© 2005 International Council for the Exploration of the Sea
Experimental verification of the acoustic characteristics of the clupeoid diel cycle in the Baltic
Sea Fisheries Institute ul. Kollataja 1, 81-332, Gdynia, Poland
*Correspondence to A. Orlowski: tel: +48 58 620 1728 ext 215; fax: +48 58 620 2831. e-mail: orlov{at}miryb.mir.gdynia.pl.
This paper describes the results of a short-term experiment using measurements of sv from the diel, spatial, clupeoid distribution in the southern Baltic. The aim of the experiment was to verify fish-behaviour characteristics measured over the period from 1995 to 2001. It was also intended to estimate the dynamics of fish behaviour over one separate diel cycle. The studies were based on a 24 h continuous integration of fish echoes using an EY500 echosounder at 38 kHz. Measurements were carried out by RV "Baltica" travelling along the sides of a square of 4 nautical miles at a constant speed of 8 knots. The South Gotland Deep was chosen for the experiment because of the greatest amplitude there of fish vertical, diel migrations within the Polish EEZ. The duration of the experiment was limited by weather conditions, but nearly 300 EDSU samples were collected. 3-D distributions of echoes were correlated to the values of coincident environmental factors of time, depth, water temperature, salinity, and oxygen level. Fish distribution compared with environmental factors is described by different macrosounding visualizations, statistical, and mathematical models. Measurements are compared with the average characteristics of fish behaviour based on data from the autumn acoustic studies between 1995 and 2001 in a wider environment of the Gotland Deep. The results strongly confirmed the instability of the diel acoustic response of fish echoes in both situations. More precise measurements during the experiment indicated the biggest column-scattering strength (Svc) increase during the sunrise period, appearing simultaneously in the whole area. A similar increase was detected by analysis of 1995–2001 data from the South Gotland Deep environment. Major emphasis is given to the explanation of the diel irregularities. Diel instability of fish acoustic response can significantly effect the results of target-strength measurements (up to 400% during the sunrise) and, as a consequence, the calibration of acoustic fish stock-assessment models.
Keywords: acoustic, Baltic Sea, diel cycle, environment, fish behaviour, herring, sprat, target strength, vertical migration
Received 27 August 2004; accepted 10 February 2005.
 |
Introduction |
|---|
 Top Introduction Material and methodsResults and discussionConclusionsReferences |
|---|
Acoustic research of fish behaviour is pursued by many acousticians and particular ICES Working Groups for two basic reasons:
- to minimize variation of acoustic stock assessments, and
- to enhance the possibilities given by the acoustic technique and associated data processing.
Examples of such activities are given in Petitgas and Levenez (1996), Nøttestad (1998), Orlowski (1998, 1999, 2000, 2001, 2003), Jech and Luo (2000), Szczucka (2000), Mayer et al. (2001), Cardinale et al. (2003), Huse and Korneliussen (2003), Nilsson et al. (2003), Ona (2003), Soria et al. (2003), Knudsen and Gjelland (2004).
Specific features of the Baltic ecosystem, characterized by low salinity (2–20 psu) and a two-layer structure, generate a necessity to develop these studies of fish behaviour and its influence on acoustic response from the fish in various ways (Orlowski, 1998, 1999, 2000, 2001, 2003; Szczucka, 2000; Cardinale et al., 2003; Nilsson et al., 2003). Using October data, for the period 1989–1999 for the Polish EEZ, the behavioural and physical effect on acoustic measurements of clupeoids in the Baltic within a diel cycle has been studied (Orlowski, 2001). This has shown the possibility of determining a significant variability of observed factors by applying different transformation methods for data collected in the period 1995–2001. Such analyses can be widely applied for monitoring relationships among biotic and abiotic factors of the marine ecosystem.
