A Scientific Review of the Potential Environmental Effects of Aquaculture in Aquatic Ecosystems - Volume 3

Near-Field Organic Enrichment from Marine Finfish Aquaculture

D.J. Wildish
Marine Environmental Sciences, Fisheries and Oceans Canada St. Andrews Biological Station, St. Andrews, New Brunswick

M. Dowd
Dept. of Mathematics and Statistics, Dalhousie University, Halifax, Nova Scotia

T.F. Sutherland
Marine Environment and Habitat Sciences, Fisheries and Oceans Canada, West Vancouver Laboratory, West Vancouver, British Columbia

C.D. Levings
Marine Environment and Habitat Sciences, Fisheries and Oceans Canada, West Vancouver Laboratory, West Vancouver, British Columbia

Executive Summary

This paper reviews the literature on near-field organic enrichment associated with intensive marine finfish aquaculture. Fish farms are a source of suspended and dissolved organic matter, originating as fish feces, excess fish feed and net-cleaning wastes. The term “near-field” is used to differentiate between local (footprint limited) and more distant (far-field) effects, and refers to effects within the sedimentary footprint and between the population and community level. Near-field effects of mariculture are bounded by the physical limits of particulate waste dispersal and sedimentation from individual cages or farms.

Particle Transpor, Dispersion and Behaviour

To investigate the biogeochemical fate of organic matter in the benthic or pelagic ecosystem, it is necessary to understand how this material is transported and dispersed from the fish farm. Organic matter is transported from fish cages to the surrounding marine environment by the action of all types of water movements in the immediate vicinity of the cages. As this material moves, it spreads out and its concentration decreases, both by dilution and sedimentation.

Canadian coastal environments where aquaculture occurs are characterized by irregular coastlines and complex topography. Consequently, they often exhibit highly structured and quite complex flow fields. In strongly stratified marine systems, dissolved material can be effectively trapped in the upper or lower parts of the water column. In an aquaculture context, stratification can be an important factor in the dispersal of organic matter from certain farms in the inner portions of fjords in British Columbia.

Observational studies and numerical modelling are used to study transport and dispersion in the coastal zone. Observational studies of mixing often rely on the deployment of drifting buoys that mimic the movement of water parcels. Field experiments releasing coloured dye into the ocean are better able to characterize dispersion from a point source, but these are often expensive, logistically difficult and environmentally sensitive. Numerical circulation models, based on a set of mathematical equations that govern fluid motion, provide a practical solution to the problem of coastal mixing.

In the water column, the behaviour of particles of organic matter is characterized by settling and deposition. Settling rate depends on the size, density and shape of particles. Stokes' law provides a basic framework for predicting the size dependence of particle settling. While discrete particles exhibit Stokes' settling behaviour, attractive forces also cause the formation of particle aggregates. Particle aggregation can effectively increase the settling flux of the original disaggregated particle mixture by repackaging small particles into larger, more rapidly sinking ones.

When particles settle from the water column, they may accumulate at the sediment-water interface or be subject to further transport and redistribution. When bottom shear stress exceeds a critical value, particles at the boundary between the sediment and water can be set in motion. Susceptibility to resuspension is set by a variety of sediment characteristics, including particle density and size, and degree of consolidation. Factors such as bioturbation have also been shown to effect erodibility of sediments (Andersen 2001). Numerical models are used to predict the temporal and spatial distribution of particulates in the marine environment.

Organic Enrichment as a Process

Organic matter flux of fish feces, waste feed and detached fouling organisms from cage surfaces, as well as natural particles, is monitored by measurements based on nitrogen or carbon in sedimentation traps. In the context of aquaculture, many authors examine carbon determined pyrolitically as a measurement of organic matter flux. However, the chemical and physical state of organic matter in sediments is not well understood. Thus, quantitative measures of total organic carbon are poor indicators of biochemical availability to biota.

The input of organic carbon to sediments, or sedimentation rate, is measured in suspended sediment traps. Reported sedimentation rates under and near salmon mariculture cages vary from 1 to 181 g C·m-2·day-1. Notwithstanding differences in environmental conditions, farm size and fish stocking densities, some of the variation in sedimentation rates is due to inherent difficulties in suspended sediment trap methodology. The benthic input from fish feces and waste feed can also be measured by mass balance calculations: based on total organic carbon accumulated in sediment, total input of fish feed, growth period and surface area of the cage footprint. However, all farm sites have some water movement and therefore require an examination of where the particles are deposited. Cromey et al. (2002) developed a particle tracking model that uses depth and particle settling rates as affected by observed currents with modelled shear velocity and turbulence.

