|
Acoustic Remote Sensing of Hydrothermal Flow
Chris
Jones
cjones@apl.washington.edu
Darrell Jackson
drj@apl.washington.edu
Collaborators:
Peter
Rona
Institute
for Marine and Coastal Sciences
Rutgers University
New Brunswick, NJ 08901
H.
Paul Johnson
School of Oceanography
University of Washington
Seattle, WA 98195
Objectives:
Our present
day ability to study hydrothermal systems is severely limited by our
inability to detect and map flow fields in a systematic manner over
large areas of the seafloor. Exploration still plays a fundamental role
in the science of seafloor hydrothermal systems. Discoveries often happen
by chance rather than by systematic surveying and remote sensing. Recent
discoveries of a new type of 'off-axis' vent system by Kelley et al.
2001 highlight the need for new instruments that are capable of exploring
the new areas of the seafloor and detecting both high and low temperature
venting.
Once a
field has been located, our ability to model hydrothermal systems and
their relation to tectonic, magmatic, oceanographic, and biological
processes is limited by the lack of tools for resolving critical spatial
and temporal scales of flow. For example, our ability to estimate heat
flux from an active ridge segment is limited by our inability to map
distributions of localized and diffusive sources of heat over length
scales that are characteristic of ridge tectonic processes. Observing
the temporal/spatial variability and partitioning of energy between
different types of flow is critical for understanding fluid circulation
in the crust, its interaction with crustal alterations, and its interaction
with biological habitat. The ability to model hydrothermal fluid circulation
after it has left the seafloor and the entrainment of surrounding water
into hydrothermally induced flow is limited by our inability to characterize
flow at the scale of the entire plume within sub-tidal time scales.
Point measurements made from a moving ROV or AUV, for example, are often
subject to variability associated with tidal cycles. Interpretation
of such aliased measurements requires well-defined forward models of
tidal flow and its interaction with complicated topography.
High-frequency
acoustic remote sensing offers an attractive method of detecting and
probing scales of hydrothermal flow that are unattainable by point sampling
methods. Two new methods of detecting and characterizing flow are being
developed: 1) scintillation thermography to detect and characterize
diffuse flow fields; and 2) plume particulate scattering to estimate
flow velocity and particulate concentrations in the high temperature
vent plumes.
Video
Clips of Flow:
Doppler
Measurements of Black Smoker Flow Velocity:
Photo of SM2000 sonar and rotator (elevation)
mechanism mounted on JASON |
As
part of the VIP2000 (Vent Imaging, Pacific) measurement effort
on the Endeavour segment of the Juan de Fuca Ridge, the feasibility
of using a multibeam sonar (Kongsberg-Simrad-Mesotech SM2000)
to determine vent flow velocities was investigated. The sonar,
operating at 200 kHz, was mounted on JASON, which maintained a
stable position on the seafloor. The sonar was pointed to provide
image slices of several different black smoker plumes. In order
to permit coherent Doppler processing, the sonar was set to transmit
two-ping bursts, and the phase difference of the resulting echo
pair was used to determine the component of velocity along the
sonar line of sight ("radial" velocity). This work was done in
collaboration with Peter Rona at Rutgers University.
|
|
Sonar
Geometry:
Electronic
beamforming provides a "slice" through the plume at each elevation
step. Each slice consists of 128 beams, 1.5° wide, spanning
a 120° sector. To produce a 3-D image of the plume, the 2-D
slices are mechanically scanned with 1° steps in elevation.
|
|
Coherent
Doppler Processing Algorithm:
The
backscattered signals from two consecutive pings are divided into
range bins along the axis of each beam. The velocity within each
range bin is found by estimating the phase of the cross correlation
between the pings. To avoid aliasing (when f>2π), phase unwrapping
is performed. First, the region of scattering from the plume is
isolated for processing (the area outlined in white). The peak amplitude
within this area is then located, and the phase is unwrapped along
the axis of the beam containing the peak (in the direction of the
arrows along axis 1). Phase is then unwrapped across the beams for
each range bin (in the direction of axis 2, moving away from axis
1 in azimuth). Multiples of 2π are added to the phase so that
it approaches zero at the edges of the processing area. The radial
velocity within each range bin is then computed as:
|
v
= vmax p / π,
where:
vmax = c/(4ft) =1.85 cm/s
is the aliasing velocity
with:
c
= sound speed,
t = time between
pings = 0.1 s
p = phase
f = acoustic frequency =
200 kHz
|
 |
Fluid
Velocity of Hydrothermal Plume:
3-D
Reconstruction of Fluid Velocity of Hydrothermal Plume:
Slices
of plume velocity are combined to reconstruct a 3-D grid of "radial"
velocity. The
3-D image consists of slices recorded over a period of 4 minutes.
