Bruce C. Heezen Graduate Research Fellowship

Transcription

1 Bruce C. Heezen Graduate Research Fellowship G. M. Purdy Lamont Doherty Earth Observatory Palisades NY Phone: (845) fax: (845) J. C. Mutter Lamont Doherty Earth Observatory Palisades NY Phone: (845) fax: (845) Award Number: N LONG-TERM GOALS Provide the nation with the highest quality expertise in marine geology and geophysics research capabilities, specifically in areas that are of current interest to the Office of Naval Research. OBJECTIVES Select and support an outstanding graduate student carrying out research of interest to the Office of Naval research APPROACH The Bruce C. Heezen Graduate research Fellowship will be awarded to an outstanding graduate student whose research focus is marine geology and geophysics, and who is a U.S. citizen or permanent resident. The student was identified from among the Lamont student body, the criteria for selection of the successful candidate combining consideration of scientific excellence and the field of research, with emphasis upon fields of current interest to the Office of Naval research. WORK COMPLETED Kori Newman graduated from Smith College in May 2003 and has just completed her third year as a graduate student in Department of Earth and Environmental Sciences at Columbia University under the mentorship of Professor William Menke. This year, Kori has pursued two research projects, descriptions of which follow: Methane venting at giant seafloor pockmarks A series of giant (kilometer scale), en-echelon, elongate pockmarks are present along the US mid- Atlantic shelf break, south of the Norfolk Canyon. Gas was imaged in the subsurface in chirp seismic profiles (Hill et al., 2004), prompting the return to the pockmarks site to study in situ gas venting.

2 Using a METS methane sensor mounted on the Autonomous Underwater Vehicle (AUV) SeaBED, continuous, we made near seafloor measurements of dissolved methane concentration to determine if gas is actively venting at the pockmarks site. In the AUV data we observe a correlation between dissolved methane concentration, salinity and temperature. Using this observation were able to develop and algorithm to correct the time lag response of the METS sensor. We notice a distinct pattern in the corrected, near seafloor methane anomaly such that methane-rich water is concentrated along the upper walls of the pockmarks and on the shelf area surrounding the pockmarks, and not at the bottoms of the pockmarks. This year, geochemical analysis of the collected water samples was completed and the results have been incorporated into the analysis of the pockmarks dataset. Hydrocast samples further illustrate the distribution of the methane anomaly. Methane-rich water is generally located between the 80 m and 110 m water depth (Fig. 1, left panel), a depth range that corresponds to the top of the pockmarks walls. The previously observed correlation between methane concentration, salinity and temperature becomes less robust in the water column, likely due to pre-existing salinity and temperature gradients. The base of the methane-rich water mass coincides with a step in the salinity and temperature profiles. Sometimes a similar step is visible at the top of the water mass. The lateral extent of the methane-rich water mass was determined by looking for this step in conductivity, temperature and depth (CTD) profiles collected both during hydrocast sampling and during the AUV s descent and ascent. Most profiles display this step. However, two profiles located further away from the pockmarks walls lack the step. From this observation and from measurement of the currents by acoustic Doppler current profiling (ADCP), we conclude that methane is venting from the top of the pockmarks walls and possibly from the shelf area surrounding the pockmarks and, due to currents, is being advected into the surrounding area (Fig. 1, right panel). Sulfate and dissolved inorganic carbon (DIC) δ 13 C profiles were evaluated for core 23P and confirm that the source of the vented methane is in the underlying sediment. The sulfate profile approaches zero at 1.5 m below the seafloor. This very shallow sulfate reduction zone is evidence of methane advection. Furthermore, the δ 13 C of the DIC approaches -30 at the base of the profile, showing that anaerobic methane oxidation (AMO) must be occurring and thus that methane is present in the sediment pore fluid. Carbon isotopes reveal the vented methane is primarily biogenic, but whether it is fossil methane or is actively being produced is still unknown. Work on this project is nearly completed. Kori is preparing a manuscript for publication and plan on submitting it withn the next few weeks. Results from this project have been presented at the 2004 and 2005 Fall AGU Meetings, a workshop for New York state teachers and at the AUVs for Scientific Applications meeting at WHOI last June. Work on this project may continue as a collaborative effort with personnel from the Navel Research Laboratory who are scheduled to dive over the pockmarks site in early November. Seismic structure and evolution of oceanic crust along the Juan de Fuca Ridge In July-August 2002 multichannel seismic (MCS) data were collected along the Juan de Fuca Ridge (JdFR). Seismic layer 2A is imaged in this dataset. It thickens off axis in the southern section in a similar manner as at the fast-spreading East Pacific Rise (EPR) and the evolution of the layer s properties, up to 4-5 Ma, correlate with sediment burial (Carbotte et al., 2002). Moho is well imaged across most of the survey and evidence of sub-crustal melt sills was found (Nedimovic et al., 2005).

