Acoustic measurement method of the volume flux of a seafloor hydrothermal plume

Measuring fluxes (volume, chemical, heat, etc.) of the deep sea hydrothermal vents has been a crucial but challenging task faced by the scientific community since the discovery of the vent systems. However, the great depths and complexities of the hydrothermal vents make traditional sampling methods laborious and almost daunting missions. Furthermore, the samples, in most cases both sparse in space and sporadic in time, are hardly enough to provide a result with moderate uncertainty. In September 2010, our Cabled Observatory Vent Imaging Sonar System (COVIS, http://vizlab.rutgers.edu/AcoustImag/covis.html) was connected to the Neptune Canada underwater ocean observatory network (http://www.neptunecanada.ca) at the Main Endeavour vent field on the Endeavour segment of the Juan de Fuca Ridge. During the experiment, the COVIS system produced 3D images of the buoyant plume discharged from the vent complex Grotto by measuring the back-scattering intensity of the acoustic signal. Building on the methodology developed in our previous work, the vertical flow velocity of the plume is estimated from the Doppler shift of the acoustic signal using geometric correction to compensate for the ambient horizontal currents. A Gaussian distribution curve is fitted to the horizontal back-scattering intensity profile to determine the back-scattering intensity at the boundary of the plume. Such a boundary value is used as the threshold in a window function for separating the plume from background signal. Finally, the volume flux is obtained by integrating the resulting 2D vertical velocity profile over the horizontal cross-section of the plume. In this presentation, we discuss preliminary results from the COVIS experiment. In addition, several alternative approaches are applied to determination of the accuracy of the estimated plume vertical velocity in the absence of direct measurements. First, the results from our previous experiment (conducted in 2000 at the same vent complex using a similar methodology but a different sonar system) provide references to the consistency of the methodology. Second, the vertical flow rate measurement made in 2007 at an adjacent vent complex (Dante) using a different acoustic method (acoustic scintillation) can serve as a first order estimation of the plume vertical velocity. Third, another first order estimation can be obtained by combining the plume bending angle with the horizontal current measured by a current meter array deployed to the north of the vent field. Finally, statistical techniques are used to quantify the errors due to the ambient noises, inherent uncertainties of the methodology, and the fluctuation of the plume structure.