During the last 50 years, eutrophication has been identified as one of the main problems affecting aquatic ecosystems (Karydis and Kitsiou, 2012). This nutrient excess has provoked massive microalgae blooms that have severe consequences in the ecosystem [hypoxia, decrease of water clarity, fisheries decline, loss of biodiversity, etc (Smolders et al., 2006)]. In marine environments, coastal waters have been more susceptible to it due to their proximity to heavily populated areas, especially in marine lagoons and other paralic environments where large water renewal time makes them more vulnerable to nutrient enrichment (Kennish and Paerl, 2010).

The Mar Menor (MM) lagoon, with its 135 km² and 645 hm³, is the largest coastal hyperhaline lagoon in the Mediterranean basin. A constant rise of chlorophyll a level has been observed in the lagoon since the 1980s (Ruiz et al., 2020; IEO-CSIC, 2023), with the threat of eutrophication beginning in the early 1990s with the nutrient increase noted when high peaks of chlorophyll a were reported from the area close to the Albujon ravine, which provides the main freshwater supply of the lagoon (Pérez-Ruzafa et al., 2005; Ruiz et al., 2020). The lagoon suffered for decades from the constant increase of nutrients coming in, together with other pollutants, until its recent ecological collapse in 2015, when a massive phytoplankton bloom deteriorated strongly the environmental conditions and erased most of the macrophytes and benthic fauna (Ruiz et al., 2020). The eradication of the natural ecological community destabilized the lagoon, reducing its resilience and hence making it more vulnerable against future perturbations, as has been observed in several ecological crises after 2015, including anoxic episodes (Álvarez-Rogel et al., 2020; Ruiz et al., 2020; IEO-CSIC, 2023).

Bivalves and the ecosystem services they provide have been used to mitigate the effects of eutrophication and recover the ecosystem functionality as a nature-based solution (Rose et al., 2014; Galimany et al., 2020). Bivalves are filter feeders that fed on suspended particulate matter such as phytoplankton, detritus, micro-zooplankton, bacteria and dissolved organic matter (Gosling, 2015), thus improving water clarity and quality by extracting nutrients and other pollutants (Ismail et al., 2014; Ayvazian et al., 2021). Moreover, bivalves such as mussels and oysters are ecosystem engineers, creating biogenic structures or reefs, which help to improve biodiversity by acting as a natural hatchery and habitat for many species and providing coastal protection (Smaal et al., 2019; zu Ermgassen et al., 2020b).

In this context, The Mar Menor Oyster Initiative (MMOI, https://noraeurope.eu/spain-the-mar-menor-oyster-initiative/) was launched in 2020 at the Native Oyster Restoration Alliance (Pogoda et al., 2019) 3rd Conference. The MMOI aims to gain knowledge on the nutrient extraction capability of the MM flat oyster (Ostrea edulis) and develop the necessary tools for restoration and bioextraction actions. The selection of the flat oyster as the bioremediation species was based on the actual presence of the species in the lagoon and the existence in the 1980s–1990s of a population over 130 million oysters (Rosíque, 1994). This population vanished through the years because of the degradation of the environmental conditions, the negative effect of the sponge Cliothosa hancocki, and the massive proliferation of the invasive macroalgae Caulerpa prolifera (Taxon Estudios Ambientales SL, 2006).

The last assessment of the state of the MM oyster beds was carried out in 2006 (Taxon Estudios Ambientales SL, 2006), and no sampling has been carried out since then. This study estimated from a partial sampling that the population had decreased by more than 10-fold compared to the 1990s. Currently and according to the observations of divers, the population has practically disappeared, so we considered that it would be very difficult for it to be able to recover on its own.

