Multi-frequency Acoustic Data Research Articles

Fine-scale vertical relationships between environmental conditions and sound scattering layers in the Southwestern Tropical Atlantic.

Ocean dynamics initiate the structure of nutrient income driving primary producers, and these, in turn, shape the distribution of subsequent trophic levels until the whole pelagic community reflects the physicochemical structure of the ocean. Despite the importance of bottom-up structuring in pelagic ecosystems, fine-scale studies of biophysical interactions along depth are scarce and challenging. To improve our understanding of such relationships, we analyzed the vertical structure of key oceanographic variables along with the distribution of acoustic biomass from multi-frequency acoustic data (38, 70, and 120 kHz) as a reference for pelagic fauna. In addition, we took advantage of species distribution databases collected at the same time to provide further interpretation. The study was performed in the Southwestern Tropical Atlantic of northeast Brazil in spring 2015 and autumn 2017, periods representative of canonical spring and autumn conditions in terms of thermohaline structure and current dynamics. We show that chlorophyll-a, oxygen, current, and stratification are important drivers for the distribution of sound scattering biota but that their relative importance depends on the area, the depth range, and the diel cycle. Prominent sound scattering layers (SSLs) in the epipelagic layer were associated with strong stratification and subsurface chlorophyll-a maximum. In areas where chlorophyll-a maxima were deeper than the peak of stratifications, SSLs were more correlated with stratification than subsurface chlorophyll maxima. Dissolved oxygen seems to be a driver in locations where lower oxygen concentration occurs in the subsurface. Finally, our results suggest that organisms seem to avoid strong currents core. However, future works are needed to better understand the role of currents on the vertical distribution of organisms.

Multiple sampling methods to develop indices of mid‐trophic levels abundance in open ocean ecosystems

AbstractMid‐trophic level (MTL) organisms play a key role in the sub‐Antarctic ecosystem food web, linking primary producers and tertiary consumers transferring energy across trophic levels. In this region, the relative abundance of MTLs has been monitored using single‐frequency acoustic data collected opportunistically in a time series of trawl surveys carried out in the austral late spring/early summer. To advance this approach, we developed a methodology that enabled us to disaggregate multifrequency acoustic data into three different MTL groups to assess their temporal and spatial patterns: mesopelagic fish, gelatinous zooplankton (excluding pyrosomes), and euphausiids. We used acoustic data and biological sampling information collected with two midwater trawl nets from a dedicated voyage, to train a classification tree algorithm. We complemented these data with footage from an in‐trawl camera system that gave us new insights on vertical distribution and presence of pelagic taxa. The testing dataset used to implement our classification algorithm was a time series of acoustic data from the sub‐Antarctic trawl survey. This approach was successful at disaggregating acoustic data and provided useful information to study the relative abundance and temporal and spatial patterns of pelagic taxa. Our analysis revealed the presence of dense shallow layers of gelatinous zooplankton which showed a statistically significant increasing trend over the time series. Although we were unable to detect pyrosomes acoustically, we encountered large biovolumes of these organisms during our biological sampling, which was supported by the in‐trawl camera system.

Decadal community structure shifts with cold pool variability in the eastern Bering Sea shelf.

A characteristic feature of the eastern Bering Sea (EBS) is a subsurface layer linked to seasonal sea ice (SSI) and defined by bottom temperatures less than 2 °C, which is termed the cold pool. Cold pool variability is directly tied to regional zooplankton and fish dynamics. Multifrequency (200 and 460 kHz) acoustic backscatter data were collected remotely using upward looking echosounders along the EBS shelf from 2008 and 2018 and used as a proxy of biological abundance. Acoustic data were coupled with bottom temperature and regional SSI data from the cold (2006-2013) and warm (2014-2018) regimes to assess the relationship between biological scattering communities and cold pool variation. Acoustic backscatter was 2 orders of magnitude greater during the cold regime than during the warm regime, with multifrequency analysis indicating a shift in the warm regime frequency-dependent scattering communities. Cold pool proxy SSI was a stronger predictor for biological scattering than bottom temperature in the cold regime, while warm regime bottom temperature and SSI were equal in predictive power and resulted in improved predictive model performance. Results suggest coupled cold pool and frequency-dependent scattering dynamics are a potential regime shift indicator and may be useful for management practices in surrounding Arctic ecosystems.

