How do mussel provenance and spat size affect mussel aquaculture performance in the South-Western Mediterranean (Algeria)?

In recent decades, aquaculture production has served as an essential source of protein for a large portion of the world's population. An ever-increasing demand for aquatic products suggests that aquaculture will continue to be one of the fastest-growing sectors for protein-based food production (FAO 2016). Therefore, the rapid development of aquaculture has increased the need for sustainable production strategies (Fuentes-Santos & Cubillo 2015). In this context, integrated multitrophic aquaculture (IMTA) has demonstrated a reduction in the environmental footprint and has increased the profitability of farming (Chatzivasileiou et al. 2022). This innovative mariculture methodology combines species cultivation from different trophic levels, simulating a natural food web (Chatzivasileiou et al. 2022). As commercial aquaculture is a profit-orientated activity, achieving control over the costs and returns is a key element in optimising production (Asche et al. 2008, 2012; Theodorou et al. 2014, 2020, 2021). It is therefore possible to analyse the performance of a given production process and to propose improvements to increase its efficiency.

According to the latest statistics from the Food and Agriculture Organization of the United Nations, mariculture and coastal aquaculture production reached 30.8 million tonnes (equivalent to 106.5 billion aquatic animals) in 2018 (FAO 2020). A significant part of its production is of bivalve origin (i.e. clams, oysters, mussels and other species) (FAO 2020). Their production has been increasing considerably with consumer demand (Prato et al. 2019). In the Mediterranean countries, they are considered a rich food source and a healthy balanced diet (Cherifi et al. 2018; Dridi et al. 2007). Their cultivation continues to play an important role in providing food for the growing world population (Landmann et al. 2019). For example, Mytilus galloprovincialis is the main species used in mussel farming on the French, Spanish, Italian, Greek, Tunisian and Algerian coasts (FAO 2016). In order to increase its production from less than 150 metric tonnes in 2013 to 7600 tonnes in 2025 (MPRH 2008), the Algerian Ministry of Fish and Fisheries Resources designed a strategic plan to sustainably develop marine aquaculture (MPRH 2008). This programme aims to establish 56 new farms of this species along the Algerian coast between 2015 and 2025.

Several scientific studies have been carried out to improve the production of mussels in offshore systems. Some authors were interested in comparing the farming production (Camacho et al. 1995; Okumus & Stirling 1998; Fuentes et al. 2009), taking into account response variables such as growth (Fernandez-Reiriz et al. 1996; Hatzonikolakis et al. 2017) and condition index (Okumus & Stirling 1998). Guillou et al. (2020) studied the commercial performance of the blue mussel in Canada. Theodorou et al. (2014) reported that the optimal farm size for mussel production of the Mediterranean mussel in Greece is a larger than 3 ha. Other researchers focussed on improving aspects of grow-out techniques in order to reduce losses, by minimising mussel thinning practices during the grow-out period (Pérez-Camacho et al. 2013), varying the seeding density (Cubillo, Peteiro, Fernández-Reiriz, et al. 2012b), grading mussels by size during the initial seeding (Lauzon-Guay et al. 2005; Supono et al. 2020), adjusting the spacing of mussel ropes (Drapeau et al. 2006) and providing a period of feeding and conditioning before transferring them to coastal farms (Sim-Smith 2006; Carton et al. 2007).

Several studies have demonstrated that PE and CI are the main tools for the commercialisation of bivalves, depending generally on endogenous and exogenous conditions (Lagade et al. 2015). They provide information about the physiological state and growth of the organisms (Andral et al. 2004), which is of great interest for harvesting purposes, e.g. to indicate the market value (Martinez et al. 2018). The physiological implications of mussel size on shell thickening or byssus attachment strength, together with the ecological implications driven by site-specific heterogeneity, will therefore play a key role in mussel performance in a widely exploited area.

