An Integrated ecological–economic modeling framework for the sustainable management of oyster farming

    An Integrated Ecological-Economic Modeling Framework for the Sustainable Management of Oyster Farming Carrie J. Byron, Di Jin, Tracey M. Dalton PII: DOI: Reference:

S0044-8486(14)00422-0 doi: 10.1016/j.aquaculture.2014.08.030 AQUA 631314

To appear in:

Aquaculture

Received date: Revised date: Accepted date:

31 January 2014 8 August 2014 18 August 2014

Please cite this article as: Byron, Carrie J., Jin, Di, Dalton, Tracey M., An Integrated Ecological-Economic Modeling Framework for the Sustainable Management of Oyster Farming, Aquaculture (2014), doi: 10.1016/j.aquaculture.2014.08.030

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT An Integrated Ecological-Economic Modeling Framework for the Sustainable Management of Oyster Farming Carrie J. Byrona*, Di Jinb, Tracey M. Daltonc

PT

a*

RI

Marine Sciences Department, University of New England, 11 Hills Beach Road, Biddeford, Maine 04005, United States [email protected] 1-207-602-2287 b

NU

SC

Marine Policy Center, Woods Hole Oceanographic Institution, 266 Woods Hole Road, Woods Hole, Massachusetts 02543, United States [email protected] c

MA

Department of Marine Affairs, University of Rhode Island, 1 Greenhouse Road, Kingston, Rhode Island 02281,United States [email protected]

D

Abstract

AC CE P

TE

Sustainable resource management requires improved understanding of complex ecological processes and the socioeconomic drivers shaping human-environment interactions. To better understand complex interconnections among ecological and economic systems, this study integrates a coastal marine ecosystem model with a model of the associated coastal economy. Through simulations of different ecological and socioeconomic scenarios, the integrated model can be used to generate predictive ecological and economic values for policy analysis, providing an opportunity for more rational and informed debate concerning sustainable marine resource development. To demonstrate utility of this integrated model, it was applied to coastal shellfish aquaculture production in Narragansett Bay, Rhode Island, US, a coastal ecological-economic system that provides important ecosystem services and contributes to the regional economy.

Keywords carrying capacity; aquaculture; bivalve; shellfish; Ecopath; IMPLAN *

Corresponding author

1

ACCEPTED MANUSCRIPT Introduction

SC

RI

PT

As the world demand for seafood products continues to expand, it is unlikely that the annual harvest of fish from wild stocks can be increased significantly. Thus, aquaculture, where practicable, now is recognized as the only means of increasing the supply of protein from seafood. Most seafood products today are traded in a competitive international market. The United States today imports about 70% of its seafood consumption, with a sizable seafood trade deficit, largely because seafood is produced inexpensively abroad. Based on ecological and market considerations, the most likely growth areas for US aquaculture include bivalve mollusks (Hoagland, et al., 2007). Thus, shellfish aquaculture will continue to be a driving force for socioeconomic and ecological change in complex coastal systems.

TE

D

MA

NU

Although the potential benefits associated with sustainable coastal aquaculture in the United States are arguably significant (Chopin, 2006), development of coastal aquaculture has been limited due to constraints in the regulatory system or lease-permitting process (Hopkins, et al., 1995), user conflicts in an inherently multi-use environment, and environmental and ecological concerns (Chu, et al., 2010; Whitmarsh and Palmieri, 2009). Improved understanding of complex ecological processes and the socioeconomic drivers shaping human-environment interactions can inform on-going policy discussions about sustainable marine resource development in general (Hughes, et al., 2005) and sustainable aquaculture development more specifically. In this study, an approach is presented for integrating a coastal marine ecosystem model with a model of the associated coastal economy to better understand complex interactions within integrated ecological-economic systems.

AC CE P

The economic-ecological modeling framework that we present is an extension of the traditional bioeconomic approach based upon simple biological growth functions (e.g., Clark 1976). Although the bioeconomic approach can involve nonlinear biological and technological interrelationships, most multispecies bioeconomic models incorporate only two species. In order to analyze systems with a large number of interacting elements, such as industries and consumers in an economy or species in an ecosystem, economists and ecologists have explored the use of linear models (e.g., IMPLAN and ECOPATH). The economic effects of different ecosystem conditions can be analyzed by linking ecological and economic models (Jin, et al., 2003) Most of the existing ecological models for shellfish aquaculture have been developed for and implemented at the production or farm scale (Bacher, 1998; Carver, 1990; Nunes, et al., 2003; Raillard, 1994), neglecting all trophic levels equal to or higher than bivalves. This approach is useful on a farm scale but ignores the impacts of aquaculture development on the stability and sustainability of the entire system. The integrated model presented here incorporates an ecosystem approach to model bivalve aquaculture that has been developed in a small number of recent studies (Byron, et al., 2011a; Byron, et al., 2011b; Jiang, 2005). Food web models can be used to examine species interactions and carrying capacity for aquaculture (Byron, et al., 2011a; Byron, et al., 2011b). It must be recognized that bivalves feeding on ambient sources of nutrients directly connects them with the food web of the body of water in which the farm is located. Changes in primary production and detrital production in the coastal environment have potential to influence bivalve production. Food web modeling can be 2

ACCEPTED MANUSCRIPT

PT

used to examine how different environmental conditions may influence standing stock biomass on the farm. Conversely, food web modeling can be used to examine different standing stock biomass levels on the rest of the ecosystem for the purpose of calculating carrying capacity of aquaculture in a system or identifying other species that may be strongly influenced by aquaculture (Byron, et al., 2011a; Byron, et al., 2011b).