Finite-amplitude sound propagation effects in volume backscattering measurements for fish abundance estimation

Scientific advice for governmental management of marine resources relies on acoustical observation methods. Quantification and identification of fish and plankton species are often achieved using multi-frequency acoustic data. Accurate measurements of backscattering cross-sections and volume backscattering coefficients are essential. Systematic errors from finite-amplitude sound propagation are demonstrated in experimental survey measurements on Atlantic mackerel using 120 kHz and 200 kHz echosounders and high power settings. Finite-amplitude signal distortion causes excess transmission loss that is not accounted for in fisheries acoustics today, other than by fixed limits on the maximum transmitted power. The demonstrated errors are of a magnitude that can seriously bias abundance estimation and species identification. It is shown how the finite-amplitude effects can be modelled and predicted quantitatively, within a framework of electroacoustic power budget equations. A method is provided to calculate related errors in echosounder calibration and oceanic measurement of acoustic volume backscattering. When accounting for finite-amplitude effects in echosounder signal processing, higher transmit powers can be used when needed to improve signal-to-noise ratio or extend measurement range. The results indicate that historical survey data can be adjusted for such errors using numerical simulations. The echosounder characteristics relevant to finite-amplitude effects can be determined by laboratory measurements.

Acoustic Assessment of the Micronekton Community on the Chatham Rise, New Zealand, Using a Semi-Automated Approach

Micronekton are a key component of the pelagic food web of the Chatham Rise east of New Zealand. The Chatham Rise is an important fishing area for hoki (Macruronus novaezelandiae), New Zealand’s largest finfish fishery, and a predator on mesopelagic fish. Four fisheries oceanographic voyages provided multi-frequency acoustic data (18, 38, 70 120 and 200 kHz) and midwater trawls, which were used to define a classification tree to separate micronektonic organisms. We carried out validation and sensitivity analyses that showed that we were able to classify pearlside (Maurolicus australis) and euphausiids. Other mesopelagic targets (mainly myctophids) were classified together based on their acoustic frequency response. Using scripting in the open-source software ESP3, we applied our classification tree to an independent time series of acoustic data from trawl surveys on the Chatham Rise between 2009 and 2018, that was not used for model development or validation. Our methodology allowed us to study temporal and spatial patterns of M. australis, euphausiids, and total backscatter in the water column. Total backscatter associated with micronekton has varied over the last ten years, with no clear trend. The abundance of euphausiids showed a significant decreasing trend over the last ten years. Abundance of M. australis also decreased since 2012, though this was not significant. This work contributes to on-going efforts to monitor and detect changes in the pelagic ecosystems.

Big data based ocean acoustic source localization via deep learning

Machine learning has shown its potential in ocean acoustic source localization when real data are available to construct the training data set (Niu et al., JASA 142, 1176–1188 (2017); Niu et al., JASA 142, EL455–460 (2017)). This paper investigates the cases suffering from deficient real observations, where big data generated from the acoustic propagation model are used as the training set and therefore deep learning models could be exploited. To handle the environmental uncertainty, the multi-frequency acoustic data are generated using an acoustic propagation model (KRAKEN) based on tens of thousands of environments. A 50-layer residual neural network model (ResNet-50) is used to address this challenging problem. The performance on simulated and experimental data demonstrates that the proposed approach is able to adapt to a variety of environments in source localization problems.

