A food-web model as a tool for the ecosystem-level management of bivalves in an Atlantic coastal lagoon

doi:10.21203/rs.3.rs-2070611/v1

Download PDF

Research Article

A food-web model as a tool for the ecosystem-level management of bivalves in an Atlantic coastal lagoon

https://doi.org/10.21203/rs.3.rs-2070611/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The Ria de Aveiro is an important coastal lagoon for wildlife in Portugal, where the production of bivalves reaches approximately 2700 tonnes annually. However, the illegal overfishing of bivalves is frequent in this lagoon, which causes critical changes in the ecosystem. In this study, using a developed food-web model (Ecopath model), the ecological carrying capacity and maximum sustained yield of bivalve filter feeders were estimated, and further increases in bivalve biomass in other species groups were investigated. The results showed that 1) the current biomass and legal catch of bivalves do not yet exceed the ecological carrying capacity (177.84 tonnes km^− 2) or the maximum sustained yield (88.92 tonnes km^− 2 year^− 1) in Ria de Averio; 2) the harvested bivalves of the maximum sustained yield represent removing from the ecosystem ~ 581 tonnes carbon (C) and ~ 83 tonnes nitrogen (N) annually, with substantial ecological and economic implications; and 3) a further increase in the biomass levels of bivalves may cause the ecotrophic efficiency of other groups to become unrealistic, potentially leading to decreases in ecosystem transfer efficiency, biodiversity and health. The results here are expected to guide the sustainable development and management of bivalves in Ria de Averio and the protection of the local environment.

Ria de Aveiro

Ecopath model

Bivalves

Ecological carrying capacity

Maximum sustained yield

Bivalve filter feeders, such as mussels, oysters and clams, play key roles in many estuarine and coastal waters, owing to their high density (Smaal et al. 2001), extensive filtration capacity (Prins et al. 1996), enhanced biodeposition (Crawford et al. 2003) and alteration of nutrient and oxygen fluxes (Mao et al. 2006). During the past 50 years, the total production of bivalve shellfish (both capture and aquaculture) has increased quickly from 1 million tonnes to 13.2 million tonnes (Dumbauld et al. 2009; Dias et al. 2021). Their commercial exploitation has led to either increased or decreased abundance in many coastal waters, raising concerns of environmental problems and sustainable development. Application of the carrying capacity in the exploitation of natural resources has been used as an approach to examine environmental sustainability (Smaal et al. 2001).

Coastal lagoons are among the most productive aquatic systems, supporting a large diversity of habitats that offer optimal niches for numerous aquatic species (Villanueva et al. 2006). Ria de Aveiro is a shallow costal lagoon situated on the northwest coast of Portugal (40°38’N, 8°45’W). This aquatic ecosystem comprises an area between 66 and 83 km² at low tide and high tide (spring tide), respectively, and is characterized by narrow channels, inner basins and large areas of mudflats (Lillebø et al. 2015). The average depth of Ria de Aveiro is approximately 1 m, except for navigation channels where dredging operations are frequently carried out (Dias et al. 2000). The production of bivalve filter feeders has reached approximately 2700 tonnes in Ria de Aveiro annually (Bueno-Pardo et al. 2018). The Manila clam, Ruditapes philippinarum, was the most abundant clam in this lagoon (Velez et al. 2015) and represents most of the national annual production of bivalves. However, currently, the exploitation of clams in Ria de Aveiro is continuously increasing, with critical changes in ecosystems and severe consequences (Velez et al. 2015). The catch limitation of clams urgently needs to be estimated based on a fishing yield that ensures sustainable harvesting.

Maximum sustained yield (MSY) is defined as the maximum production of food from the sea on a sustained basis year after year (Carmel 2011). According to Byron et al. (2011a) and Mace (2001), the maximum sustained yield is half of the carrying capacity; at that point, the growth rate of animals is highest and has a significant implication for management. Estimating the maximum sustained yield of aquatic animals is also essential to meet the objectives of maximizing yield while protecting the long-term viability of populations and ecosystems. In recent years, researchers have contributed to the understanding of population dynamics and maximum sustained yield of various kinds of fish species (Sher et al. 2012; Panhwar and Liu 2013). However, despite a huge commercial demand for bivalves, no published information is available on the maximum sustained yield of bivalves in Ria de Aveiro.

Inglis et al. (2000) divided carrying capacity into four functional categories, i.e., physical carrying capacity, production carrying capacity, ecological carrying capacity, and social carrying capacity. On the ecological level, carrying capacity is approached more holistically and defined as the stocking or farm density that causes “unacceptable” impacts on the ecosystem (McKindsey et al. 2006). A holistic approach is, therefore, important because certain levels of carrying capacity may be “unacceptable” to other system components; e.g., high stock densities lead to cascade effects within the trophic structure of the ecological system (Jiang and Gibbs 2005; Kluger et al. 2016). On the other hand, positive effects of bivalve cultivation may also be possible; e.g., the cultured species provide new habitat structures or increase the food sources (McKindsey et al. 2006; Meyer 2014). A food-web Ecopath model has been proposed to estimate the ecological carrying capacity for use in aquaculture by increasing the biomass of cultured bivalves until the ecotrophic efficiency (EE) becomes unrealistic (i.e., > 1) (see Jiang and Gibbs 2005; Byron et al. 2011b; Carrie et al. 2011). Ecopath creates a static snapshot representation of the different compartments of an ecosystem and their interactions in a specific period, while Ecosim simulates its dynamic evolution through time and analyzes the effects of different variables on the ecosystem (Christensen and Walters 2004; Han et al. 2018).

The study presented here aims 1) to implement an Ecopath model to estimate the ecological carrying capacity and maximum sustained yield of bivalve filter feeders in Ria de Aveiro; 2) to quantify clam feeding, respiration and excretion based on flow-through experiments to analyze the ecosystem role of bivalves in Ria de Aveiro; and 3) to evaluate the possible impact of a further increase in bivalve biomass on other species groups and possible ecosystemic changes with Ecosim. This research will contribute to the protection of the ecological environment and provide basic data for the sustainable management of bivalve farming in Ria de Aveiro. The results observed here could also be used for other coastal lagoons or estuaries with similar characteristics.

