Seamount morphology, distribution and structure of the Southwest Indian Ridge

Ultra-slow spreading ridges have a full spreading rate of less than 20 mm/yr, and show deviations from the fundamental characteristics identified at faster spreading rates; however, they are poorly studied compared to faster spreading ridges. In this thesis we aim to build on previous studies in order to further characterise the unique processes occurring at the ultra-slow spreading Southwest Indian ridge (SWIR). This is an exploratory study utilising novel and existing bathymetry, gravity, video imagery data to understand tectonic and volcanic processes along the ridge crest from triple junction to triple junction. We conduct a series of studies which focus on the axial valley morphology, large seamount morphology, the distribution of small seamounts and large seamounts on the ridge, and admittance studies. We show that the long wavelength trends in geophysical data, geochemical data, and axial valley morphology are well correlated. These variations reflect along-axis changes in the crustal and mantle structure, governed by the thermal structure of the ridge. Seamounts provide important morphological evidence of the volcanic and tectonic processes beneath the Earth's surface. We use detailed morphological investigations to show that volcanic and erosional processes on six large seamounts are controlled by the local faulting and melt migration mechanisms. We investigate the distribution of seamounts on the ridge using a numerical algorithm, and demonstrate a lower seamount density than faster spreading ridges. We show that the seamount population statistics vary along the ridge due to changes in the degree of partial melting, magma conduit availability, and melt focussing mechanisms; the locations of three proximal hotspots coincide with increases in the estimated magmatic flux. Finally, best fit elastic thicknesses are determined using a 3D windowed admittance technique, which was tested using synthetic data. We construct a combined convection-flexure model to account for misfits between the observed data and existing flexural theoretical models. The average elastic thickness varies between 7 km and 12 km, and we show a strong dependence on spreading rate.