The UPFLOW experiment: peeking from the sea floor to the deep mantle with a ~1,500 km aperture array of 49 ocean bottom seismometers in the mid-Atlantic

Deep upward mantle flow is key to bring volatiles to the atmosphere and to produce Earth’s largest melting events extending through hundreds of thousands of kilometres, which coincide with major extinctions and changes in the geodynamo. Without knowing upward flow, we cannot understand global mantle flow and directly link the Earth’s interior with the surface. Yet, while downward flow (at subductions zones) is well constrained by seismology, plume-like mantle upwellings that connect the deepest mantle to the surface are poorly understood. The goal of the UPFLOW project (https://upflow-eu.github.io/) is to develop new high-resolution seismic imaging approaches along with new data collection, and to use them to constrain upward flow in unprecedented detail. We conducted a large amphibian experiment in the Azores-Madeira-Canary Islands region, which is a unique natural laboratory with multiple upwellings that are poorly understood in general. UPFLOW deployed 50 and recovered 49 ocean bottom seismometers (OBSs) in a ~1,000×2,000 km² area in the Azores-Madeira-Canary Islands region starting in July 2021 for ~13 months, with an average station spacing of ~150-200 km. These data will be combined with land data from over 30 seismic stations in nearby islands. The seismic deployment and recovery involved institutions from five different countries: Portugal (IPMA, IDL, Univ. of Lisbon, ISEL), Ireland (DIAS), UK (UCL), Spain (ROA) and Germany (Potsdam University, GFZ, GEOMAR, AWI). In addition to its scientific component, the experiment also had a substantial science outreach component, with participation from children from primary schools in the UK, Germany, Portugal, Spain and Ireland, facilitated by UCL’s GeoBus project. We discuss some of the activities conducted, such as the children’s naming of the OBSs, their drawings on how they imagined the OBSs, their stories on what may have happened to the OBSs during the experiment and zoom sessions including live calls to the ship. Names chosen included Neptune, Triton, Jelly, Caesar and Thor. The deployment (28 days offshore) and recovery (35 days offshore) of the OBS instruments was carried out during two dedicated expeditions with IPMA’s research vessel NI Mário Ruivo with the strong support of its skilled and highly motivated crew. 32 OBSs were loaned from the DEPAS international pool of instruments maintained by the Alfred Wegener Institute (Bremerhaven), Germany, while other institutions borrowed additional instruments (7 from DIAS, 4 from IDL, 3 from ROA, 4 from GEOMAR). Most of the instruments have three-component wideband seismic sensors and hydrophones, but three different designs of OBS frames were used. Initial data analysis shows high-quality data, notably a substantial decrease in noise levels in the vertical component long-period data (T>~30s). We show illustrative recordings of teleseismic events, a local seismic swarm, and long-period seismic signals as well as of non-seismic signals such as whales vocalisations and ships’ noise. We discuss the lessons learned from our international collaborative expedition, as well as possible future directions.