Use of acoustic estimates of euphausiid biomass in trophic dynamics and ecosystem models of the Georges Bank/Gulf of Maine Region from 1999 to 2012

Euphausiids are a key link between primary production and higher-level predators in the Gulf of Maine, but are not well sampled during standard fisheries surveys. Multifrequency acoustic data may provide useful estimates of euphausiid distribution and biomass, as long as automated classification of acoustic backscatter is reliable and robust. Estimates of euphausiid biomass in the Georges Bank region of the Gulf of Maine were derived from annual acoustic/midwater trawl surveys from 1999 through 2012. Acoustic data were collected continuously with Simrad EK500 and EK60 echo sounders operating at 18, 38, and 120 kHz. Four different methods were used to classify euphausiids from the acoustic data: multifrequency single beam imaging, “dB-differencing” of 120- and 38-kHz volume backscatter, multifrequency z-score, and a multifrequency index. Scattering model predictions of euphausiid target strength and biological metrics were used to scale acoustic data to biomass for each classification method. Biomass estim...

Estimation of biological parameters of marine organisms using linear and nonlinear acoustic scattering model-based inversion methods.

The linear inversion commonly used in fisheries and zooplankton acoustics assumes a constant inversion kernel and ignores the uncertainties associated with the shape and behavior of the scattering targets, as well as other relevant animal parameters. Here, errors of the linear inversion due to uncertainty associated with the inversion kernel are quantified. A scattering model-based nonlinear inversion method is presented that takes into account the nonlinearity of the inverse problem and is able to estimate simultaneously animal abundance and the parameters associated with the scattering model inherent to the kernel. It uses sophisticated scattering models to estimate first, the abundance, and second, the relevant shape and behavioral parameters of the target organisms. Numerical simulations demonstrate that the abundance, size, and behavior (tilt angle) parameters of marine animals (fish or zooplankton) can be accurately inferred from the inversion by using multi-frequency acoustic data. The influence of the singularity and uncertainty in the inversion kernel on the inversion results can be mitigated by examining the singular values for linear inverse problems and employing a non-linear inversion involving a scattering model-based kernel.

Acoustic surveys of euphausiids and models of baleen whale distribution in the Barents Sea

As in many high-latitude ecosystems, euphausiids (order Euphausiacea, 'krill') play a key role in the Barents Sea by channeling energy from primary producers to fish and other zoo- plankton predators. We used multifrequency acoustic data from several recent multidisciplinary surveys to describe the spatial distribution of backscatter likely to be from euphausiids. Spatial patterns in euphausiid backscatter observed in 2010, 2011, and 2012 were correlated with verti- cally integrated euphausiid biomass collected with plankton nets, and were also broadly consis- tent with the distribution of euphausiids expected from the literature. We used the high-resolution and broad-spatial coverage of our euphausiid backscatter data to update multiple regression models of baleen (fin, humpback, and minke) whale distribution to test the hypothesis that these animals aggregated where euphausiids were abundant. After controlling for physical environ- mental factors and the densities of capelin and several other potential prey taxa, we found that fin whale densities were positively and linearly associated with euphausiid backscatter, and higher than average densities of humpback whales were found in areas with high euphausiid back - scatter. No association was found between minke whales and euphausiids. Densities of all 3 whale species were also positively associated with capelin. For fin and humpback whales, the effects of capelin and euphausiids on whale densities appeared to be principally separate and additive, although there was some evidence for a stronger effect of euphausiids at low capelin densities. In terms of their preference for euphausiids and capelin, these whale species appeared to be flexible, opportunistic predators.