In soft sediments receiving natural background levels of sedimentation (0.1 - 1.0 g C·m-2·day-1), the sediment-water interface is aerobic, and this aerobic layer contains abundant macrofauna and meiofauna. In contrast, soft depositional sediments receiving a continuous elevated level of sedimentation, for example due to fish feces and waste feed, become anoxic at the interface and develop a black, top-down sulfide layer. Some of the sulfide produced may also be oxidized by other chemoautotrophic bacteria (e.g.Beggiaotoa sp.) (Lumb and Fowler 1989). The white patches of this bacteria may be seen on the sediment surface under fish cages. Hargrave (1994) considered that flux rates of fish feces and waste feed of >1.0 g C·m-2·day-1 caused marked organic enrichment effects in net depositional sediments under salmon cages. The process of organic enrichment in farm footprints requires a varying amount of time to reach an equilibrium point, depending on oceanographic and substrate conditions. Some studies have shown that organic enrichment can be reversed if the increased levels of sedimentation are stopped (Brooks et al. 2003).

The initial effect of a sudden addition of a high flux of readily decomposable organic carbon to sediments is increased metabolism by aerobic bacteria, leading to hypoxia and anoxia causing death of the most susceptible aerobic life forms (Gray et al. 2002). In the fish cage footprint, the bulk of the metabolism is by sulfate reduction at a higher rate than at reference stations (Holmer and Kristensen 1992) and results in the loss of nitrification and denitrification pathways (Kaspar et al. 1988). The death of burrowing macrofauna leads to a rapid decline in the capability for irrigation or aerated water within the upper profile and a more rapid development of anoxia. The predominant bacteria become anaerobes, principally sulfate reducers and methanogenic bacteria. Methane makes up the bulk of the outgassing from heavily impacted farm sediments (Wildish et al. 1990), and sulfate reduction produces hydrogen sulfide, which is readily oxidized in aerobic seawater.

Organic enrichment indices may be regarded as proxies of the ecosystem to indicate extent or amount of organic enrichment. The most widely accepted index is based on macrofaunal species, abundance and biomass. Organic enrichment gradients by Pearson and Rosenberg (1978) and Poole et al. (1978) can be arbitrarily divided into four groupings based on the species and density of macrofauna present. Both of these older indices suffer because the macrofaunal or microbial surveys required to determine them are expensive in time to obtain. Other methods can be linked to spatial or temporal versions of the organic enrichment gradients based on macrofauna. They include sediment profile imaging (Rhoads and Germano 1986; Nilsson and Rosenberg 1997) and sediment geochemistry by redox and sulfide measurements (Wildish et al. 2001).

Hargrave (1994) proposed a benthic enrichment index based on the product of organic carbon and redox. The concept is to determine the sedimentation rate from the carbon present in interfacial sediments. Cranston (1994) described a geochemical method that is a direct measure of net carbon burial rates. The method is based on downcore concentrations of sulfate and ammonium, as indicators of the remineralization rates, and requires large cores and numerous, expensive chemical determinations. However, the method appears to be robust, and there is a positive linear relationship between sedimentation and carbon burial rates. Recently, Dell'Anno et al. (2002) suggested a suite of environmental variables to assess the organic enrichment status of the coastal zone in the Mediterranean sea, based on interfacial sedimentary variables.

Ecological Effects of Organic Enrichment

Numerous reviews of organic enrichment associated with many industries show two important generalities: the ecological response is complex, involving pelagic-benthic coupling and both water column and sediments; and the effects include all sources of organic matter, both natural and anthropogenic.

Different physical and biological conditions on the Atlantic and Pacific coasts of Canada result in differing ecological responses to aquaculture. In the Bay of Fundy, the salmon aquaculture industry does not usually experience severe hypoxia in seawater because of energetic tidal mixing. If hypoxia occurs, it is in sediment pore water and localized to the benthic footprint area. There are few available data on dissolved oxygen in the sediments and water column near fish farms in British Columbia. However, naturally occurring low (<4 mg·L-1) dissolved oxygen levels are present on the west coast of Vancouver Island and in Queen Charlotte Strait during the late summer and early fall (Levings et al. 2002). On the Pacific coast, fish farms are located over deeper and more diverse seabed habitats and most tenures include a mosaic of sediment types and are not characterized by homogenous level mud bottoms. Rock terraces, cliff walls and boulder fields are also under fish farms in some locations (see Levings et al. 2002 and references therein).