Diel variability of the acoustic responses of fish – the basic parameter of interest from the point of view of fish stock assessment – was identified as being associated with three main elements (Orlowski, 2000, 2001; Szczucka, 2000; Huse and Korneliussen, 2003; Ona, 2003; Knudsen and Gjelland, 2004):
- direct physical factors dependent on fish depth (pressure and temperature),
- indirect physical factors (inertial or fish-controlled changes of gas volume inside the fish body) and,
- behavioural factors (tilt-angle variability, the regulation of gas balance due to migrations with depth).
The results from different areas of the Baltic suggest different patterns of relationships attributable to local diversity of the range of factors (Orlowski, 2003). The use of data from many cruises taken together provides significant smoothing, or neutralizing from each other, of all time-dependent particular effects.
One way to avoid such problems is by the continuous observation of diel fish behaviour and correlated environmental and acoustic factors within a small area of the sea over a shorter time period (one to two diel cycles). Measurements have to be collected by sweeping the section with a constant and repeatable itinerary. Such an experiment, planned for 48 h, was carried out during an international survey of pelagic fish resources of the Baltic during October 2001. The area of the experiment was selected after analysis of all the data collected from the southern Baltic in the period 1995–2001. The main criteria in selecting the experimental area was the minimizing of the gradients of the observed parameters within it and its surroundings while the fish migration could be recorded in parallel at maximum range (more than 100 m). Such an area, with the depth of >110 m, lies within the Polish EEZ in the southern part of the Gotland Deep.
The studies were based on three-dimensional measurements of the sv distribution, representing echoes from a mixture of herring and sprat. These were correlated with coincident environmental factors: time, depth, water temperature, salinity, and oxygen level. Hydrological factors were estimated on the basis of data collected during the same survey. Fish behaviour vs. environmental factors was described by different macrosounding visualizations, and statistical and mathematical models. The basic results of the experiment were not at all similar to the characteristics estimated for the whole Polish EEZ.
The main goal of this paper is to describe fish-group diel behaviour in a representative small section of the sea over one realization of the cycle and to compare it with characteristics found during the period from 1995 to 2001 over the whole South Gotland Deep area.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResults and discussionConclusionsReferences |
|---|
Systematic acoustic surveys of the southern Baltic area have been conducted by the Polish Sea Fisheries Institute since 1981. Recording of samples 24 h a day, for each 1-nautical mile distance unit, where EDSU = 1852 m, in a slice-structured database started aboard RV "Baltica" in 1994. At that time, an EK400 echosounder with a QD echo-integrating system was used with proprietary software. Since 1998 an EY500 has been used for acoustic surveys. The frequency of both systems was 38 kHz with the same hull-mounted transducer of 7.2° x 8.0°. Calibration has been performed with a standard target in Swedish fjords in 1994–1997 and in Norwegian waters from 1998 to 2000. The cruises were carried out in October and lasted 2–3 weeks, giving a possibility of collecting samples over 1–1.5 thousand nautical miles (6 nautical miles of a transect per 1 square nautical mile approximately). Survey tracks of all cruises were on the same regular grid to obtain high comparability and the better splitting of measurements collected in succeeding years.
Biological samples were collected over the period from 1995 to 2001 by the same pelagic trawl, on average every 37 nautical miles of the transect. Fish observed during all surveys were mostly pelagic, herring, and sprat (Clupeidae). Hydrographic measurements of temperature – T, salinity – S, and oxygen level – O2, were made by a Neil-Brown CTD system. These were mostly at sample-haul positions.
Area A (South Gotland Deep) was taken into consideration in comparisons, and that area and area B, where a 24-h experiment transect was carried out, are shown in Figure 1. Area A had a surface area of approximately 4.6 thousand square nautical miles and represented in total nearly 1000, 1-nautical mile samples (an EDSU). Only samples where bottom depth exceeded 80 m were considered. The area of the experiment (B) was 290 times smaller than A, having a surface area of 16 square nautical miles and representing 298 such samples. Biological characterization of the experimental area was based on the results of one pelagic daytime haul, headline at 60 m, depth of vertical opening 20 m, carried out in the middle of the square. The weight composition of the haul was: 19.5% herring (mean length 17.59 cm and mean weight 31.78 g), 78.7% sprat (mean length 12.49 cm and mean weight 11.82 g), and 1.8% cod (mean length 48.86 cm and mean weight 1092.86 g). A CTD station was made at the haul position. The results of four catches carried out earlier during the same cruise in ICES rectangle 40G8 (containing areas of analyses) were quite similar: weight composition 34.0% herring (mean length 17.57 cm and mean weight 31.55 g), 65.9% sprat (mean length 12.70 cm and mean weight 12.15 g), and 0.5% cod (mean length 48.1 cm and mean weight 1009 g).