Distribution of Atlantic herring (Clupea harengus) in the Gulf of Maine from 1998 to 2012

Atlantic herring (Clupea harengus) in the Gulf of Maine were surveyed using fisheries-independent systematic and stratified-random surveys conducted annually by the Northeast Fisheries Science Center from 1998 to 2012 during autumn. The systematic surveys were dedicated to surveying Atlantic herring and were designed to be synchronous with pre-spawning, whereas the stratified-random surveys were broad-scale, multi-species surveys. Both types of surveys collected multi-frequency acoustic and trawl data. These surveys were analyzed to evaluate their individual effectiveness as surveys, whether results could be combined, or whether either survey could be used alone to estimate Atlantic herring abundance. With the exception of two years, more Atlantic herring in the water column were observed in the Georges Bank region during the dedicated systematic survey than later in the season by the stratified-random survey in the Georges Bank and northern Gulf of Maine regions. These observations suggest that the systematic survey was not out of synchrony with Atlantic herring spawning in the Georges Bank area, and the majority of spawning in the off-shore Gulf of Maine region occurred in the Georges Bank area. Additionally, the area backscattering time series showed no dramatic changes among surveys conducted with different vessels, suggesting that the Atlantic herring did not react substantially to changes in vessels and/or reacted similarly to the presence of any of the four vessels used.

Zooplankton and micronekton biovolume at the Mid-Atlantic Ridge and Charlie–Gibbs Fracture Zone estimated by multi-frequency acoustic survey

To examine the potential influence of the Mid-Atlantic Ridge and Charlie–Gibbs Fracture Zone on zooplankton and micronekton biovolume in the upper 200m of the water column, multi-frequency acoustic data (18, 38, 70, 120 and 200kHz) were acquired at four study sites from the RRS James Cook using hull-mounted scientific echosounders. Multi-frequency inversion techniques were employed to classify each 20m depth×500m along-track region of the water column to a zooplankton or micronekton acoustic scatterering class, such as copepod or euphausiid, and to estimate biovolume. We found a highly significant north–south (across fracture zone) difference in areal biovolume (p-value=0.01) but no significant east–west (across ridge) difference (p-value=0.07). Areal biovolume at all sites was dominated by the acoustic scatter class ‘euphausiid’, with higher biovolumes occurring in the southern stations. Our acoustic observations suggest the existence of different pelagic communities to the north and south of the SPF, with the southern community having a greater proportion of fish.

School characteristics of mesopelagic fish at South Georgia

In this paper, we describe the variation in myctophid schools characteristics at South Georgia using multi-frequency acoustic data collected annually between November 2007 and January 2012. We studied the relationship between the proximity of land and the distribution and schooling characteristics of myctophid fish. We also examined patterns in schooling behaviour in relation to oceanographic data. Fish schools were identified using a dual-frequency dB identification method (Sv120−38kHz), where negative Sv120−38kHz is indicative of gas-bearing organisms, such as swimbladdered fish. Available net data were used to provide information on the meosopelagic fish community in the region. School morphometrics (e.g. length, thickness, area) data were extracted and pooled according to their distance from the shore (0–30km, 30–60km, 60–90km, 90–120km). A total of 578 schools were detected in the survey region and there was high variation in school backscattering strength (NASC), length, height, perimeter, depth and horizontal distribution between surveys, including distinct inter-annual variation in these parameters when surveys were conducted in the same season (2009–2012). Schools were distributed predominantly on-shelf (0–500m isobaths) during some summer surveys (2007 and 2012) and predominantly off-shelf on others (>500m isobaths; 2009, 2010 and 2011). There was also evidence of bathymetric separation in the horizontal distribution of schools. Schools observed in the late-season in 2008 had the greatest NASC, largest height and the deepest depth distribution. Most schools occurred at depths between ∼60 and 300m, but they did not occupy the same water masses during each survey. Schools became progressively thinner, shallower and acoustically weaker with increasing distance to land, whilst school length and perimeter increased correspondingly. The change in fish schooling behaviour between environments could be a response to a combination of local predatory threats over short spatial/temporal scales and differences in oceanographic conditions, such as current velocity. The trend could also be evidence of spatial habitat partitioning of myctophids, with different schooling species, or different life-stages, orientating along a bathymetric gradient at South Georgia.

