Effect of Water Flow Rate on Nursing Spat Mussels, Mytilus galloprovincialis (Lamarck, 1819)

Hatchery production of mussel seeds could be a solution to ensure reliable supply and introduce opportunities for selective breeding. To overcome the prohibitively expensive cost of hatchery-reared spat, it is necessary to optimize several factors in hatchery conditions. The water flow rate is among the main parameters for regulating the growth of nursing hatchery-reared juvenile mussels. In this study, three water flow rates of 2.5, 5.0 and 10 L.min-1 were compared to investigate their effects on growth and survival of mussel spats reared in an upwelling culture system. The results highlighted that spat mussels reared at the water flow rate of 10 L min-1 showed the highest performance in length, width and live weight, while juveniles deployed at water flow rates of 2.5 and 5 L min-1 exhibited the lowest (p < 0.05). At the end of the experiment, no mortality was recorded in all treatments. The size fraction of spat mussels larger than 8 mm significantly increased (p < 0.05) at water flow rate of 10 L min-1. However, the size frequency distribution of mussels from rates of 2.5 and 5 L min-1 was found to exhibit a similar pattern at the end of study (p > 0.05). After four weeks of nursing, at water flow rates of 2.5, 5 and 10 L min-1, size fraction > 8 mm were 3.5 ± 1.57, 4.4 ± 0.99 and 43.3 ± 1.22 % respectively. However, the fraction of spat mussels smaller than 6 mm significantly increased (p < 0.05) when the rate of water flow decreased.


Introduction
Mussel production is often subject to the spread of diseases, algal blooms, predation and lack of spat (Avdelas et al. 2021).Wild seed mussels are still gathered in many countries, using ropes or shells as new collecting techniques, or by harvesting from natural beds.This last practice is criticized because of the damage of bottom habitats and the possible food short-age for shellfish-eating birds (Kamermans et al. 2013).Obtaining natural supply of spat is highly seasonal and cannot always match the increasing demand of mussel farmers (Filgueira et al. 2007;Soria et al. 2014;Campbell and Hall 2019).Nowadays, hatcheries can complement the wild seed supply and may produce seed on a year-round basis under controlled conditions.The lack of mussel seed (spat), and low is among the main causes of the EU (European Union) mussel production decrease (Avdelas et al. 2021).Hatcheries also allow the development of genetic improvement through selective breeding (Kamermans et al. 2013).
In Morocco, the mussel production achieved has not increased enough to meet demand.To date, mussel culture has relied entirely on wild seed.Due to overfishing, the mussel beds tend to disappear on the exposed rocky shores thus making the manual collection of spat less efficient (Kassila et al. 2018).The uncertain status of shellfish recruitment in the foreseeable future requires the development of methodologies to ensure the sustainability of competent seed that can be collected at the time of need (Heres et al. 2022).Indeed, seed production from hatcheries is the subject of great interest in this country.A reliable supply of seed from hatcheries will allow mussel farmers to overcome the unpredictability of seed supply, but great investments are needed to promote hatchery production at a scale that would be economically viable.
The development of efficient settlement and nursery systems remains the main challenge (Kamermans et al. 2013).Nursing of shellfish spat produced from the hatchery depends on several factors.Temperature, food ration, stocking density and water flow rate are the most factors for regulating the growth of shellfish spat in nursery conditions (Tanyaros et al. 2012).
A good knowledge of the effects of water flow on upwelling culture systems will permit the production of higher addedvalue products with high specific growth rate and low mortality.The seeds should be transferred as soon as possible to the outdoor on growing areas in order to avoid the high costs of rearing them indoors (Kamermans et al. 2013).Flow rate is also important to mitigate other issues related to overstocking (Campbell and Hall 2019).For mussel spats, literature is scarce on their growth response to the influence of hydrodynamics.Previous research has focused on broodstock conditioning, spawning, fertilization, larvae rearing, water conditions, larvae concentration, food quality, food ratio and settlement systems, but less work has been done on nursery rearing of mussel spat up to seed (Kamermans et al. 2013).In this study, we investigated the effect of water flow in nursery conditions on the growth and survival of mussel (Mytilus galloprovincialis) spats, in relation to usual and common procedures used in shellfish hatcheries.The main objective was to determine the appropriate water flow rate for mussel spats under hatchery conditions.

