technology can provide detections of fishes throughout the water column at high
spatial resolution, and precise estimates of acoustic target strengths from which to
infer sizes, it is not yet possible to infer species from the acoustic signatures alone.
This may pose a problem in diverse systems such as coral reefs, and particularly
when the objective for reef fish assessments is for the purpose of monitoring
species-specific abundances or responses to management measures. In contrast,
identifying and studying large, single-species aggregations of reef fish, such as reef
fish spawning aggregations can benefit from using split-beam echo sounders.
Large areas can be rapidly surveyed and analysis of split-beam data can provide
accurate estimates of density and abundance for large groups of fish, a task that can
be very difficult to accomplish using divers alone (Taylor et al. 2006).
The utility of these coral reef fish habitat maps to coral reef management are
several-fold. First, these maps provide a broad depiction of fish biomass distribution over a coral reef ecosystem, which will help identify regions of high and
low fish density and inform marine spatial planning and management. Second, in
regions which have not yet been surveyed visually for fish communities, the maps
can guide survey designs by identifying locations of relatively high or low fish
densities. Third, the maps can be used to help interpret fine-scale, and limitedextent, direct visual observations in a larger spatial context. Fourth, the survey
technique is rapid, repeatable and consistent, and can be conducted over a range of
spatial and temporal scales. As such, split-beam echo sounder surveys can also add
value to coral reef assessments by (1) tracking changes in fish biomass over time,
or (2) inferring movements and migrations of fishes over daily or seasonal time
scales. The value of these surveys could be particularly important when monitoring or assessing the efficacy of marine reserves or other spatial management
measures prior to and following implementation. Additionally, extending fish
habitat relationships through statistical and process-based modeling will guide
further interpretation and utility of these products.
9.3 State of the Science and Future Directions
In response to the growing need by resource managers for current, accurate and
consistent benthic habitat maps across a range of spatial scales, an international
group of scientists met to review the state of acoustic remote sensing (Anderson
et al. 2008). The group concluded that acoustic seabed classification using singlebeam, multi-beam, and sidescan systems is in its nascence and prioritized the top
ten issues in need of immediate and future attention by the international scientific
community to advance the utility of acoustic remote sensing for mapping marine
ecosystems. The main issues raised by Anderson et al. (2008) are recounted below,
including progress that has been made in the intervening years.
• Statistical versus interpretive classification: In the interest of advancing
repeatability of results, statistical classification is preferred over interpretive
246
G. Foster et al.
spatial resolution, and precise estimates of acoustic target strengths from which to
infer sizes, it is not yet possible to infer species from the acoustic signatures alone.
This may pose a problem in diverse systems such as coral reefs, and particularly
when the objective for reef fish assessments is for the purpose of monitoring
species-specific abundances or responses to management measures. In contrast,
identifying and studying large, single-species aggregations of reef fish, such as reef
fish spawning aggregations can benefit from using split-beam echo sounders.
Large areas can be rapidly surveyed and analysis of split-beam data can provide
accurate estimates of density and abundance for large groups of fish, a task that can
be very difficult to accomplish using divers alone (Taylor et al. 2006).
The utility of these coral reef fish habitat maps to coral reef management are
several-fold. First, these maps provide a broad depiction of fish biomass distribution over a coral reef ecosystem, which will help identify regions of high and
low fish density and inform marine spatial planning and management. Second, in
regions which have not yet been surveyed visually for fish communities, the maps
can guide survey designs by identifying locations of relatively high or low fish
densities. Third, the maps can be used to help interpret fine-scale, and limitedextent, direct visual observations in a larger spatial context. Fourth, the survey
technique is rapid, repeatable and consistent, and can be conducted over a range of
spatial and temporal scales. As such, split-beam echo sounder surveys can also add
value to coral reef assessments by (1) tracking changes in fish biomass over time,
or (2) inferring movements and migrations of fishes over daily or seasonal time
scales. The value of these surveys could be particularly important when monitoring or assessing the efficacy of marine reserves or other spatial management
measures prior to and following implementation. Additionally, extending fish
habitat relationships through statistical and process-based modeling will guide
further interpretation and utility of these products.
9.3 State of the Science and Future Directions
In response to the growing need by resource managers for current, accurate and
consistent benthic habitat maps across a range of spatial scales, an international
group of scientists met to review the state of acoustic remote sensing (Anderson
et al. 2008). The group concluded that acoustic seabed classification using singlebeam, multi-beam, and sidescan systems is in its nascence and prioritized the top
ten issues in need of immediate and future attention by the international scientific
community to advance the utility of acoustic remote sensing for mapping marine
ecosystems. The main issues raised by Anderson et al. (2008) are recounted below,
including progress that has been made in the intervening years.
• Statistical versus interpretive classification: In the interest of advancing
repeatability of results, statistical classification is preferred over interpretive
246
G. Foster et al.