NU

SC

RI

The integrated model is comprised of links between a food web model that includes aquaculture and an economic model of the associated coastal economy. In this study, we describe the integrated ecological-economic model and demonstrate how it can be used to assess the socioeconomic and ecological impacts of aquaculture development in a particular area. We will show that the model can be used to characterize existing economic and ecological conditions and demonstrate the potential wealth to society that may be derived from alternative scenarios of sustainable aquaculture development.

MA

Methods

TE

D

We developed an integrated modeling framework for assessing resources in a coupled ecological-economic system that was then applied to a well-studied area, Rhode Island, US, to demonstrate its potential as a decision-support tool for sustainable aquaculture development (Figure 1). The study area and methodology for each of the model components are described below.

AC CE P

Study Area

We applied the integrated ecological and economic framework to the issue of oyster aquaculture development in Narragansett Bay, Rhode Island, US (Desbonnet and Costa-Pierce, 2008), (Figure 2) where it has been argued that the capacity exists for increased aquaculture production (Byron, et al., 2011a). Currently, 0.1% of Narragansett Bay and Block Island waters and 2.0% of the coastal salt ponds along the southern coast of Rhode Island are being used for aquaculture farming. Although it is increasing, this level of farming is still far below the 5% regulatory limit set in 2008 (RI CRMP 300.11) and peak aquaculture levels of the early twentieth century when about one-third of Narragansett Bay was leased for cultivating oysters (Pietros, 2003). Today, most oysters in Narragansett Bay are grown in bags. While some stakeholders have expressed concerns about the increasing area used for aquaculture farms, others are promoting the industry’s growth and development (Lord, 2008). Results from our integrated model can inform on-going policy discussions about potential changes to the aquaculture industry in Rhode Island. We demonstrate the model in Rhode Island, however the approach can be easily extended to other areas facing similar issues related to aquaculture development. Ecological Model To simulate the effects of marine resource development, this study utilized an existing Ecopath model of Narragansett Bay (Byron, et al., 2011a) (Figure 3). Ecopath is a static mass-balance ecosystem-based modeling software that focuses on the energy transfers between trophic levels. Ecopath has been used for modeling a wide range of systems and management scenarios 3

ACCEPTED MANUSCRIPT

RI

PT

(Christensen and Pauly, 1993; Monaco, 1997; Vasconcellos, 1997). Ecopath differs from most other modeling approaches because it incorporates the full trophic spectrum of species, making it appropriate for use in estimating ecological carrying capacity and for undertaking policy analysis related to ecosystem-based management. Ecopath provides a methodology to standardize model outputs, making it easy to compare across systems. Overall, the Ecopath modeling framework provides a reasonable compromise between model simplicity (thereby enhancing transparency to users and stakeholders) and the detail of other more complex ecosystem models.

NU

SC

Ecopath operates according to two major equations: production = predation + catches + net migration + accumulated biomass + other mortality; and consumption = production + respiration + unassimilated food (Christensen et al. 2008). Expressed mathematically, these relationships are: (1)

D

MA

where Bi and Bj are biomasses of prey i and predator j; P/Bi is productivity (the production/biomass ratio); EEi is ecotrophic efficiency; Q/Bj is the food consumption per unit biomass j; DCji is the fraction of prey i in the average diet of predator j; BAi is the biomass accumulation rate for i; and Ei is the net migration of i (emigration minus immigration).

AC CE P

TE

Ecopath requires inputs for the three main parameters, B, P/B, and C/B for every defined functional group in the system (Christensen, et al., 2008). From these three parameters, one can calculate a fourth main parameter required for balancing, ecotrophic efficiency, EE, which is the proportion of the production that is utilized in the system. The final two input components that must be entered into the model for every functional group are the diet composition and fishery removals (i.e., aquaculture harvests or wild catches). Prior to balancing the model, additional non-Ecopath derived diagnostics are performed to evaluate the validity of the data. Diagnostics check and aid in balancing the model independent of Ecopath assumptions and prior to the mandatory Ecopath automated balancing routine. Diagnostic tests allow the evaluation of the cohesiveness of the data, despite natural discrepancies that occur when using myriad data sources measured across varying scales. Prebalancing diagnostics allow the modeler to look at the system holistically in lieu of the way individual data sources are collected piecemeal. When simple ecological and physiological “rules” were not met, as manifest in diagnostic tests, parameters were corrected to improve ecological integrity and validity. Diagnostic tests allowed greater control over the mass-balancing of the model by bringing the parameters closer to massbalance manually instead of relying solely on the automated Ecopath mass-balance routine (EwE version 4). For a detailed description of diagnostic tests for Ecopath, see Link (2010) and Byron (2011a; 2011b). Ecopath mass-balanced the model by slightly adjusting the input parameters within their confidence limits according to two master equations so that energy input and output were equal for each group (Christensen, et al., 2005) . Energy between groups was linked through the diet matrix. 4