In Algeria, mussel farming data are almost unavailable, apart from a study by Laama and Bachari (2018) on evaluating site suitability for the expansion of mussel farming and one by Lourguioui et al. (2017) on the environmental impacts of aquaculture. Offshore aquaculture remains a very recent activity with little research into the characterisation and improvement of farming production. Specifically, there is a lack of data on the results of mussel production optimisation due to a lack of scientific studies, monitoring and follow-up systems of mussel production in Algeria. Furthermore, there is a lack of knowledge about the effects of different spat sizes and origins on the growth of M. galloprovincialis and the biomass produced in a mussel culture system in Algeria. This work aims to acquire the information on mussel farming necessary for the management and optimisation of M. galloprovincialis in Algeria. It is the first study to develop best farming strategies for optimising the production of this species in the study area, and the only one that deals with optimally growing in a suspended culture system on a microgeographic scale. The objectives of this paper are 1) to study the main production processes governing mussel farming activity, 2) to estimate the variation of relative biomass production (RBP) in a mussel farming system and 3) to analyse biomass production gain and loss based on spat size and origin in order to understand their effect on productivity. For the analysis, we used data from Cultmare farm on three initial spat size classes and four different sources, over three years (2017–2020). The experiment was strictly conducted according to commercial culture methods to ensure that our findings from the study are applicable to improving mussel culture and contribute to the development of an IMTA methodology suitable for use in the oligotrophic Western Mediterranean.

The production generated by different combinations of stocks at a single farm site, using the traditional techniques and criteria of mussel producers, has never been evaluated in Algeria. This study investigated the growth performance of mussels grown from spat from four different sources in three different initial seeding size categories. The density was controlled and did not exceed 1200 ind/m, depending on the initial seeding size, in order to eliminate any smothering effect. After three years of mussel farming, the yield expressed as APP was inversely proportional to the initial size. It differed significantly from one stock to another, favouring small spat, with a maximum production of 4.3 kg harvested per kg seeded. These results are consistent with previous work from seeding to harvest. An average of 1.5–2.5 kg harvested per kg seeded of M. edulis was reported in the Wadden Sea, and often less than 1 kg harvested per kg seeded in Ireland (Calderwood et al. 2015; Wijsman et al. 2014). Along the Moroccan Atlantic coast, reported relative biomass production shows more variation for this species (2.8–5.7 kg harvested per kg seeded) (Idhalla et al. 2017). For M. galloprovincialis culture, a maximum of 7 kg harvested per kg seeded has been modelled (Ferreira et al. 2007a).

The production cycle of mussels depends on their growth and optimal harvesting conditions (Petersen et al. 2020), including local environmental conditions such as temperature, salinity (Bayne & Worrall 1980; Fuentes et al. 2000; Karayücel & Karayücel 2000; Waite et al. 2005), concentration of food particles (Camacho et al. 1995; Celik et al. 2009; Filgueira et al. 2008; Lök et al. 2007; Waite et al. 2005; Hatzonikolakis et al. 2017), seeding density and farming system design (Drapeau et al. 2006; Ferreira et al. 2007a; Lauzon-Guay et al. 2005; Raman-Nair & Colbourne 2003; Strohmeier et al. 2008). In fact, the mussel density on a rope will decrease as the mussels grow in size and take up more space (Bonardelli et al. 2019). In this study, the final biomass production decreased as the spat size increased. The percentage of biomass lost was greatly compensated for by the gain in biomass for small mussels, followed by a remarkable gain from the medium-sized spat. This stands in contrast to large mussels, in which there was almost no difference between the seeded and harvested weights. The growth rate of the small mussels in suspension was higher than that of the large mussels (Lauzon-Guay et al. 2005) and the growth rate was slower in older mussels (Lauzon-Guay et al. 2005; Lök et al. 2007; Seed 1969). Although smaller mussels may be more sensitive to the effects of desiccation or other stressors occurring during their transfer from harvest to the nursery site (Webb & Heasman 2006; Jenewein & Gosselin 2013) and may have higher losses (Lauzon-Guay et al. 2005), they also have a higher potential for biomass production (Petraitis 1995). The existence of high inter-individual variability in growth is a major problem for aquaculture and positions bivalves as prime candidates for size-based selective breeding programmes to increase the production of shellfish farms (Fernández-Reiriz et al. 2016). Although the difference in overall production between the four spat origin sites was not statistically significant in this study, the variation in the rates of biomass gain and loss was significant depending on the source, showing the best values for the Tlemcen mussels. Mussels from the on-site collection were not affected by a change in their living environment. They experienced the stress of the grow-out process for the first time, after attaching to the collectors of the farm. The spat from the other three sources was placed into water after remaining out of water for more than 24 hours during transport, selection and socking. Transfer-associated factors such as emersion, desiccation, temperature variations and fasting have been shown to affect the fitness, behaviour and subsequent losses of juvenile mussels (Calderwood et al. 2014; Carton et al. 2007; Theodorou et al. 2017, 2018; South et al. 2020). Contrary to our expectations, the mussels from the study site were not the most efficient in terms of biomass yield, though they did compensate with the biomass gained. The biomass loss rate was the highest from this source, testifying to their fragility.

