Final Report Summary - ZIP (Zooming In between Plates: deciphering the nature of the plate interface in subduction zones)
Subduction zones, by dragging down oceanic tectonic plates (or 'slabs') irreversibly into the mantle, have a profound impact on the lives of ~half or the world population and on our environment, including climate. Subduction indeed ranges amongst the deadliest and costliest natural hazards, having killed ~20,000 people and caused over 200 billion € loss for the single 2011 Japan earthquake. Gigantic earthquakes result from friction across the first ~40 km, and disastrous volcanic eruptions from processes occurring deeper down, beyond ~80-120 km, as slabs heat up, dehydrate and release fluids that enhance mantle melting. Subduction risk assessment requires to understand processes occurring at depths inaccessible to direct observation, right at the boundary between tectonic plates. How complex materials, mechanically and lithologically diverse, release stresses and energy through earthquakes or creep, how material and energy are transferred via fluids and melts, from seconds to millions of years, from nanometres to hundreds of kilometres, has been the prime objective of the ZIP ("Zoom In between the Plates") project.
Through dedicated multidisciplinary supervision and training programmes, with a broad, high-end scientific expertise, up-to-date laboratory facilities and a solid cross-disciplinary basis, ZIP trained 12 early-stage researchers (ESRs) and 2 experienced researchers (ERs) from 10 nationalities in 10 leading Universities and Research Centres spread across Europe (France, Germany, Italy, Spain, Switzerland, Greece) to become Europe's future scientific or science manager leaders.
At the outcome of two (ERs) or three (ESRs) years of thorough training through hands-on innovative training initiatives ("ZIP Starters and Consolidator"), dedicated scientific and transferable skills short-courses and private sector involvement, and after establishing a common scientific language and a strong group feeling, ZIP fellows and scientists fostered interactions, both in terms of scientific education and research. They have published their first joint results in peer-reviewed journals, organized several very well-attended sessions and one large-scale international conference (with > 100 participants from 17 countries), as well as developed dissemination and outreach on the ZIP website (see the seven published newsletters at http://www.zip-itn.eu/ for an overview of ZIP activities) or through the successful final ZIP exhibition.
Three process-oriented research work packages guided ZIP scientific endeavor: (1) determining the plate interface dimensions, geometry and physical properties; (2) modelling time-integrated material fluxes; (3) constraining how rock mechanics controls seismicity, mega-earthquake nucleation and rupture propagation. Buiding on the wealth of data collected and multiple modelling strategies, a number of exciting interdisciplinary (and, for some, supradisciplinary) results can be highlighted:
—Tracking and relocating earthquakes (EQ): GPS data across the Chilean subduction zone (which releases ~1/2 of the world's seismic energy) reveal the diversity of deformation transients along the plate interface: from small amplitude slow slip events, which occur as foreshocks before large EQ, to viscoelastic relaxation of the mantle following intermediate depth intraslab EQ. In the same region, the joint inversion of wide-angle and multi-channel tomography, using land and marine data, allows to map out the structure and distribution of physical properties around the plate boundary (with increasing damage towards the trench) and to refine the position and shape of the plate interface within ~500m. Precise relocation of seismicity in the Hellenic subduction helps refine tsunami wave scenarios or the extent and exact location of ruptures. Relocation of seismicity across the whole Aegean sea also constrains the 3D geometry of sinking plates with unprecedented resolution, revealing slab tears at intermediate depth and a pronounced increase in slab topography near the down dip end of the seismogenic zone (suggesting efficient return flow in the mantle above the interface).
—On the origin of the highly debated intermediate-depth (60-300 km) seismicity: although two mechanisms are generally put forward — dehydration embrittlement of the rock (i.e. fluids weaken the rock as they are released during the mineral transition from a 'blueschist' to an almost dry 'eclogite') or thermal runaway (i.e. due to thermodynamic instabilities), several independent ZIP results suggest that a third mechanism, transformational faulting (thus in both hydrous and anhydrous rocks), is more viable: (i) spatio-temporal properties of these EQ reveal offsets of the moment rate density which are independent from slab age or subduction velocity, (ii) EQ associated with dehydration reactions, reproduced for the first time in the laboratory (and detected by acoustic emissions), reveal that failure was rather caused by volume change and grain size reduction, (iii) spectacular pseudotachylytes (i.e. products of frictional melting during coseismic faulting) formed at ~80 km depth have even been identified in a dry section of the fossil Alpine subducting slab.
