The Costs of Eutrophication from Salmon Farming: Implications for Policy

:NAL Call number: HC75.E5J6 Title currently indexed. Holdings availability: :Location: NAL-Stacks :Status: CURRENTLY RECEIVED :Copy#: 1

:Holdings: la 1-57 li 1973-1999 :Holdings: 58 lb 1-4 li 2000 :Holdings: 59 Ib 1-4 li 2000 :Holdings: 60 Ib 1-2 li 2000 :Holdings: la 34-35 li 1992 :Holdings: la 38-39 li 1993 :Holdings: la 40 li 1994 :Holdings: Missing vol. 6 no. 4; vol. 7 no. 4; vol. 9 no. 4; vol. 19

nola 19 / b 1; vol. 42 no. 4

:ILL: 3719077 :Borrower: OAO :RegDate: 20001109 :NeedBefore: 20001209

:Status: IN PROCESS 20001109 :OCLC: 1786927 :Source: OCLCILL :Lender: *AGL,RLA,EJB,UDI,FZU

:TITLE: Journal of environmental management. :IMPRINT: London, New York, Academic Press. :ARTICLE: Folke, C., Kautsky, N. et a1 "the costs of eutrogh :VOL: 40 :NO: 2 :DATE: 1994

:PATRON: robert1005 :VERIFIED: OCLC ISSN: 0301-4797

:SHIP TO: NOAA Miami Regional Library at AOML 4301 Rickenbacker Cswy Miami, FL 33149 or send it to ariel 192.111.123.231 (new IP)

:BILL TO: Same :SHIP VIA: Library Rate : MAXCOST e :FAX: 305-361-4449 :E-MAIL: [email protected] :AFFILIATION: FLIN,NOAALINK

Journal of Environmental iManagernmt (1994) 40, 173-182

The Costs of Eutrophication from Salmon Farming: Implications for Policy

Carl Folke"?, Nils Kautsky" and Max Troell"

'Department of Systems Ecology, Stockholm University, S - 106 91 Stockholm, Sweden and ?The Beijer International Institute of Ecological Economics, The Royal Swedish Academy of Sciences, Box 50005, S - 104 05 Stockholm, Sweden

Received 16 December 1992

We examine the ecological and economic costs of environmental impacts caused by coastal cage farming of salmonids, in particular those associated with nutrient releases and their relation to eutrophication and toxic algal blooms. We find that the nutrient releases from a fish farm producing I00 tons of salmon correspond to those of a human settlement of 850-3200 persons. Extrapolating the results to the 200 000 tons of salmonid production by Nordic countries, the release exceeds those of the population of Stockholm, the largest city in Sweden, and even whole states, such as Lithuania. Internalizing the environmental cost of the nutrient release from salmon farms at the level of the firm reveals that the total cost of salmon production exceeds the highest price paid for salmon in the 1980s, the decade when the industry boomed. In addition, there are several other environmental impacts not internalized. It is concluded that, from the viewpoint of Swedish society, salmonid farming, as performed today in coastal waters, is not only ecologically but also economically unsustainable.

Keywords: aquaculture. fish farming, sustainability, internalizing costs, salmonids, eutrophication, algal blooms, environmental policy, polluter-pays principle.

1. Introduction

The Rio declaration on environment and development contains 27 principles for redirecting the behaviour of the world economy towards a sustainable path. It was signed by 178 states during the United Nations' conference in early June 1992. In this declaration it is stated that unsustainable patterns of production and consumption should be reduced and eliminated, and that national authorities should endeavour to promote the internalization of environmental costs and the use of economic instruments, taking into account that the polluter should, in principle, bear the cost of pollution.

One example of activities that cause environmental costs is the farming of salmonids in cages. This is an industry which has grown rapidly during the last decade (e.g. FAO, 1992). We have previously argued that the development of this industry is a clear-cut

173 03014797/94~020173 + 10 $08.00:0 0 1994 Academic Press Limited

174 Costs of eutrophication

example of an ecologically unsustainable pattern of food production (Folke and Kautsky, 1989, 1992), repeating many of the development mistakes which have taken place in agriculture, forestry and fisheries (e.g. Regier and Baskerville, 1986; Trenbath et al., 1990). Unsustainable behaviour causes environmental impacts (e.g. Ackefors, 1986; Gowen and Bradbury, 1987; Braaten et al., 1988), and may also lead to severe socio- economic conflicts (Baskerville, 1988; Shiva, 1991).

A major reason for the generation of environmental impacts is that salmon farming in cages is an intensive monoculture, a throughput system. This means that resources, collected'over large areas, are pumped into the production site, and when used up released back into the environment in concentrated forms as waste and pollutants. causing various environmental problems (Folke and Kautsky, 199 1, 1992). Throughput systems are heavily dependent on inputs of resources, both renewable and non- renewable, as well as ecological services. By quantifying this dependence, we have shown that there is instead a strong complementarity between the industry and its environment (Folke, 1988; Folke and Kautsky, 1989), that functioning ecosystems are the resource base on which the salmon farming industry ultimately depends (Constanza and Daly, 1992; Folke and Jansson, 1992).

However, there is as yet no explicit management of the feedbacks between the salmon farming and its environment, and there is no recycling of resources and waste. Instead, the characteristics of intensive salmon farming have been found to be similar to those of stressed ecosystems (Folke and Kautsky, 1992). Among characteristics typical of a monoculture are a strong dependence on external fossil-fuel-based inputs, wasteful resource use, nutrient leakage, generation of by-products that are stored or exported, which often lead to reduced productivity, and environmental degradation.

