Addressing Concerns for Water Quality Impacts
from Large-Scale Great Lakes Aquaculture

Based on a Roundtable
Co-hosted by the Habitat Advisory Board of the Great Lakes Fishery Commission
the Great Lakes Water Quality Board of the International Joint Commission

August, 1999


Known Water Quality Impacts of Great Lakes Basin Aquaculture

Aquaculture operations will have some impact on the natural environment or an "ecological footprint." For this report, ecological footprint means the extent and severity of ecological impact caused by aquaculture operations. This concept of ecological footprint has been developed by Wackernagel and Rees (1996) to address carrying capacity (i.e., the corresponding area of productive land and aquatic ecosystems required to produce the resources used, and to assimilate the wastes produced by, a defined population at a specified material standard of living).

Major water quality problems due to aquaculture can and have included:

Roundtable participants noted that an aquaculture operation can become a problem when operators attempt to grow more fish than a given area and water volume can sustain (i.e., exceed the carrying capacity). In addition, there is degradation and loss of habitat.

Aquaculture operations in Minnesota, Ontario, and Michigan have resulted in water quality impacts (Table 1). In Minnesota (i.e., mine pit lakes), the water quality improved following cessation of aquaculture operations (Appendix 10). In Michigan (i.e., Big Platte Lake), water quality conditions have improved after making improvements in hatchery operations and reducing phosphorus loadings (Appendix 7). However, recovery has not yet been documented in LaCloche/North Channel, Ontario (Appendix 9). It was noted by industry participants that several operations in Georgian Bay have been well managed and have not resulted in any substantial water quality problems (Appendix 12). Participants noted that many water quality impacts can be prevented through better assessment, siting, prediction of carrying capacity, and management of aquaculture operations for example.

Table 1. Water quality problems attributed to selected aquaculture operations in Minnesota, Ontario, and Michigan.

Characteristic or Concern Minnesota LaCloche/North Channel (Ontario) Big Platte Lake/River (Michigan)
Type of operation Caged aquaculture Caged aquaculture Hatchery
Location Inland mine pit lakes North Channel and Georgian Bay on Lake Huron Tributary to northeastern Lake Michigan
Species reared Chinook salmon and rainbow trout Rainbow trout Coho and chinook salmon
Known water quality impacts Exceedance of the NPDES* permit limit for phosphorus (33 µg/L); approximately an order of magnitude increase in water column phosphorus, nitrogen, and chlorophyll levels; increased attached algal growth Exceedance of Provincial water quality objectives for phosphorus and dissolved oxygen; reduced water transparency; presence of algal blooms; dissolved oxygen depletion over 250 ha Increased phosphorus loadings; elevated primary productivity; decreased water transparency; anecdotal reports of reductions in crayfish populations, emergent aquatic vegetation, and mayfly hatches

* National Pollutant Discharge Elimination System

Potential Long-term Impacts of Great Lakes Aquaculture

Based on a review of environmental effects of aquaculture in the U.S. by the Environmental Defense Fund, Goldburg and Triplett (1997) concluded that "when compared to the largest sources of nutrient pollution, such as municipal sewage systems, U.S. aquaculture operations have a relatively small impact on water quality." In 1998, there was an estimated production of 3,000 tonnes of rainbow trout in Georgian Bay (Lake Huron) which would have contributed an estimated 15 tonnes of phosphorus to Lake Huron that year (Appendix 9). This phosphorus loading would represent about 0.3% of the total phosphorus loading target to Lake Huron (4,360 tonnes). No lakewide effect would be expected, however, localized water quality impacts are possible.

Water quality impacts can be short- or long-term. Some degraded sites can recover relatively quickly (e.g., Minnesota mine pit lakes) where water quality improved following cessation of aquaculture operations (Appendix 10). In Michigan, water quality in Big Platte Lake did not substantially increase for approximately ten years after making improvements in hatchery operations and reducing phosphorus loadings (Appendix 7). Full recovery has not yet been documented in LaCloche/North Channel, Ontario (Appendix 9). Participants noted that many water quality impacts can be resolved through proper siting and management of aquaculture operations.

