1 Introduction

The global production of aquaculture has steadily increased during the past two decades, and it presently accounts for > 50% of the fish food supply (Fig. 1; Aksnes et al. 2017; Botta et al. 2020; FAO 2022). Specifically, the production of mollusks has seen a significant increase, with the Pacific oyster, Crassostrea gigas, emerging as the leading species (Aksnes et al. 2017; FAO 2022).

Fig. 1
figure 1

Global aquaculture production by major taxonomic groups reported in the FAO (2022)

Full size image

C. gigas live in water with temperatures of 4–24 ℃ around intertidal or shallow subtidal areas (Shatkin et al. 1997). They primarily consume phytoplankton and detritus in the water column, directly. C. gigas was globally introduced for aquaculture due to its rapid growth and high productivity (Shatkin et al. 1997; Harris 2008;). In Korea, C. gigas is typically hardened before cultivation to prevent mass mortality (Park et al. 1988). However, C. gigas cultured on farms often suffer from summer mortality syndrome, which leads to mass killing (Koganezawa 1975). The exact causes of the syndrome remain unclear, but several potential causes including hypoxia, over-maturation, and the emergence of new disease outbreaks have been proposed (Han et al. 2013).

C. gigas contribute to human well-being in various ways. First, they can supply food sources to humans. In 2020, ca. 6 × 106 tons of C. gigas were produced, accounting for ca. 8% of the total global production from aquaculture (FAO 2022). Second, oyster species can mitigate eutrophication by removing nitrogen from the water column by consuming phytoplankton and detritus, thereby helping maintain water quality (Newell et al. 2005). Newell et al. (2005) estimated that the eastern oyster, Crassostrea virginica, reefs remove ca.13,080 kg of nitrogen, corresponding to 314,836 US dollars (USD), in the Choptank River, Maryland, USA. Ferreira et al. (2007) estimated that net nitrogen removal was almost 10.7 tons per year from 0.6 ha (100 × 100 m2) of oyster farms, equivalent to the total amount of nitrogen in untreated sewage discharged by 3237 people. Last, oysters provide supporting services to marine ecosystems. Their reefs serve the habitats for other marine species and can increase fish production (Peterson et al. 2003). Oyster shells, separated from oyster reefs, stabilized sediment and reduced the wave energies affecting the erosion of salt marsh fringes (Meyer et al. 1997). Furthermore, oyster reefs can protect coastal areas against rising sea levels caused by global warming (Coen et al. 1999; Piazza et al. 2005). These benefits are called “ecosystem services” (Ehrlich and Ehrlich 1981).

Korea is the second biggest producer of cultured C. gigas (ca. 300,084 tons in 2020) following China (FAO 2022). However, Korea’s interest in cultured C. gigas has primarily focused on its role in the food supply, overlooking its other functions. This approach has impeded the proper evaluation of the coastal ecosystem, leading to confusion and conflicts in coastal development planning (Choi 2008).

Here, we evaluated the ecosystem services provided by cultured C. gigas for the first time in Korea to assist policymakers in making decisions on marine spatial planning. For this purpose, we developed a bio-economic model that evaluates the ecosystem services based on the production and total amount of nitrogen accumulated by the C. gigas.

2 Material and Methods

2.1 Study Area

We chose Hansan Bay to evaluate the ecosystem services provided by the C. gigas because the coverage of the farm area (ca. 26%) is higher than in other bays (ca. 4.7% on average) (National Federation of Fisheries Cooperatives 2023). Korea has predominantly used the submerged long-line method for culturing C. gigas since the 1960s, accounting for ca. 98% of the total oyster production (ca. 300,840 tons in 2020). This method is practiced offshore and applied to large-scale aquaculture. The long lines, made of light materials, float on the sea surface to hold the submerged lines to which oyster spats are attached (Aldon 1998). The water depth in the bay ranges from 6.5 to 27.9 m, and the monthly mean water temperatures range from 9.0 to 25.6 °C, with monthly mean salinities ranging from 29.5 to 34.2 (Park et al. 2016). Cultivation of C. gigas in Hansan Bay typically is operated from May to the following February in the area of 542 ha (Kang et al. 2000; National Federation of Fisheries Cooperatives 2023).