In this article we will examine the cost of environmental impacts caused by salmon farming, in particular those associated with nutrient releases and their relations to eutrophication and toxic algal blooms. The cost of environmental degradation from fish farming is at present not paid by the polluter, instead it is born by others in society, and society even supports environmental degradation by subsidizing the industry in many cases. Social costs of mangrove ecosystem degradation as a consequence of intensive shrimp aquaculture have been discussed and partly estimated by Hamilton et al. (1989) and La1 (1990). To our knowledge, analyses of the economics of fish farming have not included environmental costs to society of the activity, such as the cost of eutrophication of coastal areas.

With the Rio declaration in mind, we estimate and internalize the environmental cost of the nutrient releases, and, in this context, discuss the polluter-pays principle and its implications for the future of the fish farming industry. The article ends with a few policy recommendations, where we emphasize in particular the need for clear and long-term signals from decision makers to the industry that production has to be redirected from throughput to eco-efficiency, recycling and integration with the processes and functions of ecosystems.

2. Source of nutrient release from fish farms

To produce salmon in cages, feed from outside areas has to be added. Among other things, the feed contains nutrients. The release of such eutrophicating substances from fish farming is mainly connected to feeding practices and to the composition of the artificially-produced feeds. Since the raw material for fabricating the feed pellets often

C. Folke et af.

originates from other water b of organic matter to the rece

Because of the impacts as and fabricated feed composi the last 15-20 years. Lots of . partly as a consequence of str marine salmonid farms in Sc. 2.0 in 1989 and the nitrogen ( from 7.8 to 7.1 % (wet weight (Ackefors and Enell, 1990). A farming site and thereby less

Further investments and protein content in the feed an( the feed pellets. To increase g more-easily digestible low mo mental impacts further, low E

3. Changes in N/P ratios in tl

One possible and detrimental that the relationship between demand of nitrogen to phosph bodies is about 7 (by weight). valuable as food for filter-feec to reduced grazing and thus eutrophication.

Changes in N/P ratios havt number of different types o f t nitrogen-fixing blue-green a1 blooms tend to occur during p shown that high N/P ratios pol-vlepis, the Prymnesiophyce in Scandinavia, with a disast salmon industry included (Car bloom in terms of lost reveni (Folke and Kautsky, 1989).

The N/P ratios in the tota around 7.5 in 1989 (recalcula inorganic part, which is direct

It is clear that the present e will cause further increases in occurring plankton communit are generally P-limited (Schinc ments that are usually N-limite be less significant. It is even pc blooms as a consequence of nutrient load is reduced.

Hence, the plankton com; deviations from normal NIP ra

Costs of eutrophicati

3f food production (Folke an ment mistakes which have take nd Baskerville, 1986; Trenbath t

ltal impacts (e.g. Ackefors, 198( d may also lead to severe socio

1 impacts is that salmon farmini tem. This means that resources [uction site, and when used uy forms as waste and pollutants. lutsky, 199 1, 1992). Throughput :es, both renewable and non- this dependence, we have shown he industry and its environment ng ecosystems are the resource depends (Constanza and Daly,

2 feedbacks between the salmon f resources and waste, Instead, found to be similar to those of lg characteristics typical of a jil-fuel-based inputs, wasteful s that are stored or exported, mtal degradation. tal impacts caused by salmon :leases and their relations to imental degradation from fish born by others in society, and ;idizing the industry in many s a consequence of intensive ted by Hamilton et al. (1989) nics of fish farming have not as the cost of eutrophication

ialize the environmental cost )hter-pays principle and its article ends with a few policy eed for clear and long-term )n has to be redirected from the processes and functions

0 be added. Among other )phicating substances from to the composition of the 2ting the feed pellets often

175 C. Folke et af.

originates from other water bodies, the result is a net addition of dissolved nutrients and of organic matter to the receiving water body.

Because of the impacts associated with the released nutrients, the feeding techniques and fabricated feed composition in salmonid fish farming have been improved during the last 15-20 years. Lots of research and resources have been devoted to these aspects, partly as a consequence of stricter environmental policies. Feed conversion efficiencies in marine salmonid farms in Scandinavia have been reduced from 2-0-2.5 in 1974 to 1.4- 2.0 in 1989 and the nitrogen (N) and phosphorus (P) content in feeds has been decreased from 7-8 to 7.1% (wet weight) for nitrogen and 1.7 to 1.0% (wet weight) for phosphorus (Ackefors and Enell, 1990). All this has contributed to more-efficient resource use at the farming site and thereby less impacts of nutrients.

Further investments and developments are now going on to increase vegetable protein content in the feed and thus reduce the needs for fish meal and animal proteins in the feed pellets. To increase growth efficiency, there are also attempts being made to use more-easily digestible low molecular carbohydrates in the feeds, and to reduce environ- mental impacts further, low P feeds are being developed.

3. Changes in N/P ratios in the feed-toxic algal blooms

One possible and detrimental side-effect of reducing N and P content in the fish feed is that the relationship between N and P in the waste from the fish farm is altered. The demand of nitrogen to phosphorous, the N/P ratio, for plant growth in unpolluted water bodies is about 7 (by weight). An unbalanced N/P ratio will make the phytoplankton less valuable as food for filter-feeding animals (Ryther and Officer, 1981), which would lead to reduced grazing and thus result in higher algal biomass and plankton blooms, i.e. eutrophication.

Changes in N/P ratios have been mentioned as likely causes for the development of a number of different types of toxic algal blooms. It is generally assumed that blooms of nitrogen-fixing blue-green algae are stimulated by lowered N/P ratios, since these blooms tend to occur during periods of nitrogen deficit (Graneli et al., 1989). It was also shown that high N/P ratios increased the toxin production of Chiysochromulina polylepis, the Prymnesiophycean that bloomed in 1988 in the Kattegatt/Skagerack area in Scandinavia, with a disastrous economic impact on several human activities, the salmon industry included (Carlsson et al., 1990; Edvardsen et al., 1990). The cost of this bloom in terms of lost revenues for 35 salmon farms was estimated at US$7 million (Folke and Kautsky, 1989).