Model construction

Ecopath model

The Ecopath model was configured to best evaluate the current state of the ecosystem and then evaluated to estimate animal biomass levels until dramatic changes to energy fluxes occurred, which is defined as the ecological carrying capacity here (Jiang and Gibbs 2005). The model includes a set of linear equations that describe the production of each functional group:

$$\text{B}\text{i}\text{ }\text{× }\text{(}\text{P}\text{/}\text{B}\text{)}\text{i}\text{ }\text{× }\text{EE}\text{i}\text{ }\text{ }\text{=}\sum _{\text{j }\text{= 1}}^{\text{n}}\text{B}\text{j}\text{ }\text{× }\text{(}\text{Q}\text{/}\text{B}\text{)}\text{j}\text{ }\text{ }\text{× }\text{DC}\text{ij}\text{ }\text{ }\text{+ }\text{Y}\text{i}\text{ }\text{+ }\text{E}\text{i}\text{ }\text{+ }\text{BA}\text{i}$$

where B_i and B_j are the biomasses of group i and group j, respectively; (P/B)_i is the production to biomass ratio for group i or total mortality in steady-state conditions; EE_i is the ecotrophic efficiency, which is defined as the proportion of the production of group i that is used in the ecosystem; (Q/B)_j is the food consumption to biomass ratio for predator group j; DC_ij is the fraction of prey i in the stomach content of predator j; Y_i is the fishery yield of group i; E_i is the net migration rate; and BA_i is the biomass accumulation rate for group i. For each functional group i, the DC_ij and at least three of the four basic parameters are required to build the Ecopath model: B, P/B, Q/B and EE.

A trophic model of Ria de Aveiro was constructed using the software Ecopath with Ecosim (EwE) 6.5 (Christensen and Walters 2004) and was based on a previous model by Bueno-Pardo et al. (2018). The updated model of Ria de Aveiro comprises 24 functional groups, including detritus, phytoplankton, zooplankton, gastropods, annelids, cephalopods, bivalves, mysidacea, crustaceans, and fish. The input parameters of these functional groups were collected from various sources, including catch statistics, empirical relationships, samplings, and experiments studied in other references (Bueno-Pardo et al. 2018). Values for P/B (year^− 1), Q/B (year^− 1) and fleet landings (FL, tonnes km^− 2 year^− 1) were based on former estimates of Bueno-Pardo et al. (2018). The diet composition of the 24 functional groups was introduced as a diet matrix, which can be found in the reference of Bueno-Pardo et al. (2018).

Ecosim explorations

As a time-dynamic simulation model, Ecosim was used to explore the potential impacts of different bivalve densities on the Ria de Aveiro ecosystem through both direct and indirect effects. The Ecosim master equation is as follows:

$$\text{ }\text{dB}\text{i}\text{/}\text{dt }\text{= }\text{g}\text{i }\sum _{\text{j}}\text{Q}\text{ji}\text{ -}\sum _{\text{j}}\text{Q}\text{ji}\text{ + }\text{I}\text{i}\text{ - }\left(\text{M}\text{i}\text{ + }\text{F}\text{i}\text{ + }\text{e}\text{i}\right)\text{ ×}{\text{ B}}_{\text{i}}$$

where $\text{dBi}\text{/}\text{dt}$ represents the growth rate during the time interval dt of group i in terms of its biomass B_i, g_i is the net growth efficiency, Q_ji is the consumption rates calculated based on the “foraging arena” concept, where B_i are divided into vulnerable and invulnerable components (Walters et al. 1997), I_i is the immigration rate, M_i is the nonpredation natural mortality rate estimated from the EE, F_i is the fishing mortality rate, and e_i is the emigration rate.

The vulnerabilities (v) of each functional group in Ecosim, representing the flows and type of trophic control (i.e., top-down, bottom-up and mixed control) between predator and prey, were adjusted to be proportional to the trophic level (Cheung et al. 2002; Christensen and Walters 2004; Kluger et al. 2016):

v _i = 0.1515 × TL_i + 0.0485

where TL_i is the trophic level of a functional group obtained from Ecopath. The vulnerabilities (v) in Ecomsim range from 0.0 to 1.0, with 0.0 implying a bottom-up control, 1.0 serving as a top-down impact, and 0.3 describing a mixed effect.

The actual vulnerabilities (v_new) used in the computations range from 1.0 to Inf (Kluger et al. 2016) (Table 1):

Table 1

Vulnerabilities for the Ria de Aveiro Ecopath model
	Functional group	Vulnerabilities (v_i)	Actual vulnerabilities (v_new)
1	Filter-feeding birds	0.496960453	3.142587507
2	Waders	0.525989671	3.359766010
3	Piscivorous birds	0.598294712	3.968215322
4	Sardine	0.457550000	2.870035034
5	Smelts	0.507212003	3.217631155
6	Seabass	0.528838780	3.381873811
7	Detritivorous fish	0.421626320	2.642243757
8	Zooplanktivorous fish	0.510995564	3.245778157
9	Omnivorous fish	0.518056070	3.298963414
10	Piscivorous fish	0.609390572	4.070879038
11	Zoobenthivorous fish	0.514900174	3.275083863
12	Crustaceans	0.356137718	2.272486678
13	Mysidacea	0.366803015	2.328969834
14	Decapods	0.397248455	2.498051392
15	Bivalves	0.351500000	2.248354752
16	Cephalopods	0.530401351	3.394060366
17	Annelids	0.393278398	2.475325672
18	Gastropods	0.351500000	2.248354752
19	Zooplankton	0.351500000	2.248354752
20	Seagrass	0.200000000	1.586369386
21	Halophytes	0.200000000	1.586369386
22	Phytoplankton	0.200000000	1.586369386
23	Microphytobenthos	0.200000000	1.586369386
24	Detritus	0.200000000	1.586369386

ln (v_new) = 2.301985 × v_i + 0.001051

Simulation scenarios

Bivalve density expansion was simulated for a period of 20 years under three scenarios of differing bivalve biomasses. The current status (biomass levels of 142.90 tonnes km^− 2) of the Ria de Aveiro Ecopath model was set as Scenario 1 (Bueno-Pardo et al. 2018). Scenarios 2 and 3 were set based on one and two times the ecological carrying capacity biomass of bivalves as calculated by Ecopath, respectively. Scenario 2 was based on the ecological carrying capacity of bivalves as calculated by Ecopath. The bivalve biomass chosen for Scenario 3 was one that extensively exceeded the level of ecological carrying capacity. Scenario 4 was also carried out based on 10 times the ecological carrying capacity of bivalves, but the ecosystem collapsed.