ACCEPTED MANUSCRIPT

MA

NU

SC

RI

PT

As described in Byron et al. (2011a), an Ecopath model of Narragansett Bay was constructed to assess the ecological carrying capacity for aquaculture development. This food web model covers 15 species groups including a group specifically for “cultured oysters” (Figure 3). Parameter values for all species groups were based on recent field data from a variety of sources and agencies. In order to calculate carrying capacity, cultured oyster biomass and proportional cultured oyster harvest were increased in consecutive models until the system became unbalanced and no longer represented its present condition. The point just prior to any change in the system was the ecological carrying capacity. The results of the model simulation suggest that cultured oyster biomass was at 0.0095g DW m−2 in the mid-2000s and could be increased 625times to5.93g DW m−2 without exceeding the ecological carrying capacity. Assuming a conversion of 2% live to dry weight based on measurements made on oysters farmed in Rhode Island, that translates to a live weight of 0.47 t km−2 (surface area of the Bay) currently and 297 t km−2 at ecological carrying capacity (i.e., 105,435 t in the Bay). This ecological carrying capacity limit is based on food limitation at a water basin scale and not subject to grow-out techniques which influence specific farm sites within the basin. Economic Model

AC CE P

TE

D

A regional input-output (IO) model gives one an understanding of the effects of activities in one sector on all other industry sectors from which it purchases and to which it sells products. Thus, we can use an IO model to understand the economic “influence” of an industry in the study region on the other sectors of the economy to which they are linked. A static Leontief input-output model is a system of linear equations: (I - A)X = Y

(2) where I is a n  n identity matrix; A is a n  n technical coefficient (input-coefficient) matrix; X is a n  1 column vector denoting output; and Y is a n  1 column vector denoting final demand. The idea behind the model is that the output of any industry (xi, an element of X) is needed as an input in many other industries, or even in that industry itself. Therefore, the correct level of xi depends on the input requirements of all the n industries as well as final demand. The elements of A, aij, are called technical coefficients and are defined as: zij aij = xj

(3) where zij is the monetary value of the flow from sector i to sector j; and xj is the total output of sector j. The matrix (I-A) is called the technology matrix. If the technology matrix is not singular, the impact of changes in final demand (Y) on output (X) can be estimated as X = (I - A )-1 Y (4) where (I-A)-1 is called the Leontief inverse. For a comparative static analysis, we calculate the matrix of multipliers: X = (I - A )-1 (5) Y For empirical analysis, an input-output table (transactions table) includes all processing sectors 5

ACCEPTED MANUSCRIPT

RI

PT

(industries), final demand (including consumer/household purchases, private investment, government purchases, and exports), and the payments sector (value added including labor cost, capital cost, taxes, rental payments, and profit). Total industry outlays equal the value of total industry outputs. Outlays are payments made by firms for inputs and for other purposes in the payments sector. Inputs are purchased locally (within the region) or imported from outside the region. Outputs are goods or services produced by the industry. They can be consumed directly by households and others as final demand within the region, or sold to other industries as intermediate demand.

MA

NU

SC

An industry's contribution to the overall regional economy consists of three components: direct, indirect, and induced effects. In the case of aquaculture, if a farm starts its operation in the region, the associated jobs and income are the direct effects. Indirect effects are additional jobs and income growth in other industries, such as wholesale trade and truck transportation that can be indirectly credited to the aquaculture farm. The more inputs produced and purchased within the region, the greater the magnitude of the indirect effect. Finally, added jobs mean higher household income or a larger number of households in the region. Higher income leads to increased spending on food, housing, and cars. The latter are induced effects.

AC CE P

TE

D

For the study, we developed an input-output model for Rhode Island using the IMPLAN software and database (MIG, 2000). The IMPLAN database is compiled initially using county-level data from 19,171 individual industries. Related industries are grouped into 506 distinct industry “sectors,” including seafood-related sectors such as commercial fishing, seafood processing. Although aquaculture is grouped in the “animal production” sector in IMPLAN, the sector typically includes only fish farms in the data for most New England coastal counties. Linking Ecological Model with Economic Model and Model Scenarios To simulate the economic impacts of sustainable aquaculture development in Rhode Island, we use the results from the ecological modeling (i.e., the stock biomass at ecological carrying capacity) as inputs to the economic model. Three scenarios were examined: bivalve aquaculture development at (1) its current state (Beutel, 2012), (2) at 5% the surface area of water body (RICRMP, 2012), (3) ecological carrying capacity (Byron, et al., 2011a). To infer biomass from number of oysters or vice versa, some assumptions were made based on industry practices. Harvest size of an individual oyster was a minimum of 7.62 cm (3 inches). An allometric relationship between length and weight was developed from a subset of measured oysters at the project site (Y = 0.0009x2.4747). Farmers, on average, harvest 30% of their standing stock biomass. Scenario 1 was generated using a biomass of 933 t (= 280 t/0.3). Scenario 2 uses the regulatory limit for aquaculture in coastal salt ponds, which is 5% of the open water surface area (RI CMP 300.11). While the policy applies only to salt ponds at this time, we apply this level of aquaculture to Narragansett Bay for Scenario 2 (18.72 km2). Assuming a harvest level of 400.71 t/km2 on farm and 59.86 g/oyster (8.89 cm or 3.5 inches), the total harvest is 7,502 t (90 million pieces) of oysters, and the stock biomass is 25,007 t. Scenario 3 uses the shellfish stock biomass at ecological carrying capacity which is 130,037 t equating to harvest of 39,011 t. The value of output is estimated at $0.60/oyster (Table 1).

