Oceanic crust recycling centers - subduction
zones and arc complexes.
Subduction zones are convergent plate boundaries where
oceanic lithosphere gets recycled deep down into
the mantle to depths of at least 670 km (the upper-lower mantle boundary). What is the significance of this particular depth? Why are volcanic arcs associated with this process?
Presently active subduction arc-complexes:
- circum-Pacific "ring of fire".
- Malay archipelago, Java-Sumatra.
- in between South America and Antarctica - Scotia Arc.
Island arcs vs. continental arcs:
- definition based on type of crust/lithosphere the overlying plate is composed of, island arcs are embedded in oceanic crust, and continental arcs develop on the leading edge of a continental crust/lithosphere.
- they differ in the character of geologic activity
at the plate setting; e.g. the composition of arc volcanism.
- Andes as example of continental arc.
- Aleutians as example of continental arc that
becomes an island arc along strike to the west with a trapped basin behind (a rear-arc
- Tonga-Kermadec system as an example of an
- Japan as a hybrid?
- can accrete an island arc to a continental
margin or to each other (more on accretion later).
Components of subduction-arc
These are described in order from the oceanic plate being subducted across the trench and into the overriding plate. Click on the figure to the right for a schematic cross section of many of the features.
- forebulge in descending plate: subtle feature.
- trench, where the plates meet:
- outer trench wall - site of small scale normal faulting thought mechanically to be due to plate bending and the beam extrados position.
- inner trench wall
- complex topography with trench parallel ridges cut by submarine canyons.
- trench sediments;
- mainly turbidites and pelagic sediments.
- amount and nature of influenced by whether island or continental arc and 'vigor' of source terranes.
- commonly deformed even though young.
- accretionary wedge, in between the arc-trench gap:
- information on from: seismic images of, drill holes, local exposures,
- Above image from USGS site http://walrus.wr.usgs.gov/research/sopac.html. It shows a seismic image of the wedge. Note the typical well developed reflection off the top of the subducting slab. An interesting question is - what determines the taper angle of the wedge?
- character of accretionary melange:
- very complex geology.
- Franciscan melange along the California coast is a classic example, and represents an exhumed Cretaceous accretionary wedge.
- melange: pervasively deformed matrix often of muds and cherts (once pelagic oozes on the sea floor), mixed with trench fill turbidites. Within are inclusions of variable size (some larger than Durham Science Center) of basalt, gabbro, serpentine, ultramafites, sometimes blue schist and eclogite, limestone.
- tectonic mixing (folding, thrusting, cleavage formation, penetrative deformation
and mass solute transport).
- sedimentary mixing and soft-sediment deformation due to large scale mass wasting on inner trench slope.
- mud and serpentine diapirism is common, and suggest that high fluid pressures occurred, making faulting easier.
- image to the right is of melange from Washington state (http://geomaps.wr.usgs.gov/parks/noca/t3bellpass.html). Note the multiple 'surfaces' in the rock and their greenish coloration (hence the general descriptive term greenstone. This rock has been very highly sheared during low grade metamorphism.
The above two photos are from a locality along the northern California coast known as Goats Rock where the Franciscan melange is well exposed. The first image is of serpentinized material. Each smooth surface is a slip surface coated by the dark green serpentine. There are so many slip surfaces in this material that it was able to "flow" and deform easily, a mode of deformation known as cataclastic flow. The other photo shows folded and veined layers of dark gray chert, and light green sandstones composed of volcanic debris. These can be interpreted as interbedded turbidites (the sandstones), and siliceous pelagic oozes that lithified into the cherts.
This is a larger view of an outcrop at the Goats Rock site where the extremely deformed character of the material and the existence of tectonic blocks/inclusions can be seen.
This image is of a meta-basalt block within the Franciscan melange at a site known as Patricks Point. Note the sea otter for scale.
Complexly deformed ribbon chert boulder at Patricks Point. Again, the interpretation would be that these were deep sea siliceous oozes.
This is highly deformed and veined meta-mudstone (also from Patricks Point), that has been metamorphosed to something close to a phyllite. These may represent the easily deformed mud matrix that once were sea floor pelagic muds but now hosts a variety of tectonic inclusions. The abundant vein fill reminds one of the solute mass transfer that was occurring during deformation.
- Distinctive high pressure metamorphic facies in exhumed accretionary wedges (will hear more about paired metamorphic belts next week) and as inclusions in the melange: blue schists and eclogites.
- What are the larger scale internal deformation (mixing) processes?
- thin skinned thrusting and folding very
common, with foreland propagation pattern.
- Above: Seismic section across the Soloman subduction zone from the USGS site: http://walrus.wr.usgs.gov/tsunami/solomon07/ .
Note the flat lying decollement with thrust splays that define the thin-skinned toe of the accretionary wedge.