Participants noted that the long-term concerns for water quality impacts of aquaculture include short-term operational impacts and long-term impacts from accumulated nutrients, feces, and degraded habitat. A number of other potential issues were raised by participants. Limited information exists on many of these potential issues. Some participants felt that there is no evidence to suggest that there is any concern for some of these potential issues and that they should not be identified because they would be viewed by some individuals as actual problems. These potential issues raised at the roundtable, but not addressed either at the roundtable or in the report, include:

Some of these concerns have received considerable media attention outside the Great Lakes basin. They may not be relevant or applicable to the Great Lakes. Again, this report only addresses water quality impacts of large-scale Great Lakes aquaculture.

Prevention or Abatement of Water Quality Impacts

Overall, it was predicted that there will be more fish produced from caged aquaculture in the future, but probably not an order of magnitude increase in operations. Growth of land-based aquaculture is expected to be minimal due to higher costs for siting and often limited availability of suitable water quantities. Currently, there are an estimated 1,000 U.S. aquaculture producers in the Great Lakes (Appendix 5) and approximately 200 facilities in Ontario (Appendix 8). More accurate statistics on U.S. aquaculture will be available later in 1999 when the U.S. Department of Agriculture's first ever 1998 Census of Aquaculture will be completed (U.S. Department of Agriculture 1998). Globally the amount of seafood and freshwater fish consumed will increase due to an increase in population. For example, the U.S. population is expected to increase 1%. Although the 1997 U.S. per capita consumption of seafood has declined slightly to 6.6 kg (14.6 pounds) from 7 kg (15 pounds) in 1996, the total U.S. supply (landings plus imports) has remained relatively constant (Johnson 1998; Goldburg and Triplett 1997). Johnson (1998) attributes the consumption decline to supply constraint of some species and notes consumption increases in species such as salmon and catfish that are not supply constrained. Further to this, there has been a shift in consumption from wild-caught fish to aquaculture-raised fish due to depleting global wild stocks. These factors will contribute to a growing aquaculture industry.

Growth of the caged aquaculture industry in the Great Lakes is limited by the few remaining sites that are suitable using currently available technology. Key criteria for caged aquaculture site selection include a sheltered location, deep water, and good circulation.

Aquaculture within the Great Lakes is primarily limited to cage culture operations raising rainbow trout. This is primarily due to rainbow trout being more domesticated and having well established husbandry (Gord Cole, Personal Communication). Rainbow trout is a coldwater species and, as a result, the majority of aquaculture operations are located in northern areas of the Great Lakes such as Georgian Bay (Figure 1).

In Ontario, regulations state that 42 species of fish may be considered for aquaculture production (Appendix 8). Currently farming of warmwater species of fish such as walleye and yellow perch is limited by the economical viability, technology, husbandry information, and domestication necessary for farming species. In 1998, the University of Wisconsin System Aquaculture Institute in Milwaukee signed an agreement with the Red Lake Band of Chippewa Indians to study the potential of raising yellow perch at an aquaculture facility (University of Wisconsin Sea Grant 1998). The North Central Regional Aquaculture Center is also researching aquaculture production of several centrarchid species and lists the current potential for these species as moderate to high. Continued demand of high dollar value commercial fish and limitation of available coldwater sites may result in increased farming of warmwater species. Ocean cage technologies, that can be located further offshore, are not yet practical or economically feasible in the Great Lakes.

All aquaculture operations will have some impact or "ecological footprint." Concern was raised for how and who determines what changes in water and habitat quality are "unacceptable". There are currently no biologically-based, regulatory standards for making a determination of "unacceptable" water quality or habitat impacts. Therefore, there is a need for biologically-based regulatory standards for "unacceptable" water quality and habitat impacts that are applicable to aquaculture operations. Industry interests suggest that there must be a balance of economic, social-political, industrial, and administrative factors. It was suggested that any monitoring required in a permit/license should include an early warning capability (e.g., monitoring benthos or periphyton).

Figure 1. Great Lakes - St. Lawrence locator map displaying current, known, commercial cage aquaculture operations.

