Ecological Economics 29 (1999) 253 – 268

ANALYSIS

Ecosystem services generated by fish populations

Cecilia M. Holmlund *, Monica Hammer

Natural Resources Management, Department of Systems Ecology, Stockholm Uni6ersity, S-106 91, Stockholm, Sweden

Abstract

In this paper, we review the role of fish populations in generating ecosystem services based on documented
ecological functions and human demands of fish. The ongoing overexploitation of global fish resources concerns our
societies, not only in terms of decreasing fish populations important for consumption and recreational activities.
Rather, a number of ecosystem services generated by fish populations are also at risk, with consequences for
biodiversity, ecosystem functioning, and ultimately human welfare. Examples are provided from marine and
freshwater ecosystems, in various parts of the world, and include all life-stages of fish. Ecosystem services are here
defined as fundamental services for maintaining ecosystem functioning and resilience, or demand-derived services
based on human values. To secure the generation of ecosystem services from fish populations, management
approaches need to address the fact that fish are embedded in ecosystems and that substitutions for declining
populations and habitat losses, such as fish stocking and nature reserves, rarely replace losses of all services. © 1999
Elsevier Science B.V. All rights reserved.

Keywords: Ecosystem services; Fish populations; Fisheries management; Biodiversity

1. Introduction 15 000 are marine and nearly 10 000 are freshwa-

ter (Nelson, 1994). Global capture fisheries har-
Fish constitute one of the major protein sources vested 101 million tonnes of fish including 27
for humans around the world. There are to date million tonnes of bycatch in 1995, and 11 million
some 25 000 different known fish species of which tonnes were produced in aquaculture the same
year (FAO, 1997). Despite the abundance and
* Corresponding author. Tel.: +46-8-164252; fax: +46-8- variation of fish, most western fisheries focus on a
158417. few target species. Approximately 75% of the
E-mail address: cecilia@system.ecology.su.se (C.M. Holm-
world’s marine fish landings consist of 200 (
lund)

0921-8009/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 1 - 8 0 0 9 ( 9 9 ) 0 0 0 1 5 - 4
254 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

1%) known existing marine fish species (FAO, al., 1995). We also discuss some of the conse-
1997). Further, recreational fishing in lakes and quences of human impacts on fish populations,
along coasts is a major tourism activity (FAO, such as overfishing in relation to ecosystem re-
1996; Postel and Carpenter, 1997). Estimates of silience. We define resilience here as the amount
the value of fish populations for human societies of change or disruption that is required to trans-
have predominantly focused on these goods. The form a system from being maintained by one set
fact that such values are derived from ecosystems of mutually reinforcing processes to a different set
with complex interactions, and that both econom- of processes (Holling, 1973). Identifying ecosys-
ically and non-economically valuable fish popula- tem services that various fish populations are part
tions play active roles in the maintenance of these of, or generate for human societies, is one step
ecosystems and in the provision of a range of toward holistic, ecosystem-based, resource man-
ecosystem services, is seldom taken into account. agement with increased understanding of effects
Ecosystem services have been defined as ‘‘condi- on the dynamic, often unpredictable ecosystems
tions and processes through which natural ecosys- by fisheries.
tems, and the species that make them up, sustain
and fulfill human life’’ (Daily, 1997, p. 3). This
includes the life-support functions (Odum, 1989) 2. Ecosystem services
of ecosystems and nature’s capacity to provide
aesthetic and cultural quality to human life In this paper, we distinguish between two major
(Daily, 1997). categories of ecosystem services: fundamental and
In 1995, almost 70% of the world’s major demand-derived ecosystem services (Table 1). By
marine fish resources were fully- to overharvested, ‘fundamental ecosystem services’ we mean those
or depleted (World Resources Institute, 1996). that are essential for ecosystem function and re-
Capture fisheries not only reduce the abundance silience, such as nutrient cycling. These are ulti-
of targeted stocks with cascading responses in the mately a prerequisite for human existence,
food web and with consequences in other ecologi- irrespective of whether humans are aware of it or
cal and fishery dependent systems, but also impact not. Such services are often not linked to any
an array of other species, including mammals, as specific economic market value. The ‘demand-
bycatch (Dayton et al., 1995; Steneck, 1998). In derived ecosystem services’, such as recreational
addition, many nearshore ecosystems are substan- values, are formed by human values and de-
tially altered through the destruction of benthic mands, and not necessarily fundamental for the
habitats by detrimental fishing methods survival of human societies. Nevertheless, all de-
(Malakoff, 1997). Indirect effects of fishing can mand-derived ecosystem services ultimately de-
have more important impacts on aquatic ecosys- pend on natural systems and the fundamental
tem structure and function than the removal of ecosystem services provided by fish, and are not
the fish (Hammer et al., 1993; Hughes, 1994; replaceable by technological innovations.
Botsford et al., 1997; Estes et al., 1998).
In this paper, we review some of the current
knowledge about how fish populations provide 3. Fundamental ecosystem services generated by
ecosystem services for human societies, and the fish populations
relations between these services and functioning
ecosystems in different regions of the world. 3.1. Regulating ser6ices
Available literature relating to large-scale marine
systems is limited owing to the obvious difficulty Consumption of organisms by fish is a salient
of performing ecosystem experiments here. Small- feature which can regulate trophic structure and
scale freshwater ecosystems, on the other hand, thus influence the stability, resilience, and food
are better understood in terms of influences of fish web dynamics of aquatic ecosystems; moreover,
on ecosystem structure and function (Carpenter et these regulatory influences change as fish pass
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 255

from one life stage to another (Carpenter et al., ing in an increase of zooplankton, which in turn
1992; Post et al., 1997). A fish larva, feeding on increases the predation on phytoplankton (Car-
zooplankton, is as distinct ecologically from its penter et al., 1985). Although most trophic cas-
adult form as it is from its planktonic prey. cade studies have been done in calm freshwater
Piscivores (fish that eat fish), preying on environments such as lakes, some studies have
zooplanktivores (fish that eat zooplankton), can been done in freshwater streams, the brackish
exert a strong top-down control resulting in a Baltic Sea, and coral reefs (Hughes, 1994; Rud-
cascade of effects down the food chain (Fig. 1). stam et al., 1994; Deegan et al., 1997). The degree
The general mechanism is explained as a decrease of regulatory influence on foodchain relationships
by fish varies with physical and climatological
in the predation pressure on zooplankton result-
preconditions, including stream flow, temperature,
upwelling fronts, storms, seasonal variability, and
Table 1
also with nutrient content and water depth (Rud-
Major fundamental and demand-derived ecosystem services
generated by marine and freshwater fish populations stam et al., 1994; MacKenzie et al., 1996; Deegan
et al., 1997; Jeppesen et al., 1998a). Also, Carpen-
Fundamental ecosystem services ter and Kitchell (1993) suggest that regulatory
effects on species composition can be strong with-
Regulating services Linking services
out affecting the overall function of the ecosys-
Regulation of food web Linkage within aquatic tem. We illustrate these general concepts with
dynamics ecosystems several examples that follow.
Recycling of nutrients Linkage between aquatic
and terrestrial ecosystems 3.1.1. Regulating food web dynamics and nutrient
Regulation of ecosystem Transport of nutrients,
balances
resilience carbon and minerals
Redistribution of bottom Transport of energy Removal of fish with key characteristics and
substrates functions from the ecosystem may result in loss of
Regulation of carbon fluxes Acting as ecological memory resilience and in the ecosystem changing from one
from water to atmosphere equilibrium state to another (Holling, 1986;
Maintenance of sediment
Chapin et al., 1997; Grime, 1997; Tilman et al.,
processes
Maintenance of genetic, 1997) (cf. multiple stable states, e.g. Sutherland,
species, ecosystem 1974; Sheffer, 1990; Levin, 1992; Walker, 1993).
biodiversity Coral reef ecosystems in shallow coastal waters
are exposed to hurricanes, typhoons, or cyclones
Demand-derived ecosystem services which are irregularly occurring events. Thus, the
regeneration of a healthy reef system is dependent
Cultural services Information services on rapid colonization of larval recruits. Hughes
Production of food Assessment of ecosystem
(1994) showed how this recovery mechanism has
stress been hindered in Jamaican coral reefs by human
Aquaculture production Assessment of ecosystem activities. Since the 1950/1960s the Jamaican coral
resilience reefs have been chronically overfished, such that
Production of medicine Revealing evolutionary
sharks, snappers, jacks, triggerfish, groupers, and
tracks
Control of hazardous Provision of historical
a number of other target species have declined
diseases information markedly (Hughes, 1994). The loss of herbivorous
Control of algae and Provision of scientific and and predatory fish species has reduced total fish
macrophytes educational information biomass and altered the taxonomic composition
Reduction of waste
of the fish community.
Supply of aesthetic values
Supply of recreational
However, the ecological effects of this decrease
activities in biodiversity were not realized for several
decades, as the reefs appeared to be healthy with
256 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

