Gondwana subduction-modified mantle domain prevents magmatic seafloor generation in the Central Indian Ridge

The creation of oceanic crust at mid-ocean ridges is essential to understanding the genesis of oceanic plate and the evolution of the Earth. Detailed bathymetric measurements coupled with dense sample recovery at mid-ocean ridge revealed a wide range of variations in the ridge and seafloor morphologies, which cannot be simply explained by a spreading rate, but also by ridge geometry, mantle compositions and thermal structure (Dick et al., 2003 Nature; Cannat et al. 2006 Geology). It is now widely accepted that very limited magmatic activity with tectonic stretching generates oceanic core complex and/or smooth seafloor surface in the slow to ultraslow-spreading ridges, where serpentinized peridotite and gabbros are expected to be exposed associated with detachment faults (Cann et al., 1997 Nature; Cannat et al., 2006), although magmatism might be an essential role for the formation of oceanic core complexes (Buck et al., 2005 Nature; Tucholke et al 2008 JGR). A rising question is why magmatic activity is sometimes prevented during the oceanic plate formation. Ancient melting domain, that are too refractory to melt even in adiabatically upwelling to the shallow upper mantle, might cause the amagmatic spreading ridges (Harvey et al., 2006 EPSL, Liu et al.,2008 Nature). Its origin and effect on seafloor generations are, however, not well understood yet. We report an oceanic hill as an example of an ancient subduction-modified mantle domain, probably formed at continental margin of the Gondwanaland~Pangea supercontinent, existing beneath the Central Indian Ridge. This domain is the most likely to have prevented magmatic seafloor generation, resulting in creation of very deep oceanic valley and serpentine diaper (now the studied oceanic hill) at the present Central Indian ridge.