In this study, the Tlemcen mussels – followed by those from Tenes – showed more efficient growth and physiology for a commercial size, reflected by greater variation in the APP compared to the other two sources. This is in agreement with the study by Phillips (2002). The variation in growth can be explained by the nutritional status of the juveniles, as their native environment was very rich in suspended matter, due to artificial feeding and waste from the floating fish cages cultivated near the mussel lines. Thus, food supply seems to be the principal factor that can limit production in suspended culture systems (Babarro & Zwaan 2008; Navarro et al. 1991) absorption efficiencies and metabolic rates. The availability of food and the nutritional status of the mussels have been described as some of the many factors influencing production (Babarro & Zwaan 2008; Clarke 1999). Moreover, growth is also influenced by a genetically determined trait (Prieto et al. 2020). Consequently, growth between populations can vary markedly, even when the individuals are exposed to identical environmental conditions (Prieto et al. 2018). The large variation in APP can be also attributed to the genetic characteristics that make some mussels more efficient and resistant to environmental changes than others. This also may have resulted from adaptation to regional environments in which animals have developed stock-related traits of genetic variation (Camacho et al. 1995; Pace et al. 2006; Wang et al. 2012). As mussels grow, their requirements for limitied resources increase and intraspecific competition becomes more pronounced (Cubillo et al. 2012a).

In this study, the highest APP was recorded for the Ain Tagourait mussels, but this can only be due for some plots where the yield was maximum. On the other hand, the variation in global production of this source has not been efficient compared to others, meaning that inter-individual differences are present even among the same stock origin and size, and that mussel growth can vary despite all breeding and transfer conditions being the same. This hypothesis has been investigated in a few bivalves (Fernández-Reiriz et al. 2016; Tamayo et al. 2011) and is supported by physiological experiments on Mytilus galloprovincialis and Ruditapes philippinarum (Fernández-reiriz et al. 2016; Tamayo et al. 2011). At the medium seeding size, the biomass yield showed a small variation in gain, between 33% and 15% on average for all sources, in favour of those from the study site. This suggests that medium-sized mussels are less fragile than smaller ones at the same site. Additionally, the resistance was not developed at this size, as it was with the large mussels. This category of large individuals from Tlemcen and Ain Tagourait lost their initial biomass. The production seems to be balanced between the small mussels’ fragility and the large ones’ resilience and lack of adaption to new environments.

In the present study, a significant difference in the marketability indices, the STI, TW and PE was noted between three sources (the study site, Tlemcen and Tenes), in favour of the Tlemcen mussels, followed by those from the study site. This confirms that certain stocks will be more efficient than others, even at the same site, and that the mussels have as a genetic trait the ability to adapt to changes in the environment when they have good conditions in the larval state. However, the mussels from Tlemcen were genetically more robust and developed genetic plasticity, as they were born in a rich environment under IMTA conditions. They were implemented in the study site and they exhibit better production efficiency, confirmed that mussels in proximity to multiculture farms have more flesh in their shell than those growing on a typical mussel farm, as reported by Chatzivasileiou et al. (2022). The authors indicated that the areas near fish farms may be an exception for the oligotrophic Mediterranean, due to the significant amount of nutrients released from fish cages (Chatzivasileiou et al. 2022). According to Wenne et al. (2022), seascape genetic analyses suggest that a complex mix of environmental variables help explain the genetic variation in M. galloprovincialis populations within the Mediterranean Sea, which most likely reflects the complex geological history of formation, isolation and reconnection among the regional sub-basins of the Sea. These authors also analysed one population from the west of Algeria (Oran) and identified it as being intermediate between the two main groups from the Mediterranean Sea and the Atlantic Ocean. As Tlemcen is on the western coast of Algeria, mussels from this area are more likely to have similar characteristics and may be influenced by the Atlantic form of M. galloprovincialis, whereas mussels from sites east of Oran – Tenes, Tipaza (the study site) and Ain Tagourait – most likely represent populations of M. galloprovincialis that have been influenced by the Mediterranean (Simon et al. 2021; Wenne et al. 2022).

