Pierre Petitgas

Sampling variance of species identification in fisheries-acoustic surveys based on automated procedures associating acoustic images and trawl hauls

During the acoustic surveys of fish stocks, a small number of echo traces are identified to species by fishing. During data analysis, the process of echogram scrutiny leads to allocating echo-trace backscattered energies to species. While the precision of survey estimates is generally based on the spatial variation in the energy, no variance term accounts for species identification and energy allocation. In this paper, the sampling variance of species identification is developed and automated procedures are used allowing energy allocation to be carried out by a non-expert. The procedures are based on the fact that at the sampling stage trawl hauls are linked with particular acoustic images. The procedures have two steps: the classification step corresponds to species identification and the aggregation step to energy allocation. Classification is performed on the identified images and results in defining groups of images and estimating in each the sampling variability of the species identification. Aggregation is performed on non-identified images and results in post-stratifying the data. The estimation (map, abundance and variance per species) is then derived automatically and is conditioned by the post-stratification. Two approaches are followed, one based on the echo-trace characteristics making full use of the echogram (acoustic-image classification) and the other on the spatial continuity of the species composition between trawl hauls (trawl-haul classification). These methods are described and compared. The species-identification variance term is also compared to the spatial variance.

Standard protocols for the analysis of school based data from echo sounder surveys

This paper presents a set of standard extraction parameters and protocols for the use of image analysis techniques in the processing of echo sounder data. The paper includes parameters at the school, elementary distance sampling unit (EDSU) and regional levels. The school level parameters, which are mainly derived from the image analysis, fall into four main categories: positional, morphometric, energetic and environmental. At the sampling unit level (i.e. standard integration units, commonly 1 or 2.5 nautical miles), parameters used include: school structures, protocols for including layers and general scatter plus ancillary (e.g. environmental) variables. These variables are derived mostly from visual examination of the echogram and from ancillary data collected underway. Regional level parameters include those mapped from point samples (e.g. trawls) or which are available as maps. Each school thus has its own unique parameters and is associated with an EDSU and through that to regional data. We discuss the application of such databases to the analysis of echo surveys at a school level in relation to aggregation patterns (school, school cluster and population) and to changes in those aggregation patterns with stock biomass and exploitation pattern.

A method for the identification and characterization of clusters of schools along the transect lines of fisheries-acoustic surveys

The school-aggregation pattern (schools and clusters of schools) is presumed to play a significant role in determining pelagic fish-stock catchability. However, its analysis has seldom been undertaken because it requires field-behavioural data that is seldom available. Such information can now be obtained by analysing school-based data of fisheries-acoustic surveys. This paper proposes a method for doing so. The method allows for the identification of clusters of schools and the estimation of their parameters along one-dimensional, acoustic-survey transect lines. It is based on a spatial point-process approach that considers schools as point events occurring along the track sailed by a ship. More precisely, it is based on defining a maximum distance between schools in a cluster. This distance is chosen to optimize various criteria and in particular that of homogeneity concerning school location inside the clusters and school number per km. The algorithm is described and applied to a series of acoustic surveys carried out in the Bay of Biscay. The pertinence of the clusters obtained by the algorithm is evaluated by analysing which component of the spatial distribution of the schools corresponds to those clusters. This involves considering all the distances between school events and performing simulations of cluster point processes. The school clusters obtained by the proposed algorithm represent a small-range structure of a few kilometres when a longer-range structure of tens of kilometres was also present in the data.

Energy density of anchovy Engraulis encrasicolus in the Bay of Biscay

The energy density (E(D)) of anchovy Engraulis encrasicolus in the Bay of Biscay was determined by direct calorimetry and its evolution with size, age and season was investigated. The water content and energy density varied seasonally following opposite trends. The E(D) g(-1) of wet mass (M(W)) was highest at the end of the feeding season (autumn: c. 8 kJ g(-1)M(W)) and lowest in late winter (c. 6 kJ g(-1)M(W)). In winter, the fish lost mass, which was partially replaced by water, and the energy density decreased. These variations in water content and organic matter content may have implications on the buoyancy of the fish. The water content was the major driver of the energy density variations for a M(W) basis. A significant linear relationship was established between E(D) g(-1) (y) and the per cent dry mass (M(D); x): y =-4.937 + 0.411x. In the light of the current literature, this relationship seemed to be not only species specific but also ecosystem specific. Calibration and validation of fish bioenergetics models require energy content measurements on fish samples collected at sea. The present study provides a first reference for the energetics of E. encrasicolus in the Bay of Biscay.