|
The results of echo integration for each EDSU and for each depth layer were converted into values of
|
| (1) |
is the conversion constant (m2 per square nautical mile sr), sv(z) is the volume-backscattering coefficient (m–1 sr–1), and z, zi, and zi+1 are the depth and slice layer i limits (m). Owing to the vessel' draft, the hull reverberations and aeration zone caused the first layer of integration to start at 15-m depth. Each EDSU unit was characterized by a series of i values of sA (for each sub-layer), geographic position, date, time of day, and sea-bottom depth. The day and night-time index was estimated on the basis of the analysis of the fish-echo recordings. For each EDSU, parameters characterizing fish distribution as upper and lower fish-depth limits, fish main depth (depth of the "gravity centre" of sv(z) distribution), and the values of the corresponding temperature, salinity, and oxygen, were calculated with the use of the self-prepared software. The approximate models of diel variability of parameters mentioned above were estimated on the basis of the methods described by the author in 1998. Macrosounding (Orlowski, 1990, 1998) and T-macrosounding (Orlowski, 2003) visualizations that significantly complemented the description of fish-diel behaviour were applied.
The final description of the variability of fish acoustic response was based on a 2D probability function representing the distribution of data corresponding to each EDSU against column-scattering strength (Svc) and time-of-day classes. Distributions were found for both comparable cases (A and B). Numerical models of diel variability of relative fish-echoes energy (sA/
sA
) as well as the other daily modulated parameters were estimated using trigonometric polynomial functions (Clay and Medwin, 1977; Orlowski, 1998). Charts of the average values of Svc values in the experiment area for different periods of the day were also produced.
|
Results and discussion |
|---|
|
TopIntroductionMaterial and methodsResults and discussionConclusionsReferences |
|---|
Characteristics of the spatial and temporal distribution of fish within the area of the experiment (area B) are given in Figures 2 and 3. Macrosounding visualization in Figure 2A shows continuous changes of the main range of fish recordings at successive 280-nautical miles of circulation along the sides of a square. The density of dots is proportional to sv values while the period of repetition (16 nautical miles) is clearly seen in the shape of a bottom-profile pattern. The upper (magenta) layer, is closed between the minimal fish depth and the main fish depth ("gravity centre") and the lower layer (blue) is limited by the maximal fish depth. The distance on the Ox axis corresponds to the distance between the ends of each EDSU: the yellow sector at the sea surface relating to daytime and the blue to night-time. This pattern illustrates the vertical dynamics of fish distribution over the whole period of the experiment. During the day fish are mostly concentrated between 50 m and 70 m (layer of sample haul), randomly migrating towards the bottom. These were possibly caused by feeding migrations, being limited in time by low oxygen level below 80-m depth (<0.5 ml l–1). The lower limits of fish distribution depth are correlated to the salinity and oxygen gradients (see Figure 3) while the lower depth limit at night was correlated to the temperature gradient.
|
|
T-macrosounding visualizations, expressing the average night and day fish distribution (Orlowski, 2003), respectively, are shown in Figure 2B, C. The number of mile samples was limited to 214, owing to the necessary elimination of dense migrating fish schools appearing in the first 62 miles of the experiment and the transition periods. The experiment involved 24 h of continuous integration of echoes and a steam of 214 nautical miles. Each unit on the Ox axis corresponds to one square nautical mile (symmetrical stripe along the transect – see Figure 1) and contains all the samples taken by night (Figure 2B) or by day (Figure 2C). The visualization applied gives a more detailed average distribution of fish, related to particular space units of the experiment area B. Simple visual analysis shows that the vertical distribution of the fish was different between night and day, but very uniform in the whole area, B. It permits, in fact, the treatment of the experimental results as representative of the diel behaviour of the fish. Uniformity of character is confirmed by the small differences in the statistical parameters of the sA values in the four quadrants of area B (Figure 1, Table 1).