Fig. 1. Illustration of ecosystem services generated by fish populations: Fundamental ecosystem services: Regulating services include
(a) top-down effects regulating population dynamics and nutrient availability, (b) bioturbation in or near sediments, and (c) carbon
exchange. Linking services include active transport of nutrients, carbon and energy between the pelagic and (d) hard and soft
bottoms, and (e) the littoral. Linking services also include passive transport of nutrients between ecosystems when fish eggs, fry,
juveniles, adults, and carcasses are preyed on by (f) birds, and (g) mammals. Demand-derived services: Information services include
fish as indicators of (h) ecosystem health, recovery and resilience, and (i) environmental recorders. Cultural services include fish as
(j) goods, (k) for purifying water, (l) for recreation, (m) for mitigating the spread of diseases, and (n) for aquaculture.

large coral cover and high benthic diversity the previous overfishing of grazing fishes.
(Hughes, 1994). This was largely due to the high In the Baltic Sea, Atlantic cod (Gadus morhua)
abundance of one grazing echinoid Diadema antil- is among the commercially most important spe-
larum, which held in check the growth of algae on cies. Cod, although omnivorous, is also most
the reef. After the decline of fish predators and likely the most dominant piscivore in the Baltic
competitors, Diadema increased (Hughes, 1994). Sea (Rudstam et al., 1994). Approximately 70% of
However, after an intensive hurricane at the the diet of adult cod is fish of which 55% is
beginning of the 1980s the coral cover decreased Atlantic herring (Clupea harengus) and sprat
between 1980 and 1993 from 52 to 3%, while the (Sprattus sprattus) (Sparholt, 1994). Between 1980
cover of benthic macroalgae increased from 4 to and 1992 the biomass of cod in the Baltic Sea
92%, shifting the system from a coral dominated decreased from a peak of over 800 000 to less than
to an algal dominated ecosystem (Hughes, 1994). 80 000 t, owing to overfishing in combination with
Hughes (1994) argues that this shift is partly human-induced eutrophication and natural cli-
explained by the fact that in 1983, Diadema suf- matic variations (MacKenzie et al., 1996). The
fered a mass mortality due to a species-specific decline is illustrated by the Swedish harvest of cod
pathogen reducing the population by 99%, and in the Baltic Sea, which reached a maximum of
partly a result of loss of buffering capacity due to 59 000 t in 1984, and dropped to 16 000 t in 1993.
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 257

The overall decline of cod resulted in a shift in ence dynamics of nutrient-rich lakes. Changes of
major fish community composition in many parts grazing fish community, such as grass carp
of the Baltic Sea towards a relatively high abun- (Ctenopharyngodon idella), can trigger a shift be-
dance of sprat and herring (Bagge et al., 1994; tween two alternative stable states of a nutrient-
Rudstam et al., 1994; Sparholt, 1994). This has rich lake, one turbid and one clear (Jeppesen et
resulted in a shift in Baltic commercial fisheries al., 1998b).
from cod to an increased industrial fishery for
mainly zooplanktivorous herring and sprat, which 3.1.2. Regulating sediment processes
are used for fish meal production. A limited number of studies about the relation-
In freshwater systems, the feeding behavior of ship between fish, bioturbation (the physical dis-
many adult and young fishes has cascading effects turbance of sediment associated with foraging or
on population dynamics down the food web (Car- burrowing activities by consumers), and the struc-
penter et al., 1985). The feeding pattern of fishes turing of bottom conditions has been done in
can also influence the temporal availability of rivers and lakes (Fuller and Cowell, 1985; Flecker,
nutrients and the potential for algal blooms in 1992; DeVries, 1997; Gelwick et al., 1997).
nutrient-rich lakes, since fish mineralize nitrogen Salmonids cause bioturbation in streams while
and phosphorus through excretion and defeca- spawning and thereby create and maintain their
tion, thereby making these nutrients available for own habitats (Montgomery et al., 1996). The fe-
primary production (Schindler, 1992). Further, it male salmon deposits her eggs in redds, where-
has been suggested that in lakes with adult pisci- after the eggs are fertilized, and covered with a
vores, large zooplankton are more abundant than layer of gravel. The spawning activities remove
small zooplankton, while in lakes dominated by aquatic macrophytes, fine sediment particles in-
adult zooplanktivores, small-sized zooplankton cluding organic matter (Field-Dodgson, 1987),
are more abundant than large-bodied taxa and may displace invertebrates from the bottom
(Brooks and Dodson, 1965; Vanni et al., 1990). to the water column, making them more available
Large zooplankton recycle nutrients more slowly to the river fish, as suggested by Bilby et al.
and are more efficient grazers on phytoplankton (1998). Repeated salmon spawning over many
than are small zooplankton. Hence, in a nutrient- years at the same location can modify the bottom
rich lake with piscivory and large zooplankton, contour and may lead to the formation of persis-
primary production can be suppressed (Shapiro tent bedforms with dune heights of over a meter
and Wright, 1984; Vanni and Layne, 1997). This (DeVries, 1997). The stream-bed alteration may
is illustrated by the events in Lake Mendota, WI, provide suitable habitats for juvenile salmonids
USA. Here, the sudden, massive mortalities of the (Field-Dodgson, 1987), and enhances the survival
zooplanktivorous cisco (Coregonus artedii ) during of salmon embryos as the build-up of coarse
an unusually hot summer resulted in reduced pre- gravel protects the embryos from rapid stream
dation pressure on zooplankton, and increased current (Montgomery et al., 1996).
abundance of large zooplankton, and decreased Gelwick et al. (1997) describe how benthic al-
the availability of nutrients in the water column, givorous fishes in the Illinois River in OK, USA,
with resulting decreases in biomass of algae and primarily Campostoma anomalum, Campostoma
improved water clarity (Vanni et al., 1990). oligolepus, and Ozark minnow (Notropis nubilus),
Juvenile fish also have the potential to influence resuspend silt, detritus, and other particulate or-
the abundance of algae. In Lake Pääjärvi in Fin- ganic matter from the bottom into the current
land, juvenile roach (Rutilus rutilus) feed on and while feeding, and thereby maintain a rapidly
suppress benthic insects and zooplankton in the growing algal community and enhance food
littoral zone, and thereby stimulate the growth of availability for collector-filterers. Flecker (1992)
algae in the lake (Kairesalo and Seppälä, 1987). found that feeding activities on algae or detritus
Further, the capacity of top-down control of associated with sediments by a diverse fauna of
grazing freshwater fishes has been shown to influ- grazing fishes, in the stream Rio Las Marias in
258 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