Improved agreement of net and acoustical methods for surveying euphausiids by mitigating avoidance using a net-based LED strobe light system

Abstract Wiebe, P. H., Lawson, G. L., Lavery, A. C., Copley, N. J., Horgan, E., and Bradley, A. 2013. Improved agreement of net and acoustical methods for surveying euphausiids by mitigating avoidance using a net-based LED strobe light system. – ICES Journal of Marine Science, 70: 650–664. Euphausiids are known for their ability to avoid capture by standard plankton nets. Repeat sampling was conducted in the Gulf of Maine with a 1 m2 MOCNESS equipped with a light-emitting diode (LED)-based strobe light concurrent to multifrequency acoustic data collection. On cruise one, the strobe light's efficacy in reducing euphausiid net avoidance was evaluated on two horizontal tows where four of the eight nets were sampled with the strobe flashing and four with the strobe off. With the strobe light on, there was a significant increase in euphausiid abundance by a factor of 4.5 at night and by a factor of 11.0 during the day. There was also a significant increase in biovolume catch of zooplankton when the strobe light was on by factors of 2.2 at night and 5.5 during the day due to a higher abundance of 10 and 35 mm euphausiids. Euphausiids caught with the strobe light on accounted for most of the observed backscattering, and acoustic and net estimates of their abundance agreed. Similar results were obtained on cruise two. Agreement of acoustic and net estimates when using the strobe light suggests that reduced avoidance enhanced the efficiency of catching euphausiids.

Shoreward advection of phytoplankton and vertical re-distribution of zooplankton by episodic near-bottom water pulses on an insular shelf: Oahu, Hawaii

Shoreward pulses of cold, high salinity, phytoplankton-rich bottom water represent short time scale changes in nearshore hydrography and biological community structure off the leeward coast of Oahu, Hawaii. A three-week mooring deployment in Spring 2009 revealed that ‘cold pulses’ occurred over all phases of the semidiurnal surface tide, but that a statistically greater proportion occurred at low slack tide, when the typically along-shore flow rotated and briefly exhibited a more dominant onshore component. Cold pulses were more frequent and propagated farther shoreward when background water-column stratification increased. Targeted shipboard sampling of a cold pulse in Spring 2010 revealed that chlorophyll fluorescence within the cold pulse was 7 standard deviations higher than the 11-h mean outside the cold pulse, phytoplankton concentrations (cells mL−1) were up to a factor of 3 higher within the cold pulse, and phytoplankton entrained within the cold pulse were adapted to habitats with lower light levels and higher nutrient concentrations compared to ambient waters. Analysis of multi-frequency acoustic data collected during two shipboard surveys in 2009 and 2010 indicated that acoustic scattering during cold pulses was predominantly biological, dominated by 1.0–1.5mm spheroid fluid-like scatterers, both in the waters above the cold pulse and in a strong-scattering feature at the cold–pulse interface. These aggregations of larger organisms at the cold-pulse interface did not appear to migrate downwards into the phytoplankton-rich water during the active passage of the cold pulse. Observations of similar temperature events throughout the tropical Pacific, combined with our multidisciplinary findings, suggest that pulsed deliveries of phytoplankton-rich water to nearshore habitats may be regular occurrences throughout the North Pacific Subtropical Gyre.

Aggregative patterns of pre-spawning Atlantic herring on Georges Bank from 1999-2010