The N/P ratios in the total (particulate and dissolved) waste from a fish farm was around 7.5 in 1989 (recalculated from Ackefors and Enell, 1990), with the dissolved inorganic part, which is directly accessible to algae, as high as 28 on an annual basis.

It is clear that the present efforts by the fish farming industry to develop low P feeds will cause further increases in N/P ratios. The effects of these changes on naturally- occurring plankton communities are largely unknown. Freshwater environments that are generally P-limited (Schindler, 1977) will probably benefit, but in marine environ- ments that are usually N-limited (Grandi et al., 1990) the result of the reduction of P will be less significant. It is even possible that such feeds will stimulate undesired toxic algal blooms as a consequence of the increased NIP ratio. despite the fact that the total nutrient load is reduced.

Hence, the plankton community in the ecosystem can be very sensitive to any deviations from normal N/P ratios, and such changes can in fact be generated by the fish

I76 Costs of eutrophication

farms and hit back on the fish farming industry itself through, for example, oxygen deficiency and diseases killing the fish.

The alternative for changes in the N/P ratio, that is to produce low N feeds, will be much more difficult for the industry to accomplish, because this would mean that the total protein content will have to be reduced, which would inevitably lead to impaired fish production.

The current development towards use of more-easily digestible organic compounds in fish feeds as a means of increasing feed conversion efficiencies will certainly lead to increased releases of dissolved organic compounds, with largely unknown effects on the environment. Carlsson and Graneli (1993) and GranCli et al. (1985) have shown that marine phytoplankton can utilize nitrogen bound in humic acids, which are much longer molecules than the ones liberated from fish feed. It is very likely that low molecular compounds are incorporated more easily.

4. The scale of environmental impact of nutrient releases from fish farming

It is generally not understood that aquaculture systems may be ecologically completely divergent, putting entirely different demands on the environment, and affecting the ecosystem in quite different and even opposite ways (Kautsky and Folke, 1989; Kautsky et al., 1994). In the case of salmon farming, and in contrast to many other aquaculture systems, the nutrient releases from the cages represent a net addition to the environment. Excessive nutrient enrichment or eutrophication is a major environmental problem in agricultural as well as in industrialized areas, causing both qualitative and quantitative impacts in coastal areas (e.g. Rosenberg et al., 1990), which in turn affect socio- economic activities that are dependent on the quality of these areas (Folke and Kautsky, 1989; Folke et al., 1991).

It has been argued that the addition of nutrients from fish farming to the Baltic Sea as a whole is negligible in comparison with other sources of eutrophication, such as land runoff and atmospheric downfall (e.g. Ackefors and Enell, 1990). However, the relative eutrophication impact by fish farming is of course dependent on the amount of fish farmed-the more salmon farming monocultures in an area, the higher their share of pollution. For example, Alandic fish farming with a yearly production amounting to about 7000 tons at present generates more than three times and almost two times as much P and N than agricultural and municipal sources taken together, thereby being a very large contributor to the eutrophication in the archipelago of Aland, in the Baltic Sea (E. M. Blomqvist, pers. comm., Informationskontoret for skargirdens marina miljo, Mariehamn). Furthermore, a fish farm has to be regarded as a point source of pollution, and its outlets are therefore not directly comparable to non-point sources such as atmospheric downfall and river inputs of nutrients. This means that the release of eutrophicating substances has to be related in particular to the capacity of the local recipient to process them.

From an ecological perspective, nutrients from fish farming are identical to nutrients added from municipal sewage, as they have exactly the same potential to cause eutrophication problems (Person, 1992). It has been estimated that the release of nutrients by each ton of produced farmed fish in Scandinavia will have been reduced from about 31 kg P and 129 kg N in 1974 to about 9.5 kg P and 78 kg N in 1994, mainly due to changed feed composition and improved feed coefficients (Enell and Ackefors, 1991). Nutrients in sewage effluents are sometimes quantified as per capita equivalents, i.e. how many persons the untreated outlet of P and N corresponds to. On average, one

C. Folke el al.

TABLE 1. The release of nutrien cultivation cycle, in kg, in person

Nutrient release

1974

1994

P N

P N

?Based on the Swedish society’s treatment plants (see text).

TABLE 2. The estimat salmonids in Nordic

Nutrient release

1994

?Enell and Ackefor salmonic production in 12000 tons of nitrogen.

person produces 3 g P/day an nutrient releases in 1994, the I: 100 tons would amount to S cultivation season, equal the settlement of between 850 per: are used, nutrients generated persons. It may be realistic t developed world, but we wou developing world.

The production of salmc Finland, Iceland, Norway anc assumed to be 200 000 tons i results in Table 1 to the assun 1994 of 200 000 tons, and comr if the efforts to reduce the nu produced by the fish farms w( population of about 6 million. in 1994, the release would still city in Sweden, or even the PO

inhabitants) (Table 2). Obvioi bodies is substantial.