2.2 C. gigas Production in Hansan Bay

To estimate the production of C. gigas in Hansan Bay, we first tried to estimate the initial oyster population by multiplying the number of oysters per spat collector, ca. 50, with the count of spat collectors per longline, which varies between 5000 and 7500. Additionally, multiplying this with the number of longlines per ha of farms, typically ca. 20 (NFRDI 2012; Han et al. 2013; Lee et al. 2018; Oh et al. 2002). The derived initial oyster population ranged from ca. 5 to 7.5 million individuals per ha of farms.

To estimate the growth performance of C. gigas, we adopted the growth model developed by Gangnery et al. (2003). The model is as follows:

$$G_{{\text{t}}} = {\text{a}} \times F_{{\text{t}}}^{{\text{b}}} \times T_{{\text{t}}}^{{\text{c}}} \times Y_{{\text{t}}}^{{\text{d}}} ,$$
(1)
$${\text{TW}}_{{\text{H}}}\text{ = TW}_{{\text{I}}}\text{ } + \sum_{\text{t=1}}^{\text{Cultivation period in Months}}{\text{G}}_{\text{t}},$$
(2)

where Gt is the growth in grams per month, Ft is food concentration in micrograms per liter, Tt is the water temperature (°C), TWt is an individual dry tissue weight in grams, TWH is a harvested individual’s dry tissue weight in grams, TWI is an initial individual’s dry tissue weight in grams, and t is the month in the cultivation period. The parameters "a," "b," "c," and "d" were estimated through non-linear regression using R software, and the data of the variables were obtained from a past study that reported the growth of C. gigas and environmental conditions during the cultivation periods of 2008 and 2013 in Hansan Bay (Lee et al. 2018). For Ft, we used the concentration of chlorophyll-a, which represents the biomass of phytoplankton. This was based on a previous study that indicated phytoplankton as the primary food source for oysters, confirmed through stable isotope analysis (Langdon and Newell 1996).

To account for the mortality of C. gigas during cultivation, we assumed that the accumulated mortality rates of C. gigas range from 20 to 95%, taking into account the summer mass mortality event (Han et al. 2013; Hong et al. 2022; Koganezawa 1975; Luna-González et al. 2008). With these assumptions, we calculated the initial biomass and harvested biomass of C. gigas using the equations as follows:

$${\text{B}}_{{\text{I}}} = {\text{TW}_{{\text{I}}}} \times {\text{PPO}}_{{\text{I}}}$$
(3)
$${\text{B}}_{{\text{H}}} = {\text{TW}}_{{\text{H}}} \times {\text{PPO}}_{{\text{I}}} \times {\text{M}}_{{\text{a}}}$$
(4)

where BI is the initial biomass of C. gigas from farms in tons, BH is the harvested biomass of C. gigas from farms in tons, PPOI is the initial population of C. gigas cultured on farms in individuals, and Ma is the accumulated mortality rate of cultured C. gigas. Calculating the difference between BH and BI, we determined the cultured C. gigas production.

2.3 Provisioning and Regulating Services provided by cultured C. gigas in Hansan Bay

C. gigas is mainly cultured using the long-line method in Hansan Bay, making it difficult to offer oyster shells to coastal areas to prevent erosion of sediment caused by the tide. Therefore, we only evaluated the provisioning and regulation services provided by the C. gigas, excluding the supporting service.

To evaluate the provisioning service, we investigated the domestic market prices of C. gigas. According to the Korean Statistical Information Service (2022), the total market price and amount of oyster production, including shell weight, was ca. 305,614 tons and 214.88 million USD in 2021. They estimated the values based on the wet weight of C. gigas, including shell weight. As our growth model estimated the dry tissue weight of C. gigas, we assumed the ratio of dry weight to wet weight is 0.2, adjusted the total weight to the tissue weight using Eq. 5, and evaluated the market price using Eq. 6 as follows (Hyun et al. 2001; Shim et al. 2021):

$${\text{pTW}} = - 0.0559 \times {\text{Year}} + 128.29$$
(5)
$${\text{The domestic market price of a ton of }}{{C. gigas}} = \frac{{214.88 {\text{ million USD}} }}{{{\text{pTW}} \times 305,614}}$$
(6)

where pTW refers to the proportion of tissue weight in the total wet weight of C. gigas. By multiplying the result of Eq. 6 and the C. gigas production estimated from Eqs. 3 and 4, we calculated the monetary value of the provisioning service provided by cultured C. gigas in Hansan Bay.