|
Diel T-macrosounding characteristics, called here TDS (time-depth sv) are shown in Figure 3. These visualizations characterize time-variable, vertical distributions of fish in the A and B study areas. In the first case (A), we can see the average migration characteristics for the long-term period 1995–2001. In the second case (B), the fish distribution represents a 1-day period of successive EDSUs observed over the time of the experiment in October 2001. The diagrams of the mean values of T, S, and O2 at standard depths are shown to give an idea of the influence of absolute values and their gradients on fish distributions.
Fish echoes in the long-term period (Figure 3A) are more dispersed in the water column than in the experiment (Figure 3B). This can be explained by the influence of dispersion, brought about by the averaging in time and space of the values of all the factors controlling fish vertical position, e.g. light-intensity level and the absolute values of environmental gradients. Dispersion is also caused by the greater geographical size of area A, which leads to the differentiation of local diversity. The light level, owing to the influence of meteorological conditions, can also be modulated strongly in relation to astronomic time. The depth structure of the hydrographic gradients is influenced by long-term climatic variability. As a consequence, over a long-term period we can observe fish in a wider range of depths.
The pattern observed for the experiment (Figure 3B) shows stronger gradients of fish distribution in a smaller range of depths. Values from the experiment can be taken as adequate for the estimation of a single example of fish-diel-behaviour process in the time domain. In this case dispersion attributable to meteorological or geographical factors is negligible, and the environmental gradients are significantly stronger. Fish vertical distribution by night is clearly limited by temperature and oxygen gradients. The depth of fish occurrence during the day was limited by absolute value of oxygen (>0.5 ml l–1). The difference in the hour of the sunrise migration between cases A and B indicates the impact of particular differences in the light-intensity time-increase associated with meteorological conditions.
Figure 4 provides a comparison of the basic parameter measurements and their mathematical models for both the cases (A and B) considered. The diagrams are complemented by Table 2. Values of four basic factors (fish depth, T, S, and O2) were estimated at three characteristic levels: minimal, main, and maximal fish depths. The average values of all parameters within 1-h classes were applied to calculate trigonometric polynomials of the eleventh degree. The error of approximation was very low, between 0.16% and 4.34% for extreme cases. Higher degree of approximation polynomials had no positive influence on the decrease of approximation error. The diagrams show the confidence intervals of the average values at 1-h intervals. The highest dispersion within the intervals is observed during the transition periods, mostly correlated to sunrise and sunset periods. General dispersion is higher in area A, the larger the area, the longer the time.
|
|
The vertical distribution of fish by night and day was very similar in general, in both cases as shown by very small values of differences in average values of characteristic fish depths (Table 2) and similar patterns of the models. The average fish depth during the night was 33.1 ± 0.8 m in area A and 30.3 ± 1.6 m in area B. Values of the same parameter for the day were: 66.7 ± 1.9 m (A) and 67.5 ± 1.6 m (B). It must be taken into account in analysing the data presented in Table 2 in comparison with their models that transition periods were also included in night and day hours.
Detailed analysis of the variability of fish depths vs. time of day exposes a more complicated character of the fish-diel behaviour. Owing to the influence of factors producing higher dispersion of data in area A it seems reasonable to concentrate on an analysis of results collected in the experimental area B only. That conclusion underlines the importance of providing such experiments to measure the real dynamics of fish behaviour.
Starting from the moment of sunrise at 05:30, a synchronized migration of fish towards greater depths is observed. Sunrise migration from 30 m to 70 m (approximate values of fish main depth) has been observed in a very narrow 20-m layer. At 09:00 the fish schools are dispersed into two depth layers (see Figures 2 and 3). Sunset migration is less synchronized and starts at different moments for various fish depths. At first, migration towards the surface appears in the upper depth layers. The difference between a migration starting point between upper and lower fish depths can be estimated as more than 1 h.