Prior to establishing an aquaculture farm, proper site assessments and adequate modelling are essential to determine where to locate operations. Site assessments must address "carrying capacity" and site-specific characteristics (e.g., shelter, water depth, circulation, sedimentation, morphometry). Better siting criteria can be built into modelling to predict site suitability. Size/scale must be taken into account, but putting restrictions on size/scale will limit business and the economy of scale. In British Columbia, all applications for finfish licenses and leases must follow specific siting and spacing criteria in relation to other operations, wild stocks, stream mouths, and critical habitat (Salmon Aquaculture Review 1997). Goldburg and Triplett (1997) agree stating that "siting net pens in areas with strong currents or tides that flush wastes and avoiding overly dense siting of net pens can help limit problems from waste accumulation."

The development of predictive models for the Great Lakes which consider current, feed rates, depth, bottom characteristics, flow velocity, ecological sensitivity, and other factors would be useful to predict the scale of operation a site can sustain. A site selection predictive model has been developed by the provincial Ministry of Agriculture, Fisheries and Food in British Columbia (Appendix 11). Proper siting would likely have prevented the establishment of an aquaculture operation in mine pit lakes in Minnesota. Mine pit lakes were not suited for aquaculture due to the relatively unnatural state of a very deep pit with little natural vegetation, no littoral zone, and low nutrients. The input of nutrients from overfeeding and fish waste resulted in a more intensive and immediate problem than other more natural sites.

One suggestion for managing impacts was to limit the size of aquaculture operations by establishing a feed quota with specified quality. For example, after a proper site assessment and prediction of size and scale of operations, a license/permit could be issued based on a given amount of food of a known phosphorus amount or establishing phosphorus quotas. The license/permit would also require monitoring. If monitoring during initial operations showed water quality problems, there would have to be an adjustment in the feed quota. If monitoring during initial operations showed no water quality problems, then operations could continue with the given feed quota.

Advances in the development of high nutrient dense, low phosphorus fish feeds have resulted in higher feed conversion rates and a decrease in fish waste (Cho and Bureau 1997; Appendix 4). The use of high nutrient dense diets for the production of rainbow trout is key to decreasing the amount of fish waste. Improvements in feeding strategies which result in higher consumption of feed and less waste of feed have also resulted in improved water quality and higher farm productivity.

Protection under current law

In Ontario, more than 12 agencies and 30 pieces of legislation potentially regulate aquaculture (Moccia and Bevan 1996). In general, roundtable participants felt that protection tools are available, but can be improved. In Ontario, aquaculture licensing is being addressed under the Fish and Wildlife Conservation Act, 1997 ( proclaimed on January 1, 1999). The Canadian Federal Fisheries Act (Section 36) prohibits the deposit of deleterious substances into Canadian fisheries waters and the Act's definition of "deleterious" substance" can include organic waste, therapeutics, antifoulants, and pesticides. Participants felt that this vehicle is generally adequate if there are sufficient data and information available for the site. One suggested area of improvement would be to establish feed or production limits and a monitoring program in these licenses.

In the United States, National Pollutant Discharge Elimination System (NPDES) permits are used to regulate the industry. However, there is variability among the states because of differences in established numerical standards and designated uses (e.g., fish, recreation, etc.). Again, adequate data and information are essential for NPDES permits to be successful. A physicochemical and biological monitoring program should be a specific requirement in each permit. The U.S. Environmental Protection Agency (EPA) announced in February 1999 that the agency will be conducting a preliminary study of the aquaculture industry in order to gather information and determine whether national effluent guidelines will be developed for the aquaculture industry (U.S. Environmental Protection Agency 1998, 1999).

Roles and responsibilities of government and the aquaculture industry

Caged aquaculture systems differ from conventional land-based aquaculture facilities in the use of public trust resources as direct subsidy in the treatment of their wastes. Fish, wildlife, and navigable waters (including Great Lakes bottomlands) are common property of the people of the provinces and states that surround the Great Lakes. This common property right is one of the oldest public rights with roots from 13th Century English Law. Caged aquaculture operations do not have any means to treat their wastes and use Great Lakes bottomlands and aquatic biota to treat their wastes. It is very likely that the public will lose some of its established property rights and values as bottomlands decompose wastes. This will impair and decrease the fisheries value of the immediate area around the caged aquaculture operation. This loss of the public's property value from caged aquaculture wastes is a direct, but hidden, public subsidy to this industry. This is in direct contrast to land-based aquaculture facilities that are highly regulated and must use their own capital to treat their wastes prior to discharge to public trust waters of the provinces or states. As the trustee for the public's property, it is incumbent upon the provinces and states to be highly selective in allowing the use of caged aquaculture systems in Great Lakes waters and the public should be appropriately compensated for any loss of their property value.