6

ACCEPTED MANUSCRIPT Results

RI

PT

The baseline data on Rhode Island marine industries in 2006 are summarized in Table 2. While the broadly defined marine industries accounted 9.26% of the state's total output value and 12.55% of employment, the size of the aquaculture industry was small ($1.3 million in output). The commercial fishing and seafood processing industries were significantly larger ($80 million and $70 million, respectively). However, the aquaculture industry has been growing rapidly since the mid-1990s (Figure 4). In 2012, the output value reached $2.8 million with a total farmed area of 172.55 acres (0.7 km2).

MA

NU

SC

Also shown in Table 2 are output and employment multipliers for different marine industry sectors. The multipliers capture the impacts on the economy of one unit increase in the output and employment. For example, every one dollar output in the aquaculture industry generates $1.32 impacts on the Rhode Island economy. The magnitude of multiplier for aquaculture is smaller than that of commercial fishing (1.74), because, compared with harvest fisheries which use vessels and gears to catch different species (including significant amounts of finfish), bivalve aquaculture products (mostly oysters) require little processing, and capital inputs to aquaculture farming are less.

AC CE P

TE

D

The wide range of connections of aquaculture industry to other economic sectors in Rhode Island is illustrated by the indirect impacts (Table 3). These industries (e.g., real estate, wholesale trade, truck transportation, and warehousing) are linked to aquaculture by either providing supplies to or receiving inputs from shellfish farming. Table 4 lists the sectors with the largest induced impacts. These sectors are connected to the aquaculture sector indirectly through Rhode Island households and their expenditures (e.g., housing, hospitals, and food services). The economic impacts associated with the three scenarios (Table 1), in terms of output value, employment, value added, employee compensation, and indirect business taxes, are presented in Table 5. For model scenario 1, the current state in 2012, the total output impact is $3.7 million. The output value rises to $100 million in model scenario 2 with 5% surface area farmed, and to $520 million at ecological carrying capacity in model scenario 3 (Figure 5). The corresponding increase in employment in the aquaculture industry is from 121 to 16,806 jobs with the total employment impacts across the Rhode Island economy rising from 129 to 17,945 jobs (Figure 5). Note that at ecological carrying capacity, the output and employment figures for the aquaculture industry would be significantly higher than those for the harvest fisheries and seafood processing combined. The results highlight the potential economic benefits associated with aquaculture development.

Discussion Multiple linkages exist between the coastal economy and the marine ecosystem. By linking an ecosystem model with a model of a coastal economy, we demonstrate ecological and economic effects of different levels of sustainable aquaculture development. The use of two distinct models that share information at different stages maintains the complexity in each model (Bockstael, et al., 1995; Dalton, 2004). For instance, the ecosystem model incorporates the full 7

ACCEPTED MANUSCRIPT

PT

trophic spectrum of species, making it appropriate for use in estimating ecological carrying capacity, and the economic model captures linkages among all the industries in the region. By using results from the ecosystem model as input for the economic model, we developed and tested different scenarios for aquaculture development that were based on a more comprehensive understanding of the ecological and economic conditions of the area.

AC CE P

TE

D

MA

NU

SC

RI

We demonstrate the utility of the approach by applying it to a case study in the Northeast US. The aquaculture industry in Rhode Island is still quite small which is reflected in the economic impact (Table 1; Figure 5). However, previous modeling studies show great potential for aquaculture growth in the state, including Narragansett Bay (Byron, et al., 2011a; Byron, et al., 2011b; Byron, et al., 2011c). As we note, some stakeholders have expressed concerns about rapid growth in the industry while others are promoting the industry’s development (Lord, 2008). A recent initiative to plan for and manage shellfish in Rhode Island state waters has brought attention to the relevance of the 5% regulatory limit for aquaculture. Our results can inform these policy discussions by providing the potential economic impacts of increased shellfish aquaculture in Rhode Island waters. For instance, increasing the industry’s coverage to 5% of Narragansett Bay fully developed with aquaculture at maximum sustainable production rates would increase the total economic impacts from the industry by $96 million. Increasing coverage to the ecological carrying capacity level would result in $520 million in total impacts to the state. Although this potential growth may seem quite impressive given the current state of aquaculture, coastal planners and managers should understand that promoting industry growth to the ecological carrying capacity limit is risky given that static modeling techniques were used to represent a dynamic system. The ecological carrying capacity is simply meant to represent the upper limit of potential growth. Coastal planners and managers can use these estimates to evaluate economic tradeoffs between different levels of aquaculture development. As State and Federal regulatory agencies in the US actively promote sustainable aquaculture development in coastal waters (NOAA, 2011), it is useful to understand ecological and economic impacts of different aquaculture options. The integrated model presented here can be applied to other regions throughout the US where coastal planners and managers are considering tradeoffs among different uses. Future steps