Before evaluating the regulating services provided by C. gigas, we needed to estimate the amount of nitrogen removal by C. gigas. We assumed that the amount of nitrogen removed by cultured C. gigas equals the difference in accumulated nitrogen between the BI and BH. Here, we only considered nitrogen accumulation rates in the tissues because, in cultivation, C. gigas typically settle on empty shells prepared by farmers and do not invest in shell development. Based on a past study, we assumed that the nitrogen accumulation rate accounts for 7.85% of the dry tissue weight of C. gigas (Schatte Olivier et al. 2018). The derived equation is as follows:

$${\text{Amount of nitrogen removed by cultured }}{{C. gigas}} = 7.85\% \times ({\text{B}}_{{\text{H}}} - {\text{B}}_{{\text{I}}}).$$
(7)

Beseres Pollack et al. (2013) previously estimated the monetary value of regulation services provided by C. gigas using the cost of sewage treatments. Building upon this approach, we evaluated the regulation service provided by cultured C. gigas in Hansan Bay and assumed that they fully replace the denitrification process in sewage treatments. We focused on three types of sewage treatments mainly used in Korea for eliminating nitrogen through denitrification (Ministry of Environment 2019): Anaerobic–anoxic–aerobic method (A2O), modified Kist Shinwon Biological Nutrient Removal (KSBNR), and modified Ludzack Ettinger (MLE). To evaluate the effectiveness of these treatments, we assumed the annual operational cost of removing nitrogen per m3 of sewage treatments to be ca. 73.6 USD, the total volume of sewage treatments to be 0.424 × 106 m3, the annual amount of nitrogen removed by 1 m3 of sewage treatments to be ca. 7.36 kg, the total cost of constructing sewage treatments to be 534.6 million USD, and the life span of sewage treatments to be ca. 20 years (15–20 years), based on the reports by the Ministry of Environment (2019). To estimate the monetary value of the annual regulation service provided by cultured C. gigas, we used the following equations:

$${\text{dcST}}_{{\text{annual}}} = \frac{{{\text{cCST}}_{{{\text{total}}}} }}{{{\text{lsST}} \times {\text{vST}}_{{{\text{total}}}} }}.$$
(8)
$$\text{The cost for removal of }1\text{ kg of nitrogen through sewage treatments }=\frac{\text{ dcST}_{\text{annual}}+ \text{ ocNR} _{\text{annual}}}{{\text{aNR}}_{\text{annual}}},$$
(9)

where dcSTannual is the annual depreciable cost for a ton of sewage treatment, cCSTtotal is the total cost of constructing sewage treatments in USD, lsST is the life span of sewage treatments, vSTtotal is the total volume of sewage treatments, aNRannual is the annual amount of nitrogen removed per m3 of sewage treatments, and the ocNRannual is the annual operational cost of removing nitrogen by 1 m3 of sewage treatments. We calculated the annual depreciable cost per m3 of sewage treatment to be 62.9 USD using Eq. 8. Considering the annual depreciable cost, we calculated that ca. 18.55 USD is incurred annually for the removal of 1 kg of nitrogen through sewage treatments using Eq. 9. By multiplying the results of Eqs. 7 and 9, we calculated the monetary value of regulation services provided by C. gigas farm.

3 Results and Discussions

3.1 C. gigas Production in Hansan Bay

The derived growth equation (Eq. 1) was as follows:

$$G_{{\text{t}}} = 76.4240 \times F_{{\text{t}}}^{0.2462} \times T_{{\text{t}}}^{ - 2.2619} \times Y_{{\text{t}}}^{ - 0.8701} .$$

The estimated dry weight of a harvested individual of C. gigas ranged from ca. 1.502 to 2.051 g in 2009 and from ca. 1.456 to 2.004 g in 2014. All observed dry weights of an individual of C. gigas in 2008 were within the 95% confidence interval, except in October and November 2013 (Fig. 2a and b). Applying the results of Eqs. 1 to 3 and 4, we estimated that the production of C. gigas in Hansan Bay ranged from ca. – 710 to 6070 tons of dry tissue weight in 2009 and from ca. – 1492 to 5136 tons of dry tissue weight in 2014.

