Tristan da Cunha is a volcanic island in the South Atlantic located very close to the Mid-Atlantic Ridge. Generally, it is accepted to be the location of a mantle plume, which has been active at least since the breakup of Gondwana at 130 Ma, the time when the Paraná/Etendeka flood basalts were emplaced. Furthermore, it is associated with the formation of the Walvis Ridge and the Rio Grande Rise, and therefore it’s one of the key examples of a hot spot track linking a flood basalt province to an active ocean island volcano. However, global tomography models are contradicting about the origin of Tristan da Cunha: Whether it is a deep mantle plume or caused by shallow plate tectonics. To gain a better understanding, we deployed 24 broadband ocean-bottom seismometers, 26 ocean-bottom electromagnetic stations and 2 seismological land stations in January 2012 with the German research vessel Maria S. Merian. We acquired continuous seismological data for one year and recovered the instruments in January 2013. We use cross-correlated travel time residuals of teleseismic earthquakes to perform a finite-frequency tomography to resolve the P wave velocity upper mantle structure beneath the island. Here we show our preliminary results of the 3-D velocity perturbations in the upper mantle: We do not image a plume-like structure directly beneath the island. Instead we observe a low velocity region in the southwest of our array that might be related to a local mantle upwelling (mantle plume). Additionally we show the local seismicity in the Tristan da Cunha region. Chen et al. and Baba et al. will present the first results on the magnetotelluric experiment and Ryberg et al. will present the crustal structure around the Tristan da Cunha hotspot.