In this manuscript we demonstrate linkages between ecological and economic systems using an integrated modeling framework. This demonstration is meant to serve as an example for application of this framework. We also want to note additional potential of this framework which can be adapted to accommodate (1) non-linear interaction in the coastal economy, (2) social preferences and (3) anthropogenic disturbances on the ecosystem, such as climate change. We describe how nonlinear interactions, social surveys and climate change factors can be included in the framework. Non-linear interaction Although linear economic models (e.g., IO) can handle a large number of variables (industry sectors), the approach often is limited only to descriptive studies. These models do not capture 8

ACCEPTED MANUSCRIPT

RI

PT

some key nonlinear interactions in the relevant economy, such as supply and demand for goods and services, and they cannot really be used to examine economic efficiency and welfare changes. To address these issues, a computable general equilibrium (CGE) model that incorporates several marine resource development sectors could be used (Jin, et al., 2012). Extending the integrated model presented here by incorporating a CGE model would allow for estimation of normative welfare aspects of the ecological-economic system under different aquaculture development options. Socioeconomic tradeoffs

Climate Change

TE

D

MA

NU

SC

Although regional economic models can simulate market interactions among different industry sectors, these models do not directly capture some important nonmarket effects, such as stakeholders’ perceptions of different levels and types of development options. These nonmarket effects are crucial in the identification and quantification of potential impacts of aquaculture development, which is important for comparing societal impacts of different development options. For instance, in our example, it is ecologically feasible and economically beneficial to expand aquaculture development to cover 26% of Narragansett Bay, however it is not clear how Rhode Island stakeholders would respond to such a proposal. In fact, recent discussions suggest that this level of development would exceed acceptable levels in Rhode Island (CRMC, 2008; RICRMP, 2012). Extending the integrated model to incorporate stakeholders’ perceptions on different development options would help to identify more reasonable alternatives for evaluation.

AC CE P

Climate change may have many impacts on the coastal marine ecosystem, such as change in species composition, ocean acidification, or primary production, which can be captured, indirectly, by adapting the food web model. Climate change alters preferred habitats and, consequently, affects stock migrations and concentrations, and the distribution and mix of species (Grafton, 2010; Hannesson, 2007). Changes in species abundances can be explored by changing species biomasses in the food web model. Ocean acidification affects mollusks. Lower pH levels hinder the growth of calcium carbonate shells and skeletons of many marine plants and animals. Controlled experiments and modeling exercises provide some insight into how chemical changes can affect physiological and biological processes on specific species and can be explored in the food web model by changing the vital rates of shellfish (Le Quesne and Pinnegar, 2012). Lastly, there are several pathways by which phytoplankton can fuel higher trophic levels (i.e. diverted to benthos, microbial look, or directly consumed by zooplankton and fish) (Marquis, et al., 2011). Because of these different pathways, not all primary production is available for higher trophic level species. It is also not assumed that an increase in primary production will directly translate to a proportional increase in biomass at higher trophic levels due to transfer efficiency and loss of energy at each trophic level. Climate change may have numerous other impacts on the food web and can generally be explored indirectly by manipulating species biomasses and vital rates accordingly.

Conclusion

9

ACCEPTED MANUSCRIPT

RI

PT

We demonstrate a methodology for coupling human and natural perspectives in a modeling framework. It is not enough to examine resource needs, uses, and impacts from a single perspective. By acknowledging that aquaculture is an integrated ecological-economic system and creating tools to evaluate it as an entire intact system we can better understand limitations, tradeoffs, and alternative management scenarios. As additional model components are refined (as described above in “future steps”), we intend on applying this framework to alternative project areas at different scales and for other resources.

SC

We showed the aquaculture industry in Rhode Island is closely linked to a wide range of other industries, and thus, growth in aquaculture will have additional positive effects on the state economy.

MA

NU

At ecological carrying capacity, the output and employment figures for the aquaculture industry in Rhode Island would be significantly higher than those for the harvest fisheries and seafood processing combined. The results highlight the potential economic benefits associated with sustainable aquaculture development in Rhode Island.

Acknowledgements

AC CE P

TE

D

Funding for this project is provided by NSF SEES Fellows (CHE-1313963). Dave Beutel at the Rhode Island Coastal Resources Management Council and the Working Group for Aquaculture Regulations identified the need and provided the motivation for this project.