—Tracking the fate of metamorphosed slab fragments: Burial of metasediment-rich sequences to ca. 80km depth along the plate interface reveal the existence of slicing of km-scale rock bodies, stripped from the downgoing slab and stacked together at great depth (with implications for mass transfer and advection mechanisms). Analysis of trace element (B, Be, As, Sb, U, Th) and isotopic (Pb, Sr, B) composition of subducted material have allowed to fingerprint element exchange and fluid-rock interactions at these depths. Bulk rock analyses performed with the new, ZIP-derived PPP-LA-ICP-MS technique have also demonstrated the potential of fluid tracers such as B, As, Sb, W and Tl (rather than Cs, Rb, Ba), with consequences on global element cycling and fluid-rock interactions.
—Fluid release and fluid migration: In a wet section of the fossil Alpine subducting slab, multiple brittle rupture events accompanied by a progressive change of the nature of circulating fluids, shortly spaced in time (but whose coseismic origin remains uncertain), have been shown to predate a stage of massive eclogite facies fluid ingression. In a comparable setting, time-integrated fluxes were constrained for these dehydration pulses; they are being compared to 2-phase fluid flow models coupling solid rock visco-elasto-plastic rheology and fluids, with built-in porosity evolution, water loss, density changes due to dehydration, and dehydration fronts.
—Versatile numerical models: One approach has been to link multiple scales, from the long-term (inferred from fossil examples) to the short-term (as seen by GPS and seismological data for active subductions), by combining determination of paleo-stress estimates retrieved from rock samples with modelling of rock mixing (testing the influence of shearing velocity, temperature and viscosity) to quantify the bulk rheology of the subduction inter-plate boundary at a given depth range. Another approach, in 3D, successfully tested the effect of subducting seamounts on the stress field within overriding plates and potential megathrust earthquakes, showing that subducting seamounts hinder the propagation of large EQ and result in smaller, high stress-drops. Finally, unique, new algorithms have been created to model the full breadth of rheological transients, from Ma (with accurate advection and computationally stable implicit formulations) to milliseconds (with computationally fast, explicit inertial formulation), by combining visco-elasto-plastic rheology with a bulk formulation of the rate-and-state frictional formalism (i.e. the state of the art in the simulation of seismic transients).
In parallel, ZIP dissemination and outreach activities have reflected the progressive development of ZIP, the acquisition of promising results and growing ZIP achievements. ZIP fellows have been taking an active part to the dissemination of their results by participating to international conferences (European Geosciences, Goldschmidt, American Geophysical conferences,...) giving lectures and seminars at ZIP events or mini-workshops, linking up to companion international projects (e.g. the 'EFIRE' NSF project), preparing the ZIP final conference (which took place in Barcelona in April 2017), framing a ZIP special issue in a renowned journal (Lithos), publishing peer-reviewed articles. ZIP endeavor has been widely advertized in the rest of the Solid Earth Science community. Five fellows are now hired as post-docs in academia, one was hired in the private sector, five are finishing their PhDs in the next three months, and three still have a few months to finish theirs.
All ZIP results have a great potential impact and will ultimately be used to manage geohazards. Efforts to explain the seismic, tectonic and tsunamigenic processes to the general public culminated in the organization of the 10 day long ZIP exhibition in Paris (October 13-20, 2017), which attracted over 1,400 people. The virtual tour of the ZIP exhibition, which is available online (https://www.visite-virtuelle360.fr/visite-virtuelle/171110-UPMC/) may also serve as an educational toolkit for outreach in vulnerable countries.
