
the decrease
decrease inin convergence
convergence rate
rate isis
explained
explainedeither
eitherby
bythe
theincreasing
increasingload
loadofofthe
theAndes
Andesor
orby
bythe
theappearance
appearanceofofflat
flatslab
slabsegments
segmentsbeneath
beneathSouth
South
America.
America.
Here,
Here,we
wepresent
presentan
analternative
alternativeview
viewbased
basedon
onaathermomechanical
thermomechanicalself-consistent
self-consistent(gravity
(gravitydriven)
driven)model
modelofof
Nazca
Nazcaplate
platesubduction.
subduction.We
Weexplain
explainthe
thechanges
changesininthe
theconvergence
convergencerate
rateas
asaanatural
naturalconsequence
consequenceofofthe
the
Nazca
Nazcaplate
platepenetration
penetrationinto
intothe
thetransition
transitionzone
zoneand
andlower
lowermantle
mantleafter
afterlong-term
long-termoblique
obliquesubduction
subductionofofthe
the
Farallon
Farallonplate.
plate.The
Themodel
modelisisconsistent
consistentwith
withseismic
seismictomographic
tomographicimages
imagesofofthe
theNazca
Nazcaplate
platebeneath
beneathSouth
South
America.
America.Our
Ourmodel
modelalso
alsoshows
showsthat
thatthe
thepresence
presenceofofthe
theAndes
Andesdoes
doesnot
notsignificantly
significantlyaffect
affectthe
theconvergence
convergence
rate
ratebetween
betweenthe
theNazca
Nazcaand
andSouth
SouthAmerica
Americaplates.
plates.
Global reconstruction of plate velocities
(after Müller et al., 2008). Subduction in
central and northern Chile had a highly
oblique (or even parallel) component at
least in the past ~95 m.y.
The situation changed during the
Oligocene, with the abrupt increment in
convergence and the change in the
direction of subduction; i.e., more
perpendicular to the continent. At about
the same time, the Cocos and Nazca
plates formed from the splitting of the
Farallon plate.
All remnants of the old Farallon slab are found under Central and North America (Fukao et al., 2001) and
disconnected from the slab at the surface. South of Bolivian Orocline (~19°S) only the Nazca plate seems to be
found in the mantle transition zone and a small part in the shallower lower mantle.
Model
Model setup
setup
We used a version of SLIM-3D (Popov and Sobolev, 2008) with the following enhancements to simulate the
evolution of a subducting slab up to shallower lower mantle (Quinteros et al., 2010):
● elasto-visco-plastic rheology
● diffusion, dislocation and Peierls creep mechanisms
● real free surface
● olivine-spinel and spinel-perovskite phase transitions (~6% and ~8% density increase)
● gabbro-eclogite phase transition
● dynamic, low-friction and low-viscosity subduction channel
Two
Two examples
examples of
of evolution
evolution
With
With or
or without
withoutAndes?
Andes?
Slab
Slab evolution
evolution inin two
two experiments
experiments isis shown
shown below.
below.The
The experiment
experiment inin the
the left
left column
column
best
bestfits
fitsthe
theconvergence
convergencehistory
history(blue
(blueline
lineininthe
theleft
leftplot).
plot).Left
Left(ref2–40S-500)
(ref2–40S-500)isis40
40m.y.
m.y.
old
oldslab
slabwith
with+2.0
+2.0and
and−0.5
−0.5MPa/K
MPa/KofofClapeyron
Clapeyronslope
slopeatat410
410and
and660
660km
kmboundaries,
boundaries,
respectively.
respectively. Viscosity
Viscosity isis ~1021.5
~1021.5 Pa·s
Pa·s inin the
the shallower
shallower lower
lower mantle,
mantle, and
and overriding
overriding
velocity
velocityisis33cm/yr
cm/yruntil
until10
10Ma,
Ma,and
and22cm/yr
cm/yrafter
afterthat.
that.
The
Thesetup
setupfor
forthe
theexperiment
experimentshown
shownininthe
theright
rightcolumn
columnincludes
includesaa40
40m.y.
m.y.old
oldslab
slabwith
with
+2.0
+2.0and
and−1.5
−1.5MPa/K
MPa/KofofClapeyron
Clapeyronslope
slopeatat410
410and
and660
660km
kmboundaries,
boundaries,respectively.
respectively.
Viscosity
Viscosityisis~1021.7
~1021.7Pa·s
Pa·sininshallower
shallowerlower
lowermantle,
mantle,and
andoverriding
overridingvelocity
velocityisis33cm/yr.
cm/yr.