For all three scenarios, the following ecological network analysis indicators were used to describe the responses of the Ria de Aveiro ecosystem to the increase in bivalve density: 1) total system throughput, which describes the sum of all flows through the ecosystem and is identified as the “size of the ecosystem”; 2) capacity is the product of total system throughput and entropy and represents the upper limit to ascendency; 3) ascendency, which is the product of growth and development of the system; 4) Finn cycling index quantifies the relative amount of cycling in the system and is an indicator of stress, stability and structural differences; 5) average mutual information is a measure of the information regarding the network of material exchange within the system; 6) entropy describes the total number and diversity of flows in a system; 7) transfer efficiency is calculated as the ratio between the sum of the exports plus the flow transferred from one trophic level to the next and the throughput on the trophic level; and 8) Kempton's Q is a relative index of biomass diversity.

Physiological measurement

Animal collection and acclimation

Healthy R. philippinarum of three different sizes (Table S1) were collected from Ria de Aveiro, Portugal, in October 2018. After sampling, the clams were then transported to the laboratory, where they were acclimated for seven days in seawater (salinity, 30 ± 1; pH, 8.1 ± 0.1) under continuous aeration, a constant temperature of 18.0 ± 1.0°C and a photoperiod of 12:12 h (light/dark). Animals were fed daily with Algamac Protein Plus during the period of acclimation.

Clearance and egestion rates

The clearance rate was measured by a flow-through system consisting of 12 feeding chambers with square edges (after Rastrick et al. 2018; Jiang et al. 2021). These chambers were supplied with the same running seawater as in the acclimation tanks. Nine chambers were used for individual clams, and three chambers were left empty as controls. Seawater was pumped to a header tank and through the feeding chambers at a manually regulated flow rate (average: 207 mL min^− 1). Prior to the measurements, clams were taken from the laboratory maintenance tank and acclimated in the feeding chambers for at least 1 h. Water samples from the outflow of both control chambers and chambers with clams were collected simultaneously. The content of particulate organic matter (POM) (representing food concentration) was detected by collecting 700 mL seawater from each chamber and filtering through precombusted (450°C, 4 h) and preweighed Whatman GF/C glass microfiber filters. Filters were dried to determine dry weight (60°C, 24 h) and ash free-dry weight (450°C, 4 h) to establish the content of POM. The clearance rate (CR) of the clam was calculated using the following formula:

CR (L g^− 1 h^− 1) = (Q × (C₁ – C₀) / C₁) / W

where Q is the flow rate (L h^− 1), C₁ is the outflow concentration represented by the POM in the control (mg L^− 1), C₀ is the outflow concentration from each experimental chamber (mg L^− 1), and W is the dry weight of soft tissue (g) (60°C, 48 h).

Clearance rates were converted to ingestion rates of particulate carbon (POC) of the clam (mg POC g^− 1 h^− 1) using the mean temporal POC content across different seawater layers of the Ria de Averio (Cunha et al. 2003). The ingestion rates of particulate nitrogen (PON) (mg PON g^− 1 h^− 1) were evaluated using POC values and a Redfield ratio of 6.625 (Redfield 1934).

The egestion rate of the clam was estimated by using an assimilation efficiency of 0.75 in accordance with van der Veer et al. (2006), which seems to be reasonable for clams based on findings from several bivalve growth models where this estimate produced realistic weights (Pouvreau et al. 2006; Filgueira et al. 2011).

Respiration and excretion rates

The respiration rate of nine clams was determined using stop-flow respirometers (volume 370 mL). The respirometers were supplied with fully oxygenated filtered seawater. Individuals of R. philippinarum were placed into a respirometer and allowed to settle to the experimental conditions for 1 h. At the end of the experiment, the concentration of dissolved oxygen in each respirometer was kept at a relatively high level of no less than 60% of the initial dissolved oxygen concentration (based on a preliminary experiment) to avoid a negative effect on normal physiological activities. Dissolved oxygen was measured with a LabQuest® 2 multimeter (Vernier, Beaverton, USA). The oxygen consumption rate was detected by the difference between the experimental and control (without clams) respirometers. The respiration rate (RR) of the clam was calculated using the following formula:

RR (mg C g^− 1 h^− 1) = RQ × (12/32) × (DO₀ − DO_t) × V/(W × t)

where RQ is the respiration quotient of 0.85 (Bott 2006), DO₀ and DO_t are the dissolved oxygen contents (mg L^− 1) of the control and experimental respirometers, respectively, V is the respirometer volume (L), W is the dry weight of soft tissue (g) (60°C, 48 h), and t is the measuring time (h).

Clam excretion rates, primarily composed of ammonia-N, were calculated based on stoichiometry (Mayzaud and Conover 1988) using a ratio of respiration to excretion for clams of 7.83 obtained by Murphy et al. (2016).

Statistical analysis

Statistical analysis was performed using SPSS Statistics (ver. 16.0, SPSS Inc., Chicago, USA). All data are presented as the mean ± standard deviation and subjected to analysis of variance (ANOVA) followed by Duncan’s multiple range test. All differences were considered significant at P < 0.05.

Ecological carrying capacity and maximum sustained yield estimation

One target of this study was to determine the ecological carrying capacity and maximum sustained yield of bivalves in the marine ecosystem of the Ria de Aveiro. The bivalve biomass was 142.90 tonnes km^-2 and could be increased to 177.84 tonnes km^-2, after which the Ecopath model was not balanced, as the EE for microphytobenthos was > 1 (Table 2). Interestingly, the model parameters estimated for other groups were not affected by this level of bivalve biomass except for the EE for phytoplankton, microphytobenthos and detritus. Therefore, the mass-balance ecosystem model suggested that the Ria de Aveiro could support a mean bivalve filter feeder biomass of 177.84 tonnes km^-2 without significantly changing flows of other functional groups within the system, and this value could be identified as an estimation of the ecological carrying capacity, which was equivalent to a total bivalve biomass of 14760 t. Furthermore, the maximum sustained yield of bivalves was 88.92 tonnes km^-2year^-1, which was half of the ecological carrying capacity.