The general response of soft-sediment macrofaunal populations and communities to organic enrichment gradients is well established. It involves the local extinction of the resident equilibrium community, followed by the re-establishment of opportunists if conditions improve. Some species are more resistant to hypoxia than others (Diaz and Rosenberg 1995). In general, the crustaceans and echinoderms are most sensitive. Field studies suggest that, where seasonal hypoxia occurs, dissolved oxygen (DO) levels of 1 ml·L-1 begin to cause macrofaunal invertebrate mortality (Diaz and Rosenberg 1995). In areas of permanent oxygen deficiency, the benthic communities appear to be adapted to an even lower critical DO level. Although very low DO levels are seen as the major limiting factor for macrofauna, the role of H2S (and ammonia) is less clear. Where severe hypoxia is present, both may be released.

Relatively little is known about the effects of organic enrichment on ecosystem functioning. Simplistic approaches to assess ecosystem effects assume that all of the secondary benthic production that becomes anoxic or hypoxic is lost to the next trophic level of predators. However, these approaches ignore that many individual predators are adapted to prey on a specific set of a few species in an equilibrium community.

Azoic or anoxic sediments could cause a significant shift in pelagic-benthic coupling. The effect of organic enrichment in sediments is to move the system to one dominated by bacteria, ciliates and meiofauna, and where the trophic links to the next level of the food web are broken. In bays heavily occupied by salmon farms, Pohle et al. (2001) analyzed the macrofaunal community at reference stations and found significant structural change. While the cause could not be identified, the most probable explanation is that an aspect of enrichment linked to salmon farming caused changes in benthic-pelagic coupling in such a way as to exclude some species and encourage others. Although the severely hypoxic areas caused by fish farming in the Bay of Fundy are relatively small, no studies have examined benthic-pelagic coupling in the vicinity of fish farms.

Towards Predictive Models

Prediction of benthic organic enrichment from fish farms requires the application of mathematical models. Process-oriented models use mathematical frameworks that describe the major physical and biogeochemical components and the processes through which they interact. Ocean circulation models predict the transport and mixing in the coastal zone, including the dispersion of organic matter from fish farms. Comprehensive ocean models, such as the Princeton Ocean Model (Blumberg and Mellor 1987) and the CANDIE ocean model (Sheng et al. 1998), have been successfully applied to coastal waters. Sediment transport models predict the time evolution of the spatial distribution of suspended particulate matter, as well as the exchanges of material between the water column and the benthos. Diagenetic models couple water column processes to vertically resolved sediment biogeochemical models (Wijsman et al. 2000).

Empirical models use statistical descriptions of the relationships between observable quantities that are indicators for key environmental components or processes. Findlay and Watling (1997) proposed an oxygen-based framework, based on the balance between benthic oxygen supply to oxygen demand, for assessing the benthic response to organic enrichment from salmon aquaculture. Dudley et al. (2000) developed a more process-oriented approach using a transport model to estimate dispersion of wastes from a fish farm to the benthos. DEPOMOD uses a hybrid approach to model benthic enrichment effects, and includes a particle tracking model and empirical relationships between the spatial distribution of solids and changes in benthic community structure (Cromey et al. 2002). In British Columbia, Carswell and Chandler (2001) and Stucchi (in prep.) have developed particle tracking models that give estimates of size of the sediment field under and near fish farms, but these models are not yet linked with benthic biological data.

Towards Methods of Monitoring Organic Enrichment

Several variables to assess organic enrichment in seawater have been identified, including phytoplankton species and abundance matrices, dissolved oxygen concentrations, and total nutrient concentrations. Because none of these variables is accepted as the single indicator of the trophic status of seawater, multiparameter classifications have been used. For example, the OECD classification is based on chlorophyll a, plant nutrients and Secchi depths (Vollenweider and Kerekes 1982).

There are several methods to assess organic impacts in sediments. Those methods based on classical macrofaunal sampling and analysis are among the best known, although most costly. Alternative methods, such as sediment profile imaging or geochemistry, are more cost-effective. Recently developed methods, such as aerial photography, video photography and multibeam acoustics, have the potential to produce detailed and accurate maps over large areas. However, these methods require further research and groundtruthing. For benthic monitoring, the presence of predominantly hard or soft substrates will dictate the type of sampling. Other considerations include size of the ecosystem and the primary goal of the study.