Venezuela, modified the distribution and abun- structure of fish communities can regulate the
dance of resources important to insects living in carbon-fixing capacity of nutrient-rich lakes, and
the stream. In the subtropical Lake Thonotosassa, thus indirectly mediate the flux of carbon between
FL, USA, spawning beds of the cichlid fish lake and atmosphere.
(Sarotherodon aurea) disturbed 11.5% of the lit-
toral zone and modified benthic community orga-
3.2. Linking ser6ices
nization (Fuller and Cowell, 1985). These studies
suggest that riverine and freshwater fishes may
Fish generate a large number of services related
play an active role in influencing abiotic and
to their movement patterns, including daily, sea-
biotic factors during feeding or spawning.
sonal, and yearly migration patterns in lakes,
Although no literature was found of the role of
rivers, estuaries, and oceans (Polis et al., 1997).
marine fish as bioturbators, the authors speculate
Fish that are consumed also transport nutrients
that marine fish also participate in influencing the
across spatial boundaries and thereby link differ-
organized layers of microbial communities in sedi-
ent ecosystems. The function of fish as active or
ments (Service, 1997), and other life conditions
passive transporters and distributors of energy
for benthic organisms, including animals living in
and materials can enhance primary production in
sediments (Woodin, 1978; Brenchley, 1981), when
nutrient poor environments (Bilby et al., 1996;
spawning, hiding, resting or feeding in or close to
Larkin and Slaney, 1997).
bottoms. In addition, bioturbation should be of
special importance in nutrient-poor environments,
since bottom-derived nutrients released into the 3.2.1. Fish as acti6e links between ecosystems
water can be incorporated into primary produc- Several species of anadromous salmonid fishes
tion instead of stored in the sediments. migrate from marine environments where they
spend most of their lives, to natal rivers to spawn
3.1.3. Regulating carbon flux and then die, and thereby transfer nutrients and
Oceans and lakes can function as either carbon carbon (Bilby et al., 1996; Larkin and Slaney,
sinks or sources (Kling et al., 1992; Cooper et al., 1997). In North American rivers, it has been
1996). Carbon fixation increases with high nutri- shown that marine-derived carbon and nutrients
ent input and high primary production. Recent are delivered to the river through fish excretion,
whole-lake experiments comparing two nutrient- production of gametes, and fish carcass decompo-
enriched lakes in WI, USA, have linked the com- sition, and contribute to the production of algae,
position of fish communities with ecosystem insect larvae, microbial decomposers, young
carbon fixation (Schindler et al., 1997). The first salmon, and other fish in the rivers (Kline et al.,
nutrient-rich lake (Peter Lake) with zooplanktivo- 1990; Bilby et al., 1996; Larkin and Slaney, 1997).
rous fish became a carbon sink because zooplank- For example, in Snoqualmie River in WA, USA,
ton were suppressed, and primary producers 20–40% of the nitrogen and carbon in juvenile
(carbon fixers) were released from grazing pres- coho salmon (Oncorhynchus kistutch) originated
sure. The second lake (West Long Lake) with from spawning coho salmon (Bilby et al., 1996).
piscivores was shown to be a carbon source be- In Lakes Dalnee and Blizhnee, Paratunka River
cause the piscivorous fish suppressed the abun- basin, in Kamchatka, northern Russia, between
dance of zooplanktivores. This allowed the 20 and 40% of the total annual phosphorus input
zooplankton community to exert a high grazing was supplied by anadromous sockeye salmon car-
pressure on phytoplankton. Carbon fixation de- casses (Oncorhynchus nerka) (Krokhin, 1975).
creased, the concentration of dissolved carbon Further, marine-derived nutrients and organic
dioxide in the water increased due to community matter originating from salmon eggs and car-
respiration, and a net diffusion of carbon from casses have been found to stimulate biomass pro-
the water surface to the atmosphere followed. The duction up to 50 km downstream (Bilby et al.,
study by Schindler et al. (1997) illustrates that the 1996; Larkin and Slaney, 1997).
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 259

The timing of this linking service is important. 3.2.2. Fish acting as ecological memory
As shown by Cederholm (1989), and Bilby et al. By acting as energy and nutrient reservoirs, and
(1998) carcasses and eggs from migrating salmon as gene pool storage between years and ecosys-
add marine-derived nutrients to river systems tems, migrating fish link spatial and temporal
from late autumn to early spring, a period of the scales (Kairesalo and Seppälä, 1987; Cederholm,
year when other nutrient inputs, like litterfall, are 1989). These active linking qualities of fish (Sec-
scarce in these streams. Thus migrating salmon tion 3.2.1) function as ‘memory’ in the ecosystem.
provide the nutrients necessary for the spring Consider the migratory behavior of fish in con-
algal bloom, which in turn drives ecosystem pro- nection with a natural break-down or change of
duction at other trophic levels. an ecological system, for example a glaciation
over thousands of years. During the ice period,
An example of the vast spatial scale over which
migratory fish can move into adjacent ecosystems
fish can link freshwater and marine ecosystems is
that remain ice-free and stable, and thereby es-
provided by European eels (Anguilla anguilla) in
cape the ‘destructive’ ice period. As the glaciation
the Baltic Sea. Eel juveniles and adults spend
diminishes, fish can return to the altered ecosys-
most of their life in fresh or brackish waters where tem, provide nutrients, energy, and their genetic
they feed and grow before they return to their material while spawning, and also contribute to
spawning grounds in the Sargasso Sea, over 8000 the structure and function of the new ecosystem
km away, where they reproduce and then die. The while feeding and moving about. Migratory fish
eels are not presumed to feed during their migra- can be seen as an important part of the ecological
tion to the Sargasso as their alimentary tracts memory that is necessary for the renewal (build-
virtually disappear with the onset of sexual matu- up) phase of ecosystems (Holling and Sandersson,
rity (Anon., 1994). Consequently, eels in the 1996).
Baltic Sea is one example of long-distant migrat-
ing fish species that transport nutrients, carbon 3.2.3. Fish as passi6e links between ecosystems
and other substances from one part of the world’s When fed upon by other organisms, fish, in-
seas to another. cluding eggs, fry and carcasses, serve as passive
There is considerable evidence for the impor- links between aquatic, aerial and terrestrial
tance of fish as ‘mobile links’ between ecosystems ecosystems, contributing to other food webs (Fig.
at short distances, relating to their daily migrating 1). Scavengers are important as vectors of this
between feeding and resting areas. In lakes, fish linkage. For example, in three streams on the
transport and redistribute phosphorus and other Olympic Peninsula in WA, USA, carcasses of
essential nutrients between the shore, pelagic, and coho salmon (O. kisutch) constituted a food
deeper bottom zones (Carpenter et al., 1992). In source for 22 species of mammals and birds living
coral or rocky reefs, juvenile grunts (Haemulon near the river (Cederholm, 1989).
In the Baltic Sea, approximately 20 000–25 000 t
spp.), blacksmith (Chromis punctipinnis) and other
of fish are eaten by seabirds every year (Sparholt,
fishes transport substantial amounts of nutrients
1994). The birds produce phosphate-rich feces
from their feeding areas (seagrass beds or open
which are usually deposited at the seabirds’
water) to their resting areas in the reef in the form
colony sites on islands or in coastal zones and has
of fecal products (Meyer et al., 1983). Bray et al. been shown to stimulate the production of
(1981) calculated that blacksmith feces contribute macroalgae in rockpools in the Baltic Sea region
an average of 23 mg and a maximum of 60 mg of (Ganning and Wulff, 1969).
carbon per square meter per night to smaller In eastern Florida Bay, many of the seagrass
crevices in the reef. Geesey et al. (1984) showed areas are nutrient-limited (Powell et al., 1991).
that migrating blacksmith contribute to an input However, nutrients from fishes consumed by
of phosphorus and trace minerals to coral seabirds are deposited along shores as feces and
crevices. stimulate the growth of the seagrass meadows in
260 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

