2.1. The forces related to Mantle Plume encounter
The most likely candidate for the above exploration is the Reunion (Deccan) mantle plume. However as pointed out by previous studies, the Deccan plume propelled India forward towards N-NE directions (Gibbons et al 2015, Gaina et al 2015, Cande et al 2010, Cande and Stegman 2011). The plume thinned the Indian craton, and could have pushed/propelled India towards west, with anticlockwise rotation. However, the plume itself being positioned on the western continental margin of India, it seems unlikely that it would drag the subcontinent towards west. Although it is possible that there could have been a torsional effect exerted on India arising from the toroidal flows in the asthenospheric mantle surrounding the plume, which could have possibly rotated the subcontinent anticlockwise towards west. The removal of cratonic root beneath India did not support anchoring in the mantle flow. Such anchoring could have encouraged absolute torsional rotation. Further given the geodynamic positioning of the plume towards the western margin, it seems highly unlikely that the plume would entirely rotate the subcontinent towards west. If the plume were to rotate the subcontinent wholly then it would be a much larger effect something comparable to the lithospheric tilting of the Indian subcontinent (e.g., Sangode et al 2022). Thus, from all the above evidences and analogical inferences it appears to be unlikely that the Reunion plume was solely responsible for the westward drift of India during the Deccan eruptions. However, the inferences need holistic consideration of all the possible major forces for Indian plate kinematics that can be evident from the observations of the oceanic lithosphere surface.
2.2. Forces related to Mid Ocean Ridge activity
The spreading history of the western Indian ocean can be viewed as an artifact of two major spreading ridges viz., i) Mascarene Spreading centre (MSC) and ii) Central Indian ridge (CIR)
Spreading rates in the Wharton basin would not have accounted for much as it was in the waning phases. A similar predicament prevails for the Mascarene spreading centre. However, any significant contributions from the MSC would result in enhanced motion of India towards east eliminating any chances of a westward motion and the proto-CIR might possibly have played a role in this westward migration. The spreading in MSC slowed down by ~ 68.5 Ma (anomaly C31n), and a ternary rift system came into existence off the west coast of India in the form of Laxmi ridge-Gop ridge-Narmada rift. Spreading in the Northern Mascarene basin ceased completely by 62.5 Ma, the time by which Seychelles was completely stripped off of India (Bhattacharya and Yatheesh 2015). In Southern Mascarene basin the spreading stopped somewhere around ~ 60.9 Ma and the spreading centre jumped north between southern end of the Laccadive plateau and the northern boundary of the Mascarene basin to relocate itself closer to the hotspot.
Based on the relative movements between spreading ridges and hotspots, three major interactions in the Indian ocean have been identified here as described below (Small 1995).
-
Ridge migration towards the hotspot.
-
Ridge captured by the hotspot Or Ridge situated above a hotspot.
-
Ridge migrating away from a hotspot.
If the hotspot remained/existed within moderate distance of a ridge, the ridge moved through a series of ridge jumps and remained (through spreading asymmetry) above or in close proximity to the hotspot (Small 1995). This moderate distance seems to depend on the strength of the hotspot e.g., extinction of MSC and the opening of the Carlsberg ridge in response to the arrival of Reunion hotspot, that can be envisioned as 1000–1500 km ridge jump.
Prior to the Deccan event, the hotspot was located NE of the Ridge while at the Deccan event it was below the ridge and following the main phase of Deccan eruptions, it was situated southwest of the ridge. As mentioned earlier, the ridge jumping and relocation of a spreading centre took place which resulted in shift of spreading axis ~ 1000–1500 km shift towards east.
Another interesting fact that needs to be considered here is the time required for a ridge jump. It is found to be of the order of 105-106 years for fixed or migrating ridges (Mittelstaedt et al 2001). Once a ridge jump occurs, the hotspot and ridge migrate together for time periods that increase with magma flux. However, in the present case, after the ridge jump occurred, the magma flux was reduced causing decoupling/separation of the ridge from hotspot. The magmatic heat flux from the Reunion plume would have been sufficient enough to initiate the ridge jump, from which spreading in the Mascarene basin ceased. In the event if the ridge jump happens, there could be a temporary change/distortion in the existing plate kinematics. This discordance could possibly result in disfigured motions albeit for a short while, which would be long enough for India to produce the observed directional drift due to high drift rates. Once the new spreading regime along the CIR is established, restoration of the drift directions might take place and India continued its NE journey unimpeded. But, the fact that the reorganization of the spreading systems occurred around 60–62 Ma and not at 65–66 Ma, rules out the reorganization of the spreading centres as the sole cause for the observed westward drift of India.