[Fig. 1 — ZIP in a nutshell: (i) ZIP fellows; (ii) our target: the subduction plate interface, a major source of hazard and a complex physical and chemical realm; (iii) ZIP tools: ranging from state-of-the-art seismic imaging methods using both active and passive seismic data, continuous monitoring through GPS, InSAR (and permanent or mobile observatories), high-precision analytical technique (with high spatial resolution and low detection limits), and geochemical tracers tied to thermodynamic tools and petrological databases, a variety of experimental apparatus, numerical modelling with high-resolution multi-physics codes taking into account thermodynamics, fluid migration, multiple rheologies and even enabling to reproduce seismicity; (iv) sharing expertise within and outside ZIP, with external scientists (e.g. with EFIRE scientists to the right); (v) reaching out to the general public and sharing enthusiasm (e.g. ZIP exhibition)]
Through dedicated multidisciplinary supervision and training programmes, with a broad, high-end scientific expertise, up-to-date laboratory facilities and a solid cross-disciplinary basis, ZIP trained 12 early-stage researchers (ESRs) and 2 experienced researchers (ERs) from 10 nationalities in 10 leading Universities and Research Centres spread across Europe (France, Germany, Italy, Spain, Switzerland, Greece) to become Europe's future scientific or science manager leaders.
At the outcome of two (ERs) or three (ESRs) years of thorough training through hands-on innovative training initiatives ("ZIP Starters and Consolidator"), dedicated scientific and transferable skills short-courses and private sector involvement, and after establishing a common scientific language and a strong group feeling, ZIP fellows and scientists fostered interactions, both in terms of scientific education and research. They have published their first joint results in peer-reviewed journals, organized several very well-attended sessions and one large-scale international conference (with > 100 participants from 17 countries), as well as developed dissemination and outreach on the ZIP website (see the seven published newsletters at http://www.zip-itn.eu/ for an overview of ZIP activities) or through the successful final ZIP exhibition.
Three process-oriented research work packages guided ZIP scientific endeavor: (1) determining the plate interface dimensions, geometry and physical properties; (2) modelling time-integrated material fluxes; (3) constraining how rock mechanics controls seismicity, mega-earthquake nucleation and rupture propagation. Buiding on the wealth of data collected and multiple modelling strategies, a number of exciting interdisciplinary (and, for some, supradisciplinary) results can be highlighted:
—Tracking and relocating earthquakes (EQ): GPS data across the Chilean subduction zone (which releases ~1/2 of the world's seismic energy) reveal the diversity of deformation transients along the plate interface: from small amplitude slow slip events, which occur as foreshocks before large EQ, to viscoelastic relaxation of the mantle following intermediate depth intraslab EQ. In the same region, the joint inversion of wide-angle and multi-channel tomography, using land and marine data, allows to map out the structure and distribution of physical properties around the plate boundary (with increasing damage towards the trench) and to refine the position and shape of the plate interface within ~500m. Precise relocation of seismicity in the Hellenic subduction helps refine tsunami wave scenarios or the extent and exact location of ruptures. Relocation of seismicity across the whole Aegean sea also constrains the 3D geometry of sinking plates with unprecedented resolution, revealing slab tears at intermediate depth and a pronounced increase in slab topography near the down dip end of the seismogenic zone (suggesting efficient return flow in the mantle above the interface).
—On the origin of the highly debated intermediate-depth (60-300 km) seismicity: although two mechanisms are generally put forward — dehydration embrittlement of the rock (i.e. fluids weaken the rock as they are released during the mineral transition from a 'blueschist' to an almost dry 'eclogite') or thermal runaway (i.e. due to thermodynamic instabilities), several independent ZIP results suggest that a third mechanism, transformational faulting (thus in both hydrous and anhydrous rocks), is more viable: (i) spatio-temporal properties of these EQ reveal offsets of the moment rate density which are independent from slab age or subduction velocity, (ii) EQ associated with dehydration reactions, reproduced for the first time in the laboratory (and detected by acoustic emissions), reveal that failure was rather caused by volume change and grain size reduction, (iii) spectacular pseudotachylytes (i.e. products of frictional melting during coseismic faulting) formed at ~80 km depth have even been identified in a dry section of the fossil Alpine subducting slab.