10

ACCEPTED MANUSCRIPT References

AC CE P

TE

D

MA

NU

SC

RI

PT

Bacher, C., P. Duarte, J. G. Ferreira, M. Héral, O. Raillard, 1998. Assessment and comparison of the Marennes-Oléron Bay (France) and Carlingford Lough (Ireland) carrying capacity with ecosystem models. Aquatic Ecology. 31, 379-394. Beutel, D., 2012. Aquaculture in Rhode Island. 2012 Annual Status Report. Coastal Resources Management Council. http://www.crmc.ri.gov/aquaculture/aquareport12.pdf Bockstael, N., Costanza, R., Strand, I., Boynton, W., Bell, K., Wainger, L., 1995. Ecological economic modeling and valuation of ecosystems. Ecological Economics. 14, 143-159. Byron, C., Link, J., Bengtson, D., Costa-Pierce, B., 2011a. Calculating Carrying Capacity of Shellfish Aquaculture Using Mass-Balance Modeling: Narragansett Bay, Rhode Island. Ecological Modelling. 222, 1743-1755. 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. Byron, C., Bengtson, D., Costa-Pierce, B., Calanni, J., 2011c. Integrating Science into Management: Ecological Carrying Capacity of Bivalve Shellfish Aquaculture. Marine Policy. 35, 363-370. Carver, C.E.A., A. L. Mallet, 1990. Estimating the carrying capacity of a coastal inlet for mussel culture. Aquaculture. 88, 39-53. Chopin, T., 2006. Integrated multi-trophic aquaculture. What it is, and why you should care...and don't confuse it with polyculture. Northern Aquaculture. July/August, 4. Christensen, V., Pauly, D., 1993. Trophic models of aquatic ecosystems. International Center for Living Aquatic Resources Management (ICLARM) Conference Proceedings. 26, pp.390 Christensen, V., Walters, C.J., Pauly, D., 2005. Ecopath with Ecosim: a User's Guide, Fisheries Centre. November 2005 edition, University of British Columbia, Vancouver, pp. 154. Christensen, V., Walters, C.J., Pauly, D., Forrest, R., 2008. Ecopath with Ecosim version 6 User Guide. Chu, J., Anderson, J.L., Asche, F., Tudur, L., 2010. Stakeholders' perceptions of aquaculture and implications for its future: a comparison of the USA and Norway. Marine Resource Economics. 25, 61-76. Clark, C., 1976. Mathematical Bioeconomics: The Optimal Management of Renewable Resources. John Wiley & Sons, New York. CRMC, 2008. Coastal Resouces Management Council, Working Group on Aquaculture Regulations, Report on the Deliberations of the Group and Suggestions to Guilde Future Development. http://www.crmc.ri.gov/aquaculture/riaquaworkinggroup/CRMC_WG_AquaPlan.pdf Dalton, T.M., 2004. An approach for integrating economic impact analysis into the evaluation of potential marine protected area sites. Journal of Environmental Management. 70, 333349. Desbonnet, A., Costa-Pierce, B., 2008. Science for Ecosystem-Based Management: Narragansett Bay in the 21st Century. Springer, New York. Grafton, Q.R., 2010. Adaptation to climate change in marine capture fisheries. Marine Policy. 34, 606-615. Hannesson, R., 2007. Global warming and fish migrations. Natural Resource Modeling. 20, 301319. 11