Interannual Changes in Biomass Affect the Spatial Aggregations of Anchovy and Sardine as Evidenced by Geostatistical and Spatial Indicators

Geostatistical techniques were applied and a series of spatial indicators were calculated (occupation, aggregation, location, dispersion, spatial autocorrelation and overlap) to characterize the spatial distributions of European anchovy and sardine during summer. Two ecosystems were compared for this purpose, both located in the Mediterranean Sea: the Strait of Sicily (upwelling area) and the North Aegean Sea (continental shelf area, influenced by freshwater). Although the biomass of anchovy and sardine presented high interannual variability in both areas, the location of the centres of gravity and the main spatial patches of their populations were very similar between years. The size of the patches representing the dominant part of the abundance (80%) was mostly ecosystem- and species-specific. Occupation (area of presence) appears to be shaped by the extent of suitable habitats in each ecosystem whereas aggregation patterns (how the populations are distributed within the area of presence) were species-specific and related to levels of population biomass. In the upwelling area, both species showed consistently higher occupation values compared to the continental shelf area. Certain characteristics of the spatial distribution of sardine (e.g. spreading area, overlapping with anchovy) differed substantially between the two ecosystems. Principal component analysis of geostatistical and spatial indicators revealed that biomass was significantly related to a suite of, rather than single, spatial indicators. At the spatial scale of our study, strong correlations emerged between biomass and the first principal component axis with highly positive loadings for occupation, aggregation and patchiness, independently of species and ecosystem. Overlapping between anchovy and sardine increased with the increase of sardine biomass but decreased with the increase of anchovy. This contrasting pattern was attributed to the location of the respective major patches combined with the specific occupation patterns of the two species. The potential use of spatial indices as auxiliary stock monitoring indicators is discussed.

A geostatistical method for assessing biomass of tuna aggregations around moored fish aggregating devices with star acoustic surveys

Universal kriging was used to model the spatio-temporal variability in the acoustic density of tuna aggregations recorded during star echosounding surveys around moored fish aggregating devices (FADs) in Martinique (Lesser Antilles). The large-scale deterministic drift in the tuna spatial distribution was modeled using an advection–diffusion equation applied to animal grouping. Residuals from the drift were modeled as a random component with small-scale spatial correlation. An estimation variance formula was derived from this deterministic–statistical mixed model to assess the mean precision of density estimates of daytime tuna aggregation. The mean relative error obtained with our star design for daytime surveys was 24%. The methodology was applied to estimate daily maxima of tuna biomass around moored FADs during four monthly sea cruises. The daily peak of tuna biomass aggregated around moored FADs was 9 t on average (standard deviation = 4). Estimation variances for different survey designs were compared for optimizing sampling strategy. Resume:

A Geostatistical Definition of Hotspots for Fish Spatial Distributions

Research surveys at sea are undertaken yearly to monitor the distribution and abundance of fish stocks. In the survey data, a small number of high fish concentration values are often encountered, which denote hotspots of interest. But statistically, they are responsible for important uncertainty in the estimation. Thus understanding their spatial predictability given their surroundings is expected to reduce such uncertainty. Indicator variograms and cross-variograms allow to understand the spatial relationship between values above a cutoff and the rest of the distribution under that cutoff. Using these tools, a “top” cutoff can be evidenced above which values are spatially uncorrelated with their lower surroundings. Spatially, the geometric set corresponding to the top cutoff corresponds to biological hotspots, inside which high concentrations are contained. The hotspot areas were mapped using a multivariate kriging model, considering indicators in different years as covariates. The case study considered here is the series of acoustic surveys Pelgas performed in the Bay of Biscay to estimate anchovy and other pelagic fish species. The data represent tonnes of fish by square nautical mile along transects regularly spaced. Top cutoffs were estimated in each year. The areas of such anchovy hotspots are then mapped by co-kriging using all information across the time series. The geostatistical tools were adapted for estimating hotspot habitat maps and their variability, which are key information for the spatial management of fish stocks. Tools used here are generic and will apply in many engineering fields where predicting high concentration values spatially is of interest.

Indicator-Based Geostatistical Models For Mapping Fish Survey Data

Marine research survey data on fish stocks often show a small proportion of very high-density values, as for many environmental data. This makes the estimation of second-order statistics, such as the variance and the variogram, non-robust. The high fish density values are generated by fish aggregative behaviour, which may vary greatly at small scale in time and space. The high values are thus imprecisely known, both in their spatial occurrence and order of magnitude. To map such data, three indicator-based geostatistical methods were considered, the top-cut model, min–max autocorrelation factors (MAF) of indicators, and multiple indicator kriging. In the top-cut and MAF approaches, the variable is decomposed into components and the most continuous ones (those corresponding to the low and medium values) are used to guide the mapping. The methods are proposed as alternatives to ordinary kriging when the variogram is difficult to estimate. The methods are detailed and applied on a spatial data set of anchovy densities derived from a typical fish stock acoustic survey performed in the Bay of Biscay, which show a few high-density values distributed in small spatial patches and also as solitary events. The model performances are analyzed by cross-validating the data and comparing the kriged maps. Results are compared to ordinary kriging as a base case. The top-cut model had the best cross-validation performance. The indicator-based models allowed mapping high-value areas with small spatial extent, in contrast to ordinary kriging. Practical guidelines for implementing the indicator-based methods are provided.

A Review of Linear Geostatistics for Fisheries Survey Design and Stock Assessment

In the past ten years, fisheries scientists have shown great interest in geostatistical tools when analysing survey data of fish stocks for estimating population abundance. Fisheries scientists have first paid attention to the relation between a model based variance estimate and a covariance structure in their data as this enabled to reconsider survey design. The possibility to use the spatial covariance for optimizing sampling strategy has been a second motive for using geostatistics. Last, kriging has also raised interest as it provides a weighing of the data values which can be usefull when sample points are clustered.