One
Onecan
cansee
seethat
thatthe
thecombination
combinationofofaayoung
youngslab
slaband
andaaClapeyron
Clapeyronslope
slopeofof-1.5
-1.5MPa/K
MPa/K
does
doesnot
notgenerate
generateenough
enoughslab
slabpull
pulltotoallow
allowthe
theslab
slabtotopenetrate
penetratethe
theendothermic
endothermicphase
phase
transition.
transition.Thus,
Thus,even
evenififthe
thevelocity
velocitypattern
patterncould
couldbe
beconsidered
consideredsimilar
similartotothe
thereal
realone,
one,
the
thefinal
finalstate
stateprediction
predictionfrom
fromthis
thismodel
modelisisnot
notininagreement
agreementwith
withthe
thepresent-day
present-daystate
state
resulting
resultingfrom
fromseismic
seismicstudies.
studies.
We also considered the effect of the
thick crust and high topography of
the Andes on the convergence rate
between the plates. We repeated
half of our models with an orogen
similar to the Andes (70 km crust
resulted in 4–5 km of topography
high) instead of the flat Moho in the
first set of models.
The main patterns of slab
evolution are the same with or
without the presence of the
Andes. Differences in velocity with
and without the Andes were small,
even in the extreme case that includes the thicker crust since the beginning of the simulation.
ref2-40s-500: Preferred case without the Andes, as shown in the results.
Andes (10 Ma): same experiment but with the presence of an orogen like the Andes since 10 Ma.
Andes: same experiment with an orogen during the whole simulation.
What
What we
we did
did
Conclusions
Conclusions
Based
Basedon
onthis
thisreconstruction
reconstructionofofglobal
globalplate
platevelocities
velocitiesas
aswell
wellas
asseismic
seismicimages,
images,we
wesuggest
suggestthat
thatthe
thetip
tipof
of
the
theoceanic
oceanicslab
slabwas
wasstill
stillin
inthe
theupper
uppermantle
mantleunder
underthe
thecentral
centraland
andsouthern
southernparts
partsof
ofSouth
SouthAmerica
America
by
bythe
theend
endof
ofEocene.
Eocene.ItItcould
couldbe
beaaresult
resultofofaatoo-oblique
too-obliqueconvergence
convergence(not
(notenough
enoughtotoform
formaalong
longslab
slab
under
underS.America)
S.America)and/or
and/ordripping
drippingoff
offofofthe
thehead
headofofthe
theslab
slabwhile
whilesinking
sinkingininthe
thehot
hotasthenospheric
asthenosphericmantle.
mantle.
InInall
allour
ourexperiments
experimentsthe
themain
mainincrement
incrementininconvergence
convergencevelocity
velocityisisrelated
relatedtotothe
thepenetration
penetrationofofthe
thetip
tipofof
the
theslab
slabinto
intothe
themantle
mantletransition
transitionzone.
zone.This
Thisisiscaused
causedby
bythe
thedensity
densityincrement
incrementininthe
theoceanic
oceaniccrust
crustand
and
increasing
increasinglength
lengthofofthe
thehanging
hangingnegatively
negativelybuoyant
buoyantslab,
slab,as
aswell
wellas
asby
bythe
theadditional
additionalslab
slabpull
pullexerted
exerteddue
duetoto
the
thepositive
positiveClapeyron
Clapeyronslope
slopeatat410
410km
kmdiscontinuity.
discontinuity.The
Thesubduction
subductionvelocity
velocityisislater
laterreduced
reducedwhen
whenthe
theslab
slab
interacts
interactswith
withthe
thespinel/perovskite
spinel/perovskitephase
phasetransition
transitionand
andunderlaying
underlayingmore
moreviscous
viscouslower
lowermantle.
mantle.
Convergence velocity for three different
experiments compared with that
observed in nature (Sdrolias and
Mueller, 2006).
ref2–40s500(and -avg): preferred
case. Yield strength of 500 Mpa.
Overriding velocity from 3 to 2 cm/yr.
Clapeyron slope at 660 km is -0.5
Mpa/K.
ref2–40s: Yield strength of 350 Mpa.
ref-40: 3 cm/yr of overriding velocity
and Clapeyron slope of −1.5 MPa/K.