Florida Bay up to 200 m away from bird colony seabirds in Peru (Gootenberg, 1989). In 1998,
sites. Increased seagrass biomass is of economic owing to El Niño, many fish schools occurred at
and ecological importance as seagrass is the pri- depths beyond the reach of the seabirds, resulting
mary habitat for larval pink shrimp (Penaeus in the death of guano birds and other animals,
duorarum), an important prey item for fish, birds and loss of income from guano production (Line,
and humans (Powell et al., 1991). 1998).
In the more extreme environment of Marion
Island, between South Africa and Antarctica, the
excretory products of fish-feeding penguins repre- 4. Demand-derived ecosystem services generated
sent a net subsidy of nitrogen that is crucial for by fish populations
the growth of plants and peat formation (Linde-
boom, 1984). 4.1. Information ser6ices
Loss of fish from the diets of avian and terres-
trial animals can have direct socio-economic con- The features and functions of fish populations
sequences. In the Flathead River-Lake ecosystem, provide information to scientists and managers.
located in Glacier National Park in MT, USA, Due to the size and abundance of fish they are
the population of landlocked kokanee salmon (O. easily sampled research objects. Studies of the
nerka) collapsed after the invasion of opossum genetic make-up of fish, earstones (otoliths) and
shrimp (Mysis relicta) in 1981 (Spencer et al., other features inform about the life history of
1991). The opossum shrimp was deliberately in- fishes, including growth rate, age, taxonomic stud-
troduced in order to enhance the fish production. ies, identification of spawning locales, migrating
However, both opossum shrimp, and kokanee and colonization patterns, as well as environmen-
salmon living in freshwater are zooplanktivores, tal history (Campana and Neilson, 1985; Ryman
and Spencer et al. (1991) suggest that opossum and Utter, 1986). Such information is crucial for
shrimp, a voracious predator, outcompeted koka- management and our understanding of the rela-
nee salmon in only a few years. The angler har- tive importance of fish for resilient ecosystems,
vests, which up until 1985 were around 100 000 what ecosystem services they influence, how
kokanee per season, declined to zero in only a few overfished or naturally fluctuating populations
years. Long-range spatial negative effects on the should be dealt with, etc. In addition, genetic
terrestrial and avian fauna, which had depended material of fish populations serves as a source of
on salmon eggs, juveniles, adults, or carcasses, information for aquaculture production and for
resulted as well. The population of bald eagles in biological conservation programs.
the Flathead River-Lake area decreased by 96%
between 1981 and 1989, leading to a decrease of 4.1.1. Assessing ecosystem stress
visiting bird-watchers from 46 500 to 1000 persons Fish are sensitive to many stresses from para-
between 1983 and 1989. Other wildlife popula- sites or diseases to acidification. Further, due to
tions were also negatively affected by the loss of such factors as rapid growth rates, large body
kokanee: coyotes (Canis latrans), minks (Mustela sizes, habitat choice, and trophic level, many fish
6ison), river otters (Lutra canadensis), white-tailed have the capacity to bioaccumulate toxic sub-
deer (Odocoileus 6irginianus), and grizzly bears stances. It has also been suggested that response
(Ursus arctos) (Spencer et al., 1991). Hence, the by fish to stress at the population level can be
number of recreational activities in Glacier Na- identified before changes at the ecosystem level
tional Park dramatically decreased as a result of (Schindler, 1990). These features make many
altered ecosystem function and loss of ecosystem fishes suitable as early-warning signals of anthro-
services owing to loss of a fish population. pogenic stress on natural ecosystem dynamics, or
A more recent example of the importance of conversely, as indicators of ecosystem recovery
fish functioning as passive links is provided by the (Harris, 1995; Moyle and Moyle, 1995; Balk et al.,
economically valuable guano production by 1996), and of resilience (Carpenter and Cotting-
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 261

ham, 1997). In addition to the physical condition as goods depend on ecosystem services such as
of fish, fish species richness and composition, food web interactions involving fish and other
trophic composition, and abundance can be used organisms.
for monitoring human influence on water quality Cultural services generated by fish populations
(cf. Index of Biotic Integrity, e.g. Fore and Karr, are based on our preferences and refer to when
1994). For example, the distribution pattern of knowledge about the ecology of fish is used for
butterflyfishes (Chaetodontidae) in the reefs of producing renewable goods for human societies,
Hawaii and Sri Lanka has been suggested to for supplying aesthetic and recreational values,
function as indicator of disturbance caused by and for ameliorating human health.
human activities, since the abundance of but-
terflyfishes appears to be positively correlated 4.2.1. Food production
with the distribution of live coral reefs (Reese, In 1995, 139 million tonnes of fish were har-
1995; O8 hman et al., 1998). vested by capture fisheries and produced in aqua-
culture, and the average global food fish supply
4.1.2. Long-term en6ironmental recorders reached a record high level of 14 kg per capita
The present fish fauna is living witness to cli- (FAO, 1997). The production of fish in natural
matic changes in the past, a fact that gives us systems is strongly influenced by the intricate and
information about past climate. For example, the alternating prey-predator relationships between
distribution of Arctic char (Sal6elinus alpinus) in the target species and other species. Thus, ex-
Scandinavian lakes reveals a climatic pattern of a ploitation of one particular fish species ultimately
maximum water temperature of 16°C from the affects the existence of other organisms including
most recent glaciation period 10 000 years ago to other fishes. Conversely, overexploitation of fish
today (National Environmental Protection Board, can undermine the life-support systems of other
1989, p. 106). Also, the distribution of 7 000 – organisms in the same or adjacent ecosystems.
9 000-year-old bones of the warm-water living An illustrative example is provided by the
bream (Abramis brama) found in north of Sweden rapidly developing intensive shrimp farming local-
shows evidence of a climatic temperature maxi- ized in or adjacent to mangrove forests. In 1990,
mum here in the past (National Environmental approximately 25% of the world’s shrimp produc-
Protection Board, 1989, p. 106). tion came from shrimp farming (Rosenberry,
1996). Deforestation of mangroves for shrimp
4.2. Cultural ser6ices farming degrades essential habitats for both wild
shrimp and several stationary and migratory fish
Fish are generally valued for their qualities as populations in the adjacent coastal zone, resulting
goods, selected by human preferences, in the form in diminished fish and shrimp catches in capture
of food protein, fishmeal, fish oil, game fish, and fisheries (e.g. Martosubroto and Naamin, 1977;
for aquaculture production. Also, in the pharma- Sasekumar and Chong, 1987). Moreover, the
ceutical industry, certain substances from fish are coastal fishing communities are deprived of their
used in research and might become important in main basis for livelihood based on ecosystem ser-
the production of medicines. For example, a new vices provided by the mangrove ecosystem, lead-
water-soluble, broad-spectrum antibiotic class has ing to a marginalization of these communities
been found in stomach extracts from the dogfish (Primavera, 1991; Flaherty and Karnjanakesorn,
shark (Squalus acanthias) (Carté, 1996). Another 1995; Rönnbäck, this issue).
example of pharmaceutical production involving In contrast, integrated fish cultures exist in
fish concerns the extraction of the substance several countries with long traditions of using
tetrodotoxin—a potent neurotoxin — from puffer coastal resources (Mitsch and Jørgensen, 1989;
fish (Tetraodontidae), which is a valuable tool in Folke and Kautsky, 1992). Based on the princi-
the study of the human nerve cell (Higa, 1997). ples of ecological engineering, the self-organizing
However, because fish are part of ecosystems, fish ability of ecosystems is stimulated by profiting
262 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