General types of monitoring to detect organic enrichment in the marine environment are distinguished by their purpose. Geographical studies determining the limit of impact benefit primarily from synoptic survey methods, such as remote sensing of chlorophyll a in surface waters, underwater photography, video photography and acoustic surveys of soft sediments. Studies involving site comparison (treatment/reference sites), temporal trends (before/after) and practical monitoring (relative impact) can use a few alternative methods. Sediment profile images and sediment geochemical methods provide a much more cost-effective and credible alternative to the use of macrofaunal sampling and analyses for routine monitoring purposes.

Research Needs

Research is needed to provide an understanding of processes and to provide input to models and monitoring programs. Research is also needed to aid the scientific assessment of organic enrichment from marine finfish aquaculture (or near-field enrichment of marine finfish aquaculture). The following specific research needs are identified:

• Conduct sedimentology and physical and chemical oceanography studies, including coastal circulation, mixing, dispersion and transport processes in support of process models, and observation and modelling studies of water column particle dynamics.
• Conduct seasonal studies of organic enrichment (such as redox potential and sulfide) to examine ecological factors affecting organic enrichment events from salmon farming.
• Measure the availability of carbon to microbial decomposers.
• Determine effects of organic enrichment on coarse and hard substrates in British Columbia where fish farms are located over mosaics of sediment types.
• Determine effects of organic enrichment on ecosystem functioning to establish cause-effect relationships.
• Investigate how organic enrichment from aquaculture affects benthic-pelagic coupling.
• Undertake further fallowing studies in Pacific and Atlantic Canada.
• Verify models, such as Lagrangian-based particle models, sediment transport models and biogeochemical models, through collaborative efforts between modellers and field biologists.
• Predict holding capacity or assimilative limits related to the amount of local organic enrichment.
• Develop geographical survey methods, such as satellite, aerial surveillance, underwater video photography and acoustics.
• Devise new environmental monitoring methods.
• Calibrate, standardize and audit existing environmental monitoring methods.

References

Andersen, T.J. 2001. Seasonal Variation in erodibility of two temperate, microtidal mudflats. Estuar. Coast. Shelf Sci. 53: 1-12

Blumberg, A.F., and G.L. Mellor. 1987. A description of a three dimensional coastal circulation model, p. 1-16. In N.S. Heaps [ed.]. Three dimensional coastal ocean models, Coastal and Estuarine Studies (vol. 4). Am. Geophys. Union, Washington.

Brooks, K.M., A.R. Stierns, C.V.W. Mahnken, and D. Blackburn. 2003. Chemical and biological remediation of the benthos near Atlantic salmon farms. Aquaculture 219: 355-377.

Carswell, L.B., and P. Chandler. 2001. A modular aquaculture modelling system (MAMS) and its application to the Broughton Archipelago, British Columbia (BC) Coastal Engineering V Computer Modelling of Seas and Coastal Regions Editor C.A. Brebbia, Wessex Institute of Technology, UK 2001.

Cranston, R. 1994. Dissolved ammonium and sulfate gradients in surficial sediment pore water as a measure of organic carbon burial rate, p. 93-120. In B.T. Hargrave [ed.]. Can. Tech. Rep. Fish. Aquat. Sci. 1949.

Cromey, C.J., T.D. Nickell, and K.D. Black. 2002. DEPOMOD - modelling the deposition and biological effects of waste solids from marine cage farms. Aquaculture 214: 211-239.

Dell'Anno, A., M.L. Mei, A. Pusceddu, and R. Danovaro. 2002. Assessing the trophic state and eutrophication of coastal marine systems: a new approach based on the biochemical composition of sediment organic matter. Mar. Pollut. Bull. 44: 611-622.

Diaz, R.J., and R. Rosenberg. 1995. Marine benthic hypoxia: a review of its ecological effects and the behavioural responses of the benthic macrofauna. Oceanogr. Mar. Biol. Annu. Rev. 33: 245-303.

Dudley, R.W., V.G. Panchang, and C.R. Newell. 2000. Application of a comprehensive modeling strategy for the management of net-pen aquaculture waste transport. Aquaculture 187: 319-349.