- Seismic image of Nankai subduction accreationary wedge (Japan) showing the thin-skinned character and folds and thrusts above a detachment (decollement), and associated slumps. Slumping contributes to the chaotic and complex character of accretionary wedge rocks. Image from https://www.iris.edu/gallery3/research/lrsp/APPEND2 , the original Courtesy of University of Texas Jackson School of Geosciences.
- How does the accretionary wedge evolve with time?
- accretionary growth primarily through underplating
at toe (so deformation and sediments get younger towards the trench).
- addition of arc sediments.
- accretionary tectonic erosion at the basal detachment a possibility (so that wedge loses volume).
This has interesting implications for what may be carried to deeper depths.
- internal basins (e.g. Shortland Basin in above image): transient, catch turbidite
flows, can incorporate arc-derived sediments (which will be relatively immature).
Summary image from USGS site of subduction-arc complex components http://earthquake.usgs.gov/learning/glossary.php?term=accretionary%20wedge .
In class exercise - evaluation
of seismic images of accretionary complexes. In the xerox
copies of portions of seismic reflection profiles of accretionary
complexes identify as many geologic features as possible and
relate them to the subduction process. Report on the three or
four most significant features to the class.
- Brief tips on seismic interpretation:
- Be aware of surface and other multiples.
- Remember that as you go down seismic resolution
decreases - the seismic 'waters' get muddier.
- The vertical axis is travel time, not distance.
Because velocities change, usually increase with depth, distortions
can result. Structures will generally be more compressed at depth.
- Fault truncations and other point reflectors
produce a diffraction pattern that looks a bit like a upwards
pointing bullet profile. This type of seismic noise is hard to
get rid of.
- In general, sedimentary rocks have a layered
signature (with their bedding reflectors), while basement igneous
or metamorphic rocks are structureless, or at least a lot less
- Important features that can often be well
imaged: basement-cover contact, fold structures, fault structures,
The descending slab
How do we know what is happening at greater depths?
Wadati- Benioff zone:
- defined by earthquakes.
- elsewhere earthquakes in upper crust generally at depths of 30 km or less, as determined by the crustal brittle-ductile transition.
- the mechanism behind these deeper subduction-related earthquakes likely differs, may involve a phase change.
- basic assumption is that it is the colder subducted crust that the earthquakes localize in - i.e. the top of the subducting slab.
- dip of the zone is highly variable.
- 3-D geometry of can be complex.
- Image from USGS site on models of subduction slabs. Source: http://earthquake.usgs.gov/research/data/slab/.
Additional downloads that will allow you to visualize the slabs in 3-D using Surfer or GIS are available at this site.
- Possible in-class exercise: using the USGS earthquake database to produce kml files of earthquake depths and construct contours on subducting slabs. In the USGS earthquake database one can chose the option to download the requested earthquake data as a kml file, which Google Earth will directly read and display. In addition to specifying the area and time over which you like to extract the earthquake data one can also select to display by depth (which is particularly useful for this exercise). It is possible to download tens of thousands of points which tends to slow things up, and so it is useful to also select the option where you retrieve only earthquake greater than 4.5 in magnitude. One site entry into this database is at: http://earthquake.usgs.gov/earthquakes/search/ .
- First motions studies of slab earthquakes and the stresses within.
- Cross section image across the Tonga trench. The small circles with four quadrants are showing first motion studies. By convention the colored quadrant is the compressive, and so the geometry is consistent with thrust motion along the base of the edge. Note that the geometry here would infer slip within the descending plate. Image source from: http://earthquake.usgs.gov/earthquakes/eqarchives/subduction_zone/us2009ejbr/ .
The nature of the earthquakes in a descending slab can change at depth.
Thermal structure of the descending
slab, and why it subducts
What are important input parameters in a model for the thermal structure of a subducting plate?
Phase changes in the slab and slab pull:
- mantle metamorphism with depth: olivine -> spinel
- 670-780 km discontinuity:
- division between upper and lower mantle.
- lower bound of deep earthquakes.
- seismic velocity increase.
- P-T consistent with position of phase changes.
- viscosity boundary. How does it effect convection
patterns in mantle - stratified?
- P-T maps of phase changes are crucial to understanding this. Where do these
maps come from?
Consider the simple and schematic pressure-temperature
maps above of where mineral A vs. mineral B is stable for a given
composition. To the left is a temperature sensitive phase transition.
Next in line is a pressure sensitive phase transition. Third
and of interest to us is a positive slope for the phase change.
Note that for a different geothermal gradients or P-T histories,
the phase change will occur at different depths. Will the reaction
occur at a deeper or shallower depth in a subducting slab in
comparison to the surrounding mantle rock. The olivine-spinel
phase transition map basically looks like this.
What would be the consequence of elevated olivine to spinel transformation in slab due to the lower geothermal gradient within the slab?
