Trammel net selectivity studies in the Algarve (Southern Portugal), gulf of Cadiz (Spain), Basque country (Spain) and Cyclades islands (Greece) - 1999-2001. DG XIV, Ref. 98/014

Objectives and outline of the project

The objective of this project is to study trammel net gear used in the small-scale fisheries in southern Portugal, northern Spain (Bizkaia), southern Spain (Gulf of Cádiz), and Greece (Cyclades Islands).

The specific objectives are as follows:

Preliminary surveys and interviews will be carried out in order to fully characterise the different métiers in each area, to describe fishing strategies, and to obtain estimates of the quantities of trammel net gear in use.

Parallel selectivity experiments will be carried out by four teams in the four different areas. In all four areas the experimental nets will consist of three mesh sizes of the small mesh inner panel (e.g. 60, 70, and 80 mm stretched mesh), and two mesh sizes of the larger mesh outer panels (e.g. 300 and 400 mm stretched mesh), giving a total of six possible combinations. The mesh sizes to be used in each area will depend on the characteristics of the gear used locally. Commercial fishing boats will be hired and normal fishing practices will be followed. Selectivity trials will take place over a 12 month period, with 10 sets per season, giving a total of 40 sets in each area.

In comparison with other fixed or static gears, relatively little is known of the species and size selectivity of trammel nets and of their impact on living resources. Thus, these studies will contribute new and important information for the management of multi-species small-scale fisheries. The proposed research is a natural continuation of a series of studies in Portugal, Spain and Greece on fixed gears used in small-scale coastal fisheries.

Summary for non-specialists:

Trammel nets are fixed gear, generally fished on the bottom, that are composed of a small mesh inner panel and two large mesh outer panels. The netting material can be of several types, with monofilament and multifilament trammel being the most widely used. The netting panels are hung loosely and the objective to catch fish, cephalopods and crustaceans primarily by entangling in the pocket formed when a fish/cephalopod/crustacean forces the small mesh inner panel through one of the meshes of the outer panel.

Trammel nets are widely used throughout the world, particularly in small-scale fisheries. Unlike gillnets, few studies have been carried out on trammel nets and little is known of their size-selective properties. The objectives of this study were the following: 1) to identify and characterise the main trammel net metiers in the Cantabrian Sea (Basque Country, Spain), the Algarve (southern Portugal), the Gulf of Cádiz (Spain) and the Cyclades Islands (Greece), 2) to construct experimental trammel nets based on the most important métier in each area and to carry out fishing trials with these nets, 3) to quantify species composition, catch rates, discard rates and method of capture on a seasonal and a depth basis, and 4) to estimate the size selectivity parameters for the most important species.

Questionnaire surveys were carried out in each of the four areas. These allowed us to identify up to 9 different métiers in each area and to identify the characteristics (material used to construct the nets and method of construction) as well as to obtain information on vessel and crew characteristics, fishing grounds and fishing strategies.

Experimental trammel nets using 3 mesh sizes for the inner panel and 2 mesh sizes for the outer panel were constructed in each area, giving a total of 6 trammel net types for each area. Commercial fishing vessels were chartered for carrying out fishing trials. Members of each team went on board each trip in order to sort the catch according to the six trammel net types and to record information on fishing grounds, depth, number of fleets, times of setting and hauling, method of capture and discarding practices. All organisms that were caught were identified, measured and weighed. At least 40 fishing trials were carried out in each area, over all four seasons in the case of the Cantabrian Sea, the Algarve and the Cyclades Islands. In the case of the Gulf of Cádiz, fishing took place in only two seasons because of the seasonal nature of the trammel net fishery.

A total of 271, 360, 185 and 185 km of trammel nets were fished in the Cantabrian Sea, the Algarve, the Gulf of Cádiz and the Cyclades Islands respectively. Species diversity was very high, with 79, 128, 63 and 79 different species caught in the Cantabrian Sea, the Algarve, the Gulf of Cádiz and the Cyclades Islands. Overall, 17041, 16574, 8178 and 9619 fish/crustanceans/cephalopods were caught in the four areas, with the great majority of the catch consisting of fish.