To achieve economically-viable and environmentally-sustainable aquaculture there must be a shared responsibility between government and the aquaculture industry. All stakeholders should help design the rules and government should oversee enforcement. In practice, state and provincial governments have responsibility for fish management, including aquaculture. The industry wants reasonable government regulations to prevent water quality impacts from aquaculture.

Governments need to facilitate good management. For example, governments can develop processes that ensure a shared responsibility for adequate site assessments, monitoring, development of a phosphorus budget, and establishment of feed quotas. There also needs to be alternative dispute resolution mechanisms. For example, advisory committees have been successful in some states. Michigan experience has shown that aquaculture needs to stay out of the courts.

Most aquaculture operations in the Great Lakes are relatively small with limited resources. Government could fill a value-added role by sharing information on research and development, and by promoting technology transfer. One suggestion was to give governments authority and responsibility to ensure that "best management practices" are employed in all aquaculture operations.

It was suggested that the use of commercial equipment as per the recommendations of manufacturers should be encouraged. In addition, experienced and trained staff are required to carry out operations. In general, aquaculture facilities that are well sited and well managed to minimize environmental effects will be more sustainable in the long-term and will be more economically-viable.

Loss of habitat and the precautionary approach

In general, governments cannot ensure no loss of habitat from new Great Lakes caged aquaculture operations. There will undoubtedly be loss of physical, chemical, and biological habitat. There will be an ecological impact. However, this loss may be temporary if fallowing is practiced. Concern was raised for:

Such questions highlight the need to define what level of impact is acceptable. Before siting an aquaculture operation we must address "carrying capacity" and cumulative effects (e.g., via a wasteload allocation for phosphorus). Adequate monitoring will be a key to addressing loss and degradation of habitat. Industry representatives noted that any industry has environmental impacts, some of which are only temporary. Governments must make sure that any ecological impacts from caged aquaculture operations are temporary and confined. The practice of mandatory fallowing should be given consideration with the issuance of a permit/license.

Based on existing evidence, most governments cannot justify a moratorium on aquaculture operations due solely to the loss of habitat, but caution is warranted with respect to biochemical oxygen demand, total suspended solids, nitrogen, and phosphorus. What is needed is a reasonable and objective approach. Other issues raised which must be addressed within a precautionary approach include:

Monitoring strategies

Monitoring is very important and an essential part of management. All historical data must be compiled and used to help establish a monitoring program. Where governments do not have the resources to monitor all lakes and operations, partnerships can be considered among government, industry, and citizen groups. Concern was raised regarding whether agencies will trust industry or citizen data. There will need to be standardization of monitoring protocols and analytical methods, along with good quality assurance and quality control. Governments must be willing to work with lake associations and industries.

One good example of a partnership is the aquaculture operation located in Manitowaning Bay on Georgian Bay in Ontario. Extensive water quality surveys were performed by industry to ensure the best possible water quality at the farm site.

There is a particular need to document critical habitats, such as fish spawning and nursery habitat, prior to siting. Any partnership for monitoring should address water quality, benthos, phytoplankton, structure and function of the fish community, and habitat requirements. Experience has shown that monitoring programs must be site specific and address both local and lakewide concerns. All data collected must be relevant and useful.

Research needs

A sound scientific knowledge base is essential to address issues related to Great Lakes aquaculture. We will probably never have a complete knowledge base and there will always be research needs. In 1994, the United States aquaculture industry received an estimated $60 million (U.S.) in financial assistance from the federal government, most of which was for research (Goldburg and Triplett 1997). Presented below are some research needs that participants felt should be addressed to help make decisions about siting aquaculture operations and proper management: