The Geological Setting of Hydrothermal Vent Sites on Gakkel Ridge

In 1998 and 1999, the Science Ice Exercises (SCICEX) mapped the fine-scale textures of the flanks and axial valley of Gakkel Ridge, the slowest-spreading mid-ocean ridge on Earth (full-spreading rates <1.33 cm/yr). Sidescan data collected during the SCICEX expeditions showed the locations and distribution of lightly sedimented volcanic flows and faults including a small volcano near 85°N, 85°ºE associated with >250 teleseismic events that occurred in 1999 [Müller and Jokat, 1999; Edwards et al., 2001]. During the 2001 Arctic Mid-Ocean Ridge Expedition (AMORE), hydrothermal plume reconnaissance conducted during rock sampling operations revealed evidence of abundant hydrothermal venting on the Gakkel Ridge [Edmonds et al., 2003]. Comparison of the plume distributions with multibeam bathymetry data collected during AMORE showed that vent plumes were closely associated with topographic highs located inside the axial valley, with the largest and highest-temperature plume coinciding with the 85°E volcano. We describe the geological setting of vent plumes discussed in Edmonds et al. [2003] by integrating water column information from the AMORE program with detailed textural data from the SCICEX surveys and develop predictions for locations where hydrothermal venting is likely to occur on ultra-slow spreading mid-ocean ridges. Our efforts focus on five hydrothermal sites identified by Edmonds et al. [2003] and Baker et al. [2004] (7.5°E, 37°°E, 43°E, 55°E and 85°E). We co-register the observed plume distributions with interpretative maps showing the locations of tectonic and volcanic features such as faults and reflective lava flows in order to characterize the hydrothermal sites. These results are compared with similar interpretative products for the 69°E region and other sites where plume signals were observed but the hydrothermal activity could not be localized based on the strength of the hydrothermal signals or the occurrence of "near-field" signatures such as distinct layering or temperature anomalies. Finally, we examine reflective regions not associated with plumes, either because no hydrothermal anomalies were detected or the regions were not hydrographically surveyed during the AMORE program. Comparison of the different classes of vent sites are used to produce an improved model for predicting where hydrothermal venting might be observed during future Gakkel Ridge expeditions.