A comparison of the general characteristic parameters for the Ria de Aveiro ecosystem before and after bivalve biomass reached ecological carrying capacity is shown in Table 3. The mean trophic level of the catch decreased from 2.20 to 2.17, and thus, the gross efficiency increased from 0.38% to 0.45%. The sum of all exports decreased from 672.78 to 391.17 tonnes km^-2 year^-1, while the total system throughput increased from 19363.12 to 19807.02 tonnes km^-2 year^-1. Of the total system throughput, 60% flowed into detritus, 37% went to consumption and respiration, and 2% to fishery catch. However, when bivalves surpassed the ecological carrying capacity in the Ria de Aveiro, the sum of all exports decreased significantly, mainly resulting from overgrazing on phytoplankton.

Table 2

Changes in Ecopath mode of Ria de Aveiro when estimating ecological carrying capacity of bivalves

Multiplier	Biomass (tonnes km^-2)	Catch (tonnes km^-2year^-1)	Mass-balance changes in model
1.0 (Current condition)	142.90	33.63	Balances
Ecological carrying capacity
1.1	157.19	36.99	Balances
1.2	171.48	40.36	Balances
1.244506648	177.84	41.85	Balances
1.244576627	177.85	41.86	Microphytobenthos EE=1.000028

Table 3

Ecological indices of Ria de Aveiro under different scenarios of bivalve biomass

	Value 1 (Current condition)	Value 2 (Ecological carrying capacity)	Value 3 (Ecological carrying capacity*2)	Units
Sum of all consumption	4702.39	5226.49	7894.09	tonnes km^-2 year^-1
Sum of all exports	672.78	391.17	-1042.25	tonnes km^-2 year^-1
Sum of all respiratory flows	2112.41	2394.03	3827.42	tonnes km^-2 year^-1
Sum of all flows into detritus	11875.54	11795.34	11437.21	tonnes km^-2 year^-1
Total system throughput	19363.12	19807.02	22116.49	tonnes km^-2 year^-1
Sum of all production	12315.54	12348.39	12515.56	tonnes km^-2 year^-1
Mean trophic level of the catch	2.20	2.17	2.20
Gross efficiency (catch/net p.p.)	0.38	0.45	0.40	%
Total primary production/total respiration	5.49	4.85	3.03
Net system production	9492.62	9211.00	777.61	tonnes km^-2 year^-1
Total primary production/total biomass	0.88	0.88	0.87
Total biomass/total throughput	0.68	0.67	0.61	year^-1
Total biomass (excluding detritus)	13201.45	13236.39	13414.23	tonnes km^-2

Ecological network analyses under different scenarios of bivalve biomass

The size of the ecosystem increased due to the increase in bivalve biomass from scenarios 1 to 2, which can be seen from the elevated values of the total system throughput, capacity, and ascendency (Fig. 1). The Finn cycling index and ascendency increased dramatically due to the introduction of large bivalve biomass quantities. The network of material exchange, as indicated by average mutual information, decreased during the first two years of simulation in scenario 3 and increased volatility thereafter, with values higher than the initial values for scenarios 1 and 2. However, a further increase in bivalve biomass in scenario 3 had a negative effect on the diversity of flows, as described by the entropy of the ecosystem. Rapid increases in transfer efficiency were observed in the three scenarios during the first year, and then the values decreased. Interestingly, the transfer efficiency in scenario 1 was higher than that in scenarios 2 and 3.

Changing the bivalve biomass to ecological carrying capacity resulted in an increase in the total system biomass and caused changes in the model outputs of the other groups (Fig. 2). Bivalve predators, i.e., crustaceans, decapods, cephalopods and annelids, increased, but their prey, i.e., phytoplankton and microphytobenthos, decreased slightly. Furthermore, the expansion of bivalves caused the detritus biomass to increase.

Physiological activities

Fig. 3 and Fig. 4 show the physiological rates (ingestion, respiration, excretion and egestion) of three different sizes. It was obvious that the ingestion, respiration, excretion and egestion rates of R. philippinarum varied dramatically with different body sizes.

Clearance rates of clams ranged from 0.51 to 1.16 L g^-1 h^-1, with a mean value of 0.76 L g^-1 h^-1 (Fig. 3a). R. philippinarum in the small size group (mean shell length, 3.85 ± 0.46 cm) had a significantly higher clearance rate than those in the middle size group (mean shell length, 4.69 ± 0.19 cm) and the large size group (mean shell length, 5.31 ± 0.26 cm) (P < 0.05). Similar to the clearance rate, the average ingestion rates of POC and PON of clams were 5.22 mg POC g⁻¹h⁻¹and 0.79 mg PON g⁻¹h⁻¹ (Fig. 3b, c). Moreover, the egested POC rate ranged from 0.88 mg POC g⁻¹h⁻¹in the larger size group and 2.00 mg POC g⁻¹h⁻¹in the small size group, with an average rate of 1.30 mg POC g⁻¹h⁻¹; the egested rate of PON ranged from 0.13 to 0.30 mg PON g⁻¹h⁻¹,with a mean value of 0.20 mg PON g⁻¹h⁻¹(Fig. 4).

Respiration rates ranged from 0.11 to 0.16 mg carbon (C) g⁻¹ h⁻¹, with an average value of 0.15 mg C g⁻¹ h⁻¹ (Fig. 5a). In the small size and large size groups, significantly higher respiration rates were observed compared with those of the middle size group (P < 0.01). Similarly, the excretion rate ranged from 0.01 to 0.02 mg nitrogen (N) g⁻¹ h⁻¹, with an average value of 0.02 mg N g⁻¹ h⁻¹ (Fig. 5b). No significant difference was detected between the small size and large size groups (P > 0.05).

