Melting processes in the mantle have a key role in plate tectonics and in the most colossal phenomena in the Earth, like large igneous provinces, mantle plume upwellings, and the constant growth of the planet's tectonic plates. In this study we use the geochemical and petrological evidence preserved in ophiolites, large igneous provinces, and intraplate seamounts to understand causes, timing and implications of melting in these different tectonic environments.

We studied melting at extensional environments, in mid-ocean ridges and back-arc basins, preserved in ophiolites. The Santa Elena Ophiolite in Costa Rica comprises a well-preserved fragment of the lithospheric mantle that formed along a paleo-spreading center. Petrological models of fractional crystallization suggest deep pressures of crystallization of >0.4 GPa for most of the samples, in good agreement with similar calculations from slow/ultra-slow spreading ridges and require a relatively hydrated (~0.5 wt% H2O) MORB-like source composition. Our findings suggest a complex interplay between oceanic basin and back-arc extension environments during the Santa Elena Ophiolite formation.

Secondly, we analyzed large igneous provinces and their mechanisms of formation. As the surface expression of deep mantle processes, it is essential to understand the time frames and geodynamics that trigger these massive lava outpourings and their impact to life in the planet. We analyze the record and timing of preserved fragments of the Pacific Ocean Large Igneous Provinces to reconstruct the history of mantle plume upwellings and their relation with a deep-rooted source like the Pacific Large Low Shear Velocity Province during the Mid-Jurassic to Upper Cretaceous.

Lastly, we explore the occurrence of low-volume seamounts unrelated to mantle plume upwellings and their geochemical modifications as they become recycled inside the mantle, to answer questions related to the nature and origin of upper mantle heterogeneities. We present evidence that an enriched mantle reservoir composed of recycled seamount materials can be formed in a shorter time period than ancient subducted oceanic crust, thought to be the forming agent of the HIMU mantle reservoir endmember. A "fast-forming" enriched reservoir could explain some of the enriched signatures commonly present in intraplate magmas not related with an active mantle plume upwelling.