My main interest lies in thermochemical evolution of our planet Earth. To achieve this I study crust & mantle architecture using various potential field data, passive seismology data and techniques, and relate this to mineral physics.
I am also interested in geodynamics, particularly the end of a subduction zone (oceanic) where continental collision starts. This geological setting is very interesting because a lot of interesting and exotic processes could happen here i.e. slab delamination, slab tear, slab-break off.
I am equally interested in earthquake sources, which have a direct impact on society, and their tectonic significance particularly in continental collision zones, which are yet to be fit in the theory of plate tectonics realm.
Our recent work has highlighted the existence of a critical crustal thickness, controlled by its thermochemical state and resultant thermomechanical state. Simply put, if the continental crustal thickness is above this value it is weaker because of the higher-than-average radiogenic heat production (RHP), and is prone to diffusive deformation as observed in the continent-continent collision zones. The value of this critical crustal thickness matches the global average thickness of the continental crust. This allowed us to understand RHP as a dissipative source to modulate the deformation of continental lithosphere in response to the plate-boundary forces as a driving mechanism ultimately controlled by the secular cooling of the mantle (i.e., generation of new plates in the mid-oceanic-ridges following by cooling and recycling at subduction zones producing perturbative plate-boundary forces).
In this work, focused on the Alps, we use the present-day 3D architecture of the upper mantle as an input to forward geodynamic simulations to reconcile present-day deformation (i.e., seismicity, GNSS derived surface velocity field, SKS derived mantle flow)
Imaging the present-day physical state, temperature and composition, of the lithosphere asthenosphere is crucial to understand the dynamics of the lithosphere. Given the highly coupled nature of processes operating in solid Earth and implied non-uniqueness, originating either from the scarce datasets or the differential sensitivity of these datasets to the physical properties, images of the Earth's interior are often a non-unique solution. Integrating multiple datasets where they complement each other in an internally self-consistent manner significantly helps a) to narrow down the solution space and b) to have an integrated view of the Earth's interior. Integrated geophysical-petrological modelling to derive the physical state of the lithosphere explaining multi-disciplinary datasets in a self-consistent thermodynamic framework has been introduced in an approach called LitMod (Afonso et al. 2008). However, incorporating anomalies, e.g., subducting slabs, mantle upwelling, and delaminated lithosphere, in the sublithospheric mantle has been challenging. We recently improved this approach (LitMod2D_2.0, Kumar et al. 2020), allowing us to quantitatively model the thermochemical nature of such anomalies in the sublithospheric mantle. LitMod2D_2.0 is open and free to use for the community (https://github.com/ajay6763/LitMod2D_2.0_package_dist_users).
Using this method, we studied the lithosphere and upper mantle structure in the Western Mediterranean to discuss the role of the proposed geodynamic models in shaping the present-day physical state (Kumar et al. 2021). We found that subducted slabs beneath Eastern Betics in Southern Spain and Kabylies in Northern Africa, are of fertile chemical composition compared to the foreland lithospheres. The apparent orientation of these slabs allowed us to infer that the evolution of this region was governed by the subduction of the opposite dipping and retreating lithosphere of fertile composition similar to that of present-day oceanic lithosphere. Further, we found that at present these slabs are not attached to the overlying lithosphere and could have produced ~1000 m uplift during break-off.
This work highlighted the role of the arc normal heterogeneities in the crust in modulating the deformation, observed in the co-seismic time-scale during the 2015 Gorkha earthquake, where an ~N-S striking structure (i.e., lateral ramp) influenced the rupture propagation. In this work, we also developed an open-source tool for the community to perform back-projection of high-frequency sources during the co-seismic phase (https://github.com/ajay6763/earthquake_bp_obspy). We plan to develop this tool further to deploy online and in real-time monitoring of big earthquakes to produce rapid source solutions (order of minutes) for disaster management agencies.
We studied the earthquake distribution in space and the style of deformation (i.e., fault-plane solutions) in the Eastern Himalaya, bounded by continental collision to the north and Indo-Burman subduction to the east (Kumar, Mitra, and Suresh 2015). We found that the convergence between India and Eurasia in this part of the Himalayas is partitioned, resulting in a distributed active deformation away from the plate boundaries. This distributed deformation is accommodated in thrust (Himalayan front, north of Shillong Plateau) and strike-slip deformation (Kopili fault zone, Bengal Basin). The strike-slip component of the deformation along the Kopili fault zone is observed to be extending north beneath the Himalayan front as well as south beneath the Indo-Burman ranges. Interestingly, we found that most of the earthquakes in the Bengal Basin, south of the Shillong plateau, showed strike-slip deformation which is aligned along the strike of the Indian passive margin. Hence, we speculated that fault systems that developed during the separation of India from Antarctica are now being activated in a strike-slip fashion under present-day overall N20E convergence. The whole of NE India is an interaction zone of two kinds of plate convergence, continent-continent collision to the North and subduction to the east. However, what kind of force balance would give rise to such an observed configuration of deformation, is not clear at this point.