Carbon and nitrogen budget

Clams are one of the most representative bivalve species in the Ria de Aveiro ecosystem, and their physiological activities were used to estimate the C and N fluxes associated with bivalves in the entire system. When the bivalve biomass reached the ecological carrying capacity, the annual fluxes of C and N associated with bivalves in the Ria de Aveiro are shown in Fig. 6. Of the POC and PON ingested by the bivalves, 75% was assimilated and 25% was egested as feces or pseudofeces into the sediments. Subsequently, approximately 3.6% ~ 3.0% of assimilated POC and PON are respired and excreted by bivalves.

If the harvest biomass for bivalves was the value of maximum sustained yield (i.e., 88.92 tonnes km^-2 year⁻¹), the annual harvest would account for 581 tonnes C year⁻¹ and 83 tonnes N year⁻¹, respectively, and within this, approximately 481 tonnes C year⁻¹ and 24 tonnes N year⁻¹ would be stored in the bivalve shells.

Here, a food-web Ecopath model was used to determine the ecological carrying capacity for studying the potentially ecosystemic effects of a further increase in bivalve biomass on other species groups. The ecological carrying capacity was calculated following the method of EE (> 1) from the Ecopath model used by many researchers (Jiang and Gibbs 2005; Byron et al. 2011b; Carrie et al. 2011). For example, Jiang and Gibbs (2005) predicted the production and ecological carrying capacity of bivalve shellfish culture (corresponding to a biomass level of 65 tonnes km^− 2) in the Golden and Tasman Bays, and Carrie et al. (2011) calculated the ecological carrying capacity of oyster culture (corresponding to a biomass level of 297 tonnes km^− 2) using mass balance modeling in Narragansett Bay. Our study suggested that the current bivalve biomass level in the Ria de Aveiro is 142.90 tonnes km^− 2 and could be increased to 177.84 tonnes km^− 2 without exceeding the ecological carrying capacity. Although the estimated carrying capacity exceeds that found in the Golden and Tasman Bays (comparatively oligotrophic areas) (Jiang and Gibbs 2005), it is similar to the level of the Seine Estuary, with similar characteristics of shallow water depth and high primary production (Rybarczyk and Elkam 2003).

Accordingly, the maximum sustained yield of bivalves is estimated to be 88.92 tonnes km^− 2 year^− 1, and at that point, its growth rate is at its highest (Mace 2001). Bueno-Pardo et al. (2018) reported that the current artisanal fleet and leisure landings (only referring to legal catching) of bivalves in the Ria de Aveiro were 33.63 tonnes km^− 2 year^− 1, much lower than the estimated maximum sustained yield, suggesting that levels of fishing could increase while remaining sustainable.

A further increase in bivalves has impacts on the ecosystem of the Ria de Aveiro, as demonstrated by various ecological indicators, such as total system throughput, ascendency, Finn cycling index, Kempton’s Q, entropy, capacity, and transfer efficiency (Fig. 1). From scenarios 1 to 2, the total system throughput, representing the system’s size and energy flow (Hermosillo-Núñez et al. 2018), increased from 19363.12 to 19807.02 tonnes km^− 2 year^− 1, and consequently, the elevated ascendency was observed, improving the systems’ resistance to disturbances. Mature systems have been reported to have higher values in cycling accompanied by lower entropy (Odum 1969; Kluger et al. 2016), and thus, the increase in bivalve biomass up to 177.84 tonnes km^− 2 would lead to an increase in system maturity and stability. Similarly, the increase in bivalve biomass to ecological carrying capacity also indicates a more developed, mature and healthy system. However, a further increase in biomass to the level of scenario 3 had a negative effect on the biodiversity and health of the ecosystem, as shown by decreases in the indicators of Kempton’s Q and capacity. Moreover, an increasing level of bivalves causes a decrease in transfer efficiency, suggesting a less efficient transport of energy between different trophic levels and more energy being stored in detritus. Bivalve biomass continued expanding to 10 times the ecological capacity, which would eventually lead to the collapse of the Ria de Aveiro ecosystem (biomass of each functional group trends toward 0 with increasing simulation time; data not shown). Furthermore, the current and ecological carrying capacity states of the ecosystem in the following 20 years were compared using Ecosim models. As a major food source for bivalve mollusks, phytoplankton species are an important limiting factor for the biomass of bivalves; however, in the three scenarios evaluated, phytoplankton biomass was found to never be below 30% of the original value. It is possible that the v_new value of bivalves used for the simulations was constant (i.e., 2.25), thereby limiting the increase in feeding pressure on phytoplankton, and it may also be related to the relatively low value of EE (i.e., ~ 0.52), indicating the potential growth of the bivalve filter feeders without exhausting the phytoplankton (Kluger et al. 2016).

The increase in bivalve biomass has a bottom-up control on its predators, such as crustaceans, decapods, cephalopods and annelids, increasing the biomasses of these groups. However, interestingly, as the major competitors with bivalves for food, the zooplankton group was impacted only slightly during the simulations (Fig. 2). Here, zooplankton are mainly preyed upon by small fishes, such as sardine and zooplanktivorous fish. The increase in bivalves may lead to a decrease in the biomass of small fish groups through the increase in carnivorous fish, thereby reducing the predation pressure on the zooplankton (Kluger et al. 2016). This result is quite different from the studies of Jiang and Gibbs (2005) and Carrie et al. (2011), who assumed that bivalves would be completely replaced by zooplankton in the system.

Bivalve filter feeders perform a suite of ecosystem services. By filtrating phytoplankton, bivalves increase water transparency and may help prevent the occurrence of harmful algal blooms. In the Ria de Aveiro, clam clearance rates ranged from 0.51 to 1.16 L g^− 1 h^− 1, with a mean value of 0.76 L g^− 1 h^− 1 (Fig. 3a). With this filtering capacity, the number of days required by the bivalve community (reaching its ecological carrying capacity) to filter the entire seawater of the Ria de Aveiro was approximately 7 days. Dame and Prins (1998) and Murphy et al. (2016) reported that the bivalve population would control the phytoplankton if the time taken by them to filter the entire system was shorter than the turnover time of phytoplankton/primary production and the residence time of water mass. Here, Ria de Aveiro has relatively shorter phytoplankton turnover times (~ 1.7 days) (Bueno-Pardo et al. 2018; Frankenbach et al. 2020) and greatly varied system residence times (0 ~ 15 days) (Dias et al. 2007), implying that it might support the ecological carrying capacity of bivalves associated with a rapid clearance time (~ 7 days).