Costs of eutrophication

If through, for example, oxygen

s to produce low N feeds, will be ecause this would mean that the ould inevitably lead to impaired

ly digestible organic compounds efficiencies will certainly lead to I largely unknown effects on the li et al. (1985) have shown that nic acids, which are much longer very likely that low molecular

from fish farming

may be ecologically completely nvironment, and affecting the ltsky and Folke, 1989; Kautsky ’ast to many other aquaculture et addition to the environment. i o r environmental problem in th qualitative and quantitative , which in turn affect socio- ese areas (Folke and Kautsky,

fish farming to the Baltic Sea If eutrophication, such as land , 1990). However, the relative ndent on the amount of fish Tea, the higher their share of rly production amounting to nes and almost two times as ken together, thereby being a elago of Wland, in the Baltic or skargirdens marina miljo, s a point source of pollution, non-point sources such as

j means that the release of to the capacity of the local

...

ing are identical to nutrients e Same potential to cause .imated that the release of tvia will have been reduced md 78 kg N in 1994, mainly cients (Enell and Ackefors, d as per capita equivalents, sponds to. On average, one

C. Folke et al. 177

TABLE 1. The release of nutrients from the production of 100 tons of salmon in cages over a cultivation cycle, in kg, in person equivalents, and their cost to Swedish society, in SEK (1 US$ is

about SEK 6)

Nutrient release kg Person c o s t to equivalents society?

80 000 N 12 900 3200 645 000

25 000 N 7800 1950 400 000

1974 P 3100 2800

1994 P 950 850

?Based on the Swedish society‘s willingness to pay for removing nitrogen and phosphorus in sewage treatment plants (see text).

TABLE 2. The estimated release of nutrients from the total production of salmonids i n Nordic countries in 1994 of 200 000 tons, in tons and in

person equivalents

Nutrient release Tons? Person equivalents

1994 P 1900 1.7 million N 15 600 3.9 million

?Enell and Ackefors (1991) estimate the nutrient release by the Nordic salmonic production in 1994 to be 1000-1200 tons of phosphorus and to 10 000- 12 000 tons of nitrogcn.

person produces 3 g P/day and 11-12 g N/day (Torell, 1977). Based on the assumed nutrient releases in 1994, the nutrient effluents from one fish farm with a production of 100 tons would amount to 950 kg P and 7800 k g N and would, thus, taken over a cultivation season, equal the nutrients in untreated sewage produced by a human settlement of between 850 persons and 1950 persons (Table 1). If the figures from 1974 are used, nutrients generated by a 100-ton fish farm would correspond to 2800-3200 persons. It may be realistic to use the 1994 estimate for salmon production in the developed world, but we would argue that this “best” estimate is less realistic in the developing world.

The production of salmonids in Nordic countries (Denmark, Faeroe Islands, Finland, Iceland, Norway and Sweden) was 190000 tons in 1989, and conservatively assumed to be 200 000 tons in 1994 (Enells and Ackefors, 1991). Extrapolating the results in Table 1 to the assumed total production of salmonids in Nordic countries in 1994 of 200 000 tons, and comparing them with human nutrient production, we find that if the efforts to reduce the nutrient leakage would not have taken place the nutrient produced by the fish farms would have corresponded to those produced by a human population of about 6 million. Even if the presumed nutrient reduction has taken place in 1994, the release would still exceed those of the population of Stockholm, the largest city in Sweden, or even the population of whole nations (e.g. Lithuania has 3.7 million inhabitants) (Table 2). Obviously, the release of nutrients from fish farms into water bodies is substantial.

t .Isa

178 Costs of eutrophication

5. Internalizing environmental costs of eutrophication

According to a declaration in 1990 by the prime ministers of the states in the Baltic Sea region, the load of pollutants, nutrients included. to coastal waters is to be reduced by 50%. Large amounts of money are spent on reducing outlets of nutrients from point sources to coastal waters. A major measure to reduce municipal sewage effluents has been to build sewage treatment plants. A step towards the achievement of the 50% objective has been taken by the Swedish parliament, in a decision stating that the nitrogen release from new treatment plants to coastal areas of the Baltic proper and Kattegatt is to be reduced to this level. The cost of the implementation is estimated to about SEK 4 billion. The marginal cost, that is the increase in total cost when the nitrogen load is decreased by 1 kg, varies between SEK 50-100 per kg nitrogen reduction (Fleischer et al., 1989). These values can be interpreted as the Swedish society's willingness to pay at the margin for such nitrogen reduction. The cost for phosphorus reduction is SEK 20-30 per kg (The Swedish Environment Protection Agency, Stock- holm, pers. comm.). Using the cost for phosphorus and the lower cost for nitrogen reduction. the eutrophication cost to Swedish society from a fish farm producing 100 tons of salmon would then be at least SEK 425 000-725 000 (Table 1). This estimate covers parts of the environmental costs generated by the fish farming industry to society. or the external costs are referred to by economists.

As we stressed in the introduction, the polluter-pays principle and internalization of environmental costs are of high priority in the Rio declaration. What would the production cost look like for the industry if the fish farmer would have to pay the external costs to society of the nutrient releases he causes; that is. if the cost of eutrophication of salmon farming would be internalized?

During the 198Os, when the fish farming industry boomed, the highest price paid to the farmer for salmonids cage cultured in Sweden was about SEK 30 per kg. Because of the enormous expansion of farmed salmonid production. not only in Scandinavia but worldwide, there has been a considerable drop in prices. In 1991, the farmer's price per kg cage cultured rainbow trout and Atlantic salmon in Sweden varied from SEK 17- 23 per kg, while the costs of production at the level of the firm were SEK 19-23 per kg for rainbow trout and SEK 27 per kg for Atlantic salmon (Vattenbrukarnas Riksfor- bund. pers. comm.), of which more than 80% are variable costs (Cedrins, 1989). Obviously, there was not much to be gained from farming salmonids in 1991, and as a consequence there have been lots of bankruptcy in the fish farming industry.