Atlantic herring (Clupea harengus ) in the offshore regions of the Gulf of Maine migrate each fall from their feeding grounds to the northern portion of Georges Bank to spawn. The Northeast Fisheries Science Center’s (NEFSC) herring acoustic survey has taken advantage of this behavior by conducting systematic surveys of the pre-spawning fish each year since 1999. Multi-frequency acoustic and midwater trawl data were collected along transects oriented perpendicular to bathymetric contours. Acoustic backscatter was analyzed to describe the aggregative patterns (e.g., size, location in the water column, and spatial and temporal distribution) of Atlantic herring during these surveys and regression trees were used to examine the aggregation characteristics. The positional variables of distance to spawning grounds and vertical location in the water column were the primary characteristics for describing pre-spawning aggregations. Secondary to these were the temporal variables of diel and survey timing, and the morphological characteristic of aggregation area. Lower numbers of aggregations were observed close to the herring spawning grounds but with higher acoustic energy than larger numbers of aggregations observed further from the spawning grounds but smaller in size and lower in acoustic energy. Most aggregations were in the lower portion of the water column, but those that were in the upper portion of the water column had higher acoustic energy. Consistently throughout the decade, 90% or more of herring aggregations were located within 40 nautical miles of their spawning grounds. The regression tree method provided valuable insight to the data series where it highlighted spatial and temporal patterns and was an effective way to quantitatively summarize relationships.

Acoustic determination of the distribution of fish and krill across the Scotia Sea in spring 2006, summer 2008 and autumn 2009

Multifrequency acoustic backscatter data were examined from transects at eight stations across the Scotia Sea, from the South Orkneys to the north west of South Georgia. These transects were repeated in austral spring (November 2006), summer (December 2007/January 2008) and autumn (March 2009). A dB identification window ( S v120-38 ) identified two types of scatterers dominating the acoustic backscatter; a positive S v120-38 indicative of Antarctic krill ( Euphausia superba) and a negative S v120-38 indicative of mesopelagic fish with swim bladders, such as myctophids. Acoustic backscatter attributed to schooling myctophid fish was greater at stations in the north of the Scotia Sea across all seasons and found to increase significantly with increasing mean sea temperature and absolute dynamic height. A schools analysis identified more fish schools at the stations with high fish acoustic backscatter. The number of fish schools increased, were distributed further south, and occurred deeper from spring through to autumn. Krill density, calculated using a target strength model, was similar in spring (27.7 g m −2) and summer (27.2 g m −2), but lower (2.2 g m −2) in autumn 2009. An analysis of swarm characteristics showed that during the spring cruise the high mean krill density resulted from a small number of swarms containing a large number of small krill, particularly in the south of the survey region. In contrast in the summer cruise the high mean krill density resulted from a large number of swarms containing a low number of large krill. Finally, in autumn the low krill density resulted from a small number of swarms containing a small number of large krill. Krill swarms were located at a mean depth of 58 m in spring, 94 m in summer and 123 m in autumn and were generally further apart, deeper, contained fewer krill and smaller in the north of the survey region. Large-scale shifts in the latitudinal and vertical distribution of krill and schooling mesopelagic fish in the Scotia Sea were observed. This seasonal reorganisation in both horizontal and vertical distribution will alter the role of these prey items in the ecosystem.

Interpretation of multi-frequency acoustic data: Effects of fish orientation

One goal of fisheries acoustics is to develop objective classification or identification methods to automate allocation of acoustic backscatter to species. Classification schemes rely on consistent relationships for successful apportionment of acoustic backscatter to species. A method is developed that compares frequency-dependent volume backscatter from an acoustical survey of Atlantic herring (Clupea harengus) to investigate the potential for classifying herring. Predicted backscattering patterns by a Kirchhoff-ray approximation are used to explain the observed relationships and evaluate the potential for classification of multi-frequency data. Combining predicted backscatter with observations of the frequency-dependent volume backscatter gave approximately 40% classification success, which is not sufficient for survey purposes. However, this method highlighted potential consequences that fish orientation may have on classification schemes and density and abundance estimates. This method of comparing multi-frequency volume backscatter appears to be beneficial for detecting behavioral changes by groups of fish, which may be used to select target strength values for density or abundance estimates. Utilizing predicted target strengths from numerical or analytical solutions or approximations, appropriate target strengths could be selected and would provide more accurate estimates of fish density and abundance.