Bivalves have also been suggested as an indirect way to control eutrophication by extracting nutrients, called “nutrient bioextraction” (Bricker et al. 2014; Murphy et al. 2016). Some researchers have argued that this removal of nutrients should be assessed based on the harvest and regenerated N of bivalves to analyze the net impacts of bivalve activities on ecosystems (Nizzoli et al. 2005; Nizzoli et al. 2011; Murphy et al. 2016). However, others pointed out that the nutrients recovered in the sediments associated with biodeposition by bivalve farming ultimately come from primary production in the ecosystem and remain locked within the sediment (Stadmark and Conley 2011). In the Ria de Aveiro, approximately 83 tonnes N year^− 1 would be removed from the ecosystem through bivalve harvesting (when it reached the maximum sustained yield), which was similar to results reported by Stadmark and Conley (2011). PON was ingested by bivalves and egested as feces and pseudofeces. These biodeposits could be decomposed by the bacteria and further produce nitrogen gas through the processes of mineralization, nitrification and denitrification, which is also an important way to remove N. Here, approximately 2 tonnes year^− 1 would be removed through denitrification annually.

Furthermore, bivalves can significantly alter carbon fluxes in marine ecosystems, leading to carbon sequestration or storage. Bivalves convert the dissolved CO₂ into their CaCO₃ shells. This shell material can provide a long-term carbon sink and is considered a way to offset the increasing carbon in the atmosphere (Emery and New Egypt 2015; Murphy et al. 2016). Here, our study of the Portuguese population of Manila clam suggests that harvested bivalves of the maximum sustained yield represent the removal of ~ 481 tons of mineralized carbon annually (Fig. 6). As described earlier, the overall effect of bivalves on the carbon cycle is still controversial, as they can produce CO₂ during their own metabolic processes, similar to other animals. However, researchers recently reported that the CO₂ released by bivalves could support the photosynthesis of phytoplankton or macroalgae and could exclusively be recycled in marine ecosystems (Jiang et al. 2015). In many large-scale shellfish mariculture areas, such as Sanggou Bay, the annual uptake of dissolved inorganic carbon was up to 1.39 × 10⁵ t, indicating their high capacities of CO₂ absorption (Jiang et al. 2015). Therefore, further research is needed on the complex role of bivalves in the carbon cycle.

The developed Ecopath model successfully estimated the ecological carrying capacity (i.e., 177.84 tonnes km^− 2) and maximum sustained yield (i.e., 88.92 tonnes km^− 2 year^− 1) of bivalve filter feeders in Ria de Averio. Indeed, the harvested bivalves of the maximum sustained yield represent the removal of ~ 581 tonnes C and ~ 83 tonnes N annually, with substantial ecological and economic implications. Simulations of a further bivalve biomass increase suggest that exceeding the biomass level of 177.84 tonnes km^− 2 may cause the EE of other groups to become unrealistic, potentially leading to decreases in the ecosystem transfer efficiency, biodiversity, and health. The results can be used to guide the sustainable management of bivalves in Ria de Averio and the protection of the local environment.

Author contribution Weiwei Jiang: Conceptualization, Methodology, Writing - original draft, Writing - review & editing, Funding acquisition. Francesca Coppola: Investigation, Data curation, formal analysis, Writing - review & editing. Zengjie Jiang, Rosa Freitas and Yuze Mao: Methodology, Writing - review & editing, Funding acquisition. Zhijun Tan, Jinghui Fang, Jianguang Fang, Junwei WangandYitao Zhang: Investigation, Formal analysis, Writing - review & editing.

Funding The research was supported by the Natural Science Foundation of Shandong Province (ZR2021QD035), Key Programme for International Cooperation on Scientific and Technological Innovation, Ministry of Science and Technology (2017YFE0118300), National Natural Science Foundation of China (41761134052), Central Public-interest Scientific Institution Basal Research Fund, CAFS (2020TD50), Young Taishan Scholars Program of Shandong Province (tsqn201909166), and China Agriculture Research System of MOF and MARA. This work was also financially supported by the project ASARISAFE–Safety and sustainable management of valuable clam product in Portugal and China funded by the National Funds through the Portuguese Science Foundation (FCT). Francesca Coppola benefited from a PhD grant (SFRH/BD/118582/2016) given by FCT, supported by FSE and Programa Operational Capital Humano (POCH) e da União Europeia.

Conflict of Interest The authors declare no competing interests.

Ethics approval All applicable international and/or institutional guidelines for the care and use of animals were followed by the authors.

Consent to participate All authors agreed to participate in this research.

Consent for publication All authors agreed to publish this paper.