Fig. 2
figure 2

The dry tissue weights of C. gigas were measured in two periods, 2008–2009 (a) and 2013–2014 (b), in Hansan Bay, Korea. The red triangles on a dashed red line denote the dry tissue weights estimated by Gangnery’s growth model, with 95% confidence intervals. The black circles on a black line denote the dry tissue weights estimated by the Scope for Growth model, and the blue crosses the observed dry tissue weights

Full size image

We validated our growth model by comparing it with the Scope for Growth (SFG) model applied to Hansan Bay in a previous study (Lee et al. 2018). Our results indicated that, during the period of 2008, Gangnery’s growth model (r2 = 0.991) was more suitable than the SFG model (r2 = 0.984). However, during the period of 2013, the SFG model (r2 = 0.971) was more suitable than Gangnery’s growth model (r2 = 0.861). Our findings indicate that Gangnery’s growth model is as reliable as the SFG model in estimating the dry tissue weights of C. gigas during cultivation, because of no significant difference in comparison with observed dry tissue weights (Fig. 2a and b). This suggests that Gangnery’s growth model alone is sufficient for estimating growth when food consumption data are unavailable (Tables 1, 2).

Table 1 Accumulated mortality rates of cultured C. gigas reported by past studies
Full size table
Table 2 The chemical compositions of oysters
Full size table

3.2 Provisioning and Regulating Services provided by cultured C. gigas in Hansan Bay

Using Eqs. 5 and 6, we estimated that the pTW was ca. 15.3, and the domestic market price of a ton of C. gigas was ca. 4586 USD in 2021. Multiplying with C. gigas production in Hansan Bay, we estimated that the monetary value of provisioning services ranged from ca. −3260 to 27,839 thousand USD in 2009 and from ca. – 6845 to 23,552 thousand USD in 2014. The estimated monetary value of the regulation service in Hansan Bay ranged from ca. – 1035 to 8839 thousand USD in 2009 and from ca. – 2173 to 7478 thousand USD in 2014 using Eqs. 7 and 9. The results indicate a decline in monetary value in 2014 compared to 2009. This decline was attributed to a decrease in the growth performance of C. gigas in 2014 (ca. two- to threefold increase in initial weight) compared to 2009 (ca. five- to sevenfold increase in initial weight). Considering the surface area of farms in Hansan Bay, cultivating a ha of C. gigas farms has a potential value ranging from ca. – 16.6 to 67.6 thousand USD.

To validate our evaluation of the monetary value of the regulation service, we compared our estimate with past studies in USD per kg of nitrogen removal. Our estimate (ca. 18.55) was slightly lower than that of Newell et al. (2005) (ca. 24.07), but much lower than an estimate from Ferreira et al. (2007) that adopted the Farm Aquaculture Resource Management (FARM) model (ca. 206.11).

The discrepancy between our and the FARM models may be attributed to several factors. First, the FARM model estimated that a ha of oyster farms removes ca. 3 tons of nitrogen, significantly greater than our estimates, ca. – 0.216 to 0.879 tons. The FARM model was developed based on the polyculture method known as integrated multi-trophic aquaculture (IMTA), where the various species are cultured together. Reid et al. (2013) and Chopin et al. (2010) documented the advantage of IMTA regarding oxygen concentration and food supplies. In contrast, C. gigas in Korea are typically cultured using intensive monoculture, making them more vulnerable to hypoxia and food limitation, which lead to a decline in oyster production (Park et al. 2018). Second, while our model assessed the monetary value directly from the estimated nitrogen removal in kg, the FARM model utilized the sewage treatment costs per inhabitant and assumed the annual amount of nitrogen emitted by a person in the evaluation (Galvao et al. 2005). These differences may explain, at least partially, the discrepancies.

3.3 Comparison with Alternative Species

According to the Korean Statistical Information Service (2022), Korea produced ca. 16,000 tons of Korean rockfish, Sebastes schlegelii, using cage culture, which corresponds to ca. 149 million USD. In 2021, the total area of cage culture in Korea was 98 ha, in which the cage culture of S. schlegelii accounted for ca. 53.8 ha. To compare with the value of C. gigas farms, we evaluated and assumed that the cost of feed was ca. 2,782 thousand USD per ha of cage culture for S. schlegelii. At first glance, it appears that cage cultures of S. schlegelii are more valuable than C. gigas farms (ca. – 16.6–67.6 thousand USD per ha). However, Korea is planning to substitute formula feed (ca. 1.5 USD per kg) for raw feed (ca. 0.37 USD per kg) to manage the marine ecosystem. We estimated that replacing raw feed with formula feed will cause a loss of ca. 850 thousand USD per ha, because of the difference in the feed efficiency (Myeong et al. 1997). Despite the advantage of S. schlegelii in commercial value, environmental concerns suggest that C. gigas will be more preferable than S. schlegelii for cage culture.