2.3. Forces related to Subduction
Establishment of a new subduction zone would definitely change the force balance in a plate kinematic framework. However, as there was already a massive pre-existing subduction system existing along the Northern margin of the Indian plate i.e., the Neo-Tethyan subduction system, any new subduction system would prefer to be in the proximity of the established system. Along with this mother system there existed some other smaller intra-oceanic subduction systems along the northern margin of India, these systems contributed to the northward drift as well. The challenge is to how exactly did the existing subduction systems all of a sudden exerted an unprecedented westward pull over the Indian plate, and why would this pull stop shortly after the Deccan episode thereafter resuming the old drift direction. Therefore, the major questions that need to be addressed here are as follows:
-
Which subduction zone was responsible for this short westward drift?
-
What caused an increase in the slab pull, that could possibly overpower the Reunion plume push?
-
Why did this slab pull decrease suddenly or become non-existent at the end of Deccan episode?
Many previous studies have proposed for an intraoceanic subduction system to have initiated at the Deccan episode owing to the Reunion plume (Jagoutz et al 2016, Gnos et al 1998, Rodriguez et al 2019, 2020,Pandey et al 2019, Gaina et al 2015). However, despite the extensive research that has been conducted, there have been no successful candidates so far. Although 3 systems do appear in the vicinity of India (Fig. 2) at the Deccan episode, they are still surrounded by controversies regarding their age and origin. All the three are aligned along the western margin of India, this makes them worthy contenders.
-
Bela-Waziristan-Muslim Bagh-Kabul subduction system (Gaina et al 2015, Gnos et al 1998, Mahmood et al 1995).
-
Amirante Trench system (Rodriguez et al 2019, 2020, Plummer 1996).
-
Laxmi Basin Fore Arc sequence (Pandey et al 2019)
Rodriguez et al (2020) hypothesized that the Reunion plume in its pre-Deccan state at ~ 90–100 Ma formed an arcuate subduction system that broke into 3 distinct limbs, viz., The Northern, North-western and Southern limbs. Of these the North-western branch drifted off to form the Troodos and Semail ophiolites, the northern branch drifted to form subduction systems resulting in the formation of ophiolites along the Indus suture zone; and the Southern branch drifted southward along the NW frontier of India to form the string of ophiolites scattered along Bela-Waziristan-Muslim Bagh-Kabul ophiolite belt.
It is interesting to note here that Gnos et al (1998) proposed the age of subduction initiation at ~ 70 − 68 Ma and the age of ophiolite emplacement at ~ 66 − 65 Ma for the Bela ophiolite. Gnos et al (1998) proposed that the Reunion plume was responsible for initiation of subduction along the NW margin of India. However, as interesting it may appear, we are left to wonder if the subduction initiated at ~ 70 Ma then why did the change in course occur at C29r and that too for a very short time spanning ~ 700–800 ka. The ophiolites in general depict obduction due to higher drift rates, their ages above restricted to 66 − 65 Ma are to be explained. What happened after C29r, and why did the westward drift stop so abruptly if the subduction system was active? If the subduction along Bela belt caused this drift, why did it not continue after the Deccan episode? These questions beg certain answers that the Bela system fails to provide. We are therefore forced to look beyond this system for a possible answer.
Pandey et al (2019) detected Fore Arc Basalt (FAB) and Boninite like signatures from the basalts in the Laxmi basin off the west coast of India (Fig. 3). These basalts although did not entirely exhibit Boninite or FAB signatures, they however displayed eerily similar trace element signatures. This led them to concur that the crust in the Laxmi basin was relict of a failed subduction system which could not be sustained owing to rapid slab rollback within the proposed subduction zone and the collisional events along the Northern subduction margin. This could have been the possible subduction system we are looking for; however, the extent of the active margin and duration of the event make it rather unsuitable for the scale of tectonic disturbance we are dealing here. In addition to this, the proposed site of subduction initiation lies very close to the continental margin of India. This makes it unsuitable for accommodating the displacement and the rotational torques arising out of the plume induced rotation.
There was another important Neo-Tethyan system along the Indian plate active during the Cretaceous which could have had a profound effect on the drift of India after the Deccan event. This was the Sunda subduction system active along the north-eastern margin (McCourt et al 1996, Gibbons et al 2015, Crameri et al 2020). The Sunda subduction system experienced a subduction lull from 75 − 65 Ma during the accretion of the Woyla terrane, following which there was a pause in subduction along the entire margin of the Woyla arc for a period of ~ 10 Ma (Muller et al 2016). This period of quiescence came to an end after ~ 65 Ma which is precisely the Deccan event (Fig. 4). The reactivation of subduction in the Sunda system enhanced the NE slab pull on the Indian plate thereby causing the NNE drift of India. This might explain the sudden switch in plate movement towards NE, and also to some extent contribute towards the cessation of NW drift of India. However, this still leaves the origins of the Northwest drift unresolved.
The Amirante ridge trench system in the Indian ocean (Fig. 1) between the latitudes 3-100S presents an interesting arcuate feature in the above context. The origins of this arcuate body have long been debated. The curvature resembles that of an arcuate subduction zone, while there are no significant island arc rock assemblages associated with it, that could conform its origin as a subduction zone. However, in the current study we propose that the Amirante represents a failed system that could not be sustained despite being originated in the close vicinity of an active margin.