Thus, the main challenge faced by MMOI was the limited availability of individuals to study and to use in restoration/bioextraction actions, which implied the need to produce hatchery seed using aquaculture techniques. In terms of biosecurity, the best recommendation for seed production is to use local broodstock and water (Fitzsimons et al., 2020; zu Ermgassen et al., 2020c). With the purpose of using local breeders, Albentosa et al., (2023) collected some adult oyster specimens from the lagoon in order to determine their potential nutrient extraction capacity, species identity, and genetic diversity and their proximity to other flat oyster populations in Europe. This study concluded that the species was O. edulis, which exhibited a genetic diversity similar to those described for other European oyster beds and were genetically related to the Mediterranean oysters near the MM (Albentosa et al., 2023). After this initial study (Albentosa et al., 2023), the MMOI focused on building an experimental oyster hatchery for seed production using water from the lagoon itself. The use of seawater from the MM to supply the hatchery may have represented an additional challenge to the inherent difficulties of flat osyter seed production (Colsoul et al., 2021). The studies focused on flat oyster production are limited (Colsoul et al., 2021) and challenging because of several factors such as mortalities associated to larval metamorphosis, unexplained larval mass mortality events, and the presence of pathogens which can provoke large mortalities, like Bonamia ostreae (zu Ermgassen et al., 2020a; Gray et al., 2022; zu Ermgassen et al., 2023).

This study presents the results of the RemediOS project, which constitutes a proof of concept for the production of flat oyster seed from broodstock collected from the MM using water from the lagoon itself.

The broodstock cleared 38.9% of the diet, therefore giving the diet ingested as 6.43% of the DW of the individuals. Five adults died during the experiment period.

A total of 29 swarming events were observed. A swarming event was considered when each batch of larvae obtained from a tank in a single day could be related to one or several oysters. The first release of larvae was observed on March 21 from two adults collected 10 days before. Swarming events were recorded from March to mid-June when broodstock was returned to the sea. The water temperature during this time ranged between 15°C and 25°C ( Figure 1 ). Five of the swarming events were not counted due to management circumstances. From the other 24 events, 57.5 million larvae were obtained (1.6 million larvae per adult) ( Table 1 ). A total of 11.8 million larvae (20.5%) were cultured in the hatchery, while the rest of them were discarded due to the lack of space in the facilities. Considering the total of larvae cultured, 5.1 million larvae (43.6%) reached competence or pediveliger stage, when eye-spot and/or active foot was observed, and 0.68 million (5.8%) survived through metamorphosis, reaching post-larvae stage or spat and attaching effectively to the substrate ( Table 1 ). The mean settlement rate was 13.3%.

The three batches cultured at high larvae density in March (13.4 ± 1.0 larvae mL^-1) showed a lower survival of 23% compared to the larvae cultured at a lower density (5.5 ± 0.9 larvae mL^-1), which showed a survival of 73%. A negative significant correlation of -0.79 was observed between larval density and survival (df = 12, p < 0.001).

Larvae at release exhibited a size of ≃170 µm and reached maximum length after 16–20 days with ≃290 µm (growth rate = 5.54 µm day^-1) ( Figure 2 ). In total, 50% of eye-spot and/or active foot was usually observed between 17 and 20 days, with a minimum of 15 days and a maximum of 22 days. During the culture period, the ingestion rate increased together with larval size at up to a maximum of 0.5 µg of organic matter (OM) larvae^-1 day^-1 on day 17, when a decline was observed (data related to four batches) ( Figure 2 ).

In June 2022, the hatchery accounted a total of 398,317 seed individuals with a size range between 0.5 and 6 mm ( Figure 3 ). The total number of seeds decreased drastically over the following months with different mortalities related always to the individual in the lower size class. The 30,000 seeds maintained in September did not show any additional mortality during the following months ( Figure 3 ). By November 2022, the seed size ranged from 4 to 15 mm.

The present work showed the viability of producing seed using the population of flat oyster inhabiting the MM. In the Spanish Mediterranean Sea, the flat oyster can start spawning at temperatures of 12°C to 13°C (Bromley et al., 2016), which agrees with the first releases of larvae observed in March with a water temperature of 15°C. In fact, Cano et al. (1997) observed the first larvae in the MM plankton at 14°C. Larval release stopped at a temperature of 25°C at the beginning of June, and by the end of June, adults were returned to the Mar Menor. Similarly, Abellán Martinez et al. (1989) observed the first larvae in April, a little later than during the present work, and detected larvae until August. Therefore, it would be possible that more larvae could be obtained during the summer if the broodstock would have been maintained. The fertility observed for the MM broodstock, with 1.5 million of larvae per oyster, coincides with the observations reviewed by Colsoul et al. (2021) for an oyster size of 90 mm.

