Earthquakes and ophiolite emplacement in the Molucca Sea Collision Zone, Indonesia

Abstract

The Molucca Sea in eastern Indonesia is the site of an active arc‐arc collision, and among its interesting features is the exposure of ophiolitic rocks on islands of the Talaud‐Mayu Ridge (TMR), a largely submarine ridge formed of highly deformed rocks that bisects the collision zone. To explore the relationship between earthquakes and uplift of the ophiolite, the centroid depths and fault plane solutions of 18 large earthquakes occurring in the past 20 years beneath the TMR are constrained by inversion of their teleseismic, long‐period P and SH waveforms. Centroid depths range from 16 to 36 km, but uncertainties allow a range of 10 to 45 km. All events show thrust faulting; nodal planes strike parallel to the NNE trending TMR and dip 40±9° to the ESE and 53±9° to the WNW. Published seismic refraction and gravity data support the inference that the WNW dipping, steeper nodal planes are the fault planes. Dips do not change with depth, indicating either that the fault flattens below 40 km depth or that the steeply dipping fault penetrates the entire thickness of the lithosphere. I conclude that the Molucca Sea ophiolite is being lifted by high‐angle thrust faults (45° to 60°) that extend at least 15 km into the upper mantle. Summing seismic moment tensors and assuming uniformly distributed deformation suggest that the closure across the Molucca Sea collision zone may account for 14% to 59% of the Pacific‐Eurasia convergence vector or 15% to 63% of the Philippine‐Eurasia convergence. Moment tensor sums provide estimates of crustal thickening rates beneath the TMR of 7 to 20 mm/yr and these imply uplift rates of 2 to 6 mm/yr if the increase in crustal thickness is isostatically compensated. Hence the ophiolite and an enormous volume of mélange are being lifted higher than the flanks of the island arcs and will be emplaced on them by gravity. I suggest that ophiolites occur in this tectonic setting because the concave down shape of the subducted oceanic lithosphere beneath the Molucca Sea increases its effective buoyancy, preventing it from sinking out of the way of the encroaching island arcs. This case suggests a family of convergent margin settings in which normal oceanic lithosphere is buoyed up so that ophiolites can be stripped from it and emplaced on a continental margin or island arc. These settings include those where (1) slab geometry restricts asthenospheric flow away from the convergent margin, (2) subduction of younger, less dense lithosphere decreases its negative buoyancy, (3) a passive continental margin resists subduction, or (4) the presence of a thick thermal boundary layer beneath a passive continental margin inhibits asthenospheric flow necessary to accommodate the sinking oceanic lithosphere.