Intellectual Merits: The Gakkel Ridge provides rich research opportunities as the slowest spreading end member of the global system of ocean ridges, with the spreading rate decreasing progressively towards the east. As the spreading rate decreases, the thickness of the overlying lithosphere increases. This phenomenon permits examination and possibly resolution of two major questions pertinent to the formation of the ocean crust. First, is what the relative importance of lithospheric thickness and mantle temperature on the genesis of ocean ridge basalts? If ocean ridge basalts record mantle temperature, they will provide a valuable record of current and past mantle temperature variations. Second, are ocean ridge basalt isotope and trace element compositions controlled by preferential sampling of a "veined mantle" component at small extents of melting? Resolution of this question has implications for interpretations of the entire trace element and isotope record of oceanic volcanics. The Arctic Mid-Ocean Ridge Expedition (AMORE) in 2001 produced the first high resolution map of the ridge and basement rocks from over 200 stations. Preliminary analyses for major elements, trace elements, and strontium, neodymium, and lead isotopes demonstrate the importance of an unexpected discovery-the existence of a mantle domain boundary that occurs in a "sparsely magmatic zone" part way along the ridge, where magmatism drops to zero and peridotites are emplaced at the spreading axis. Samples to the west are similar to the Indian Ocean, while those to the east are similar to the Pacific. Despite these complexities, the predicted signal of low extents of melting in the eastern region is clear in both major and trace elements. Coherent data sets on the large suite of samples available will permit a separation of variables-the relative roles of mantle composition, spreading rate, and mantle temperature. The Principal Investigators will undertake a major analytical and modeling program: to complete the geochemical program to precisely define the location and sharpness of the mantle domain boundary, to model the melting process using existing and new major and trace element data, to model the trace element and isotope evolution of both domains, and to make detailed comparisons of the Arctic Basin signal to other ocean basins. The new data and modeling will permit a clear comparison between the effects of increasing lithospheric thickness and the global variations observed elsewhere. Are they distinguishable, and do these differences accord with model predictions? Important new approaches will include using hafnium isotopes to evaluate the importance of garnet in the melting process, and laser ablation ICP-MS analysis of melt inclusions to gain insights into processes of melting and melt segregation and how they vary along the ridge. The new maps and well located samples also give the opportunity to test models of magmatic segmentation, and whether clearly defined and isolated volcanic centers are the result of mantle heterogeneity, focusing of mantle flow, or melt focusing of uniform flow. This work will continue the collaboration with Dr. Peter Michael, University of Tulsa, who will also do analytical work on volatile elements, with Dr. David Graham, Oregon State University, who is measuring noble gasses, and with colleagues working on the peridotite samples recovered from this region. Broader Implications: This work addresses two of the most significant issues pertinent to the ocean crust and mantle-the distribution of mantle temperature and the nature of mantle heterogeneity, which have broad importance across many disciplines. It has had, and will continue, to have an important component of education and outreach. Students were involved in the sea-going expedition, the work will involve participation by both graduate and undergraduate students at two institutions, and there will be a series of public lectures to expose more of the community to the historic discoveries made possible by the new U. S. ice-breaking capability.