Materials and Methods
The experiment was carried out from 27 October to 29 November 2021 at the AMSA Shellfish Nursery, National Institute of Fisheries Research, Tétouan, Morocco.The experiment was designed under a semi-closed recirculating system to determine the optimum water flow rate for nursing of mussel spats.The system consisted of water pump, a rectangular 1,200-L fiberglass tank for food storage, and 9 sets of cylindrical PVCs (50 cm in diameter) used as nursing units (Fig. 1).Each cylindrical PVC was kept in the 150-L cylindrical fiberglass tanks containing 140 L of 20-µm filtered seawater and drilled at 30 cm from the bottom.A screen of 1000-µm was fixed to the bottom of each nursing unit.During the experiment, water from the rectangular tank was pumped into the fiberglass tanks where the PVC units were placed.The water was injected so that it-upwelled into each PVC unit and then drained into the rectangular tank.An adjustable valve maintained a steady rate of water flow in each PVC unit.Three water flow rates were used: 2.5, 5 and 10 L min -1 in triplicate, and the water exchange (100%) was carried out every two days.
The Mytilus galloprovincialis juvenile used in this experiment had a mean (± SD) shell length (antero-posterior mesaurement), shell width (dorso-ventral measurement), shell width and individual wet weight of 3.23 ± 0.27 mm, 2.26 ± 0.30 mm and 1.8 ± 0.5 mg, respectively (n=150 spats).Linear dimensions of individuals were measured to the nearest 0.001 mm with the Profile projector (AAQ Pro 2, JFE Advantech Co., Ltd, Japan), and mean live weight was determined to the nearest 0.1 mg with a precision laboratory balance.Each PVC unit contained 10 000 spat mussels produced at AMSA hatchery, equal to a stocking density of 5 spats cm -2 .Spat were fed daily a mixture of Chaetoceros calcitrans and Tetraselmis suecica (1:1 ratio) at a rate of 20 cells µl -1 .This ratio is given in number of cells, i.e. 10 cells µl -1 of each species.During the experiment, the estimates of the cell density microalgae is made using Malassez haemocytometer with an optical microscope (40×magnification).These estimates were done daily in culture tanks for each species to determine the food rations.
Temperature and salinity were not regulated over the experiment period (ambient conditions) except for food ration, stocking density and food condition.Indeed, water quality parameters over the experiment period were as follows: salinity 36.3-36.6 ppt, temperature 19-20°C, pH 8.1-8.2, and dissolved oxygen 6.38-6.89mg L -1 .These parameters were measured weekly using multiparameter probe (AAQ-PRO2, JFE Advantech Co., Ltd, Japan).
Shell length, shell width and wet weight were determined weekly by taking random subsamples (n=30 spats) from each replicate.The length and width growth rates (in mm/ day) were calculated by the difference between the biometric values at the beginning of the experiment and the end of each week.The daily specific growth rate (SGR) was then evaluated for an individual as SGR = (In(Wf ) − In (Wf ))∕t , where Wf = final individual weight, Wi = Initial individual and t=duration in days (Malouf and Bricelj 1989).At the end of the study, subsamples (n=150 spats) from each nursing unit were graded according to three size classes: < 6 mm, 6-8 mm and > 8 mm, using sieves with mesh sizes of 6 and 8 mm.The number of surviving animals was also evaluated in these subsamples to determine mortalities.
One-way ANOVA was used to test the effect of water flow rate on the growth of M. galloprovincialis spat.The assumptions of normality and homogeneity of variance were previously tested with Shapiro Wilk and Levene tests respectively.The difference among treatment means was analyzed using Duncan's multiple range in the SPSS 21.0 software.Significance levels were set at p < 0.05.Data are presented as means ± standard deviation (n = 3 replicates).

Results
Highest increase in both length and width was observed (p < 0.05) at the water flow rate of 10 L min -1 by the 2 nd week (Figs. 2 and 3).However, there were no significant difference (p > 0.05) of mean shell length and width among rates of 2.5 and 5 L min -1 over the study period.After four weeks of rearing, at water flow rates of 2.5, 5 and 10 L min -1 , mean shell lengths were 5.04 ± 0.34, 5.15 ± 0.11, and 6.54 ± 0.38 mm with mean shell widths of 3.02 ± 0.15, 3.25 ± 0.18 and 4.21 ± 0.20 mm respectively (Figs. 2 and 3).
Furthermore, significantly greater growth rate (GR) in length and width was observed in mussels reared at 10 L min -1 compared to those reared at 2.5 and 5 L min -1 , from week 2 to week 4 (p < 0.05, Table 1).At the end of the experiment, at water flow rates of 2.5, 5 and 10 L min -1 , mean GR in shell length were 0.065 ± 0.004, 0.068 ± 0.004, and 0.118 ± 0.007 mm d -1 with mean GR in shell width of 0.027 ± 0.004, 0.035 ± 0.003 and 0.070 ± 0.009 mm d -1 respectively.
Moreover, the results highlighted that mean SGR in weight differed significantly among treatments (p < 0.05), with the highest increase at the rate of 10 L min -1 by the fourth week (Fig. 4).However, SGR was not significantly different (p > 0.05) between rates of 2.5 and 5 L min -1 .At the end of experiment, no mortality was recorded in all treatments.The size fraction of spat mussels larger than 8 mm significantly increased (p < 0.05) at water flow rate of 10 L min -1 (Fig. 5).However, size frequency distribution of mussels from rates of 2.5 and 5 L min -1 was found to exhibit similar pattern at the end of study with no significant difference (p > 0.05) among them.After four weeks of nursing, at water flow rates of 2.5, 5 and 10 L min -1 , size fraction > 8 mm were 3.5 ± 1.57, 4.4 ± 0.99 and 43.3 ± 1.22 % respectively.However, the fraction of spat mussels smaller than 6 mm significantly increased (p < 0.05) when the rate of water flow decreased (Fig. 5).

