Konstantinos I. Stergiou

Feeding habits and trophic levels of Mediterranean fish

The estimation of fractional trophic levels(TROPHs) is essential for the management offisheries resources as well as for quantifyingthe ecosystem effects of fishing. We gatheredall available information concerning thefeeding habits of 332 fish stocks, belonging to146 species, 59 families and 21 orders,throughout the Mediterranean Sea, and estimatedtheir TROPH values. The latter ranged from 2.0to 4.5 and the following functional trophicgroups were identified: (a) pure herbivores (TROPH = 2.0–2.1, mean = 2.02, SD = 0.03),which were very rare and represented by Siganus luridus, Siganus rivulatus andSarpa salpa, all of which feed on red,brown, green and blue-green algae; (b)omnivores with a preference for vegetablematerial (2.1 < TROPH < 2.9, mean = 2.5,SD = 0.12), but feeding on other prey, such assponges, hydrozoans, anthozoans, polychaetes,ostracods, isopods, amphipods and copepods. This type of omnivore was very rare among thecases reviewed; (c) omnivores with a preferencefor animal material (2.9 < TROPH < 3.7,mean = 3.4, SD = 0.19) feeding on a wide variety ofprey (e.g., algae, foraminifera, brachyurans,balanoids, ascidians, amphipods,appendicularians, annelids, isopods,gastropods, cnidarians, ophiurids, polychaetes,cladocerans, mysids, euphausids, fish larvae,cephalopods). They were the most numerous andwere mainly represented by species of thefamilies Blenniidae, Bothidae, Centracanthidae,Gobiidae, Labridae, Lotidae, Macrouridae,Mullidae, Ophidiidae, Soleidae, Triglidaeand Engraulidae; and (d) carnivores witha preference for large decapods, cephalopodsand fish (3.7 < TROPH < 4.5). They werethe next most abundant group among the casesreviewed. They were mainly represented byspecies of the families Dalatiidae, Lophiidae,Scombridae, Scyliorhinidae, Synodontidae,Torpedinidae, Merlucciidae, Xiphiidae andZeidae. This group was divided into twosubgroups: one exhibiting a preference fordecapods and fish (3.7 < TROPH < 4.0,mean = 3.85, SD = 0.09) and another one exhibitinga preference for fish and cephalopods (4.0 <TROPH < 4.5, mean = 4.38, SD = 0.12). Potentialtop carnivores in the Mediterranean Sea werealso identified. Although such a grouping isprovisional, and subject to revision when moredatasets for other species, habitats and areas(especially from the Southern MediterraneanSea, which was underrepresented in ourcompilation) become available, it may serve asa basis for the maintenance of trophic levelbalance in the Mediterranean Sea. This willset the basis for an ecosystem approach to themanagement of the Mediterranean Sea. Acrossspecies, TROPH values increased asymptoticallywith maximum reported body length. The samewas also true for 9 out of 11 stocks, for whichdiet composition data were available for morethan three length classes. The within- andamong-species asymptotic relationships betweenTROPH and body length are discussed within theframework of foraging behavior and preyselection theory. Finally, based on thesampling characteristics and methodology usedin the studies reviewed, some recommendationsfor future stomach content studies are alsoprovided.

Weight-length relationships for 33 fish species in Greek waters

Abstract Weight-length regressions are presented for 33 fish species caught during 1992–1993 in the South Euboikos Gulf, Aegean Sea. Samples were collected with beach seine and gill and trammel nets. The values of the exponent b in the weight-length relationship W = aLb ranged from 2.320 to 3.521 and the median value was 2.987, whereas 50% of the values ranged between 2.840 and 3.140. The application of these regressions should be limited to the observed length ranges.

Editorial note on weight-length relations of fishes

Weight-length relations of fishes are useful for estimation of biomass from length observations, e.g., infisheriesorconservationresearch.Hereweprovidesomeguidancetoauthorsofsuchpapers,inordertofacil- itatethepublicationandreviewprocess.

PREDATOR AND PREY BODY SIZES IN MARINE FOOD WEBS

Knowledge of relationships between predator size and prey size are needed to describe interactions of species and size classes in food webs. Most estimates of predator and prey sizes have been based on dietary studies and apply to small numbers of species in a relatively narrow size range. These estimates may or may not be representative of values for other groups of species and body sizes or for other locations. Marine predator and prey size data associated with published literature were identified and collated to produce a single data set. If predator or prey length of mass were not measured in the original study, the length or mass was calculated using length–mass relationships. The data set consists of 34 931 records from 27 locations covering a wide range of environmental conditions from the tropics to the poles and for 93 types of predator with sizes ranging from 0.1 mg to over 415 kg and 174 prey types with sizes from 75 pg to over 4.5 kg. Each record includes: predator and prey scientific names, c...

Impacts of data quantity on fisheries stock assessment

Abstract. Understanding population dynamics of fish stocks is essential in developing optimal fisheries management strategies. This is often obtained through fitting mathematical models to information/data collected from the fisheries to estimate vital fisheries parameters and their uncertainties. The quantity of fisheries information, often positively related to economic and social values of fisheries, is one of the most important factors influencing the quality of fisheries parameter estimation. We classify the data quantity into two categories: diversity of the information, defined as the number of sources from which the information about a fisheries variable is collected, and number of observations made for a given fisheries variable. Using an abalone fishery as an example, we demonstrate the importance of data quantity in stock assessment and management. Deficiency in data quantity tends to yield biased assessment of the status of fisheries stock and increase the uncertainty in stock assessment, subsequently complicating the identification of an optimal management strategy. This study suggests that there is a need to determine the relative importance of different types of fisheries data for stock assessment in allocating sampling effort to ensure that the most critical information is collected.

Gill net and longlines fisheries in Cyclades waters (Aegean Sea): species composition and gear competition

Abstract Species composition and gear competition for the small-scale fisheries in the Cyclades waters (off Naxos Island, Aegean Sea, Hellas) was studied based on samples collected during 1997–1998 with gill nets of different mesh sizes and longlines with hooks of different sizes. Comparisons used univariate and multivariate techniques based on species compositions by number and weight and the commercial/total catch ratio in terms of weight and number. The results of the various techniques were in agreement and suggested that the different gear/size/season combinations fall into two groups (all gill nets and all longlines), which differed from each other in terms of species composition, species diversity and dominance. In contrast, the commercial/total catch ratio did not differ either within and between gears. The implications of the results for the management of the Mediterranean fisheries resources are also discussed.

Small-scale fisheries in the South Euboikos Gulf (Greece): species composition and gear competition

Abstract In the present study, we compared the species composition and gear competition of the small-scale fisheries in the South Euboikos Gulf, Greece, from samples collected during 1992–1993 by beach seine (8 mm mesh size, bar length) and gill and trammel nets of different mesh sizes (gill nets: 17, 19, 21 and 23 mm, bar length; trammel nets: 19, 24 and 30 mm, bar length). The sampling depth ranged from 18 to 60 m. Comparisons used univariate and multivariate techniques based on species compositions and the commercial/total ( C T ) ratio in terms of weight and number. The results of the various techniques were in close agreement, complemented each other, and suggested that the eight gears used may be aggregated into three main groups: (a) 30 mm trammel net (group I), which is highly selective when compared with all remaining gears; (b) all remaining gill and trammel nets (group II); and (c) beach seine (group III), which is an active unselective gear. These groups differed from each other in terms of species composition and, generally, diversity increased and C T ratio decreased from group I to group III. The results indicate that banning the beach seine could be essential for conservation of demersal and inshore biodiversity.

Prediction of the Mullidae fishery in the eastern Mediterranean 24 months in advance

Abstract The following stochastic, seasonal AutoRegressive Integrated Moving Average (ARIMA) model is presented (1 − 0.679 B 1 ) (1 − B 1 ) (1 − B 12 ) X t = (1 − 0.951 B 1 ) (1 − 0.564 B 12 ) e t where: X t = value at month t ; B p is a backward shift operator that is used as follows, B p X t = X t - p ; and e t = error term at month t , to forecast 24 months into the future the catches of Mullidae in Greek waters. The model is fitted to the 1964–1984 period, forecasts are produced for 1985 to 1986 and are compared with the actual catches observed during that period. Monthly catches are predicted reasonably accurately within an absolute percentage error (APE) range of 0–24.9% and a mean APE of 10.6% (7.5% for 1985 and 12.7% for 1986). Total observed annual catches in 1985 and 1986 were 3750 t and 4714 t respectively, while the model predicted 3993 t and 4189 t (APE 4.2 and 11.1%, respectively). As Mullidae resources are highly overfished in Greek waters, accurate forecasts can be used by resource managers for the preseasonal adjustment of Mullidae fishing mortality.

Comparative fixed gear studies in the Cyclades (Aegean Sea): size selectivity of small-hook longlines and monofilament gill nets

Fishing trials with monofilament gill nets and longlines using small hooks were carried out at the same fishing grounds in Cyclades (Aegean Sea) over 1 year. Four sizes of MUSTAD brand, round bent, flatted sea hooks (Quality 2316 DT, numbers 15, 13, 12 and 11) and four mesh sizes of 22, 24, 26 and 28 mm nominal bar length monofilament gill nets were used. Significant differences in the catch size frequency distributions of the two gears were found for four out of five of the most important species caught by both the gears (Diplodus annularis, Diplodus vulgaris, Pagellus erythrinus, Scorpaena porcus and Serranus cabrilla), with longlines catching larger fish and a wider size range than gill nets. Whereas longline catch size frequency distributions for most species for the different hook sizes were generally highly overlapped, suggesting little or no differences in size selectivity, gill net catch size frequency distributions clearly showed size selection, with larger mesh sizes catching larger fish. A variety of models were fitted to the gill net data, with the lognormal providing the best fit in most cases. A maximum likelihood method was also used to estimate the parameters of the logistic model for the longline data. Because of the highly overlapped longline catch size frequency distributions parameters could only be estimated for two species. This study shows that the two static gears have different impacts in terms of size selection. This information will be useful for the more effective management of these small-scale, multi-species and multi-gear fisheries.

Size selectivity of diamond and square mesh cod-ends for Nephrops norvegicus in the Aegean Sea

Abstract Trawl cod-end mesh selection for Nephrops norvegicus was studied in two experiments in the western Aegean Sea using the covered cod-end method (cover: diamond-shaped, 10 mm). The following cod-end mesh sizes were used: 14 mm (knot-to-knot) diamond-shaped (14D), presently operated in the Hellenic commercial trawl fishery; 20 mm square-shaped (20S); 20 mm diamond-shaped (20D), this being the mesh size proposed by the European Union (EU) for the entire Mediterranean. Selectivity estimates were derived from pooling the data over all hauls and sampling dates. The results showed that the proportion of the population retained was higher for 14D than for 20D and 20S (0.99, 0.79 and 0.84, respectively). The direct estimation of the 50% retention length for 14D was not possible because of the very small number of specimens escaping through that cod-end. The results also suggested that 20S has a higher and sharper selectivity than 20D (higher 50% retention length and selection factor, lower selection range) and retains fewer undersized individuals than 20D. Yet, the 50% retention lengths for 20D and 20S were both lower than the length at 50% maturity, a fact indicating that even the use of 20S and 20D may provide little opportunity for reproduction.