Isosurfaces of velocity are illustrated for a plume at Grotto,
Main Endeavour Field, in July 2000.
|
|
Simulation
to Test Coherent Doppler Algorithm:
|
Point acoustic scatterers are randomly distributed to simulate
simple plume structure.
Points
are assigned vertical velocities with velocity decreasing outwardly
from core.
Two
acoustic "snapshots" are generated with points in two slightly
different positions. |
Example of Proper Functioning of Phase
Unwrapping:
Example
of Partial Failure of Phase Unwrapping:
The
simulated velocity measurement shows streaks of obviously improper
values due to phase jumps of 2π
|
ACOUSTIC SCINTILLATION THERMOGRAPHY:
Acoustic
scintillation thermography (AST) is a recently developed method for
detecting and mapping hot water at the seafloor. Diffuse flow is typically
difficult to detect because fluid velocities and temperature are low,
suspended particulate matter is absent, and discharge occurs in small
irregular patches. AST exploits the same principle as the human eye
for detecting hot water - the scintillation of a wave (acoustical or
optical) as it passes through a turbulent flow field and scatters off
the underlying seafloor. The use of acoustics, however, allows the extension
to ranges unattainable by optics. The AST technique uses the echo-to-echo
decorrelation of the bottom backscattered signal from consecutive scans
of the seafloor to detect weak fluctuation in the index of refraction
of the water near the seafloor. The temporal changes in the water (i.e.,
turbulent mixing) cause slight changes in the integrated path lengths
of an acoustic ray as it propagates back and forth between the receiver
and the seafloor creating a measurable phase-coherent decorrelation
of the scattered signal between pings. For diffuse flow, where the plume
is generally concentrated near the seafloor, this decorrelation is a
measure of the temperature and velocity fluctuations in the near bottom
boundary layer, providing an extremely sensitive and robust detection
tool for identifying areas of flow (Jones et al. 2000, Rona et al. 1997).
Several recent experiments have shown that diffuse flow can be systematically
mapped over large (kilometer scale) areas using the AST method on an
ROV platform.
Thermal
Grid Project:
NSF
OCE-9911523, Direct and indirect measurement of the thermal budget of
two large hydrothermal systems on the Juan de Fuca, January 2000 - December
2002,
PI's: H.P. Johnson and S.L. Hautula (Univ. of Washington, Oceanography)
, C.D. Jones (Univ. of Washington, Applied Physics Laboratory)
During the summers of 2000 and 2001 systematic measurements of heat
flux from a segment of the Juan de Fuca Ridge between the Main Endeavour
field and High Rise Field were made using a recently developed acoustic
survey method (AST) and multiple ground truth point measurements of
temperature and velocity in areas of diffuse flow. An areas of the ridge
segment approximately 2500 meters along the axis of the ridge and 250
meters across axis was successfully mapped using the Simrad SM2000 sonar
mounted on the ROV Jason. A small area of the AST survey at Clam Bed
diffuse flow field is shown in Figure 2. During this field program,
the method of using acoustic remote sensing to detect diffuse flow from
a moving ROV to map relatively large areas of the seafloor was further
developed. Thermister array devices and acoustic flow meters (MAVs)
were used to make systematic point measurement of selected diffuse flow
areas. Measurements were made to access the fine-scale spatial variability
of temperature within a diffuse flow patch, the turbulent boundary layer
thickness associated with diffuse flow, and measurements of the effects
of tidal currents on the boundary layer. In the second and third year
of this study, the acoustic measurements will be combined with the ground
truth measurements to estimate heat flow from low temperature diffuse
flow within the surveyed area of the ridge segment and to produce a
map of the spatial distribution of diffuse flow areas.
PUBLICATIONS:
AGU
2000 Fall Meeting, Presentation No. 0561C-03
|