Our
Ourexperiments
experimentsshow
showthat
thatthe
theincrease
increaseand
andlater
laterdecrease
decreaseof
ofthe
theconvergence
convergencevelocity
velocitybetween
betweenthe
the
Farallon-Nazca
Farallon-Nazcaand
andSouth
SouthAmerica
Americaplates
platesmight
mightbe
beexplained
explainedby
bythe
thenatural
naturalevolution
evolutionof
ofaaslab,
slab,aa
large
largepart
partof
ofwhich
whichisissubducting
subductingfor
forthe
thefirst
firsttime
timeinto
intothe
themantle
mantletransition
transitionzone
zoneand
andthe
thelower
lower
mantle
mantlebelow
belowthis
thispart
partof
ofthe
theSouth
SouthAmerica
Americaplate.
plate.The
Theslab
slabfirst
firstaccelerates
acceleratesdue
duetotothe
theincreased
increasedslab
slab
pull
pullininthe
themantle
mantletransition
transitionzone,
zone,and
andthen
thenslows
slowsdue
duetotothe
theresistance
resistanceofofthe
the670
670km
kmphase
phasetransition
transitionand
and
highly
highlyviscous
viscouslower
lowermantle.
mantle.However,
However,the
thepresence
presenceofofthe
theAndes
Andeson
onthe
theoverriding
overridingplate
platedoes
doesnot
notexplain
explain
the
thestrong
strongreduction
reductionininthe
thesubduction
subductionvelocity.
velocity.Our
Ourresults
resultsare
areininfull
fullagreement
agreementwith
withthe
theseismic
seismicimages
imagesofof
the
theregion.
region.
References
References
Iaffaldano, G., Bunge, H.-P., and Dixon, T. H., 2006. Feedback between mountain belt growth and plate convergence, Geology
●
Iaffaldano, G., Bunge, H.-P., and Dixon, T. H., 2006. Feedback between mountain belt growth and plate convergence, Geology
34(10),
893–896.
34(10),
893–896.
●Fukao, Y., Widiyantoro, S., and Obayashi, M., 2001. Stagnant transition slabs in the upper and region lower mantle, Reviews of
●Fukao, Y., Widiyantoro, S., and Obayashi, M., 2001. Stagnant transition slabs in the upper and region lower mantle, Reviews of
Geophysics
31(3),
291–323.
Geophysics
31(3),
291–323.
●Mueller, R. D., Sdrolias, M., Gaina, C., and Roest, W. R., 2008. Age, spreading rates, and spreading asymmetry of the world's
●Mueller, R. D., Sdrolias, M., Gaina, C., and Roest, W. R., 2008. Age, spreading rates, and spreading asymmetry of the world's
ocean
crust,
Geochemistry
Geophysics
Geosystems
9,9,Q04006.
ocean
crust,
Geochemistry
Geophysics
Geosystems
Q04006.
●Popov, A. A., Sobolev, S. V. (2008), Slim3d: A tool for three-dimensional thermomechanical modeling of the lithospheric
●Popov, A. A., Sobolev, S. V. (2008), Slim3d: A tool for three-dimensional thermomechanical modeling of the lithospheric
deformation
with
elasto-visco-plastic
rheology,
Physics
ofofthe
Earth
Interiors,
171,
55–75,
doi:
10.1016/j.pepi.2008.03.007.
deformation
with
elasto-visco-plastic
rheology,
Physics
the
Earth
Interiors,
171,
55–75,
doi:
10.1016/j.pepi.2008.03.007.
●Quinteros, J, Sobolev, S. V., 2013. Why has the Nazca plate slowed since the Neogene?, Geology, In Press,
●Quinteros, J, Sobolev, S. V., 2013. Why has the Nazca plate slowed since the Neogene?, Geology, In Press,
doi:10.1130/G33497.1.
doi:10.1130/G33497.1.
●Quinteros, J., Sobolev, S. V., Popov, A. A. (2010), Viscosity in transition zone and lower mantle. Implications for slab penetration,
●Quinteros, J., Sobolev, S. V., Popov, A. A. (2010), Viscosity in transition zone and lower mantle. Implications for slab penetration,
Geophysical
Research
Letters,
37,
L09,307,
doi:10.1029/2010GL043140.
Geophysical
Research
Letters,
37,
L09,307,
doi:10.1029/2010GL043140.
●Somoza, R., 1998. Updated Nazca (Farallon) - South America relative motions during the last 40 My: implications for mountain
●Somoza, R., 1998. Updated Nazca (Farallon) - South America relative motions during the last 40 My: implications for mountain
building
buildingininthe
thecentral
centralAndean
Andeanregion,
region,Journal
JournalofofSouth
SouthAmerican
AmericanEarth
EarthSciences
Sciences11(3),
11(3),211–215.
211–215.
●
Bottom images show the final state of the experiments (potential temperature in
foreground) compared with the seismic tomographic image (background) modified from
Liu et al. (2003).