1,561
Views
0
CrossRef citations to date
0
Altmetric
Review Article

Analogue modelling of subduction initiation: a review and perspectives

ORCID Icon, , &
Pages 2123-2165 | Received 21 Mar 2023, Accepted 14 Oct 2023, Published online: 27 Oct 2023
 

ABSTRACT

Subduction initiation (SI) is crucial for understanding the driving mechanisms of plate tectonics, the strength, and heterogeneity of the lithosphere, as well as the interaction of various driving/resisting forces. However, where and how subduction zones form remains confusing and contentious because of limited observational data. Geodynamic modelling provides a data-independent, reproducible way to test different hypotheses of SI. Most numerical models of SI have been reviewed by the geodynamic community in the past; however, little attention was paid to analogue modelling works. In this study, we have reviewed analogue models of SI according to the driving mechanisms (induced or spontaneous), then to whether or not SI occurs near pre-existing subduction zones, and finally to the tectonic settings of SI, which are usually different-typed lithospheric weak zones, including lithospheric-scale inclined faults, locally thinned zones, transform boundaries, passive margins, island arcs, back-arc basins/spreading centres, and plume head margins. Induced SI and strain localization cannot occur within the intact plate unless some weak zones exist, whose strengths are largely affected by their tectonic styles. The lithospheric-scale inclined fault is the weakest tectonic setting; the locally thinned zone also has relatively low strength; the transform boundary and the passive margin are strong. Spontaneous SI of oceanic lithosphere is usually triggered by extra loads, such as collapsed continental crust and flooding plume materials. The negative buoyancy of oceanic lithosphere will then take over and drive self-sustaining subduction. Modelling results may explain why a young oceanic plate sometimes underthrusts beneath an old one and why transference SI still does not happen after long-lasting India-Asia collision (~55 Ma). According to the experiences and deficiencies of the past works, future analogue modelling works should focus on transform collapse, subduction transference, case studies modelling, non-cylindrical modelling, realistic model setups of weak zones, and quantitative evaluation/analyses of SI.

GRAPHICAL ABSTRACT

Acknowledgments

We thank revising suggestions from the editor Robert Stern. We also appreciate two anonymous reviewer for the constructive comments that helped to improve this manuscript. was mapped using the Generic Mapping Tools (GMT) software (Wessel et al. Citation2019).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/00206814.2023.2272273

Additional information

Funding

The work was supported by the National Natural Science Foundation of China [42025601, 92258303, 42076080]