from the natural functions of fish and other spe- term ecosystem effects of introductions of exotic
cies for an efficient and environmentally sound fish species, has been much disputed and is not yet
production of fish and other goods. Materials and clear (DeMelo et al., 1992).
energy in the system are managed so that little or
none go to waste, and goods can be harvested 4.2.3. Supplying aesthetic 6alues
without degrading the resource base or services on Fish in public aquaria, wild species in tropical
which the goods depend (Folke and Kautsky, reefs, in crowded streams during spawning, or in
1992). One of the most developed systems of lakes and along coasts, generate highly valued
ecological engineering is the integrated fish pond aesthetic services (Moyle and Moyle, 1995). In
culture in China which has existed for more than economic terms, the global aquarium industry is
3000 years (Jingsong and Honglu, 1989). These estimated to generate seven billion US dollars per
managed food webs consist of primary producers year (Moyle and Moyle, 1995). In Stockholm,
(plants and phytoplankton), various levels of con- Sweden, annual salmon and trout stockings in the
sumers including the harvested carp species, and city streams enhance the aesthetic qualities of the
decomposers, that recycle the nutrients. city, with the incentive of increasing tourism, as
Fish are also used as a management tool to the salmonids splash about and are relatively
enhance rice production in rice paddies. For ex- easily caught (Holmlund, 1996).
ample, in the Philippines several fish species are In the short term all of these aesthetic services
cultivated in rice fields (Halwart et al., 1996). The may not seem to depend on functioning ecosys-
exact functional role of fish in rice paddies how- tems. However, in the long-term many of these
ever is not yet fully understood. Halwart et al. services, such as stocking of hatchery-reared fish,
(1996) suggest that juvenile Nile tilapia (Ore- depend on evolving fish populations contributing
ochromis niloticus) and common carp (Cyprinus with their genetic diversity.
carpio) act as biological control agents by preying
on rice pests, such as damaging arthropods and 4.2.4. Impro6ing human health
snails. Besides the obvious improvement of human
health in terms of food protein, and medicine
4.2.2. Supplying recreational 6alues supply, fish are used in management to mitigate
During this century, sport fishing of wild and vector-borne diseases like schistosomiasis and
stocked game fishes in lakes, rivers, and along malaria. Mosquitofish (Gambusia spp.), for exam-
coasts has become one of the most popular recre- ple, feed on and control aquatic disease bearing
ational activities internationally (FAO, 1996). In- invertebrates and plants in tropical climates (Mar-
timate contact with nature while fishing is claimed chall and Maes, 1994; Moyle and Moyle, 1995).
to be one of the major incentives for sport fishing
(Schramm and Mudrak, 1994). The increasing
demand for game fish and suitable fishing- and 5. Discussion
swimming-areas, is in conflict with the decreasing
water quality owing to other human activities. Human societies benefit in numerous ways from
Biomanipulation (Shapiro and Wright, 1984; ecosystem services generated by fish populations.
Jørgensen and Jørgensen, 1989) is a management Fish are part of food chain dynamics, nutrient
tool used for improving water quality in nutrient- cycling, and ecosystem resilience. Their mobility
rich lakes, based on the idea of top-down food within the nested set of temporal and spatial
web control (Section 3.1), by removing fish or cycles of ecological systems enhances the func-
stocking piscivorous fish in order to suppress algal tional importance of fish as ecological memory in
blooms (Benndorf et al., 1988), or stocking her- the form of energy, nutrients, genetic reserves,
bivorous fish such as grass carp to suppress vege- and information. Fish also generate employment,
tation (Lodge et al., 1998). However, the actual function as a genetic library for possible future
role of fish in biomanipulation, as well as long use in medicine and aquaculture, stimulate human
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 263

interest in nature, and provide aesthetic and recre- However, fish stocking as compensation for de-
ational values. Certain ecosystem services gener- clining fish resources often creates artificial sys-
ated by fish populations are also used as tems dependent on continuous inputs of reared
management tools, for example, in enhancing rice fish. Further, regular stocking practices may mask
production (Tilapia, carp), mitigating diseases in natural discontinuities and irregularities, diluting
tropical zones (mosquitofish), mitigating algal feed-back signals between ecological and social
blooms (pike Esox lucius, pike-perch Lucioperca systems. Owing to natural dynamics, and the time
sandra), mitigating growth of lake vegetation lag between stocking and ecological effects, it is
(grass carp), and indicating ecosystem stress difficult to foresee long-term cascading effects of
(butterflyfishes). stocking practices in the ecosystem (Mooney et
However, increasing fishing pressure, pollution, al., 1995; Holling et al., 1998). In many cases,
habitat destruction, introduction of substituting stocking practices that seem to be successful in the
exotic species, and other stress factors continue to short term, have been found to cause dramatic
exert strong pressure on fish populations around changes in the long term, such as depletion of
the world (Malakoff, 1997). Capture fisheries ap- other economically valuable species, changes in
propriate a substantial part (8%) of the global nutrient balances, or biodiversity decline (Krueger
primary production in the sea, and require 24 – and May, 1991; Waples, 1991; McKaye et al.,
35% of upwelling and continental shelf produc- 1995).
tion (Vitousek et al., 1986; Pauly and Christensen, Thus, stocking practices often overlook the
1995). Further, the focus on high yields of a few functional roles of fishes as embedded in natural
species at the top of the food web by current ecosystems during their life cycle. The stocked
fishery management has moved fish communities individuals rarely compensate for the loss of all
towards a composition dominated by lower ecosystem services. One example is provided by
trophic levels and forced fisheries to fish lower the case of Baltic salmon (Salmo salar). Large-
down the food web, as in the case of the Baltic scale yearly stocking of reared Baltic salmon since
Sea fisheries (Hammer et al., 1993; Pauly and the 1950s to compensate for loss of natural
Christensen, 1995). The human-induced direct spawning due to hydro-power construction in
and indirect degradation of common fisheries re- Baltic rivers has resulted in an overwhelming
sources might cause impacts at the ecosystem (90%) dominance of hatchery-reared fish in the
level, jeopardizing the fundamental and demand- Baltic salmon population, seriously eroding the
derived ecosystem services generated by fish with genetic diversity. The reared salmon smolts have
consequences for biodiversity, and ecosystem re- further been shown to have a higher mortality
silience (Naeem et al., 1994; Perrings et al., 1995). rate than wild fish (Lindroth, 1984). Ecosystem
Loss of ecosystem services generated by fish popu- services generated by spawning adults and their
lations can have unexpected negative economic eggs, larvae and juveniles, and salmon carcasses,
consequences for human societies, as in the case have been largely lost, including inputs of marine-
of kokanee salmon in Glacier National Park derived organic matter and nutrients to streams
(Spencer et al., 1991). Gradual loss of resilience and adjacent terrestrial ecosystems, with unknown
due to human activities may be dramatically man- consequences.
ifested as a rapid change to an alternate stable The establishment of nature reserves is another
state, as exemplified in the Jamaican coral reef common tool used for conservation or restoration
system, documented by Hughes (1994). purposes. However, isolated reserves cannot re-
The search for substitutes for declining com- place the multitude of fish populations continu-
mercially important fish populations will most ously evolving and generating ecosystem services
likely increase next century. A common practice (Folke et al., 1996; Hughes et al., 1997). Naturally
during the 20th century has been to stock hatch- reproducing fish populations in geographically
ery-produced indigenous or exotic species into and temporally separated systems evolve under
lakes and open water systems (Welcomme, 1988). pressures of natural selection. Over large areas
264 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

