The Bay of Biscay (Fig. 1) is an ecosystem in which the fishing fleets from eight EU member states are active (France, Spain, Germany, Denmark, United Kingdom, Ireland, the Netherlands and Portugal). In 2013, the French and Spanish fleets accounted for 93% of the total catches in this area (ICES 2014).

Many extended and diverse communities of commercial species can be found in the Bay of Biscay and surrounding waters (ICES 2014). In the demersal group, the most important commercial species are hake, megrim, anglerfish and sole. These species account for approximately 50% of the fish biomass.

Hake is important due to its abundance and economic value. It can live for as long as 20 years and reach a length of 140 cm and a weight of 15 kg. It reaches its sexual maturity at around three to four years of age. It is usually found at a depth of between 75 and 400 m. It tends to stay close to the seabed in daytime, leaving it to swim up the water column only at night.

There are two (management) stocks of hake in Atlantic EU waters. The northern stock (the stock targeted by the fleet analysed) is found in the North Sea, Skagerrak, on the Atlantic coast of the UK and Ireland, and in the Bay of Biscay.

The management of hake in the Bay of Biscay is based on a TAC and quota system. An MCRS of 27 cm is in place for the Atlantic waters (30 cm for in Kattegat and Skagerrak the MCRS is of 30 cm and in the Mediterranean of 20). Hake is normally a part of a mixed fishery with other demersal species such as anglerfish and megrim and pelagic species such as mackerel and horse mackerel.

The Spanish bottom pair trawlers operating in the Bay of Biscay are an exception to this multispecies characteristic. In the years 2011-2013, this fleet consisted of 10 vessels (5 fishing units, each consisting of two paired vessels) operating exclusively in the Bay of Biscay, with an average length of 38 m. Their average freezer storage capacity was 8600 (4300 per vessel) 12-kg boxes (the characteristic box type used by this fleet). On average, their fishing effort was distributed among 50 trips (per fishing unit and year). They captured approximately 8% of the total northern stock and 25% of hake catches in the Bay of Biscay.

The vessels in this group use a very high vertical-opening bottom net (MMS of 100 mm) and target mainly hake. This species accounts for 90% of the landings and approximately the same proportion of income (Fig. 2) of the bottom pair-trawler fleet. It can therefore be considered a single-species fleet; this means that under LO, the effort is not going to be limited by any other species. However, the effort constraint might arise from the fact that approximately 5% of the catches of hake (in weight) are individuals under the MCRS. In the past, the fish smaller than the MCRS were discarded. Furthermore, according to Rochet et al. (2014), 99% of hake discards consist of individuals under the MCRS. Under the provisions of the LO, from 2016 onwards, such fish must be landed and counted against the quota.

Before assessing the impact of a change in the MMS, the extent of the size change must be chosen. The five skippers who took part in the survey requested that the MMS be increased from 100 to 120 mm.

It should also be noted that this selectivity level is specific for hake, and the changes do not affect the catches of other species. This might be a problematic assumption in the multispecies fisheries, but the fleet analysed here targets only hake, so; this assumption should not significantly affect the results.

To simulate the selectivity change, it was considered that the catchability (q) can be decomposed into a product of the selectivity of the gear used and a parameter that incorporates the vulnerability, accessibility and availability of the fish, as described in the paper of Arreguín-Sánchez (1996). Mathematically,

where S_a,m,s stands for the selectivity of fish of age a related to the MMS; ms and r_a stand for the factors affecting catchability and are not related to the MMS. If the catchability and selectivity at age a for a given MMS are known, r_a can be calculated by applying Equation 1. It can be used afterwards to estimate the hypothetical catchability for the MMS, for which age selectivity is known.

Selectivity of the fish length for pair trawlers with 100-mm MMS has been estimated by The Spanish Oceanography Institute (IEO 2006). In the same study, the selectivity of 80-mm MMS was also provided. A logistic selection curve (Eq. 2) was fitted to the results for the 100-mm MMS.

where a and b are the parameters to be estimated. The results of the estimation were a=–6.53 and b=0.2. This curve has the property that the length for 50% retention r(L₅₀) is such that r(L₅₀)=0.5 and therefore L₅₀ =–a/b.

However, there are no studies providing the length selectivity for hake for pair trawlers with 120-mm MMS. Furthermore, several factors affect the size selection of the towed fishing gears for a given mesh size. These are the spatial and seasonal variations (Ozbilgin and Wardle 2002), gear design, netting materials and twine diameters (Herrmann 2005). Vessel-level factors also affect the cod-end selectivity (Tschernij and Holst 1999). All these factors explain the large variability in the results of size selection experiments with towed fishing gears.