Bott TL (2006) Primary productivity and community respiration. In: Hauer FR, Lamberti GA (eds) Methods in stream ecology, 2nd edn. Academic Press, California, pp 663-690
Bueno-Pardo J, García-Seoane E, Sousa AI, Coelho JP, Morgado M, Frankenbach S, Ezequiel J, Vaz N, Quintino V, Rodrigues AM, Leandro S, Luis A, Serôdio J, Cunha MR, Calado AJ, Lillebø A, Rebelo JE, Queiroga H (2018) Trophic web structure and ecosystem attributes of a temperate coastal lagoon (Ria de Aveiro, Portugal). Ecol Model 378:13-25. https://doi.org/10.1016/j.ecolmodel.2018.03.009
Byron C, Link J, Costa-Pierce B, Bengtson D (2011a) Calculating ecological carrying capacity of shellfish aquaculture using mass-balance modeling: Narragansett Bay, Rhode Island. Ecol Model 222:1743-1755. https://doi.org/10.1016/j.ecolmodel.2011.03.010
Byron C, Link J, Costa-Pierce B, Bengtson D (2011b) Modeling ecological carrying capacity of shellfish aquaculture in highly flushed temperate lagoons. Aquaculture 314:87-99. https://doi.org/10.1016/j.aquaculture.2011.02.019
Carmel F (2011) All the fish in the sea: maximum sustainable yield and the failure of fisheries management. University of Chicago Press, Chicago
Crawford CM, Macleod CKA, Mitchell IM (2003) Effects of shellfish farming on the benthic environment. Aquaculture 224:117-140. https://doi.org/10.1016/S0044-8486(03)00210-2
Cheung WL, Watson R, Pitcher TJ (2002) Policy simulation of fisheries in the Hong Kong marine ecosystems. In: Pitcher T, Cochrane K. (eds.) The use of ecosystem models to investigate multispecies management strategies for capture fisheries. Fish Centre Res Rep 10: 46-54
Christensen V, Walters CJ (2004) Ecopath with Ecosim: methods, capabilities and limitations. Ecol Model 172:109-139. https://doi.org/10.1016/j.ecolmodel.2003.09.003
Cunha MA, Almeida MA, Alcântara F (2003) Ectoenzymatic activity and glucose heterotrophic metabolism in a shallow estuary (Ria de Aveiro, Portugal): Influence of bed sediments and salt marshes. Acta Oecologica 24:S97-S107. https://doi.org/10.1016/S1146-609X(03)00014-6
Dame RF, Prins TC (1998) Bivalve carrying capacity in coastal ecosystems. Aquat Ecol 31:409-421.
Dias JM, Abrantes I, Rocha F (2007) Suspended particulate matter sources and residence time in a mesotidal lagoon. J Coastal Res SI 50: 1034-1039
Dias JM, Lopes J, Dekeyser I (2000) Tidal propagation in Ria de Aveiro lagoon, Portugal. Phys Chem Earth B 25:369-374. https://doi.org/10.1016/S1464-1909(00)00028-9
Dias JM, Pereira F, Picado A, Lopes CL, Pinheiro JP, Lopes SM, Pinho PG (2021) A comprehensive estuarine hydrodynamics-salinity study: Impact of morphologic changes on Ria de Aveiro (Atlantic Coast of Portugal). J Mar Sci Eng 9:234. https://doi.org/10.3390/jmse9020234
Dumbauld BR, Ruesink JL, Rumrill SS (2009) The ecological role of bivalve shellfish aquaculture in the estuarine environment: A review with application to oyster and clam culture in West Coast (USA) estuaries. Aquaculture 290:196-223. https://doi.org/10.1016/j.aquaculture. 2009.02.033
Emery K, New Egypt N (2015) Coastal bivalve aquaculture carbon cycling, spatial distribution and resource use in Virginia, USA and Baja California, Mexico. M.Sc. Dissertation, University of Virginia.
Filgueira R, Rosland R, Grant J (2011) A comparison of scope for growth (SFG) and dynamic energy budget (DEB) models applied to the blue mussel (Mytilus edulis). J Sea Res 66:403-410. https://doi.org/10.1016/j.seares.2011.04.006
Frankenbach S, Ezequiel J, Plecha S, Goessling JW, Vaz L, Kühl M, Dias JM, Vaz N, Serôdio J (2020) Synoptic spatio-temporal variability of the photosynthetic productivity of microphytobenthos and phytoplankton in a tidal estuary. Front Mar Sci 7: 170. https://doi.org/10.3389/fmars.2020.00170
Han D, Yong C, Zhang C, Ren Y, Xu B, Ying X (2018) Evaluation of effects of shellfish aquaculture and capture fishery on a semi-closed bay ecosystem. Estuar Coast Shelf Sci 207:175-182. https://doi.org/10.1016/j.ecss.2018.04.005
Hermosillo-Núñez BB, Ortiz M, Rodríguez-Zaragoza FA, Cupul-Magaña AL (2018) Trophic network properties of coral ecosystems in three marine protected areas along the Mexican Pacific Coast: Assessment of systemic structure and health. Ecol Complex 36:73-85. https://doi.org/10.1016/j.ecocom.2018.06.005
Inglis GJ, Hayden BJ, Ross AH (2000) An overview of factors affecting the carrying capacity of coastal embayments for mussel culture. NIWA Client Report CHC00/69, Christchurch, New Zealand
Jiang W, Gibbs MT (2005) Predicting the carrying capacity of bivalve shellfish culture using a steady, linear food web model. Aquaculture 244:171-185. https://doi.org/10.1016/j. aquaculture.2004.11.050
Jiang W, Wang X, Rastrick SPS, Wang J, Zhang Y, Strand Ø, Fang J, Jiang Z (2021) Effects of elevated pCO₂ on the physiological energetics of Pacific oyster, Crassostrea gigas. ICES J Mar Sci. https://doi.org/10.1093/icesjms/fsab139
Jiang Z, Fang J, Mao Y, Han T, Wang G (2013) Influence of seaweed aquaculture on marine inorganic carbon dynamics and sea-air CO₂ flux. J World Aquacult Soc 44:133-140. https://doi.org/10.1111/jwas.12000
Jiang Z, Li J, Qiao X, Wang G, Bian D, Jiang X, Liu Y, Huang D, Wang W, Fang J (2015) The budget of dissolved inorganic carbon in the shellfish and seaweed integrated mariculture area of Sanggou Bay, Shandong, China. Aquaculture 446:167-174. https://doi.org/10.1016/j. aquaculture.2014.12.043
Kluger LC, Taylor MH, Mendo J, Tam J, Wolff M (2016) Carrying capacity simulations as a tool for ecosystem-based management of a scallop aquaculture system. Ecol Model 331:44-55. https://doi.org/10.1016/j.ecolmodel.2015.09.002
Lillebø AI, Stålnacke P, Gooch GD (2015) Coastal lagoons in Europe: Integrated water resource strategies. International Water Association, London, UK
Mace PM (2001) A new role for MSY in single-species and ecosystem approaches to fisheries stock assessment and management. Fish Fish 2:2-32. https://doi.org/10.1046/j.1467-2979. 2001.00033.x