If we base the external cost of coastal eutrophication on the predicted fish farm releases of nutrients in 1994, it would be SEK 4-43 per kg salmon. Internalizing this cost, i.e. adding it to the production cost at the level of the firm, would increase the production cost to SEK 21-273 per kg rainbow trout and to SEK 3 1-3 1.5 for salmon. With the costs of eutrophication internalized, the latter production cost exceeds the highest price paid for farmed salmonids in the 1980s, the decade when profitability of salmonid farming in cages was the highest.

6 . Discussion

Environmental effects are often caused by intensification, expansion, decreased distances between rearing sites and through the exploitation of feed resources. Such impacts aegatively affect other sectors of socio-economic significance, such as recreational activities in coastal areas, commercial fisheries and sports fisheries.

C. Folke et ul.

The international commui degradation is that polluters (

national authorities should stri In this article we have internalii farming in cages, the cost o production cost that exceeded internalizing only one cost can behaves in a sustainable fashio

Ideally, not only the exte salmonid farming (some of whi in the cost of production. This of the firm and thereby consic salmonids. Other environment: effects. are those associated WI

heavily used in the Norwegian use of antibiotics in fish farmi practice and agriculture in N Furthermore, and partly due t spreading of diseases and paras species included. Farmed salm migrate to rivers with natural 1: genetically fine-tuned adaptatio the environmental problems ge are strikingly similar to thost Trenbath et al., 1990).

It should be mentioned tha benefits of salmonid farming, : areas with emigration. This is many environmental costs of fis monoculture industry. It is clez have not been recognized in PO

The emphasis within the ,

environmental impacts is only contribute to the generation of also requires an even larger sup society. It is not sensible to take industry on it, not in a world 1

1994). Instead, the future for environmentally benign, that fol and that in particular redirects t synergy with the processes of ecc production that really are ecolc that direction (Yan and Yao, 1'

However, financial incentivl technological transition possible internalizing environmental cost for the development of more Cornwell, 1992; Stavins and W1 environmental costs will be inter

Costs of eutrophication

‘rs of the states in the Baltic Sea astal waters is to be reduced by outlets of nutrients from point municipal sewage effluents has s the achievement of the 50% in a decision stating that the areas of the Baltic proper and implementation is estimated to icrease in total cost when the )-I00 per kg nitrogen reduction :ted as the Swedish society’s :tion. The cost for phosphorus :nt Protection Agency, Stock- d the lower cost for nitrogen 3m a fish farm producing 100 SO00 (Table 1). This estimate sh farming industry to society,

rinciple and internalization of jeclaration. What would the rmer would have to pay the iuses; that is, if the cost of

ned, the highest price paid to lut SEK 30 per kg. Because of not only in Scandinavia but

n I99 1, the farmer’s price per lweden varied from SEK 17- firm were SEK 19-23 per kg

i (Vattenbrukarnas Riksfor- iable costs (Cedrins, 1989). salmonids in 199 1, and as a

1 farming industry. I on the predicted fish farm g salmon. Internalizing this .he firm, would increase the to SEK 3 1-3 1.5 for salmon. x-oduction cost exceeds the lecade when profitability of

jansion, decreased distances :d resources. Such impacts ance, such as recreational sheries.

C. Folke et al. 179

The international community has argued that a major cause of environmental degradation is that polluters do not pay for the pollution they cause, and therefore national authorities should strive to promote the internalization of environmental costs. In this article we have internalized one of the environmental costs derived from salmonid fxming in cages, the cost of eutrophication. Internalizing this cost resulted in a production cost that exceeded the highest price paid for salmon in the 1980s. Hence, internalizing only one cost can in some cases be sufficient to reveal whether an industry behaves in a sustainable fashion or not, also in economic terms.

Ideally, not only the external costs of eutrophication but all external costs of salmonid farming (some of which would be very difficult to estimate) should be included in the cost of production. This would of course increase the production cost at the level of the firm and thereby considerably reduce or eliminate the profitability in farming salmonids. Other environmental costs, besides eutrophication and its direct and indirect effects, are those associated with the use of chemicals and medicines. Antibiotics are heavily used in the Norwegian fish farming industry, and it has been predicted that the use of antibiotics in fish farming by 1990 would far exceed the totals used in general practice and agriculture in Norway (Solbe, 1987, cited in Beveridge et al., 1991). Furthermore, and partly due to the eutrophied environment, there are outbreaks and spreading of diseases and parasites, and impacts on natural flora and fauna, commercial species included. Farmed salmonids escaping the cages is a large problem since they migrate to rivers with natural populations, and may infest these stock and change their genetically fine-tuned adaptations to the specific conditions of their home rivers. In fact, the environmental problems generated by the throughput salmon cage culture industry are strikingly similar to those generated by intensive agriculture (e.g. Smil, 1987: Trenbath et al., 1990).

It should be mentioned that, in addition to the external costs, there are “external” benefits of salmonid farming, such as local benefits of maintenance of settlements in areas with emigration. This is probably a major reason why society has accepted the many environmental costs of fish farming, and in fact even subsidized this unsustainable monoculture industry. It is clear, however, that the magnitude of environmental costs have not been recognized in policy.

The emphasis within the aquaculture sector on improving the feed to reduce environmental impacts is only a marginal and short-term solution, and may in fact contribute to the generation of toxic algal blooms as discussed earlier in this article. I t also requires an even larger support from the fossil-fuel-based infrastructure of modern society. It is not sensible to take such a support for granted in the long run and base an industry on it, not in a world which faces biophysical limits (Rees and Wackernagel, 1994). Instead, the future for the industry lies in developing technologies that are environmentally benign, that focus on eco-efficiency and recycling (Schmidheiny, 1992), and that in particular redirects the present unsustainable behaviour towards working in synergy with the processes of ecosystems. The real challenge is to develop alternatives of production that really are ecologically sustainable. Indeed, progress is being made in that direction (Yan and Yao, 1989; Troell et al., unpublished).