3.4 Supporting Decisions in Compensation for Disasters and Development

The Korean government provides compensation to aquaculture entrepreneurs who lose more than 50% of their fish and shellfish by disaster, accident, and development. The most common example of a disaster that affected this industry is an oil spill, for which International Oil Pollution Compensation (IOPC) Funds compensate. The IOPC Funds calculate financial losses by comparing production before and after oil damage. Similarly, the Korean government determines the compensation amount for developments based on the business results of the past three years and land prices (Jang 2005).

However, relying solely on past business results may not always accurately reflect future earnings. Therefore, a clear and transparent method is required for calculating, as it can affect the livelihoods of workers in affected industries. Our model is expected to provide a more comprehensive approach to determining financial losses and compensation amounts by estimating future earnings.

4 Problems, Limitations, and Future Works of the Study

While our growth model is well suited for estimating the growth of C. gigas in Hansan Bay, its applicability in other regions is still uncertain. Because we developed the model based on specific parameters from Hansan Bay, it may not accurately reflect conditions in other areas. To address this, we need to improve the model for other regions. The most effective method for enhancing the model is to collect extensive growth data of C. gigas cultured in various regions. The model's parameters were calculated based on observed weight values using non-linear regression. This implies that incorporating numerous observed data from different environmental factors and periods will enhance the reliability of the model’s parameters, resulting in more realistic predictions on the growth of oysters in Korea.

Our approach to estimating nitrogen removal considers only the amount of nitrogen assimilated in the tissue of C. gigas and the nitrogen removed from the marine ecosystem through the harvest of C. gigas, which is called “bioextraction” (Donnelly et al. 2023). However, C. gigas remove nitrogen from the water column in more complicated ways. Donnelly et al. (2023) illustrated the ecological process associated with oysters that improve water quality by removing nitrogen, showing that the C. gigas produced “biodeposits” such as feces, which stimulate denitrification in marine ecosystems. In addition, they documented the various factors that influence the ability of oysters to remove nitrogen. Our approach currently has limitations in considering such factors. Furthermore, in evaluating regulation service, we didn’t consider the salvage value derived from sewage treatments due to the lack of empirical cases. Therefore, we plan to develop the nitrogen removal model by considering all of the pathways by which C. gigas removes nitrogen when empirical cases regarding salvage value for the sewage treatments are available.

Our analysis suggested that the monetary value of ecosystem services provided by C. gigas in Hansan Bay is highly dependent on the assumed mortality rates of oysters, leading to great variability and uncertainty in the estimates. Past studies have mostly reported the mortalities of cultured C. gigas as accumulated mortalities, because oysters cultured by the long-line method were submerged until harvest, making it difficult for researchers to estimate the monthly mortalities. Therefore, we could not consider the monthly variation in mortality. To provide a more accurate and reasonable estimation, we need to cooperate with oyster farm industries in Korea to estimate monthly mortalities of cultured C. gigas in future studies.

5 Conclusions

We have developed a bio-economic model to evaluate the ecosystem services provided by cultured C. gigas in Hansan Bay. Our results indicated that Gangnery’s growth model, employed for estimating oyster growth, is a suitable replacement for other growth models in cases where food consumption data are limited. Additionally, we found that the nitrogen removal model can be applied when a growth model of oysters is available. However, our results suggest that the potential value offered by C. gigas in Hansan Bay is much lower compared to other sites where IMTA is used for cultivation. This finding implies that IMTA may be a more suitable method for culturing C. gigas in terms of the regulating services provided by C. gigas. Such insights may have some limitations, but we expect that our approach and methods will provide a useful foundation for future studies and applications to assist policymakers in making decisions between conservation and development, and in determining financial compensation for disaster, accident, and development in coastal and oceanic areas.