Distribution, abundance and acoustic properties of Antarctic silverfish ( Pleuragramma antarcticum) in the Ross Sea

Antarctic silverfish ( Pleuragramma antarcticum) is a key link between plankton and the community of top predators in the shelf waters of the Ross Sea. In spite of their abundance and important role in Antarctic food chains, very little is known of many ecological and biological aspects of this species. A combined trawl and acoustic survey of silverfish was carried out on the western Ross Sea shelf during the New Zealand International Polar Year Census of Antarctic Marine Life research voyage on R.V. Tangaroa in February–March 2008. Multi-frequency acoustic data (12, 38, 70, and 120 kHz) allowed discrimination of silverfish marks from those of krill and other associated species. Mark identification was achieved using targeted midwater trawls. Additional midwater and demersal trawls were carried out at randomly selected locations over the shelf as part of the core biodiversity survey. Silverfish were widely distributed over the Ross Sea shelf. Adult silverfish tended to form layers at 100–400 m depth and were sometimes present close to the bottom, where they were frequently caught in demersal trawls shallower than 500 m. A weak layer at about 80 m depth was associated with juvenile silverfish of 50–80 mm standard length. Acoustic backscatter strength from both silverfish and krill marks increased with increasing frequency (i.e., was highest at 120 kHz), which is characteristic of species without an air-filled swimbladder. Acoustic target strengths (TS) for silverfish at 12, 18, 38, 70, and 120 kHz were estimated from anatomically detailed scattering models based on computed tomography (CT) scans of frozen specimens. The relationship between TS and fish length at 38 kHz was sensitive to estimates of density and sound speed contrast within the fish, especially for small specimens (less than 110 mm SL). Our best estimate of the acoustic biomass of silverfish in the study area was 592 000 t (95% confidence interval 326 000–866 000 t). However, the biomass of juvenile silverfish was highly uncertain due to large differences between TS model results.

ACOUSTIC PROPERTIES OF INDIAN MACKEREL (RASTRELLIGER KANAGURTA), YELLOWTAIL SCAD (ATULE MATE) AND JAPANESE SCAD (DECAPTERUS MARUADSI) BASED ON FREQUENCY RESPONSE ANALYSIS

Multi-frequency acoustic data collected during ex situ target strength measurements in Catba Island, Vietnam was used to calculate frequency response for Japanese scad, Indian mackerel and Yellowtail scad. Total of 168 echotraces of Yellowtail scad, 291 of Japanese scad and 53 of Indian mackerel were generated for frequency response analysis. Results indicated significantly different in r(f) of three species (Post Hoc Tukey Test, p < 0,001). A discriminant function analysis used frequency response at 38, 120 and 200 kHz and backscattering cross section at 38 kHz as independent variable was able to differentiate between species. Approximatly of 87% echotraces of these species was classified correctly. Frequency responses at 38 kHz, 120 kHz and backscattering cross section at 38 kHz were the most important independent variables. This study revealed that Japanese scad, Yellowtail scad and Indian mackerel could be identified acoustically; a finding may be important for acoustic surveys for small pelagic fish in Vietnam.

Proposals for the collection of multifrequency acoustic data

Abstract Korneliussen, R. J., Diner, N., Ona, E., Berger, L., and Fernandes, P. G. 2008. Proposals for the collection of multifrequency acoustic data. – ICES Journal of Marine Science, 65: 982–994. Acoustic surveys are used to estimate the abundance and distribution of many fish species, and have been based traditionally on data collected at a single acoustic frequency. Although it has been known for some time that the use of additional frequencies can provide information on the nature of the acoustic target, the knowledge and technology required to combine the so-called “multifrequency data” in an appropriate manner has been limited. The use of several transducers of different frequencies is now common on board research vessels and fishing vessels, so multifrequency data are often collected. In order for these data to be combined appropriately, their physical and spatial characteristics from each frequency should be as similar as possible. We detail the requirements deemed necessary to collect multifrequency data in an appropriate manner. They can be stringent and may not always be achievable, so we also consider the consequences of combining acoustic data originating in transducers with varying degrees of spatial separation and with different beam widths.