ACCEPTED MANUSCRIPT

AC CE P

TE

D

MA

NU

SC

RI

PT

Hoagland, P., Kite-Powell, H., Jin, D., Schumacher, M., Katz, L., Klinger, D., 2007. Economic sustainability of marine aquaculture, A report to the Marine Aquacultuer Task Force. Marine Policy Center, Woods Hole Oceanographic Institution. Hopkins, J.S., Devoe, M.R., Holland, A.F., 1995. Environmental impacts of shrimp farming with special reference to the situaitno in the ocntinental United States. Estuaries. 18, 25-42. Hughes, T.P., Bellwood, D.R., Folke, C., Steneck, R.S., Wilson, J., 2005. New paradigms for supporting the resilience of marine ecosystems. Trends in Ecology and Evolution. 20, 380-386. Jiang, W., Mark T. Gibbs, 2005. Predicting the carrying capacity of bivalve shellfish culture using a steady, linear food web model. Aquaculture. 244, 171-185. Jin, D., Hoagland, P., Dalton, T.M., 2003. Linking economic and ecological models for a marine ecosystem. Ecological Economics. 46, 367-385. Jin, D., Hoagland, P., Dalton, T.M., Thunberg, E.M., 2012. Development of an integrated economic and ecological framework for ecosystem-based fisheries management in New England. Progress in Oceanography. 102, 93-101. Le Quesne, W.J.F., Pinnegar, J.K., 2012. The potential impacts of ocean acidification: scaling from physiology to fisheries. Fish and Fisheries. 13, 333-344. Link, J.A., 2010. Adding rigor to ecological network models by evaluating a set of pre-balanced diagnostics: A plea for PREBAL. Ecologcial Modelling. 221, 1580-1591. Lord, P., 2008. Limit aquaculture to 5 percent R.I. waters, experts say. Providence Journal. March 2, 2008. Marquis, E., Niquil, N., Vézina, A.F., Petitgas, P., Dupuy, C., 2011. Influence of planktonic foodweb structure on a system's capacity to support pelagic production: an inverse analysis approach. ICES Journal of Marine Science: Journal du Conseil. 68, 803-812. MIG, 2000. Minnesota IMPLAN Group, Inc. IMPLAN Professional Version 2.0, Stillwater, MN. Monaco, M.E., Robert E. Ulanowicz, 1997. Comparative ecosystem trophic structure of three U.S. mid-Atlantic estuaries. Marine Ecology Progress Series. 161, 239-254. NOAA, 2011. NOAA Marine Aquaculture Policy. http://www.nmfs.noaa.gov/aquaculture/docs/policy/noaa_aquaculture_policy_2011.pdf. Nunes, J.P., Ferreira, J.G., Gazeau, F., Lencart-Silva, J., Zhang, X.L., Zhu, M.Y., Fang, J.G., 2003. A model for sustainable management of shellfish polyculture in coastal bays. Aquaculture. 219, 257-277. Pietros, J.M., Michael A. Rice, 2003. The impacts of aquacultured oysters, Crossostrea virginica (Gmelin, 1791) on water column nitrogen and sedimentation: results of a mesocosm study. Aquaculture. 220, 407-422. Raillard, O., A. Ménesguen, 1994. An ecosystem box model for estimating the carrying capacity of a macrotidal shellfish system. Marine Ecology Progress Series. 115, 117-130. RICRMP, 2012. The State of Rhode Island Costal Resources Management Program (as amended), http://www.crmc.ri.gov/regulations/RICRMP.pdf. Vasconcellos, M., Mackinson, Steven, Sloman, Katherine, Pauly, Daniel, 1997. The stability of trophic mass-balance models of marine ecosystems: a comparative analysis. Ecological Modelling. 100, 125-134. Whitmarsh, D., Palmieri, M.G., 2009. Social acceptability of marine aquaculture: The use of survey-based methods for eliciting public and stakeholder preferences. Marine Policy. 33, 452-457.

12

ACCEPTED MANUSCRIPT Table 1. Model Scenarios: Rhode Island Aquaculture at Different Production Scales

AC CE P

TE

D

MA

NU

SC

RI

PT

Percent of Farmed Area Output Surface Area km2 (t) Scenario 1: Current condition (2012) 0.19% 0.70 280 Scenario 2: 5% surface area 5.00% 18.72 7,502 Scenario 3: Ecological carrying capacity 26.00% 97.35 39,011

13

Value (2012 $) 2,822,734 75,681,448 393,543,532

ACCEPTED MANUSCRIPT Table 2: Rhode Island Marine Industry Output, Employment, and Multipliers

% of State Total †

Output†

Employ ment‡ 2,195 252 57 916 2,374 1,601

SAM Multiplier* 1.22 5.10 1.07 1.81 1.64 2.41

106.8 75.3 2,083.5 310.9 45.6 393.3

1.41 1.44 1.55 1.49 1.61 1.51

246 222 40,126 3,884 424 4,603

2.43 2.34 1.25 1.33 1.54 1.39

2,339.1

1.35

18,628

1.38

SC

RI

80.2 70.4 1.3 184.6 554.1 523.2

PT

SAM Multiplier* 1.74 2.02 1.32 1.51 1.36 1.60

NU

MA

Real Estate Marine Industry Totals State Totals

D

Shipping Water Quality Tourism

TE

Shipbuilding

IMPLAN Sector Fishing Seafood product preparation and packing Aquaculture** Ship building and repairing Boat building Search, detection, and navigation instruments Water transportation Water, sewerage and other systems Food services and drinking places Hotels and motels, including casino hotels Other accommodations Other amusement, gambling, and recreation industries Real Estate

AC CE P

Broad Industry Fisheries

6,770.7 73,136.6

75,628 602,830

9.26%

12.55%

$ millions (2006) Number of employees * SAM multipliers capture the direct, indirect, and induced effects where the induced effect is based on information in the social account matrix (SAM). ** Output and employment figures are from Beutel (2012); multipliers are for IMPLAN sector 13 includes aquaculture, fish hatcheries and other animal production. ‡

14

ACCEPTED MANUSCRIPT Table 3: Indirect Output Impacts of Aquaculture on Other Industry Sectors† IMPLAN Code 431 390 394 427 10 30 430 13 449 392 43 499 400 160 422 425 428 240 18 438 111 485 493 437 483

Industry Sector

AC CE P

TE

D

MA

NU

SC

RI

PT

Real estate Wholesale trade Truck transportation Insurance carriers All other crop farming Power generation and supply Monetary authorities and depository credit intermediation* Aquaculture Veterinary services Rail transportation Maintenance and repair of nonresidential buildings Other State and local government enterprises Warehousing and storage Pharmaceutical and medicine manufacturing Telecommunications Nondepository credit intermediation and related activities* Insurance agencies, brokerages, and related Metal can, box, and other container manufacturing Agriculture and forestry support activities Accounting and bookkeeping services Other leather product manufacturing Commercial machinery repair and maintenance Civic, social, professional and similar organizations Legal services Automotive repair and maintenance, except car washes Others Total † Direct impact is the output value in 2012 ($2,822,734). *Banking

