Continuous near-bottom gravity measurements made with a BGM-3 gravimeter in DSV Alvin on the East Pacific Rise crest near 9°31′N and 9°50′N

A Bell BGM-3 gravimeter has been used to collect continuous, underway, near-bottom (3- to 10-m altitude) gravity measurements from the deep-diving submersible DSV Alvin during surveys on the East Pacific Rise (EPR) crest near 9 degrees 31'N and 9 degrees 50'N. Closely spaced (20- to 30-m) gravity measurements were made along transects up to 8 km long in both regions. Repeatability of measurements made at the same location on different dives is similar to 0.3 mGal. Along-track spatial resolution of anomalies is similar to 130-160 m, with the limiting factors being precision and sampling rate of the pressure gauge depth data used to calculate vertical accelerations of the submersible. The average upper crustal density of the ridge crest determined from the relationship between depth and free-water gravity anomalies varies greatly between 9 degrees 31'N and 9 degrees 50'N. Average upper crustal densities of 2410 kg/m(3) for the 9 degrees 50'N area and 2690 kg/m3 for the 9 degrees 31'N area were calculated. The different densities are not due to differing geometry of the Layer 2A-2B boundary or a regional cross-axis gravity gradient. Differences in porosity of the shallow crustal rocks, or a difference in the proportion of low density extrusives to higher-density dikes and sills within Layer 2A in these two areas, are the likely causes of the different upper crustal densities. Bouguer gravity anomalies near the EPR axis are primarily small amplitude (0.5-2 mGal), are a few hundred meters across, and appear to be lineated parallel to the axis. Larger-amplitude Bouguer anomalies of up to 4 mGal were found at a few locations across the crestal plateau and are associated with pillow ridges composed of lavas which are clearly younger than the surrounding seafloor. These ridges have distinct chemical compositions compared to lavas from the axial summit collapse trough (ASCT) at the same latitude. Probable sources of the 0.5- to 2-mGal anomalies observed on the summit plateau include areas of collapsed and fissured terrain and dike swarms feeding melt through Layer 2A to the surface. A grid survey of the ridge axis near 9 degrees 50'N shows Bouguer anomalies lineated along the axis, suggesting that dike swarms do contribute to the observed Bouguer anomalies. The along-axis continuity of the gravity anomalies is disrupted at a 75-m offset of the ASCT, suggesting that shallow feeders of lava to the surface may be segmented on a finer scale than the deeper crustal magmatic system. This initial study confirms the ability to conduct high-resolution, near-bottom, continuous gravity measurements from Alvin. It also provides important information on how the shallow crustal structure of a fast spreading mid-ocean ridge develops and how it varies with the surface morphology.