The distribution of fish by night is also not stabilized in time. During the first period fish are dispersed within the wider layer (
30 m). Just before migration, the layer of fish becomes very thin (less than 10 m), and this phenomenon was observed clearly in both cases A and B. Similar phenomena are indicated by Nilsson et al. (2003).
The variability of temperature with time, at fish depths also varying in time, shows unexpected patterns. During the night, when the fish are stabilized in a small depth range, the chosen temperature range is the biggest (6.61–11.61°C in area A, 7.18–10.80°C in area B). During the day the most suitable temperature range (Table 2) is very small (4.87–5.96°C in area A, 5.08–5.11°C in area B), and there is wide dispersion of fish over a greater depth range. Fish-layer thickness is smallest during the process of migration through temperature gradients.
Diel variability of salinity at fish depths is opposite. By night salinity is within a very small range (6.84–7.23 psu), corresponding to the upper layer of the Baltic. In daylight, the fish disperse vertically within depths characterized by a relatively wide range of salinity (7.67–9.76 psu).
Salinity and oxygen patterns in relation to depth are negatively correlated, and this has a direct influence on oxygen patterns, as shown in the lowest panel of Figure 4. For both these factors high dispersion of values by day are comparable. The lowest oxygen level (0.79 ± 0.07 ml l–1) during the experiment (area B) was the only parameter that was quite different from that found in area A (2.22 ± 0.24 ml l–1). In Figure 2A fish were migrating towards the bottom, to the layers of low oxygen content for short periods only.
Daytime patterns of fish depth, salinity, and oxygen level show similarities in two distinct sub-periods: the period between sunrise and 10:00, and the period between 1000 h and sunset. During the first sub-period, the fish are dispersed in a narrow depth layer. It appears that the first sub-period is an adaptation phase, while the second one can be treated as a phase of full fish activity.
Finally, the factors describing the acoustic properties of fish-reflected signals in both options A and B of the studies were examined, and the results are given in Figure 5. The upper panel shows the characteristic diel modulation of echoes received by the echosounders via a 2D probability function P(t, Svc) representing the distribution of empirical acoustic data (Svc per EDSU), separated in Svc and time classes. The distributions were calculated for two separate situations: 1477 EDSU from the period 1995 to 2001 for area A, and 147 EDSU from a 24-h period selected from the experiment data relating to area B. The diel modulation of P(t, Svc) distributions is very clearly seen in both cases and related to the same time intervals viz. sunrise, daytime, sunset, and night-time. In the case of A dispersion of the Svc values is much wider, attaining 20 dB during the daytime and 14 dB during the night-time. Mean values of Svc were higher by night than by day. Some characteristic transformations of the distribution occur at sunrise and sunset, respectively. The transformation coinciding with sunrise is marked by the strongest relief. In the case of B very similar phenomena are observed. They were very clearly marked simply because the situation was only observed within one 24-h period in a small section of sea (one cycle). The most important thing to notice is a similarity of both patterns of diel variability of fish echoes. Clearly, the structure of acoustic response of the fish is strongly correlated, with the different times of a day influencing both average fish depth and the values of hydrologic factors at the characteristic depth of the fish.