15

Value (2012 $) 141,732 63,083 37,437 28,407 25,065 22,780 19,551 18,437 18,336 17,350 13,687 13,226 12,800 8,106 8,004 7,952 7,447 7,391 7,011 6,473 6,028 5,681 5,051 4,565 4,539 63,366 573,515

ACCEPTED MANUSCRIPT Table 4: Induced Output Impacts of Aquaculture on Other Industry Sectors† IMPLAN Code 509 467 465 481 431 427 390 430 401 160 410 462 468 405 466 426 422 499 30 425 470 437 483 428 412

Industry

AC CE P

TE

D

MA

NU

SC

RI

PT

Owner-occupied dwellings Hospitals Offices of physicians, dentists, and other health practitioners Food services and drinking places Real estate Insurance carriers Wholesale trade Monetary authorities and depository credit intermediation* Motor vehicle and parts dealers Pharmaceutical and medicine manufacturing General merchandise stores Colleges- universities- and junior colleges Nursing and residential care facilities Food and beverage stores Other ambulatory health care services Securities, commodity contracts, investments Telecommunications Other State and local government enterprises Power generation and supply Nondepository credit intermediation and related activities* Social assistance- except child day care services Legal services Automotive repair and maintenance- except car wash Insurance agencies, brokerages, and related Nonstore retailers Others Total † Direct impact is the output value in 2012 ($2,822,734). *Banking

16

Value (2012 $) 49,652 25,897 21,456 18,548 16,743 11,025 10,997 9,800 7,639 7,170 6,561 6,217 5,823 5,705 5,555 5,533 5,389 5,221 3,988 3,867 3,776 3,688 3,338 2,915 2,894 86,096 335,501

ACCEPTED MANUSCRIPT Table 5. Economic Impacts of Aquaculture Development (2012$) Indirect 573,515 5.2 327,802 135,342 36,696

75,681,448 3,231.9 9,335,112 9,182,566 1,166,602

Scenario 3: Carrying Capacity Output Value 393,543,520 Employment* 16,805.9 Value Added 48,542,584 Employee Compensation 47,749,344 ** Indirect Business Taxes 6,066,328

3,731,750 128.7 881,695 576,819 99,842

15,376,743 138.4 8,788,823 3,628,709 983,867

8,995,250 80.7 5,515,540 2,654,041 526,423

100,053,439 3,451.0 23,639,475 15,465,316 2,676,891

79,959,064 719.9 45,701,881 18,869,286 5,116,107

46,775,299 419.4 28,680,805 13,801,014 2,737,401

520,277,890 17,945.3 122,925,271 80,419,647 13,919,835

SC

NU

MA

AC CE P

TE

D

Scenario 2: 5% surface area Output Value Employment* Value Added Employee Compensation Indirect Business Taxes**

Total

335,501 3.0 205,716 98,989 19,634

RI

2,822,734 120.5 348,177 342,487 43,511

Induced

PT

Direct Scenario 1: Current state Output Value Employment* Value Added Employee Compensation Indirect Business Taxes**

* Number of employee ** Indirect business taxes consist of excise taxes, property taxes, fees, licenses, and sales taxes paid by businesses. These taxes occur during the normal operation of businesses but do not include taxes on profit or income.

17

ACCEPTED MANUSCRIPT Figures

SC

Figure 2. Study Area: Narragansett Bay, Rhode Island, US.

RI

PT

Figure 1. Integrated Ecological and Economic Model Framework for Sustainable Aquaculture Development. The coastal marine ecosystem is represented by a food web model. The coastal economy is represented by a regional economic model. These two model components interact with each other to make ecological and economic predictions that can be used for decision making in the management of shellfish aquaculture. The arrows represent how information is shared between nodes.

NU

Figure 3. Food web of Narragansett Bay. Each node represents a functional group of species and each line represents transfer of energy between groups. The size of the node indicates quantity of biomass and the thickness of the line indicates amount of energy flow. Horizontal lines indicate trophic levels.

MA

Figure 4. Aquaculture Development in Rhode Island. The aquaculture output value (black solid line) and the area farmed (grey dashed line) increased over the past decade. Data for this figure taken from Beutel (2012).

AC CE P

TE

D

Figure 5. Growth Potential for Aquaculture in Rhode Island. Output in 2012 dollars (black bar) and the number of people employed in industries associated with aquaculture (grey bar) increase by one or two orders of magnitude from the current (2012) condition (Beutel, 2012) and aquaculture development to 5% of surface area (RICRMP, 2012) or to ecological carrying capacity (Byron, et al., 2011a), respectively.

18

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 1

19

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 2

20

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC CE P

TE

D

Figure 3

21

AC CE P

Figure 4

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

22

AC CE P

Figure 5

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

23

ACCEPTED MANUSCRIPT

Highlights

TE

D

MA

NU

SC

RI

PT

Interdisciplinary approach to management of shellfish aquaculture Interconnections between ecology and society for holistic analyses of aquaculture Modeling framework for understanding trajectories of change amid global challenges

AC CE P

  

24