and long time periods, population diversity in- Knowledge about ecosystem services also en-
creases, and thereby enables species better to con- hances our comprehension of the non-linear na-
tinuously adapt to changing environments ture of ecosystem behavior, as well as bridges
(Myers, 1997). Diversity of fish populations pro- between the temporal and spatial scales of man-
vides a buffer for uncertainty and secure resilience agement on the one hand and ecological systems
during ecosystem perturbation and safeguards the on the other (Hammer et al., 1993; Ludwig et al.,
future health of natural and social systems (Folke 1993; Holling and Sandersson, 1996). This may
et al., 1996; Dayton, 1998). Thus, allowing certain help us to organize the system of fish resource
populations to be degraded with the underlying governance including property rights regimes, de-
assumption that this is compensated for by con- cision-making processes, and collaboration be-
serving other populations in distant ecosystems, tween local and central levels, in identifying
or in nature reserves, fails to recognize that fish beneficiaries of ecosystem services, and in dealing
populations are uniquely adapted to a relatively with resource conflicts (Ostrom, 1990; Hanna,
narrow range of environmental conditions, and 1996; Miller et al., 1997; Ostrom, 1998).
are necessary for sustaining the function and re-
silience of particular lakes or coastal areas, and
ultimately the economy of local human Acknowledgements
communities.
Current western fishery institutions need to rec- We thank Karin Limburg for her constructive
ognize the interdependency between human soci- comments on an earlier draft of this paper. We
eties and fish populations and develop adaptive also thank Steve Carpenter and two anonymous
resource use patterns (UNEP, 1998). This is of reviewers for valuable comments that have im-
fundamental importance in light of the growing proved the quality of this paper. The work was
human population, rapid transformation of the funded by the Swedish Council for Forestry and
Earth’s ecosystems, and increasing fishing pres- Agricultural Research, SJFR, and the Swedish
sure. Management needs to be directed towards Council for Planning and Coordination of Re-
an multi-hierarchial, ecosystem-based approach search, FRN.
which links the actions of populations to ecosys-
tem properties, including functional diversity and
resilience (Vitousek, 1990; Schultze and Mooney, References
1994; Lawton and Jones, 1995; Carpenter and
Turner, 1998). Identifying and taking into ac- Acheson, J.M., Wilson, J.A., Stenbeck, R.S., 1998. Managing
count what ecosystem services the fish depend on chaotic fisheries. In: Berkes, F., Folke, C., Colding, J.
and what services the fish generate, we believe, is (Eds.), Linking Social and Ecological Systems. Cambridge
University Press, Cambridge, pp. 390 – 413.
one step in that direction. Such ecosystem-based
Anon., 1994. f-Fakta om fisk, fiske och fiskevård.
management approaches, relying on both scien- Ålutplanteringar. PM: no.10, Fiskeriverket. (in Swedish).
tific and traditional ecological knowledge, have Bagge, O., Thurow, F., Steffensen, E., Bay, J., 1994. The
been identified, and exemplified by a growing Baltic cod. Dana 10, 1 – 28.
body of case studies of sustainable local fisheries, Balk, L., Larsson, Å., Frölin, L., 1996. Baseline studies of
biomarkers in the feral female perch (Perca flu6iatilits) as
e.g. integrated fish cultures in China (Jingsong
tools in biological monitoring of anthropogenic substances.
and Honglu, 1989; Berkes, 1995; Berkes and Mar. Environ. Res. 42, 203 – 208.
Folke, 1998). Less emphasis is put on controlling Benndorf, J., Schultz, H., Benndorf, A., Unger, R., Penz, E.,
how much fish that can be harvested. Instead, Kneschke, H., Kossatz, K., Dumke, R., Hornig, U.,
regulations focus on when, where, and how to Kruspe, R., Reichel, S., 1988. Food-web manipulation by
enhancement of piscivorous fish stocks: long-term effects in
fish, taking into account the spatial and temporal
the hypertrophic bautzen reservoir. Limnologica 19, 97 –
life-supporting systems of fish (Johannes, 1981; 110.
Wilson et al., 1994; Hammer, 1995; Acheson et Berkes, F., 1995. Indigenous knowledge and resource manage-
al., 1998; Johannes, 1998). ment systems: a native Canadian case study from James
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 265