|
The lower part of Figure 5 shows mathematical models of relative values of fish acoustic response (sA/
sA), normalized to mean value per day. The models are expressed by trigonometric polynomials of seventh degree, with the approximation error equalling 14.7% (area A) and 5.7% (area B). Confidence limits of the average of sA/sA values within selected time intervals are marked. The models calculated for different areas show very similar patterns. In general, they are also similar to those presented in Orlowski (2000, 2001). Those were calculated on the basis of all measurements made in the whole Polish EEZ in the period 1989–1999. In both cases, the echoes are stronger during the night. Similar conclusions on the increase of fish target strength during the night hours were given by Ona (2003) and Szczucka (2000). As the average depth of the Polish EEZ is approximately 67 m, the measurements (Orlowski, 2000, 2001) relate to a significantly shallower area than those analysed by those two authors. The most characteristic time interval (Figure 5, lower panel) corresponds to sunrise. In both cases, the effective value of sA/sA for this period is increasing strongly (two to three times). A similar increase was reported earlier (Orlowski, 2000, 2001), but it was not so strongly evident. A significant increase of fish echoes at sunrise requires an explanation. Helfman et al. (1997) suggest "when fish descend, the volume of the gas bladder decreases due to increasing pressure, the fish must add gas to maintain neutral buoyancy". We can say: it is also necessary to avoid the danger of squeezing the gas bladder and risking total loss of buoyancy. Using such a simple physical explanation we can interpret an increase of fish acoustic response by behavioural reaction, in advance of the downward migration. The moment of this migration is regulated by the astronomic factor, i.e. the level of light. Attaining a threshold value starts off an early fish-body adaptation that allows a longer period migration. Clupeoids as physostomes have good dynamic control of their gas-bladder volume. They are even able to release extensive gas bubbles (Nøttestad, 1998) for some adaptation purposes. Multiple vertical migrations are characteristic of the day period. All such migrations can produce an instability of fish-body characteristics from the point of view of the acoustic response. Such instabilities closely related to vertical migrations are described by many authors (Orlowski, 2000, 2001; Szczucka, 2000; Huse and Korneliussen, 2003; Ona, 2003; Knudsen and Gjelland, 2004). Vertical migrations of fish schools during the day, when looking for food, will produce a similar effect on fish target strength. Fish schools migrating towards the bottom produce a much higher acoustic response than those moving to the surface, and when migrating horizontally fish will keep near-constant acoustic properties. As fish-school migrations during the day are not synchronized by an external factor, such as at sunrise or sunset, and are distributed randomly in time it is difficult to estimate the error of target-strength variability that they cause. |
Conclusions |
|---|
|
TopIntroductionMaterial and methodsResults and discussionConclusionsReferences |
|---|
The results presented here stem from the application of a wide range of approaches and complementary methods of visualization used by Jech and Luo (2000), Mayer et al. (2001), and Soria et al. (2003). Petitgas and Levenez (1996) presented the first studies related to the transition from single to multi-beam applications and introduced geomatics and 3D-visualization software to assess fish stocks and enhance knowledge of pelagic fish schools. Their suggested visualizations could be useful for direct ecological studies, based on acoustic semi-tomography, or to define sub-areas where numerical models presented in the paper can be searched or verified.
In relation to this study though from the results of different forms of acoustic studies of clupeoid diel behaviour in the Baltic Sea it can be concluded that:
- short-term, local acoustic experiments on diel fish behaviour allow for effective enhancement and verification of data collected during a series of standard acoustic cruises,
- the combination of methods (acoustic, environmental and biological sampling) applied in parallel enhances the ability to identify and characterize important local differences in environmentally conditioned fish behaviour, estimated on long- and short-term scales,
- the application of a series of different macrosounding, multidisciplinary visualizations effectively provides the details of fish behaviour in particular conditions,
- each vertical migration of fish schools is followed by a consequential time variability of its acoustic response,
- fish target-strength variability is different between migration to the surface and migration towards the bottom,
- fish-behaviour studies should be changed as a result of these results to minimize the errors in acoustic measurements (i.e. associated with vertical migrations) that produce significant distortion of the fish stock-assessment process.
|
References |
|---|
|
TopIntroductionMaterial and methodsResults and discussionConclusionsReferences |
|---|
-
Cardinale M., Casini M., Arrhenius F., Hakansson N. (2003) Diel spatial distribution and feeding activity of herring (Clupea harengus) and sprat (Sprattus sprattus) in the Baltic Sea. Aquatic Living Resources 16:3283–292.[CrossRef][Web of Science]
Clay C.S. and Medwin H. (1997) Acoustical Oceanography: Principles and Applications(John Wiley & Sons, New York) ch. 4, 122–150.