Discussion
The present study demonstrated that spat mussels reared at the water flow rate of 10 L min -1 showed the highest performance in length, width and live weight while juveniles deployed at water flow rates of 2.5 and 5 L min -1 exhibited the lowest.Optimal water flows are desirable because they regulate food supply, improve water quality and enhance dispersal of biodeposits (Campbell and Hall 2019).In agreement with previous studies (Malouf and Breeze 1977;Manzy et al. 1986), this study demonstrated that higher food supply (adjusted by flow rate) was positively correlated with increased growth rates.In addition, water flows can mitigate detrimental culture conditions, such as low dissolved oxygen levels (Weber et al. 2009;Campbell and Hall 2019).Higher water flows resulting in higher filtration rates of oysters have been reported by Lam and Wang (1990) and by Tanyaros (2000).However, above a certain level, an additional energy may be required to grade food particles and reduce feeding efficiency considering the presence of unusable suspended solids (Tanyaros et al. 2012).Excessive water flow can produce differential pressures among inhalant and exhalant openings resulting in feeding inhibition (Wildish et al. 1987).Thus, it seems promising to study the flow effects in the higher range only to determine inhibiting flows.
Generally, greater understanding of the hydrodynamic effect on upwelling systems allows for the management and optimization of food concentration, temperature, spat size and stocking density (Campbell and Hall 2019).Previous studies reported the negative effects of overstocking on growth of many bivalve species (Maguire and Burnell 2001;Velasco et al. 2009;Tanyaros et al. 2012).In this study, a suitable stocking density of 5 seed cm -2 was applied to avoid the negative effects of overstocking.
This study demonstrated that among the treatments tested, a sufficient water flow of 10 L min -1 was suitable for nursing hatchery-reared juvenile mussels (M.galloprovincialis) in a semi-closed recirculation system.Moreover, it was not expected that a water flow rate above 10 L min -1 would give highest SGR for juvenile mussels.Higher flows increase spat growth but could be limited by feeding inhibition and required energy inputs (Pfeiffer and Rush 2001;Appleyard andDealteris 2002). Tremblay et al. (2020) acknowledged that increasing flow rate also affects positively the physiological condition of scallop (Pecten maximus and Placopecten magellanicus) spat by a higher accumulation of total fatty acids in neutral lipid fractions.
A greater fraction of larger (> 8 mm) spats was set up at the water flow rate of 10 L min -1 , leading to a substantial competition for microalgae and oxygen with smaller juveniles (< 6 mm), which would be accompanied by higher levels of excretory products.Despite this, that resulted in Fig. 5 Size fraction of spat mussels after nursing (four weeks) at different flow rates (in%).Mean ± SD any mortality at the highest water flow rate.Mussel seed is usually considered suitable for transfer to ongrowing sites at a size up to 0.5 cm shell length.Tanyaros et al. ( 2012) reported that higher competition for food resulted in lower survival rates (85 ˜98%) of spat oysters (Crassostrea belcheri) at water flow rates ranged from 0.5 to 4 L min -1 .Unfortunately, literature is scarce on mussel feeding, survival and growth response to the influence of flow rate in hatchery conditions.
There was a differential response to flow rate among shellfish species in upwelling culture systems (Campbell and Hall 2019).This provide them an adequate environment with higher quality food and fewer costs of indoor rearing, since this practice requires pumping and heating of sea water and an expensive production of several species of microalgae.Indeed, we believe that high flows up to 10 L min -1 may increase energy costs.Moreover, hatcheries are the most expensive method to produce spat, with the price of mussels often being too low to make it economically viable (Kamermans et al. 2013;Carrasco et al. 2015;Figueroa and Dresdner 2016).In commercial hatcheries, the spat should be transferred as soon as possible to the outdoor on-growing area, when they reached a certain size (shell length 5-10 mm), since indoor rearing requires energy costs (pumping) and highly expensive production of microalgae.Given that the mean shell lengths reached 4.7 ± 0.3, 4.8 ± 0.2 and 6.36 ± 0.2 mm from week 3, at water flow rates of 2.5, 5 and 10 L min-1 respectively, the indoor period may be shortened by one week using the high flow-rate tested (10 L/ min).Nevertheless, further studies are encouraged in mussel culture to determine optimum flow rates at different stocking densities in relation to oxygen supply and waste removal.