Mao Y, Zhou Y, Yang H, Wang R (2006) Seasonal variation in metabolism of cultured pacific oyster, Crassostrea gigas, in Sanggou Bay, China. Aquaculture 253:322-333. https://doi.org/10.1016/j.aquaculture.2005.05.033
Mayzaud P, Conover R (1988) O:N atomic ratio as a tool to describe zooplankton metabolism. Mar Ecol Prog Ser 45:289-302. https://doi.org/10.3354/meps045289
McKindsey CW, Thetmeyer H, Landry T, Silvert W (2006) Review of recent carrying capacity models for bivalve culture and recommendations for research and management. Aquaculture 261:451-462. https://doi.org/10.1016/j.aquaculture.2006.06.044
Meyer S (2014) Evaluating the impact of scallop (Argopecten purpuratus) cultivation on the benthic infaunal commnity in Sechura Bay, Peru. M.Sc. Dissertation, University of Bremen.
Mistri M, Munari C (2012) Clam farming generates CO₂: A study case in the Marinetta lagoon (Italy). Mar Pollut Bull 64:2261-2264. https://doi.org/10.1016/j.marpolbul.2012.07.010
Murphy AE, Emery KA, Anderson IC, Pace ML, Brush MJ, Rheuban JE (2016) Quantifying the effects of commercial clam aquaculture on C and N cycling: An integrated ecosystem Approach. Estuar Coast 39:1746-1761. https://doi.org/10.1007/s12237-016-0106-0
Nizzoli D, Welsh DT, Bartoli M, Viaroli P (2005) Impacts of mussel (Mytilus galloprovincialis) farming on oxygen consumption and nutrient recycling in a eutrophic coastal lagoon. Hydrobiologia 550:183-198. https://doi.org/10.1007/s10750-005-4378-9
Nizzoli D, Welsh DT, Viaroli P (2011) Seasonal nitrogen and phosphorus dynamics during benthic clam and suspended mussel cultivation. Mar Pollut Bull 62:1276-1287. https://doi.org/10.1016/j.marpolbul.2011.03.009
Odum EP (1969) The Strategy of Ecosystem Development: An understanding of ecological succession provides a basis for resolving man’s conflict with nature. Science 164:262-270. https://doi.org/10.1126/science.164.3877.262
Panhwar SK, Liu Q (2013) Population statistics of the migratory hilsa shad, Tenualosa ilisha, in Sindh, Pakistan. Journal Appl Ichthyol 29:1091-1096. https://doi.org/10.1111/jai.12134
Pouvreau S, Bourles Y, Lefebvre S, Gangnery A, Alunno-Bruscia M (2006) Application of a dynamic energy budget model to the Pacific oyster, Crassostrea gigas, reared under various environmental conditions. J Sea Res 56:156-167. https://doi.org/10.1016/j.seares.2006.03.007
Prins TC, Smaal AC, Pouwr AJ, Dankers N (1996) Filtration and resuspension of particulate matter and phytoplankton on an intertidal mussel bed in the Oosterschelde estuary (SW Netherlands). Mar Ecol Prog Ser 142:121-134. https://doi.org/10.3354/meps142121
Rastrick SPS, Collier V, Graham H, Strohmeier T, Whiteley NM, Strand Ø (2018) Feeding plasticity more than metabolic rate drives the productivity of economically important filter feeders in response to elevated CO₂ and reduced salinity. ICES J Mar Sci 75:2117-2128. https://doi.org/10.1093/icesjms/fsy079
Redfield AC (1934) On the proportions of organic derivatives in sea water and their relation to the composition of plankton. In: Daniel RJ, editor James Johnstone Memorial Volume, University Press, Liverpool pp 176-192
Rybarczyk H, Elkam B (2003) An analysis of the trophic network of a macrotidal estuary: the Seine Estuary (Eastern Channel, Normandy, France). Estuar Coast Shelf Sci 58:775-791. https://doi.org/10.1016/S0272-7714(03)00184-7
Sher KP, Liu Q, Khan F, Siddiqui PJA (2012) Maximum sustainable yield estimates of ladypees, Sillago sihama (Forsskål), fishery in Pakistan using the ASPIC and CEDA packages. J Ocean Univ China 11:93-98. https://doi.org/10.1007/s11802-012-1880-3
Smaal A, Stralen Mv, Schuiling E (2001) The interaction between shellfish culture and ecosystem processes. Can J Fish Aquat Sci 58:991-1002. https://doi.org/10.1139/f01-026
Stadmark J, Conley D (2011) Mussel farming as a nutrient reduction measure in the Baltic Sea: Consideration of nutrient biogeochemical cycles. Mar Pollut Bull 62:1385-1388. https://doi.org/10.1016/j.marpolbul.2011.05.001
Bricker SB, Rice KC, Bricker OP (2014) From headwaters to coast: Influence of human activities on water quality of the potomac river estuary. Aquat Geochem 20:291-323. https://doi.org/10.1007/s10498-014-9226-y
Tang Q, Zhang J, Fang J (2011) Shellfish and seaweed mariculture increase atmospheric CO₂ absorption by coastal ecosystems. Mar Ecol Prog Ser 424:97-104. https://doi.org/ 10.3354/meps08979
van der Veer HW, Cardoso JFMF, van der Meer J (2006) The estimation of DEB parameters for various Northeast Atlantic bivalve species. J Sea Res 56:107-124. https://doi.org/ 10.1016/j.seares.2006.03.005
Velez C, Figueira E, Soares A, Freitas R (2015) Spatial distribution and bioaccumulation patterns in three clam populations from a low contaminated ecosystem. Estuar Coast Shelf Sci 155:114-125. https://doi.org/10.1016/j.ecss.2015.01.004
Villanueva MC, Lalèyè P, Albaret JJ, Laë R, Morais LTD, Moreau J (2006) Comparative analysis of trophic structure and interactions of two tropical lagoons. Ecol Model 197:461-477. https://doi.org/10.1016/j.ecolmodel.2006.03.016
Walters C, Christensen V, Pauly D (1997) Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Rev Fish Biol Fisher 7:139-172. https://doi.org/10.1023/A:1018479526149
Zhou Y, Yang HS, Liu SL, He YZ, Zhang FS (2002) Chemical composition and net organic production of cultivated and fouling organisms in Sishili Bay and their ecological effects. J Fish China 26:21-27 (In Chinese). https://doi.org/10.1007/s11769-002-0041-9

No competing interests reported.

SupplementaryTables.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

A food-web model as a tool for the ecosystem-level management of bivalves in an Atlantic coastal lagoon

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Model construction

Simulation scenarios

Physiological measurement

Statistical analysis

Results

Ecological carrying capacity and maximum sustained yield estimation

Ecological network analyses under different scenarios of bivalve biomass

Physiological activities

Carbon and nitrogen budget

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1