3 The results from this study have motivated us to attempt 2D waveform tomography on the same dataset to investigate the evolution of the oceanic crust generated along the JdFR. Using the velocity profiles determined during the processing of the MCS data, the velocity structure of the upper crust will be fixed during 2D travel time tomography to better resolve the deeper structure. The model resulting from this analysis will be the input model used in the waveform inversion. Since waveform tomography examines the entire waveform, the resulting image should be of better quality and higher resolution than the travel time model. Work on this project is in its early stages. Kori is currently being trained in the processing of 2D MCS data using Promax. Since the seismic lines have already been processed for the sediments and underlying crust, she is working on processing reflections in the water column to see if water column structure can be imaged in a similar way to that of Holbrook (2003). Once familiarity is gained with Promax she will begin work on the travel time and waveform tomography. Processing the water column MCS data yielded interesting results. The processed line is ~130 km long, beginning near the western flank of the Endeavour segment of the Juan de Fuca Ridge and extending eastward. The uppermost section of the water column is visible in the image because the direct water wave was successfully filtered out during processing. This distinguishing our image from previously published images where the upper ~200 m are obscured by the high amplitude, low frequency direct wave. In our first finalized section we image an entire eddy and find evidence that the presence of the ridge affects the water column structure (Fig. 2). The thermocline beneath the eddy extends down to a maximum depth of about 750 m and the eddy is bounded on top by a concave down reflector. This geometry suggests that the eddy could be an intrathermocline feature, formed around a lens of water that has intruded along the thermocline. The westernmost edge of the eddy coincides with the onset of the bathymetric expression of the ridge. The location of the eddy with respect to the ridge may be explained by observations by Deirdre Byrne (2000) who noticed that eddies whose paths intersect ridges will travel along the axis of the ridge and cross the ridge at a local bathymetric low. More lines are being processed to determine if similar features exist in these areas further south, and to examine what effect the topography of this spreading center has on circulation in the upper water column. Future Research: 3D seismic imaging of the internal structure of the magmatic-hydrothermal system at the East Pacific Rise The EPR is one of the fastest spreading centers on Earth and a highly dynamic magamatic, hydrothermal and biological system. An upcoming expedition to the ridge will collect 3D MCS data to create a geometrically accurate image of the magmatic-hydrothermal system of the ridge. In particular, the geometries of the axial magma chamber (AMC), the related magma system and its subsurface distribution and the porosity of fluid content of the magma bodies are of interest. In April and May 2007, Kori will participate in the cruise to the EPR. After the cruise she will work on processing the data using a prestack depth migration approach with the goal of producing one of the most accurate images of the ridge structure to date. Once processing is completed amplitude versus offset or angel of incidence (AVO/AVA) analysis will be performed to extract the physical properties of the deep, sub-amc part of the ridge system.

4 Other Accomplishments Kori presented the results of the pockmarks project in the form of a talk at this year s Fall AGU Meeting, for which she won an Outstanding Student Paper Award in the Ocean Sciences section. In addition, she passed the oral certifying examination at Columbia University this April. REFERENCES Byrne, D.A., From the Agulhas to the South Atlantic: Measuring Inter-ocean Fluxes, Ph. D. thesis., 181 pp., Columbia Univ., New York. Carbotte, S.M., R Detrick, G. Kent, J.P. Canales, J. Diebold, A. Harding, M. Nedimovic, D. Epstein, J. Cochran, E. Van Akren, J. Dingler and A. Jacobs (2002), A Mulit-Channel Seismic Investigation of Ridge Crest and Ridge Flank Structure Along the Juan de Fuca Ridge, Eox Trans. AGU, 83 (47), Fall Meet. Suppl., Abstract T72C-07. Hill, J.C., N.W. Driscoll, J.K. Weissel and J.A. Goff (2004), Large-scale elongate gas blowouts along the U.S. Atlantic margin, J. Geophys. Res., 109, B09101, doi: /2004jb Holbrook, W.S., P. Páramo, W. Pearse and R.W. Schmitt (2003), Thermohaline Fine Structure in an Oceanographic Front from Seismic Reflection Profiling, Science, 301 (8 August), Nedimovic, M.R., S.M. Carbotte, A.J. Harding, R.S. Detrick, J.P. Canales, J.B. Diebold, G.M. Kent, M. Tischer and J.M. Babcock (2005), Frozen magma lenses below the oceanic crust, Nature, 436 (25 August), , doi: /nature03944.

5 Figure 1. Left panel: Data from hydrocast 4. Salinity and temperature profiles are plotted as black and gray lines, respectively. Dissolved methane concentrations, measured in collected water samples are plotted as points. The base and top of the methane-rich water layer occur at depts. Corresponding to steps in the salinity and temperature profiles. Right panel: Schematic diagram of methane venting at the pockmarks. Depth scale is the same as in left panel, vertical exaggeration is approximately 5X. The bold line is the seafloor, the dashed line is the location of the hydrocast sampling, the shaded area represents the location of methanerich water and black arrows display the area of active methane venting. White arrows show the general water current patterns; a southerly current dominates with east-west motion occurring as a result of the tidal cycle. Methane-rich fluids are venting along the upper section of the pockmarks and are advected due to currents. No active venting is occurring at the base of the pockmarks.

6 Figure 2. A multichannel seismic (MCS) image of the water column structure across the Endeavour segment of the Juan de Fuca Ridge. The vertical scale is in depth below the sea surface and the horizontal scale is in CMPs, with a distance between CMPs of 6.25 m. The seafloor is located at a depth of ~2600 m. The imaged eddy is in the upper 750 m of water below the sea surface, with the edge of the structure located near the ridge. A uniform water mass may be imaged in the eddy between CMPs This would be the intruded lens of water around which an intrathermocline would form.