The larvae survival rates obtained during April (97%) are comparable to the survival rates of 99% obtained by Robert et al. (2017) at 25°C and using a diet with Chaetoceros neogracile and T. lutea. In March, a lower survival rate (28%, Table 1 ) was observed compared to those of other months (>70%, Table 1 ), which could be related to the higher larval density in some of the tanks (5.5 vs. 13.4 larvae mL^-1). During this month, the tanks with a lower larval density had a survival rate of 65%, while the tanks with a higher density had a survival rate of 23%. The presence of contaminants such as heavy metals and detergents can provoke higher mortalities in larvae Colsoul et al. (2021). However, a carbon filter was used, which helps to remove certain contaminants, and different survival rates were observed for batches cultured at the same time. It seems unlikely to think that the concentration of those contaminants could significantly change in seawater in a short period of time. Another explanation could be the presence of bacteria such as Vibrio sp., which can also cause large mortalities in oyster cultures (Helm et al., 2004). These bacteria could originate in non-treated broodstock, microalgae production, and the larval culture (Colsoul et al., 2021) and be responsible of the different survival rates observed in the different batches.

Larval growth rate (5.54 µm day^-1) was in the middle of the usual growth rates reported for the species in other hatcheries, with a range observed to be between 1.4 and 11 µm day^-1 for different conditions (González-Araya and Robert, 2018; Jacobs et al., 2020; da Costa et al., 2023). Despite the fact that the growth rate was similar to those reported in some of the cited works, the pelagic larval period was much longer (a total of 17–20 days) when compared to those of other hatcheries, where periods below 14 days were observed (Gonzalez Araya et al., 2012; Mesías-Gansbiller et al., 2013; González-Araya and Robert, 2018). However, longer periods have been reported as up to 38 days (Jacobs et al., 2020).

The settlement rates were low compared to the 68% obtained by Robert et al., (2017), with a higher rate during March at 21.4%. In the following months, the survival rate was even lower, with 6.5% and 7.1% obtained for April and May, respectively. This difference could be related to different factors. One explanation could be that the first larvae obtained in March came from field fertilization, where the food available may have contributed to a higher quality of these spawns. In fact, the higher survival rate during settlement was only observed for the first two larval batches (34% of survival vs. <5% during the rest of March). On the other hand, the lower density of larvae during the two initial batches (≃500 larvae cm^-2) compared to the rest of them (≃1,000 larvae cm^-2) could lead to a higher survival rate.

The seed obtained in the hatchery suffered a great mortality during the initial months of life, with a decrease of almost 400,000 individuals to 30,000 individuals in 3 months. The mortality was always observed in the smaller size classes, as have been observed for other bivalves such as Mytilus edulis, which could be associated to inherent higher metabolic rates (Rodhouse et al., 1984). Moreover, space constraints in the hatchery forced us to prioritize the well-being of larger individuals, which resulted in rearing smaller size classes at high densities, and thus it may have impacted negatively on spat survival (Lauzon-Guay et al., 2005). The presence of pathogenic Vibrios could have also played a role in these mortalities, affecting the smaller and weaker individuals (Dubert et al., 2017).

For both larval and seed production, specific diets could be used in order to improve the growth and survival. During the present work, phytoplankton production was affected by the implementations of new protocols in the recently built hatchery. Therefore, diets were given according to the availability of the different microalgae instead of prioritizing a specific mix of species—for example, a specific diet of C. neogracile and T. lutea is recommended to increase the larval survival rate and the settlement rate (Robert et al., 2017; da Costa et al., 2023).

The objective of obtaining seed using MM broodstock was achieved at the first trial. This new stock of oyster seed will be used in future actions of the MMOI to evaluate the bioremediation capacity of the oysters and to carry out a restoration pilot. However, further advances could be done to improve the survival and development of oyster larvae, specially to increase the settlement success, such as improving the diet quality using a more controlled combination of microalgae and the use of lower densities during culture.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical review and approval was not required for the study on animals in accordance with the local legislation and institutional requirements.

SH wrote the first draft of the manuscript, organized the database, and conducted data analysis. FC and MA elaborated the conception and design of the study and provided support with data interpretation. MA provided financial support. SH, AH, and MA performed the experiment. All authors contributed to the article and approved the submitted version.