9.2.4 Split-Beam Application
Coral reef resource users and managers have need for highly resolved and highly
detailed maps of the benthic habitats; however, management mandates also
include the inhabitants (i.e., the fishes and invertebrates) that occupy the habitats.
Surveys of inhabitants of reef communities have relied on direct visual observations at fine spatial scale using scuba divers, remotely operated vehicles (ROVs),
drop cameras, or extractive methods such as traps and nets. These surveys provide
highly detailed data on species composition, but can be costly and are limited in
maximum depths, environmental conditions (e.g., sea conditions, light levels,
visibility), and overall extent. Additionally, high spatial variation in fish densities
or community assemblages can result in difficulties in capturing trends or spatial
patterns from which to interpret impacts from natural or anthropogenic impacts.
Adding split-beam echo sounder surveys to existing coral reef habitat survey
platforms is a modest investment that will provide significant payoff in terms of
adding value and interpretive power to habitat maps and related products from
coral reef benthic habitat maps.
Similar to commercially available bottom- and fish-finders, scientific split-beam
echo sounders can be used to detect fishes in the water column and near-bottom at
high vertical and horizontal resolution. Unlike most commercial echo sounders,
however, these scientific digital echo sounders have the capability of acquiring and
storing digitized echo return data for later analysis. In moderate depths (\100 m),
short pulse lengths (0.1–0.3 ms) can result in vertical resolutions of\20 cm, while
high pulse repetition (ca. 5–10 Hz) typically results in numerous echo returns as
fish pass through the acoustic beam.
Target tracking algorithms accumulate repeated echo returns from individual
fish, from which attributes are calculated for each target (Fig. 9.10). Individual fish
are identified and attributed a target strength based on the intensity of the returning
echo, which can then be converted to length using a generalized relationship.
Position of individual fish targets include the range from transducer, based on the
time delay of the acoustic echo return, relative horizontal position within the
acoustic beam, determined from phase differencing in the split-beam quadrants,
and a geographic coordinate from the ship’s positioning system.
When fish are in dense schools or aggregations, individual fish tracks are
indiscernible. Instead, the total acoustic energy that is returned from the insonification of the school is assumed to represent the sum of the acoustic energies of
the individual fish. In this way, the acoustic energy is integrated over the school
and the density of fish is estimated following theories of echo-integration (Simmonds and MacLennan 2005). Additional metrics can be derived from fish schools
and aggregations, such as size, spatial structure and average acoustic energy
return. Fish density is calculated for a discrete segment by weighting individual
fish by their position within the acoustic beam, accounting for the higher probability of detecting a fish at greater range from the transducer as the beam becomes
242
G. Foster et al.
Coral reef resource users and managers have need for highly resolved and highly
detailed maps of the benthic habitats; however, management mandates also
include the inhabitants (i.e., the fishes and invertebrates) that occupy the habitats.
Surveys of inhabitants of reef communities have relied on direct visual observations at fine spatial scale using scuba divers, remotely operated vehicles (ROVs),
drop cameras, or extractive methods such as traps and nets. These surveys provide
highly detailed data on species composition, but can be costly and are limited in
maximum depths, environmental conditions (e.g., sea conditions, light levels,
visibility), and overall extent. Additionally, high spatial variation in fish densities
or community assemblages can result in difficulties in capturing trends or spatial
patterns from which to interpret impacts from natural or anthropogenic impacts.
Adding split-beam echo sounder surveys to existing coral reef habitat survey
platforms is a modest investment that will provide significant payoff in terms of
adding value and interpretive power to habitat maps and related products from
coral reef benthic habitat maps.
Similar to commercially available bottom- and fish-finders, scientific split-beam
echo sounders can be used to detect fishes in the water column and near-bottom at
high vertical and horizontal resolution. Unlike most commercial echo sounders,
however, these scientific digital echo sounders have the capability of acquiring and
storing digitized echo return data for later analysis. In moderate depths (\100 m),
short pulse lengths (0.1–0.3 ms) can result in vertical resolutions of\20 cm, while
high pulse repetition (ca. 5–10 Hz) typically results in numerous echo returns as
fish pass through the acoustic beam.
Target tracking algorithms accumulate repeated echo returns from individual
fish, from which attributes are calculated for each target (Fig. 9.10). Individual fish
are identified and attributed a target strength based on the intensity of the returning
echo, which can then be converted to length using a generalized relationship.
Position of individual fish targets include the range from transducer, based on the
time delay of the acoustic echo return, relative horizontal position within the
acoustic beam, determined from phase differencing in the split-beam quadrants,
and a geographic coordinate from the ship’s positioning system.
When fish are in dense schools or aggregations, individual fish tracks are
indiscernible. Instead, the total acoustic energy that is returned from the insonification of the school is assumed to represent the sum of the acoustic energies of
the individual fish. In this way, the acoustic energy is integrated over the school
and the density of fish is estimated following theories of echo-integration (Simmonds and MacLennan 2005). Additional metrics can be derived from fish schools
and aggregations, such as size, spatial structure and average acoustic energy
return. Fish density is calculated for a discrete segment by weighting individual
fish by their position within the acoustic beam, accounting for the higher probability of detecting a fish at greater range from the transducer as the beam becomes
242
G. Foster et al.