However, financial incentives should be created by governments to make such a technological transition possible. A major advantage of using economic instruments and internalizing environmental costs of salmon farming would be that incentives are created for the development of more environmentally benign technologies (Costanza and Cornwell, 1992; Stavins and Whitehead, 1992). Clearly, political decisions stating that environmental costs will be internalized would give strong signals to redirect production,

180 Costs of eutrophication

and would induce the technological shift. A tax could be one such instrument. However, when the environmental cost is relatively high, an immediate internalization would presumably cause collapse of the industry, because there is no way that a firm can switch technology from one day to another. What is needed is that policy makers give clear and long-term signals that internalization will in fact take place, gradually or at full cost at a later date. It does, however, require foresight and courage among politicians and decision makers.

7. Conclusions

We have previously argued that coastal salmonid farming in cages is an intensive monoculture and an ecologically unsustainable food-production system (e.g. Folke and Kautsky, 1992). The results in this article also point out that the present behaviour of the industry is economically unsustainable from society’s perspective. The internalization of social cost of nutrient release from fish farming reveals that the cost of producing salmon exceeds the highest price paid for farmed salmon in the 1980s, the decade when the industry boomed. In addition, there are other environmental impacts not internalized in the production cost.

So far, the attempts of the industry to reduce its eutrophication impacts on the environment are based on conventional approaches. Investments are made in “end of the pipe” type solutions and in improvements of the mix of resource inputs of the feed. Few attempts have been made to restructure the production itself. The future for an ecologically, as well as economically, sustainable salmon farming industry is to be found in production systems that in reality combines environmental concern and economic development. There is an urgent need for real innovation towards production systems that work in synergy with ecosystems, such as the suggested integrated coastal culture of seaweeds-mussels-salmonids (Folke and Kautsky, 1992; Kautsky et al., 1994; Troell rt al., unpublished). To fulfill the goals of the Rio conference a major challenge for the fish farming industry is to develop such systems, and for policy makers to provide the correct signals to make the transition possible.

We have received valuable comments on the manuscripts from Edna Graneli, Department of Marine Ecology, University of Lund, Folke Giinther, Department of Systems Ecology, Stock- holm University, and Ing-Marie Gren and Karl-Goran Maler at the Beijer International Institute of Ecological Economics, Stockholm. The study was partly financed through grants from the Swedish Agency for Research Cooperation with Developing Countries (SAREC) and the Swedish Council for Forestry and Agricultural Research (SJFR).

References Ackefors, H. (1986). The impact on the environment by cage-farming in open water. Journal of Aqua. Trop. 1.

Ackefors, H. and Enell, M. (1990). Discharge of nutrients from Swedish fish farming to adjacent sea areas,

Baskerville, G. L. (1988). Redevelopment of a degraded forest ecosystem. Ambio 17, 314-322. Beveridge, M. C. M., Phillips, M. J. and Clarke, R. M. (1991). A quantitative and qualitative assessment of

wastes from aquatic animal production. In Aquaculture and Water Quality (D. Brune and J. R. Tomasso, eds), pp. 5 0 6 5 3 3 . Advances in World Aquaculture, Vol. 3. Baton Rouge: The World Aquaculture Society.

Braaten, B., Poppe, T., Jacobsen, P. and Maroni, K. (1988). Risks for self-pollution in aquaculture; evaluation and consequences. In Eficiency in Aquaculture Production: Disease Control, (E. Grimaldi and H. Rosenthal, eds), pp. 139-166. Milano: Edizioni del Sole 24 Ore.

Carlsson, P. and GranCli, E. (1993). Availability of humic bound nitrogen for coastal phytoplankton, Estuarine, Coastal and Shelf Science 36, 433447.

25-33.

Ambio 19, 28-35.

C. Folke et al.

Carlsson, P., GranCli, E. and Olsson behind Chrysochromulina polylep Edler and D. Andersson, eds), pi

Cedrins, R. (1989). Vattenbrukets ek Constanza, R. and Cornwell, L. (195

Costanza, R. and Daly, H. E. (1992; 37-46.

Edvardsen, B., Moy, F. and Paasche grown at different levels of sele Sundstrom, E. Edler and D. And

Enell. M. and Ackefors, H. (1991). h Adjacent Sea Areas. ICES Repor:

FAO. (1992). Aquaculture Producti Department, FAO.

Fleischer, S., Andrtasson, I.-M., Ho Markan vandning- vat tenk valitet : er tration Board Report 1989: 10. (I

Folke, C. (1988). Energy economy 0:

525-537. Folke, C. and Jansson, A. M. (1992)

fisheries and aquaculture. In Socit B. Aniansson, eds), pp. 69-87. Dc

Folke, C. and Kautsky, N. (1989). The

Folke, C. and Kautsky, N. (1991). Ecc

34, 12-20, 42.

18, 234-243.

Strategies & Aquaculture Waste University of Guelph Press.

Folke. C. and Kautsky, N. (1992). Ac Coastal Management 17, 5-24.

Folke, C., Hammer, M. and Jansson, Baltic Sea region. Ecological Econ

Grantli, E., Edler, L., Gedziorowsk: Prorocentrum minimum (Pav.) J. S G. Baden, eds), pp. 201-206. Ams

Granili, E., Carlsson, P., Olsson, P., C poisoning: the last ten years of p h B1oom.y- Causes and Impacts of R Bricelj and E. J. Carpenter, eds), E

GranBli, E., Wallstrom, K., Larsson, 1 production in the Baltic Sea area.