Relatively few species dominated the catches in numbers in all four areas. In the Cantabrian Sea trials Solea vulgaris (19%), Trisopterus luscus (12%), Scomber scombrus (9%) and Trachinus draco (8%) dominated the catches. In the Algarve, Scomber japonicus (21%), Sepia officinalis (17%), Microchirus azevia (12%) and Trachinus draco (6%) were the most important species in the trammel net catches. Sepia officinalis (43%), Solea senegalensis (8%), Sardina pilchardus (7%) and Torpedo torpedo (7%) accounted for most of the catch in the Gulf of Cádiz. In the Cyclades Islands, Mullus surmuletus (15%), Pagellus erythrinus (14%), Diplodus annularis (10%) and Scoraena porcus (9%) were the four numerically most important species.

Significant numbers of species and proportions of the catches were discarded in each area. Overall, 65, 105, 46 and 32 species were entirely or partly discarded in the Cantabrian Sea, Algarve, Gulf of Cádiz and Cyclades Islands respectively. The overall discard rate in terms of catch in numbers ranged from 15% for the Cyclades to 49% for the Algarve, with the high discard rate for the latter due largely to Scomber japonicus. The main reasons for discarding were as follows: 1) species of no or low commercial value (e.g. chub mackerel, Scomber japonicus; Torpedo ray, Torpedo torpedo), 2) commercial species that were damaged or spoiled (e.g. hake, Merluccius merluccius), 3) undersized commercial species (e.g. angler fish, Lophius piscatorius ) and 4) species of commercial value but not caught in sufficient quantities to warrant sale (e.g. sardine, Sardina pilchardus). The number of discarded species and discard rates generally decreased with increasing inner panel mesh size.

The size of the outer panel had little or no effect on the catch rate, discard rate or the sizes that were caught. In contrast, catch rates and discard rates generally decreased with increasing inner panel mesh size. Also, although for many species the catch size distributions were highly overlapped, indicating that the different trammel nets were not size selective, in some cases the distributions for the different inner panel mesh size were clearly different. Thus, for species were there was evidence of size selectivity, this was largely or entirely due to the mesh size of the inner panel.

Most of the distributions were skewed to the right, showing that trammel nets catch a wide size range for most species. Furthermore, many distributions were bi-modal, with two peaks. This suggests that fish were caught in more than one way. In fact analysis of the method of capture showed that while entangling was by far the most important method, there were many species where fish were either entangled and wedged or entangled and gilled.

Selectivity models were fitted to the catch size frequency data for the most abundant species in each area. Unlike the case of gillnets were entangling is unimportant, uni-modal (one peak) models generally gave poor fits to the trammel net catch frequency data. In fact, the bi-modal model was the best for most species. However, in the case of species such as the cuttlefish (Sepia officinalis) where only entangling was important, the bi-modal model also did not fit the data. Instead, size selectivity was more adequately described using the logistic model, which is normally used for trawl or other bag-type gear. This suggests that entangling by formation of a "pocket" of the small mesh inner panel passing through the larger mesh outer panel, ensures that above a certain size all individuals are retained and there is no loss of the larger individuals.

The multivariate analysis showed that there were significant differences in the species composition catches and the relative abundance of the most important species in each area, particularly in the Cantabrian Sea and the Gulf of Cádiz. In addition to the seasonal effect, depth was also an important factor. Fishing trials took place at depths to 80 m in the Cantabrian Sea and the Cyclades Islands, 100 m in the Algarve and to 20 m in the Gulf of Cádiz. For many of the most important species, catch rates in numbers per 1000 m of trammel net differed according to depth stratum, showing that different species have different depth distributions.

This project has provided important new information on trammel net fisheries in four areas in Portugal, Spain and Greece. This information will be useful for the improved understanding, management and conservation of the multi-species, multi-gear fisheries that are characteristic of the small-scale fisheries in southern European waters.

The report is available in PDF under request