This study uses gravity data to investigate crustal structure in the region of the Galapagos triple junction where the westward propagating Cocos-Nazca rift (CNR) approaches the East Pacific Rise (EPR) and forms the northern boundary of the Galapagos micro plate. Shipboard and global gravity data are analyzed from 104˚W to 96˚W and 0˚ to 4˚N. In May 2018, the high-resolution gravity data were collected along ship tracks that run across the entire width of the Galapagos gore from the tip of the CNR at ~101.7˚W to 98.5˚W. Residual mantle Bouguer anomaly (RMBA) was calculated by removing the effects of water-crust, crust-mantle, and lithospheric cooling from the free-air anomaly (FAA). We also calculated a model of gravity-derived crustal thickness by downward continuation of the RMBA, as well as a model of non-isostatic topography by removing the topographic effects of thermal subsidence and crustal thickness variations. The results reveal several distinctive features in gravity and crustal structure: (1) The eastern flank of the EPR has systematic shallower topography and more negative RMBA than the conjugate western flank, reflecting regional density variations. (2) On the eastern flank of the EPR, the region south of the Galapagos gore is associated with more negative RMBA than the conjugate region to the north, possibly reflecting closer proximity to the Galapagos hotspot in the southern region. (3) The first ~100 km behind the propagating CNR tip (~101.7˚W to 100.8˚W) is associated with more positive RMBA (up to ~35 mGal) than the CNR rift between ~100.8˚W and 98.5˚W, suggesting locally thinner crust (up to ~1.5 – 2 km). East of 98.5˚W along the CNR, RMBA decreases gradually towards the Galapagos hotspot. (4) A region of local high topography on the southern boundary of the Galapagos microplate, where fresh basalts were sampled, is associated with negative RMBA centered at ~101.6˚W and 1.3˚N, indicating local relatively thick crust. (5) Within our study area, the CNR crust shows shallower average off-axis topography and more negative average RMBA than the EPR crust of corresponding age, which is consistent with a model of isostatic compensation of average thicker CNR crust than the surrounding EPR crust, possibly reflecting Galapagos hotspot effects.

The equatorial region of the slow-spreading Mid-Atlantic Ridge is characterized by several major transform faults, which are some of the longest on Earth. Among them, the St. Paul Transform system (SPTS) is a complex group of four transform faults, bounding three short intra-transform segments with total offset of 630 km. The northernmost transform is the 200 km-long, 30 km-wide Atoba Ridge, which represents a major topographic feature that rises above sea level at the St. Peter and St. Paul islands (SPSPA). This push-up ridge formed from transpressive stresses along several transform fault step-overs and restraining bends, uplifting mantle rocks at a rate of ~1.5 mm/yr. Moderate-sized earthquakes (>4.0 Mw) have been located by global teleseismic networks along the SPTS and near region. These earthquakes are recorded at large epicentral distances, and include raypaths that travel within the upper mantle (Pn and Sn phases). Pn velocity estimates can help to understand the dynamics of upper mantle structure around of the transform faults. Here, waveforms recorded over ~6 months of 2012 by two autonomous hydrophones moored north and south of SPTS (EA-2 and EA-8), and a seismographic station installed on SPSPA island (ASPSP station) are examined. These data allow us to make Pn velocity estimates from 32 earthquakes that occurred in the SPTS region from 1.5º S to 4.5º N. Pn wave velocities are typically thought to be 8.0–8.2 km/s in upper mantle, however we identify Pn velocities ranging from 7.5 to 9.0 km/s. The slower velocities (7.5-8.0 km/s) are from ray paths oriented parallel to the ridge axis and could be explained by elevated mantle potential temperature and the presence of melt. Ray paths passing through the transform fault system have Pn velocities from 8.1 to 9.0 km/s, indicating that upper mantle conditions are strongly affected by the presence of the crustal fault system. We will also compare our velocity estimates to global shear-wave tomographic models of the upper mantle. Hence it is our goal to show that the availability of autonomous hydrophones and a single island seismic station can be used to make rare estimates of Pn velocities, as well as provide insights into upper mantle structure, in this remote part of the Atlantic Ocean.