Gowen, R. J. and Bradbury, N. B. ( 1 review. Oceanography and Marine

Hamilton, L. S., Dixon, J. A. and Mill and of the sea. In Ocean Yearbook Chicago: The University of Chical

Kautsky, N. and Folke, C. (1989). Mai

Kautsky, N., Troell, M. and Folke environmental improvement in ope (C. Etnier and B. Guterstam, eds),

Lal, P. N. (1990). Conservation or I

Occasional Paper 1 1. Environment Person, G. (1992). Eutrophication res

experiences. In Nulritionul Strateg, Ontario: University of Guelph Pre:

Rees, W. and Wackernagel, M. (191 requirements of the human econom Sustainability (A. M. Jansson, M. Press.

Regier, H. A. and Baskerville, G. L. ( exploitive development. In Sustainr pp. 75-101. Cambridge: Cambridge

Rosenberg, R., Elmgren, R., Fleisck eutrophication case studies in Swec

Biota 5, 1-1 1.

Costs of eutrophicatiol

)e one such instrument. However mmediate internalization would 'e is no way that a firm can switch that policy makers give clear and lace, gradually or a t full cost at a courage among politicians and

.rming in cages is an intensive oduction system (e.g. Folke and that the present behaviour of the rspective. The internalization of iat the cost of producing salmon he 1980s, the decade when the mtal impacts not internalized in

eutrophication impacts on the vestments are made in "end of : of resource inputs of the feed. iction itself. The future for an farming industry is to be found mental concern and economic n towards production systems ed integrated coastal culture of Kautsky et al., 1994; Troell et e a major challenge for the fish I makers to provide the correct

n Edna Graneli, Department of nent of Systems Ecology, Stock- the Beijer International Institute

lanced through grants f rom the ntries (SAREC) a n d the Swedish

)pen water. Journal ofAqua. Trop. 1.

1 fish farming to adjacent sea areas.

I. Ambio 17, 314322. tathe and qualitative assessment of .a/ity (D. Brune and J. R. Tomasso, :e: The World Aquaculture Society. pollution in aquaculture; evaluation re Control, (E. Grimaldi and H.

rogen for coastal phytoplankton.

C. Folke et al. 181

Carlsson, P., Graneli, E. and Olsson P. (1990). Grazer elimination through poisoning: one of the mechanisms behind Chrysochromulina polylepis bloom. In Toxic Marine Phytoplankton (E. Graneli, B. Sundstrom, E. Edler and D. Anderson, eds), pp. 116-122. New York: Elsevier.

Cedrins, R. (1989). Vattenbrukets ekonomi och finansiering. Foreningsbanken. Constanza, R. and Cornwell, L. (1992). The 4P approach to dealing with scientific uncertainty. E m ' . , I I onment

Costanza, R. and Daly, H. E. (1992). Natural capital and sustainable development. Conservation Biology 6, 3746.

Edvardsen, B., Moy, F. and Paasche, E. (1990). Hemolytic activity in extracts of Chrysochromulina po/ylepis grown at different levels of selenite and phosphate. In Toxic Marine Phytoplankton (E. Graneli, B. Sundstrom, E. Edler and D. Anderson, eds), pp. 284289. New York: Elsevier.

Enell, M. and Ackefors, H. (1991). Nutrient Discharges from Aquaculture Operations in Nordic Countries into Adjacent Sea Areas. ICES Report CM 1991/F:56. ICES: Copenhagen.

FAO. (19921. Aauaculture Production. F A 0 Fisheries Circular No. 8 15, Revision 4. Rome: Fisheries

34, 12-20, 42.

\ ,

Department, i'AO. Fleischer, S., Andreasson, I.-M., Holmgren, G., Joelsson, A,, Kindt. T., Rydberg. L. and Stibe, L. (1989).

Markanvandnirw-vattenkvalitet: en Studie i Laholmsbuktens Tillrinningsomrdde. Halland County Adminis- " tration Board Report 1989: 10. (In Swedish.)

Folke, C. (1988). Energy economy of salmon aquaculture in the Baltic Sea. Environmental Managemem 12,

Folke, C. and Jansson, A. M. (1992). The emergence of an ecological economics paradigm: examples from fisheries and aquaculture. In Society and the Environnzent: A Swedish Research Perspective (U. Svedin and B. Aniansson, eds), pp. 69-87. Dordrecht: Kluwer.

Folke, C. and Kautsky, N. (1989). The role of ecosystems for a sustainable development of aquaculture. Ambio

Folke, C. and Kautsky, N. (1991). Ecological economic principles for aquaculture development. In Nutritional Strategies & Aquaculture Waste (C. B. Cowey and C. Y. Cho. eds), pp. 207-222. Guelph, Ontario:

525-531.

18, 234-243.

University of Guelph Press. Folke. C. and Kautskv. N. (1992). Aquaculture with its environment: prospects for sustainability. Ocean and _. , \ I ~

Coastal Management 17, 5-24. Folke, C., Hammer, M. and Jansson, A. M. (1991). The life-support value of ecosystems: a case study of the

Baltic Sea region, Ecological Economics 3, 123-137. Graneli, E.. Edler, L., Gedziorowska, D. and Nyman, U. (1985). Influence of humic and fluvic acids on

Prorocentrum mininzuwi (Pav.) J. Schiller. In Toxic Dinufagellates (D. M. Anderson, A. W. White and D. G. Baden, eds), pp. 201-206. Amsterdam: Elsevier.