Bay. In: Hanna, S., Munasinghe, M. (Eds.), Property Cooper, D.J., Watson, A.J., Nightingale, P.D., 1996. Large
Rights in a Social and Ecological Context. Case Studies decrease in ocean-surface CO2 fugacity in response to in
and Design Applications. The Beijer International Institute situ iron fertilization. Nature 383, 511 – 513.
of Ecological Economics and The World Bank, Washing- Daily, G.C. (Ed.), 1997. Nature’s Services: Societal Depen-
ton, DC, pp. 99 – 109. dence on Natural Ecosystems. Island Press, Washington,
Berkes, F., Folke, C. (Eds.), 1998. Linking Social and Ecolog- DC, 392 pp.
ical Systems. Cambridge University Press, Cambridge, 459 Dayton, P.K., 1998. Reversal of the burden of proof in
pp. fisheries management. Science 279, 821 – 822.
Bilby, R.E., Fransen, B.R., Bisson, P.A., 1996. Incorporation Dayton, P.K., Thrush, S.F., Agardy, M.T., Hofman, R.J.,
of nitrogen and carbon from spawning coho salmon into 1995. Environmental effects of marine fishing. Aquat.
the trophic system of small streams: evidence from stable Cons. Mar. Fresh. Eco. 5, 205 – 232.
isotopes. Can. J. Fish. Aquat. Sci. 53, 164–173. Deegan, L.A., Peterson, B.J., Golden, H., McIvor, C.C.,
Bilby, R.E., Fransen, B.R., Bisson, P.A., Walter, J.K., 1998. Miller, M.C., 1997. Effects of fish density and river fertil-
Response of juvenile coho salmon (Oncorhynchus kisutch) ization on algal standing stocks, invertebrate communities,
and steelhead (Oncorhynchus mykiss) to the addition of and fish production in an Arctic river. Can. J. Fish. Aquat.
salmon carcasses to two streams in southwestern Washing- Sci. 54, 269 – 283.
ton, USA. Can. J. Fish. Aquat. Sci. 55, 1909–1918. DeMelo, R., France, R., McQueen, D.J., 1992. Biomanipula-
Botsford, L.W., Castilla, J.C., Peterson, C.H., 1997. The man- tion: hit or myth? Limnol. Oceanogr. 37, 192 – 207.
agement of fisheries and marine ecosystems. Science 277, DeVries, P., 1997. Riverine salmonid egg burial depths: review
509 – 515. of published data and implications for scour studies. Can.
Bray, R.N., Miller, A.C., Geesey, G.G., 1981. The fish connec- J. Fish. Aquat. Sci. 54, 1685 – 1698.
tion: a trophic link between planktonic and rocky reef Estes, J.A., Tinker, M.T., Williams, T.M., Doak, D.F., 1998.
communities. Science 214, 204–205. Killer whale predation on sea otters linking oceanic and
Brenchley, G.A., 1981. Disturbance and community structure: nearshore ecosystems. Science 282, 473 – 476.
an experimental study of bioturbation in marine soft-bot- FAO, 1996. Fisheries and Aquaculture in Europe: Situation
tom environments. J. Mar. Res. 39, 767–790. and Outlook in 1996. FAO, Fisheries Department, Rome.
Brooks, J.L., Dodson, S.I., 1965. Predation, body size and ONLINE. Available: http://www.fao.org/fi%2A/publ/
composition of plankton. Science 150, 28–35. circular/c911/c911%2D2.htm
Campana, S.E., Neilson, J.D., 1985. Microstructure of fish FAO, 1997. Review of the state of world fishery resources:
otoliths. Can. J. Fish. Aquat. Sci. 42, 1014–1032. marine fisheries. FAI Fisheries Circular No. 920 FIRM/
Carpenter, S.R., Cottingham, K.L., 1997. Resilience and C920. FAO, Fisheries Department, Rome. ONLINE.
restoration of lakes. Conservation Ecology, 1: ONLINE. Available: http://www.fao.org/WAICENT/FAOINFO/
Available: http://www.consecol.org. FISHERY/publ/circular/
Carpenter, S.R., Kitchell, J.F., 1993. Simulation models of the Field-Dodgson, M.S., 1987. The effect of salmon redd excava-
trophic cascade: predictions and evaluations. In: Carpen- tion on stream substrate and benthic community of two
ter, S.R., Kitchell, J.F. (Eds.), The Trophic Cascade in salmon spawning streams in Canterbury, New Zealand.
Lakes. Cambridge University Press, Cambridge, pp. 310– Hydrobiology 154, 3 – 11.
331. Flaherty, M., Karnjanakesorn, C., 1995. Marine shrimp aqua-
Carpenter, S.R., Turner, M., 1998. At last: a journal devoted culture and natural resource degradation in Thailand. En-
to ecosystem science. Ecosystems 1, 1–5. viron. Manage. 19, 27 – 37.
Carpenter, S.R., Kitchell, J.F., Hodgson, J.R., 1985. Cascad- Flecker, A.S., 1992. Fish trophic guilds and the structure of a
ing trophic interactions and lake productivity-fish preda- tropical stream: weak direct versus strong indirect effects.
tion and herbivory can regulate lake ecosystems. Ecology 73, 927 – 940.
BioScience 35 (1998), 634–639. Folke, C., Kautsky, N., 1992. Aquaculture with its environ-
Carpenter, S.R., Cottingham, K.L., Shindler, D.E., 1992. Bi- ment: prospects for sustainability. Ocean Coastal Manage.
otic feedbacks in lake phosphorous cycles. Tree 7, 332– 17, 5 – 24.
336. Folke, C., Holling, C.S., Perrings, C., 1996. Biological diver-
Carpenter, S.R., Chisholm, S.W., Krebs, C.J., Shindler, D.W., sity, ecosystems, and the human scale. Ecol. Appl. 6 (4),
Wright, R.F., 1995. Ecosystem experiments. Science 269, 1018 – 1024.
324 – 327. Fore, L.S., Karr, J.R., 1994. Statistical properties of an Index
Carté, B.K., 1996. Biomedical potential of marine natural of Biological Integrity used to evaluate water resources.
products. BioScience 46, 271–286. Can. J. Fish. Aquat. Sci. 51, 1077 – 1087.
Cederholm, C.J., 1989. Fate of coho salmon (Onchorhynchus Fuller, A., Cowell, B.C., 1985. Seasonal variation in benthic
kisutch), carcasses in spawning streams. Can. J. Fish. invertebrate recolonization of small-scale disturbance in a
Aquat. Sci. 46, 1347 –1355. subtropical Florida lake. Hydrobiology 124, 211 – 221.
Chapin III, F.S., Walker, B.H., Hobbs, R.J., Hooper, D.U., Ganning, B., Wulff, F., 1969. The effects of bird droppings on
Lawton, J.H., Sala, O.E., Tilman, D., 1997. Biotic control chemical and biological dynamics in brackish water rock-
over the functioning of ecosystems. Science 277, 500–504. pools. Oikos 20, 274 – 286.
266 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