—Tracking the fate of metamorphosed slab fragments: Burial of metasediment-rich sequences to ca. 80km depth along the plate interface reveal the existence of slicing of km-scale rock bodies, stripped from the downgoing slab and stacked together at great depth (with implications for mass transfer and advection mechanisms). Analysis of trace element (B, Be, As, Sb, U, Th) and isotopic (Pb, Sr, B) composition of subducted material have allowed to fingerprint element exchange and fluid-rock interactions at these depths. Bulk rock analyses performed with the new, ZIP-derived PPP-LA-ICP-MS technique have also demonstrated the potential of fluid tracers such as B, As, Sb, W and Tl (rather than Cs, Rb, Ba), with consequences on global element cycling and fluid-rock interactions.
—Fluid release and fluid migration: In a wet section of the fossil Alpine subducting slab, multiple brittle rupture events accompanied by a progressive change of the nature of circulating fluids, shortly spaced in time (but whose coseismic origin remains uncertain), have been shown to predate a stage of massive eclogite facies fluid ingression. In a comparable setting, time-integrated fluxes were constrained for these dehydration pulses; they are being compared to 2-phase fluid flow models coupling solid rock visco-elasto-plastic rheology and fluids, with built-in porosity evolution, water loss, density changes due to dehydration, and dehydration fronts.
—Versatile numerical models: One approach has been to link multiple scales, from the long-term (inferred from fossil examples) to the short-term (as seen by GPS and seismological data for active subductions), by combining determination of paleo-stress estimates retrieved from rock samples with modelling of rock mixing (testing the influence of shearing velocity, temperature and viscosity) to quantify the bulk rheology of the subduction inter-plate boundary at a given depth range. Another approach, in 3D, successfully tested the effect of subducting seamounts on the stress field within overriding plates and potential megathrust earthquakes, showing that subducting seamounts hinder the propagation of large EQ and result in smaller, high stress-drops. Finally, unique, new algorithms have been created to model the full breadth of rheological transients, from Ma (with accurate advection and computationally stable implicit formulations) to milliseconds (with computationally fast, explicit inertial formulation), by combining visco-elasto-plastic rheology with a bulk formulation of the rate-and-state frictional formalism (i.e. the state of the art in the simulation of seismic transients).
In parallel, ZIP dissemination and outreach activities have reflected the progressive development of ZIP, the acquisition of promising results and growing ZIP achievements. ZIP fellows have been taking an active part to the dissemination of their results by participating to international conferences (European Geosciences, Goldschmidt, American Geophysical conferences,...) giving lectures and seminars at ZIP events or mini-workshops, linking up to companion international projects (e.g. the 'EFIRE' NSF project), preparing the ZIP final conference (which took place in Barcelona in April 2017), framing a ZIP special issue in a renowned journal (Lithos), publishing peer-reviewed articles. ZIP endeavor has been widely advertized in the rest of the Solid Earth Science community. Five fellows are now hired as post-docs in academia, one was hired in the private sector, five are finishing their PhDs in the next three months, and three still have a few months to finish theirs.
All ZIP results have a great potential impact and will ultimately be used to manage geohazards. Efforts to explain the seismic, tectonic and tsunamigenic processes to the general public culminated in the organization of the 10 day long ZIP exhibition in Paris (October 13-20, 2017), which attracted over 1,400 people. The virtual tour of the ZIP exhibition, which is available online (https://www.visite-virtuelle360.fr/visite-virtuelle/171110-UPMC/) may also serve as an educational toolkit for outreach in vulnerable countries.
[Fig. 1 — ZIP in a nutshell: (i) ZIP fellows; (ii) our target: the subduction plate interface, a major source of hazard and a complex physical and chemical realm; (iii) ZIP tools: ranging from state-of-the-art seismic imaging methods using both active and passive seismic data, continuous monitoring through GPS, InSAR (and permanent or mobile observatories), high-precision analytical technique (with high spatial resolution and low detection limits), and geochemical tracers tied to thermodynamic tools and petrological databases, a variety of experimental apparatus, numerical modelling with high-resolution multi-physics codes taking into account thermodynamics, fluid migration, multiple rheologies and even enabling to reproduce seismicity; (iv) sharing expertise within and outside ZIP, with external scientists (e.g. with EFIRE scientists to the right); (v) reaching out to the general public and sharing enthusiasm (e.g. ZIP exhibition)]