Image of gravity anomalies associated with part of the Aleutian subduction zone. That is pertinent here is the broad backarc gravity low. The long wavelength of the low suggests a deeper origin, and one interpretation is that it is due to the elevated olivine to spinel transformation that occurs in the slab. Image from and more information at: http://pubs.usgs.gov/of/2000/ofr-00-0365/report.htm .
Miscellaneous remaining questions to ponder
Why do subduction zones produce the biggest earthquakes?
Why is the trace of the arc and trench arcuate in form (dimple
What is the eventual fate of the slab material? It is likely different for the different parts
of the slab, crust vs. lithospheric mantle. There is no reason
why some of the later when warmed up doesn't join the mantle convection
How does subduction initiate (Stern & Gerya 2017)?
How does it usually conclude?
Volcanic arc and rear arc activity and tectonism: covered
Expanding earth hypothesis: for those who accept seafloor spreading, but not subduction.
- Most interesting as a case study in the history of science of how a minority paradigm can persist.
- Remember earlier paradigm that earth had shrunk due to cooling.
- Link to overview of hypothesis.
- Abundant evidence idea does not work for the earth, but for other planets?
Bebout, G. E., Scholl, D. W., Stern, D. W., Wallace, L. M., and Agard, P., 2018, Twenty Years of Subduction Zone Science: Subduction Top to Bottom 2 (ST2B-2); GSA Today, 28, 4-10 - http://www.geosociety.org/gsatoday/science/G354A/article.htm. Review article and good place to start.
Dickinson, W.R., 1973, Widths of Modern Arc-Trench Gaps Proportional
to Past Duration of igneous activity in associated magmatic arcs:
Journal of Geophysical Research, v. 78, p. 3395-3417.
After a concise description of major petro-tectonic elements,
it describes how accretionary wedges may grow with time, noting
exceptions suggesting loss. This mental framework initiated a
mini-paradigm and much fruitful research.
Hacker, B. R., S. M. Peacock, G. A. Abers, and S. D. Holloway, Subduction factory, 2003, Are intermediate-depth earthquakes in subducting slabs linked to metamorphic dehydration reactions?, J. Geophys. Res.,108 (B1), 2030, doi:10.1029/2001JB001129. The mechanisms that cause deep earthquakes are different that those in the shallow crust (<30 km or so).
Kerr, R., 1986, Sinking slabs puncture layered mantle model;
Science, v. 231., p. 548-49.
This captures some of the debate as to how far subducted slabs
descend into the mantle and argues against 'conventional wisdom'
of a floor at 650-670 km. This has Implications for understanding
large scale recycling and convective behavior. Lots more recent
thought on this.
Kimura, G. & Ludden, J., 1995, Peeling oceanic crust in
subduction zones; Geology, v. 23, p. 217-220.
describes ideas on how material is transferred from ocean plate
to the accretionary wedge. Discusses implications for recycling
of volatiles in arc-trench system.
Moore, J.C. & Silver, E.A., 1987, Continental margin tectonics:
Submarine accretionary prisms: Review of Geophysics, v. 25, p.
Platt, J. P., 1986, Dynamics of orogenic wedges and the upllift
of high-pressure metamorphic rocks; GSA Bulletin, v. 97, p. 1037-1053.
one of a series of articles discussing how alternating periods
of contraction and thickenting vs. extension and thinning could
occur within accretionary wedges and produce uplift of deep cold
rocks. This article is particularly well illustrated.
Schweller, W., Kulm, L., Prince, R., 1981, Tectonics, Structure,
& Sedimentary Framework of the Peru-Chile Trench: GSA Memoir
154, p. 323-350.
Excellent detailed look at trench and trench slope processes for
this subduction zone.
Silver, E.A., Moore, C.J., 1978, The Molucca Sea Collision
Zone, Indonesia; Journal of Geophysical Research, v. 83., 1681-91.
Describes an incipient arc-arc collision due to two facing subduction
Stern, B. & Gerya, T., 2017, Subduction initiation in nature and models: A review; Tectonophysics, https://ac.els-cdn.com/S0040195117304390/1-s2.0-S0040195117304390-main.pdf?_tid=15e5ebe7-6466-4e85-9995-9f180d125c82&acdnat=1536776017_bc77473ae03d70f4e107f303427963d1
Toksoz, M.N., 1975, The Subduction of the Lithosphere; Scientiric
American, Nov. issue.
This is a nice summary article but with good depth of information,
a good starting point in reading on subduction zones. It emphasizes
deep mantle processes.
Von Huene, R. & Scholl, D.W., 1991, Observations at convergent
margions concerning sediment subduction, subduction erosion, and
the growth of continental crust: Reviews of Geophysics, v. 29,
Course materials for Plate Tectonics, GEOL
3700, University of Nebraska at Omaha. Instructor: Harmon D. Maher Jr.. Material without attributions may be used for non-profit educational
purposes with appropriate attribution of authorship. Otherwise
please contact author.