HR: 1340h
AN: B13A-0184    [Abstracts]
TI: Inferences of Particle Size and Composition From Video-like Images Based on Acoustic Data: Grotto Plume, Main Endeavor Field
AU: * Bemis, K G
EM: bemis@rci.rutgers.edu
AF: Institute of Marine and Coastal Sciences, Rutgers University, 71 Dudley Rd, New Brunswick, NJ 08901 United States
AU: Rona, P A
AF: Institute of Marine and Coastal Sciences, Rutgers University, 71 Dudley Rd, New Brunswick, NJ 08901 United States
AU: Santilli, K
AF: Center for Advanced Information Processing and the Department of Electrical and Computer Engineering, Rutgers University, Piscataway, NJ 08854 United States
AU: Dastur, J
AF: Center for Advanced Information Processing and the Department of Electrical and Computer Engineering, Rutgers University, Piscataway, NJ 08854 United States
AU: Silver, D
AF: Center for Advanced Information Processing and the Department of Electrical and Computer Engineering, Rutgers University, Piscataway, NJ 08854 United States
AB: Optical and acoustic scattering from particles in a seafloor hydrothermal plume can be related if the particle properties and scattering mechanisms are known. We assume Rayleigh backscattering of sound and Mie forward scattering of light. We then use the particle concentrations implicit in the observed acoustic backscatter intensity to recreate the optical image a camera would see given a particular lighting level. The motivation for this study is to discover what information on particle size and composition in the buoyant plume can be inferred from a comparison of the calculated optical images (based on acoustic data) with actual video images from the acoustic acquisition cruise and the IMAX film "Volcanoes of the Deep Sea" (Stephen Low Productions, Inc.). Because the geologists, biologists and oceanographers involved in the study of seafloor hydrothermal plumes all "see" plumes in different ways, an additional motivation is to create more realistic plume images from the acoustic data. By using visualization techniques, with realistic lighting models, we can convert the plume image from mechanical waves (sound) to electromagnetic waves (light). The resulting image depends on assumptions about the particle size distribution and composition. Conversion of the volume scattering coefficients from Rayleigh to Mie scattering is accomplished by an extinction scale factor that depends on the wavelengths of light and sound and on the average particle size. We also make an adjustment to the scattered light based on the particles reflectivity (albedo) and color. We present a series of images of acoustic data for Grotto Plume, Main Endeavour Field (within the Endeavour ISS Site) using both realistic lighting models and traditional visualization techniques to investigate the dependence of the images on assumptions about particle composition and size. Sensitivity analysis suggests that the visibility of the buoyant plume increases as the intensity of supplied light increases, the particle size decreases, and the particle reflectivity increases. However, decreasing the particle size (and thus increasing the extinction scale factor) results in a wider, less defined plume and increases the relative importance of the acoustic background noise; the best fit of our calculated optical images to the character of actual video images of the bottom few meters of the plumes (the acoustic data volume is 55 m tall) suggests that average particle size is fairly large ($\sim$1000 $\mu$m) in the buoyant plume. This suggests that existing data on particle size distributions underestimates the average particle size; the best explanation is the breakup of aggregates of particles during collection and filtering of water samples (no in situ measurements exist). We also investigate the effects of particle color on plume color by using models based on data collected by Feely et al (1987), Walker and Baker (1988), and Mottl and McConachy (1990). Highly reflective particles result in result in sharper-edged plumes suggesting that pyrite (albedo $\sim$0.6) and chalcopyrite (albedo $\sim$0.3) are the dominant particle compositions. This study shows that plume particles in the buoyant plume are probably larger than previously suspected and a predominance of pyrite and chalcopyrite is necessary to explain the high reflectance of black smoker plumes.
UR: http://www.caip.rutgers.edu/~kls93/realplume/index.html
DE: 4275 Remote sensing and electromagnetic processes (0689)
DE: 3015 Heat flow (benthic) and hydrothermal processes
DE: 3035 Midocean ridge processes
DE: 3094 Instruments and techniques
DE: 0669 Scattering and diffraction
SC: Biogeosciences [B]
MN: 2004 AGU Fall Meeting