To overcome these difficulties, the percentage change in L₅₀ between 80- and 100-mm MMS was calculated. It was used as a proxy of the L₅₀, keeping the shape of the curve (parameter b of Eq. 2) constant.

The results showed that the L₅₀ was 22.6 cm with an 80-mm MMS 34.6 cm with a 100-mm MMS. Consequently, it could be inferred that the L₅₀ for a 120-mm MMS was 40.8 cm. This last value was within the range of the expected L₅₀ (from 22 to 43 cm, according to ICES [2015]).

To fit the results into the age-structured dynamics of the simulation model, the selectivity-at-length curves were transformed to age using the Von Bertalanffy growth model (the Stock Synthesis III assessment model; Methot and Wetzel 2013) employed by the ICES assessment working group for hake (ICES 2014). The results of applying this procedure are presented in Figure 3.

A model coupled in all its dimensions (economic, biological and social) is required to perform the impact assessment of the change in the MMS. Economic results are related to the stock productivity, which can change depending on the retention pattern. Moreover, the stock productivity is also related to fishing effort. For example, if one fleet stops fishing, the overall productivity of the stock will change, simply because of the average selectivity changes.

The model used was FLBEIA (Garcia et al. 2013, Prellezo et al. 2016), developed in R (R Core Team 2014). The model projects several fleets and the likely effects on stock dynamics under different management scenarios.

Selectivity changes are likely to alter the productivity of the stock. Even if the removal quantities are similar, their different age compositions could change the age distribution in the stock. In particular, young individuals tend to have higher growth rates (Mellon-Duval et al. 2010). Thus, if the average retention size is increased, the productivity of the stock is likely to rise. This hypothesis was tested in the simulation.

Figure 4A shows no differences in the evolution of the stock for the MMS of 100 and 120 mm. This is because the selectivity change only affects 8% of the hake catches. Thus, the increase in the MMS from 100 to 120 mm did not have a significant impact on the stock biotic potential.

Though the evolution of SSB and fishing mortality did not change during the simulation (Fig. 4B), there were some changes in total catches of hake and in the size distribution of these catches (Fig. 5B, C).

In this fleet, the necessary effort to catch its share of hake (i.e. the quota) was not constrained by the capacity. As a result of the obligation to retain all the catches, extra storage was required (10% increase in the number of boxes). During an average trip, these fishing units catch approximately 30 t of fish (2500 boxes), with a maximum of 59 t (4900 boxes). The refrigeration capacity of one vessel is around 50 t (4000 boxes); the fishing unit has a refrigeration capacity of 100 t. This shows that the refrigeration storage capacity (meeting the safety requirements) does not limit the fishing effort within one trip.

Thus, the number of trips required to catch the TAC share were the same for 100- and 120-mm MMS. It also meant that the overall catches of hake were the same for these two mesh sizes (Fig. 5A). However, there was a difference between size compositions of the catches. The 120-mm MMS gear produced smaller catches of individuals below the MCRS (Fig. 5C). For the overall simulation period (2016–2020), these catches were reduced by 37%. Figure 5C shows that the landings of hake that could be sold for direct human consumption increased by 1.4% when 120-mm MMS were used compared with the 100-mm MMS. Furthermore, as the size structure of these landings was affected, the average price for 120-mm MMS catches was 0.5% higher than the price of catches conducted using 100-mm MMS gear. The use of 120-mm MMS raised the gross revenues (ex-vessel prices multiplied by the landings) by 2% (Fig. 5D).

However, fishing with 120-mm MMS nets required an increased level of effort (Fig. 6) because of the lower overall retention of the net. To obtain catches of similar size to those of 100-mm MMS, the effort had to be increased on average by 4.5% (for the same number of vessels for the period 2016–2020).

Figure 5A shows that the catches of hake are the same for the two MMSs. The landings of hake destined for human consumption are higher for 120- than for 100-mm MMS (Fig. 5B). Therefore, the gross revenues will also be higher. However, as can be seen in Figure 6, this result is obtained at the cost of an increased fishing effort, which implies an increase in variable costs. To determine which of these two factors has a stronger effect, two economic indicators were used: gross value added (GVA) and gross profit. GVA quantifies the value that a fishery is adding to the economy. It is comprised of the compensations received by the crew and by the capital (gross profit). The results for these two indicators are presented in Figure 7A and B.

The simulations showed that GVA was larger for 120-mm MMS than for 100-mm MMS. This causes an increase in the gross revenue, which is higher for 120-mm MMS net use (Fig. 5D). However, variable costs of the effort required to obtain these landings (crew costs, fuel costs and other variables costs, Table 1) were also higher for the larger MMS. The overall result shows that the increase in revenue is sufficient to compensate for the extra costs, generating an overall increase in the GVA of 1.5% per year. The crew compensation will also increase because they receive a percentage of the gross revenue. Overall, the use of 120-mm MMS resulted in a crew share increase of 2% per year.

