2013Geophysical Journal InternationalOpen access

A numerical investigation of continental collision styles

Reza Khabbaz Ghazian, Susanne Buiter

Open full text 20 citations

Abstract

Continental collision after closure of an ocean can lead to different deformation styles: subduction of continental crust and lithosphere, lithospheric thickening, folding of the unsubducted continents, Rayleigh-Taylor (RT) instabilities and/or slab break-off.We use 2-D thermomechanical models of oceanic subduction followed by continental collision to investigate the sensitivity of these collision styles to driving velocity, crustal and lithospheric temperature, continental rheology and the initial density difference between the oceanic lithosphere and the asthenosphere.We find that these parameters influence the collision system, but that driving velocity, rheology and lithospheric (rather than Moho and mantle) temperature can be classified as important controls, whereas reasonable variations in the initial density contrast between oceanic lithosphere and asthenosphere are not necessarily important.Stable continental subduction occurs over a relatively large range of values of driving velocity and lithospheric temperature.Fast and cold systems are more likely to show folding, whereas slow and warm systems can experience RT-type dripping.Our results show that a continent with a strong upper crust can experience subduction of the entire crust and is more likely to fold.Accretion of the upper crust at the trench is feasible when the upper crust has a moderate to weak strength, whereas the entire crust can be scraped-off in the case of a weak lower crust.We also illustrate that weakening of the lithospheric mantle promotes RT-type of dripping in a collision system.We use a dynamic collision model, in which collision is driven by slab pull only, to illustrate that adjacent plates can play an important role in continental collision systems.In dynamic collision models, exhumation of subducted continental material and sediments is triggered by slab retreat and opening of a subduction channel, which allows upward flow of buoyant materials.Exhumation continues after slab break-off by reverse motion of the subducting plate ('eduction') caused by the reduced slab pull.We illustrate how a simple force balance of slab pull, slab push, slab bending, viscous resistance and buoyancy can explain the different collision styles caused by variations in velocity, temperature, rheology, density differences and the interaction with adjacent plates.

Open-access reader

About this research paper

What this paper is about

Continental collision after closure of an ocean can lead to different deformation styles: subduction of continental crust and lithosphere, lithospheric thickening, folding of the unsubducted continents, Rayleigh-Taylor (RT) instabilities and/or slab break-off.We use 2-D thermomechanical models of oceanic subduction followed by continental collision to investigate the sensitivity of these collision styles to driving velocity, crustal and lithospheric temperature, continental rheology and the initial density difference between the oceanic lithosphere and the asthenosphere.We find that these parameters influence the collision system, but that driving velocity, rheology and lithospheric (rather than Moho and mantle) temperature can be classified as important controls, whereas reasonable variations in the initial density contrast between oceanic lithosphere and asthenosphere are not necessarily important.Stable continental subduction occurs over a relatively large range of values of driving velocity and lithospheric temperature.Fast and cold systems are more likely to show folding, whereas slow and warm systems can experience RT-type dripping.Our results show that a continent with a strong upper crust can experience subduction of the entire crust and is more likely to fold.Accretion of the upper crust at the trench is feasible when the upper crust has a moderate to weak strength, whereas the entire crust can be scraped-off in the case of a weak lower crust.We also illustrate that weakening of the lithospheric mantle promotes RT-type of dripping in a collision system.We use a dynamic collision model, in which collision is driven by slab pull only, to illustrate that adjacent plates can play an important role in continental collision systems.In dynamic collision models, exhumation of subducted continental material and sediments is triggered by slab retreat and opening of a subduction channel, which allows upward flow of buoyant materials.Exhumation continues after slab break-off by reverse motion of the subducting plate ('eduction') caused by the reduced slab pull.We illustrate how a simple force balance of slab pull, slab push, slab bending, viscous resistance and buoyancy can explain the different collision styles caused by variations in velocity, temperature, rheology, density differences and the interaction with adjacent plates.

Why it matters

OpenAlex reports 20 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Continental collision after closure of an ocean can lead to different deformation styles: subduction of continental crust and lithosphere, lithospheric thickening, folding of the unsubducted continents, Rayleigh-Taylor (RT) instabilities and/or slab break-off.We use 2-D thermomechanical models of oceanic subduction followed by continental collision to investigate the sensitivity of these collision styles to driving velocity, crustal and lithospheric temperature, continental rheology and the initial density difference between the oceanic lithosphere and the asthenosphere.We find that these parameters influence the collision system, but that driving velocity, rheology and lithospheric (rather than Moho and mantle) temperature can be classified as important controls, whereas reasonable variations in the initial density contrast between oceanic lithosphere and asthenosphere are not necessarily important.Stable continental subduction occurs over a relatively large range of values of driving velocity and lithospheric temperature.Fast and cold systems are more likely to show folding, whereas slow and warm systems can experience RT-type dripping.Our results show that a continent with a strong upper crust can experience subduction of the entire crust and is more likely to fold.Accretion of the upper crust at the trench is feasible when the upper crust has a moderate to weak strength, whereas the entire crust can be scraped-off in the case of a weak lower crust.We also illustrate that weakening of the lithospheric mantle promotes RT-type of dripping in a collision system.We use a dynamic collision model, in which collision is driven by slab pull only, to illustrate that adjacent plates can play an important role in continental collision systems.In dynamic collision models, exhumation of subducted continental material and sediments is triggered by slab retreat and opening of a subduction channel, which allows upward flow of buoyant materials.Exhumation continues after slab break-off by reverse motion of the subducting plate ('eduction') caused by the reduced slab pull.We illustrate how a simple force balance of slab pull, slab push, slab bending, viscous resistance and buoyancy can explain the different collision styles caused by variations in velocity, temperature, rheology, density differences and the interaction with adjacent plates.

Key concepts: Geology, Asthenosphere, Subduction, Lithosphere, Continental collision, Collision zone, Crust, Eclogitization

Related papers

Back to paper searchBrowse research topicsOriginal source
A numerical investigation of continental collision styles — Research Paper | ScholarLens