Findlay, R.H., and L. Watling. 1997. Prediction of benthic impact for salmon net-pens based on the balance of benthic oxygen supply and demand. Mar. Ecol. Prog. Ser. 155: 147-157.

Gray, J.S., R.S-S. Wu, and Y.Y. Or. 2002. Effects of hypoxia and organic enrichment on the coastal environment. Mar. Ecol. Prog. Ser. 238: 249-279.

Hargrave, B.T. 1994. A benthic enrichment index, p. 79-91. In B.T. Hargrave [ed.]. Modelling benthic impacts of organic enrichment from marine aquaculture. Can. Tech. Rep. Fish. Aquat. Sci. 1949.

Holmer, M., and E. Kristensen. 1992. Impact of marine fish cage farming on metabolism and sulfate reduction of underlying sediments. Mar. Ecol. Prog. Ser. 80: 191-201.

Kaspar, H.F., G.M. Hall, and A.J. Holland. 1988. Effects of sea cage salmon farming on sediment nitrification and dissimilatory nitrate reductions. Aquaculture 70: 333-344.

Levings, C.D., J.M. Helfield, D.J. Stucchi, and T.F. Sutherland. 2002. A perspective on theuse of performance based standards to assist in fish habitat management on the seafloor near salmon net pen operations in British Columbia. DFO Can. Sci. Advis. Secretar. Res. Doc. 2002/075.59 p.

Lumb, C.M., and S.L. Fowler. 1989. Assessing the benthic impact of fish farming, p. 75-78. In J. McManus and M. Elliott [eds.] Developments in estuarine and coastal study techniques. Olsen and Olsen, Fredensborg, Denmark.

Nilsson, H.C., and R. Rosenberg. 1997. Benthic habitat quality assessment of an oxygen stressed fjord by surface and sediment profile images. J. Mar. Sys. 11: 249-264.

Pearson, T.H., and R. Rosenberg. 1978. Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanogr. Mar. Biol. Ann. Rev. 16:. 229-311.

Pohle, G., B. Frost, and R. Findlay. 2001. Assessment of regional benthic impact of salmon mariculture within the Letang Inlet, Bay of Fundy. ICES J. Mar. Sci. 58: 417-426.

Poole, N.J., D.J. Wildish, and D.D. Kristmanson. 1978. The effects of the pulp and paper industry on the aquatic environment. CRC Crit. Rev. Environ. Control 8: 153-195.

Rhoads, D.C., and J.D. Germano.1986. Interpreting long-term changes in benthic community structure: a new protocol. Hydrobiologia 142: 291-308.

Sheng, J., D.G. Wright, R.J. Greatbatch, and D.E. Dietrich. 1998. CANDIE: A new version of the DieCAST ocean circulation model. J. Atmosphere. Ocean. Technol. 15: 1414-1432.

Stucchi, D., Sutherland, T.A. and C.D. Levings. In prep. Near-Field Depositional Model for Finfish Aquaculture Waste. In B.T. Hargrave [ed.]. Handbook of environmental chemistry, Springer-Verlag, Berlin.

Vollenweider, R.A., and J.J. Kerekes. 1982. Eutrophication of Waters. Monitoring, Assessment and Control.OECD, Paris. 164 p.

Wijsman, J.W.M., P.M.J. Herman, J.J. Middleburg, and K. Soetaert. 2002. A model for early diagenetic processes in the sediments of the continental shelf of the Black Sea. Estuar. Coast. Shelf Sci. 54: 403-421.

Wildish, D.J., H.M. Akagi, and N. Hamilton. 2001. Sedimentary changes at a Bay of Fundy salmon farm associated with site fallowing. Bull. Aquacult. Assoc. Can. 101-1: 49-56.

Wildish, D.J., B.T. Hargrave, and G. Pohle. 2001. Cost effective monitoring of organic enrichment resulting from salmon mariculture. ICES J. Mar. Sci. 58: 469-476.

Wildish, D.J., V. Zitko, H.M. Akagi, and A.J. Wilson. 1990. Sedimentary anoxia caused by salmonid mariculture wastes in the Bay of Fundy and its effects on dissolved oxygen in seawater, p. 11-18. In R.L. Saunders [ed.]. Proceedings of Canada-Norway finfish aquaculture workshop, Sept. 11-14, 1989. Can. Tech. Rep. Fish. Aquat. Sci. 1761.