The STI and the marketability indices showed significant differences in the length of individuals between small and medium seeded mussels, in favour of the small spat, and confirmed that growth is faster in small individuals, slowing as size increases. Also, their rapid adaptation to environmental changes makes them more efficient. However, the larger mussels were resistant to a different environment at the beginning of their transfer, especially those in the advanced stage of the reproductive cycle. Stressing them at this stage causes a significant part of their body reserves to be lost. This can be clearly seen in the changes in CI and PE and in the evolution of the TW of the mussels across the initial size categories. Differences in shell morphology – the shape or thickness of bivalve molluscs – are influenced by key environmental parameters such as food competition, substrate type, crowding, temperature, wave impact and the activity of predators (see Alunno-Bruscia et al. 2001; Beadman et al. 2003; Steffani & Branch 2003; Valladares 2010). Some growers use the stress method of shaking socks every month. This method develops resistance in the mussels, which leads to a longer life. This stress slows down the growth in the flesh and encourages the mussels to use their reserves to thicken and harden their shells, as well as to manufacture byssal threads, which reduces the loss rate at the end of the cycle (Personal communication. M. Khoudja, mussel grower in the centre of the Algerian coast).

Capital expenses have not been included in this work (e.g. buildings, boats, vehicles, equipment and land travel), nor have expenses related to licences, taxes, insurance and quality analysis, as they are too variable and depend on the governance of the farm. These expenses should be included by individual producers in order to obtain a true manufacturing estimate. However, our productivity and revenue estimates for each mussel stock have been standardised on the basis of spat supply from a long line of collectors, so that commercial performance is linked to the characteristics of the origin of the stocks and their biological characteristics, density and spat size.

This study focuses on the aquaculture of M. galloprovincialis in the central Algerian coast, and it contributes to the knowledge of mussel farming. After investigating mussel production according to seeding size and spat origin, it was found that production was greatest in the small seeded mussels. Mussel origin is also an important parameter for selecting the optimal stock. Mussels from sites east of Oran – Tenes, Tipaza (the study site) and Ain Tagourait – most likely represent populations influenced by the Mediterranean, in comparison to those from Telemcen, which were influenced by the Atlantic form of M. galloprovincialis. They have also developed genetic traits for environmental sensitivity and seem more robust, as they were born in a rich environment under IMTA conditions.

Mussel farming does not seem to be closely monitored, as no strategic management system has been established to ensure efficient cultivation. According to several aquaculture farmers who were asked about the strategy of management, all of them confirmed that they do not perform any studies to follow their production. Their cultivation does not take into account the physiological aspects of the organisms or the factors that influence their growth and productivity. As no data collection or statistical analysis are available at any institution, including the Ministry of Fisheries and Fish Resources, we recommend the creation of a database of aquaculture. As Cultmare is the first farm to analyse the production of each lot of mussels from its beginning, we encourage other farmers to perform this kind of study and to use IMTA methodology, which is suitable for use in the oligotrophic Western Mediterranean. In order to improve knowledge on the biology and production cycle of Mytilus galloprovinciallis, it is important to establish a production calendar to better manage mussel farming. It is also recommended to use spat originating near fish farms because of the environmental enrichment in nutrients, which helps ensure that the spat is suitable at the beginning of the mussel production. Moreover, the construction of new hatcheries specialised in spat quality for aquaculture in Algeria is essential to ensure better larval nutrition and optimal production by shellfish farmers. Also, further studies on shell fragility and spat resilience to new environments will complement and enrich the existing knowledge and improve aquaculture practice in Algeria; for instance, we suggest investigating how shaking socks can impact mussel culture and improve shell thickness. This can be performed by studying the structure and molecular composition of the byssus from M. galloprovencialis and analysing environmental parameters related to shell thickness and mussel genetic plasticity that act as stressors and positively affect mussel production.