Helfman G.S., Colette B.B., Facey D.E. (1997) The Diversity of Fishes(Blackwell Science, Oxford) pp. 1–528.
Huse I. and Korneliussen R. (2003) Diel variation in acoustic-density measurements of over-wintering herring (Clupea harengus L.). ICES Journal of Marine Science 57:4903–910.[CrossRef]
Knudsen H.P. (1990) The Bergen echo integrator: an introduction. ICES Journal of Marine Science 47:167–174.
Knudsen F.R. and Gjelland K.Ø. (2004) Hydroacoustic observations indicating swim-bladder volume compensation during the diel vertical migration in coregonids (Coregonus lavaretus and Coregonus albula). Fisheries Research 66:2–3337–341.[CrossRef][Web of Science]
Jech J.M. and Luo J. (2000) Digital echo visualization and information system (DEVIS) for processing spatially explicit fisheries-acoustic data. Fisheries Research 47:115–124.[CrossRef][Web of Science]
Mayer L., Yanchao L., Melvin G. (2001) 3D visualization for pelagic-fisheries research and assessment. ICES Journal of Marine Science 10:1–10.
Nilsson L.A., Thygessen U.H., Lundgren B., Nielsen B.F., Nielsen J.R., Beyer J.E. (2003) Vertical migration and dispersion of sprat (Sprattus sprattus) and herring (Clupea harengus) schools at dusk in the Baltic Sea. Aquatic Living Resources 16:3317–324.[CrossRef][Web of Science]
Nøttestad L. (1998) Extensive gas-bubble release in Norwegian spring-spawning herring (Clupea harengus) during predator avoidance. ICES Journal of Marine Science 55:1133–1140.
Ona E. (2003) An expanded target-strength relation for herring. ICES Journal of Marine Science 60:3493–499.
Orlowski A. (1990) Macrosounding as a new concept in the study of marine ecosystems. Colloque de Physique, c2 51:n269–72.
Orlowski A. (1998) Acoustic methods applied to fish-environmental studies in the Baltic Sea. Fisheries Research 34:3227–237.[CrossRef][Web of Science]
Orlowski A. (1999) Acoustic studies of spatial gradients in the Baltic: implication for fish distribution. ICES Journal of Marine Science 56:561–570.
Orlowski A. (2000) Diel dynamic of acoustic measurements of Baltic fish. ICES Journal of Marine Science 57:1196–1203.
Orlowski A. (2001) Behavioural and physical effect on acoustic measurements of Baltic fish within a diel cycle. ICES Journal of Marine Science 58:1174–1183.
Orlowski A. (2003) Acoustic semi-tomography in studies of the structure and function of the marine ecosystem. ICES Journal of Marine Science 60:1392–1397.
Petitgas P. and Levenez J.J. (1996) Spatial organization of pelagic fish: echogram structure, spatio-temporal condition, and biomass in Senegalese waters. ICES Journal of Marine Science 53:2140–147.
Soria M., Bahri T., Gerlotto F. (2003) Effect of external factors (environment and survey vessel) on fish-school characteristics observed by echosounder and multibeam sonar in the Mediterranean Sea. Aquatic Living Resources 16:3145–157.[CrossRef][Web of Science]
Szczucka J. (2000) Acoustically measured, diurnal vertical migration of fish and zooplankton in the Baltic Sea – seasonal variations. Oceanologia 42:15–17.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Loland, M. Aldrin, E. Ona, V. Hjellvik, and J. C. Holst Estimating and decomposing total uncertainty for survey-based abundance estimates of Norwegian spring-spawning herring ICES J. Mar. Sci., October 1, 2007; 64(7): 1302 - 1312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zwolinski, A. Morais, V. Marques, Y. Stratoudakis, and P. G. Fernandes Diel variation in the vertical distribution and schooling behaviour of sardine (Sardina pilchardus) off Portugal ICES J. Mar. Sci., July 1, 2007; 64(5): 963 - 972. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||