Graneli, E., Carlsson, P., Olsson, P., Sundstrom, B., Grantli, W. and Lindahl. 0. (1989). From anoxia to fish poisoning: the last ten years of phytoplankton blooms in Swedish marine waters. In Novel Phytoplankton Bloom-Causes and Impacts of Recurrent Brown Tides and Other Unusual Blooms (E. M. Cosper, V. M. Bricelj and E. J. Carpenter, eds), pp. 407428. New York: Springer-Verlag.

Graneli, E., Wallstrom, K., Larsson, U., Graneli, W. and Elmgren, R. (1990). Nutrient limitation of primary production in the Baltic Sea area. Ambio 19, 142-151.

Gowen, R. J. and Bradbury, N. B. (1987). The ecological impact of salmonid farming in coastal waters: a review. Oceanography and Marine Biology Annual Review 25, 563-575.

Hamilton, L. S., Dixon, J. A. and Miller, G. 0. (1989). Mangrove forests: an undervalued resource of the land and of the sea. In Ocean Yearbook 8 (E. Mann Borgese, N. Ginsburg and J. R. Morgan, eds), pp. 254-288. Chicago: The University of Chicago Press.

Kautsky. N. and Folke, C. (1989). Management of coastal areas for a sustainable development of aquaculture. Biota 5 , 1-1 1.

Kautsky, N., Troell. M. and Folke, C. (1994). Ecological engineering for increased production and environmental improvement in open sea aquaculture. In Ecological Engineering for Wastewater Treatment (C. Etnier and B. Guterstam, eds), Chelsea: Lewis Publishers.

Lal, P. N. (1990). Conservution or Conversion of Mangroves in Fiji: an Ecological Economic Analyses. Occasional Paper 11. Environment and Policy Institute, East-West Center, Hawaii.

Persson, G. (1992). Eutrophication resulting from salmonid fish culture in fresh and salt waters: Scandinavian experiences. In Nutritional Strategies & Aquaculture Waste. (C. B. Cowey and C. Y. Cho, eds). Guelph, Ontario: University of Guelph Press.

Rees, W. and Wackernagel, M. (1993). Appropriated carrying capacity: measuring the natural capital requirements of the human economy. In Investing in Natural Capital: The Ecological Economic Approach to Sustainability (A. M. Jansson, M. Hammer, C. Folke and R. Costanza, eds). Covelo, California: Island Press.

Regier, H. A. and Baskerville, G. L. (1986). Sustainable redevelopment of regional ecosystems degraded by exploitive development. In Sustainable Development of the Biosphere (W. C. Clark and R. E. Munn, eds). pp. 75-101, Cambridge: Cambridge University Press.

Rosenberg, R., Elmgren, R., Fleischer, S., Jonsson, P., Persson, G. and Dahlin, H. (1990). Marine eutrophication case studies in Sweden. Ambio 19, 102-108.

182 Costs of eutrophication ! Ryther, T. H. and Officer, C. B. (1981). Impact of nutrient enrichment on water uses. In Estuaries and Nutrients

Schindler, D. (1977). Evolution of phosphorus limitation in lakes. Science 195, 260-262. Schmidheiny, S. (1992). Changing Course: A Global Business Perspective on Development and the Environment.

Shiva. V. (1991). The Violence of the Green Revolution: Third World ARriculture, Ecology and Politics. London:

(B. J. Neilson and L. E. Cronin, eds), pp. 247-261. Clifton, New Jersey: Humana Press.

Cambridge, Massachusetts: MIT Press.

Zed Books.’ Smil. V. (1987). EnerPv. Food, Environment: Realities, Mvths, Options. Oxford: Clarendon Press , \ /

Solbe. J. F. de L. G. (1987). Report of the European inland Fisheries Advisory Comnzission Working Part?> on Fish Farm Efluents. Medmenham, U.K.: Water Research Centre.

Stavins, R. N. and Whitehead, B. W. (1992). Dealing with pollution: market-based incentives for environmen- tal protection. Environment 34, 19.

Torell, L. (1977). Pollutants from Swedish municipal and industrial outlets into the Baltic Sea. Ambio Special Report 5, 213-218.

Trenbath, B. R., Conway, G. R. and Craig, I. A. (1990). Threats to sustainability in intensified agricultural systems: analysis and implications for management. In Agroecdogy: Researching the Ecological Basi.s.for Sustainable Agriculture ( S . R. Gliessman, ed.), pp. 337-365. New York: Springer-Verlag.

Troell, M., Kautsky, N. and Folke, C. Integrated aquaculture as a means of reducing the problems of monocultures. Department of Systems Ecology, Stockholm University. Unpublished manuscript.

Yan, J. and Yao, H. (1989). Integrated fish culture management in China. In Ecological Engineering: an introduction to Ecotechnolo~y (W. J. Mitsch and S. E. Jorgensen, eds), pp, 375408. New York: Wiley Interscience.

Journal of’ Environmental Mana

Using Image-capture Te Urban/Forest Interface:

Rebecca L. Johnson

Department of Forest Resour Oregon 97331, U.S.A.

Mark W. Brunson

Depurtment of Forest Resoun Utah 84322, U.S.A.

and Takashi Kimura

Japan Highway Public Corpo Tokyo 100, Japan

Received 16 December 1992

Urban expansion to the edge of for resource managers as home c management. Strategies to resoh all participants are equally infori alternatives. In cases where sceni “image-capture” technology (I@ base, especially when experiment being considered. This paper des, home owners in Oregon viewed 1 their own homes if various silvici forest. The study sought to evalu localized impacts of timber mana land owners would express willin showed promise, although geners cumbersome. A majority of land scenic-protection measures that v adjacent properties.

Keywords: urbaniforest interface, easements.

03014797/94/020183 + 13 SOS.OO/O