The other component of GVA is the capital compensation, which can be illustrated using profit as an indicator. Profit does not change when the 120-mm MMS is used (Fig. 7A), because a third of the extra gross revenue goes, through the crew share, to the crew. Thus, from the point of view of the capital owner, it is not worth increasing the MMS, or at least there are no financial incentives for this change.

The productivity of fishing effort should also be considered. Capital productivity, measured in terms of the gross profit per fishing day, was 5% lower for 120-mm MMS than for 100-mm MMS (Fig. 7C). Crew productivity (crew compensation per fishing day) was also lower (2.5%) for the larger MMS (Fig. 7D). Though the total crew compensation is higher when using this MMS, it is obtained by increasing the number of fishing days, with a reduced compensation per day.

In summary, the results show that increasing the MMS from 100 to 120 mm does not change the stock biotic potential, at least when this increase is implemented in the fleet studied. However, the productivity of the effort is reduced, in terms of both catches per unit of effort and the GVA and profit per unit of effort. The total capital compensation was unchanged, so there were no private economic incentives to implement the MMS change. The MMS increase produced more hake landings above MCRS and higher average ex-vessel prices of hake, which created an increase in revenues and GVA.

As pointed out by Graham et al. (2007), a gear design solution might be more acceptable to fishers than a spatial or temporal closure. The answers obtained from the interviews with the skippers are in agreement with this notion. The MMS increase, as a way to improve the selectivity of the fishing gear, is a measure that can be implemented when the choke effect is caused by the size composition and not by the species conflict. However, increasing the selectivity of the gear is not straightforward. For example, in the same fishing area, the inclusion of a top square mesh panel has not produced significant changes in the size composition of the catches (Alzorriz et al. 2016). Furthermore, the results from the simulation show no capital-based incentives to increase the MMS.

Before an economic impact assessment, an evaluation of the effects of a change in selectivity on the biotic potential of the stock should be conducted. This evaluation should be integrated and coupled with the economic analysis. The result of the simulation presented here shows that, from the fishery community point of view, there would be no evolutionary change due to the new selection pattern. The study of Hilborn and Minte-Vera (Hilborn and Minte-Vera 2008) shows that selectivity is not the main aspect of the growth of the stocks. The results of the current simulation confirmed this observation; however, it must be taken into account that only 8% of the fishing mortality was considered.

The simulation did not deal with the impact on the ecosystem. Gear-related conservation measures assume that the escaping fish survive and support the relevant population. However, there are no reliable estimates of post-capture survival of hake (Suuronen and Sardà 2007). The overall impact on the ecosystem is not easy to anticipate. Neither the changes in the ecosystem productivity (Sardà et al. 2015) nor the effects of these changes on the fleet productivity (Garcia et al. 2011, 2012) predict ecological perturbations in a precise way. Further research in this field is urgently needed.

In the analysed fleet, an increase in MMS reduced the efficiency of the system (in terms of catch per unit of effort). The desired effect of the regulation is to change the size distribution by reducing the catches of individuals under MCRS. This change should also increase the average price. This was, indeed, the case in our simulation, although this result was the consequence of the change in the MMS and not its cause (Asche et al. 2015). Nevertheless, this result might create an incentive for using gears with increased selectivity.

In the simulation presented here, a rise in selectivity implies a lower retention of individuals below the MCRS of hake. The overall effect is that more of the landings are destined for human consumption but at the cost of increased fishing effort. When more effort is exerted, two possible limits have to be considered. The first limit is a capacity ceiling (no more fishing days are available, or no more vessels can be made). We did not observe this problem in the current simulation. The second potential limit is associated with the species caught, of which some might be constrained by the LO. In the case analysed, the fleet is “almost” a single-species fleet, so such situation is unlikely. The results show that the fleet using an increased-MMS gear needs more effort to catch their hake quota. No onboard storage limits were observed. However, the storage problems might occur in different fleets, and an increase in the number of trips might be necessary to land the same amount of fish for human consumption.

The observed 2% increase in gross revenues clearly differs from the predicted general 10% to 15% reduction (Villasante et al. 2016). This discrepancy can be explained by the single-species character of the analysed fleet. However, we also observed an increase in the required fishing effort and, hence, a rise in the variable costs. Overall, the private financial results (profits of the capital owner) are not improved by using a larger MMS. This result has been anticipated by Skonhoft et al. (2012), whose study concludes that a policy enforcing a more selective fishing gear can either reduce or increase the total profitability of the fishery.

Finally, the results showed that the GVA and human consumption hake supply increased after the enlargement of the gear MMS. This means that society has the potential capacity to compensate the productivity loss of the owners of the labour and the capital.