Geesey, G.G., Alexander, G.V., Bray, R.N., Miller, A.C., Jeppesen, E., Lauridsen, T.L., Kairesalo, T., Perrow, M.R.,
1984. Fish fecal pellets are a source of minerals for inshore 1998a. Impact of submerged macrophytes on fish-
reef communities. Mar. Ecol. Prog. Ser. 15, 19–25. zooplankton interactions in lakes. In: Jeppesen, E.,
Gelwick, F.P., Stock, M.S., Matthews, W.J., 1997. Effects of Søndergaard, M., Søndergaard, M., Christoffersen, K.
fish, water depth, and predation risk on patch dynamics in (Eds.), The Structuring Role of Submerged Macrophytes in
a north-temperate river ecosystem. Oikos 80, 382–398. Lakes. Springer, New York, pp. 91 – 114.
Gootenberg, P. (Ed.), 1989. Between Silver and Guano: Com- Jeppesen, E., Søndergaard, M., Søndergaard, M., Christof-
mercial Policy and the State in Postindependence Peru. fersen, K., 1998b. The Structuring Role of Submerged
Princeton University Press, Princeton, NJ 234 pp. Macrophytes in Lakes. Springer, New York.
Grime, J.P., 1997. Biodiversity and ecosystem function: the Jingsong, Y., Honglu, Y., 1989. Integrated fish culture man-
debate deepens. Science 277, 1260–1261. agement in China. In: Mitsch, W.J., Jørgensen, S.E. (Eds.),
Halwart, M., Borlinghaus, M., Kaule, G., 1996. Activity pat- Ecological Engineering. Wiley, New York, pp. 375 – 408.
tern of fish in rice fields. Aquaculture 145, 159–170. Johannes, R.E., 1981. Worlds of the Lagoon: Fishing and
Hammer, M., 1995. Integrating ecological and socioeconomic Marine Life in Palau District of Micronesia. University of
feedbacks for sustainable fisheries. In: Hanna, S., Munas- California Press, Berkeley.
inghe, M. (Eds.), Property Rights in a Social and Ecologi- Johannes, R.E., 1998. The case for data-less marine resource
cal Context: Case Studies and Design Applications. The management: examples from tropical nearshore finfisheries.
Beijer International Institute of Ecological Economics and Tree 13, 243 – 246.
The World Bank, Washington, DC, pp. 141–149. Jørgensen, S.E., Jørgensen, L.A., 1989. Ecotechnological ap-
Hammer, M., Jansson, A.-M., Jansson, B.-O., 1993. Diversity proaches to the restoration of lakes. In: Mitsch, W.J.,
change and sustainability: implications for fisheries. Ambio Jørgensen, S.E. (Eds.), Ecological Engineering. Wilson,
22, 97 – 105. New York, pp. 357 – 374.
Hanna, S., 1996. Property rights, people, and the environment. Kairesalo, T., Seppälä, T., 1987. Phosphorus flux through a
In: Costanza, R., Segura, O., Martinez-Alier, J. (Eds.),
littoral ecosystem: the importance of cladoceran zooplank-
Getting Down to Earth. Practical Applications of Ecologi-
ton and young fish. Int. Rev. Ges. Hydrobiol. 72, 385 – 403.
cal Economics. Island Press, Washington, DC, pp. 381–
Kline, T.C., Goering, J.J., Mathisen, O.A., Poe, P.H., Parker,
393.
P.L., 1990. Recycling of elements transported upstream by
Harris, J.H., 1995. The use of fish in ecological assessment.
runs of Pacific salmon: I. d 15N and d 13C evidence in
Aust. J. Ecol. 20, 65 –80.
Sashin Creek, southeastern Alaska. Can. J. Fish. Aquat.
Higa, T., 1997. Coral reefs: mines of precious substances.
Sci. 47, 136 – 144.
Trop. Coasts 4, 3 – 4.
Kling, G.W., Kipphut, G.W., Miller, M.C., 1992. The flux of
Holling, C.S., 1973. Resilience and stability of ecological sys-
CO2 and CH4 from lakes and rivers in arctic Alaska.
tems. Ann. Rev. Ecol. Syst. 4, 1–23.
Hydrobiology 240, 23 – 36.
Holling, C.S., 1986. The resilience of terrestrial ecosystems:
Krokhin, E.M., 1975. Transport of nutrients by salmon mi-
local surprise and global change. In: Clark, W.C., Munn,
R.E. (Eds.), Sustainable Development of the Biosphere. grating from the sea into lakes. In: Hasler, A.D. (Ed.),
Cambridge University Press, Cambridge, pp. 292–317. Coupling of Land and Water Systems. Springer, New
Holling, C.S., Sandersson, S., 1996. Dynamics of (dis)harmony York, pp. 153 – 156.
in ecological and social systems. In: Hanna, S., Folke, C., Krueger, C.C., May, B., 1991. Ecological and genetic effects of
Mäler, K.-G. (Eds.), Rights to Nature: Ecological Eco- salmonid introductions in North America. Can. J. Fish.
nomic, Cultural and Political Principles of Institutions for Aquat. Sci. 48, 66 – 77.
the Environment. Island Press, Washington, DC, pp. 57– Larkin, G.A., Slaney, P.A., 1997. Implications of trends in
85. marine-derived nutrient influx to south coastal British Co-
Holling, C.S., Berkes, F., Folke, C., 1998. Science, sustainabil- lumbia salmonid production. Fisheries 22, 16 – 24.
ity, and resources management. In: Berkes, F., Folke, C. Lawton, J.H., Jones, C.G., 1995. Linking species and ecosys-
(Eds.), Linking Social and Ecological Systems. Cambridge tems: organisms as ecosystem engineers. In: Jones, C.G.,
University Press, Cambridge, pp. 342–362. Lawton, J.H. (Eds.), Linking Species and Ecosystems.
Holmlund, C., 1996. Fish stocking - consistent with sustain- Chapman and Hall, London, pp. 141 – 150.
able management of the natural resources? Incentives and Levin, S., 1992. The problem of pattern and scale in ecology.
effects of fish stocking in the County of Stockholm. M.Sc. Ecology 73, 1943 – 1967.
thesis, Department of Systems Ecology, Stockholm Uni- Lindeboom, H.J., 1984. The nitrogen pathway in a penguin
versity, Stockholm, 13, 108 pp. rookery. Ecology 65, 269 – 277.
Hughes, T.P., 1994. Catastrophes, phase shifts, and large-scale Lindroth, A. (Ed.), 1984. De svenska laxsmoltutsättningarna i
degradation of a Caribbean coral reef. Science 265, 1547– O8 stersjön (med komplement av P.-O. Larsson). Vattenfall
1551. Vällingby. (in Swedish).
Hughes, J.B., Daily, G.C., Ehrlich, P.R., 1997. Population Line, L., 1998. Peru, epicenter of El Niño, fears for its wildlife.
diversity: its extent and extinction. Science 278, 689–692. The New York Times, 19 May 1998.
 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268 267

Lodge, D.M., Cronin, G., Donk, E.V., Froelich, A.J., 1998. Odum, E.P. (Ed.), 1989. Ecology and Our Endangered Life-
Impact of herbivory on plant standing crop: comparisons Support Systems. Sinauer Associates, Massachusetts, 283
among biomes, between vascular and non-vascular plants, pp.
and among freshwater herbivore taxa. In: Jeppesen, E., O8 hman, M.C., Rajasuriya, A., Svensson, S. The use of but-
Søndergaard, M., Søndergaard, M., Christoffersen, K. terflyfishes (Chaetodoritidae) as bioindicators of habitat
(Eds.), The Structuring Role of Submerged Macrophytes in structure and human disturbance. Ambio (in press).
Lakes. Springer, New York, pp. 149–174. Ostrom, E. (Ed.), 1990. Governing the Commons. The Evolu-
Ludwig, D., Hilborn, R., Walters, C., 1993. Uncertainty, tion of Institutions for Collective Action. Cambridge Uni-
resource exploitation, and conservation: lessons from his- versity Press, Cambridge, 279 pp.
tory. Ecol. Appl. 3, 547–549. Ostrom, E., 1998. Coping with tragedies of the commons.
MacKenzie, B., St. John, M., Wieland, K., 1996. Eastern Workshop in Political Theory and Policy Analysis, Centre
Baltic cod: perspectives from existing data on processes for the Study of Institutions, Population, and Environmen-
affecting growth and survival of eggs and larvae. Mar. tal Change, Indiana University, Bloomington, 40 pp.
Ecol. Prog. Ser. 124, 265–281. Pauly, D., Christensen, V., 1995. Primary production required
Malakoff, D., 1997. Extinction on the high seas. Science 277, to sustain global fisheries. Nature 374, 255 – 257.
486 – 488. Perrings, C.A., Mäler, K.-G., Folke, C., Holling, C.S.,
Marchall, B., Maes, M., 1994. Small Water Bodies and their Jansson, B.-O., 1995. Biodiversity conservation and eco-
Fisheries in Southern Africa. FAO, Rome CIFA technical nomic development: the policy problem. In: Perrings, C.A.,
paper, no. 29. Mäler, K.-G., Folke, C., Holling, C.S., Jansson, B.-O.
Martosubroto, P., Naamin, N., 1977. Relationship between (Eds.), Biodiversity Conservation. Kluwer, Dordrecht, pp.
tidal forests (mangroves) and commercial food production 3 – 21.
in Indonesia. Mar. Res. Indonesia 18, 81–86. Polis, G.A., Anderson, W.B., Holt, R.D., 1997. Toward an
McKaye, K.R., Ryan, J.D., Stauffer, J.R., Perez, L.J.L., Vega, integration of landscape and food web ecology: the dynam-
G.I., Berghe, E.P. v. d., 1995. African tilapia in Lake ics of spatially subsidized food webs. Annu. Rev. Ecol.
Nicaragua. BioScience 45, 406–411. Syst. 28, 289 – 316.
Meyer, J.L., Schultz, E.T., Helfman, G.S., 1983. Fish schools: Post, D.M., Carpenter, S.R., Christensen, D.L., Cottingham,
an asset to corals. Science 220, 1047–1048. K.L., Kitchell, J.F., Schindler, D.E., Hodgson, J.R., 1997.
Miller, K., McNeely, J.A., Salim, E., Miranda, M., 1997. Seasonal effects of variable recruitment of a dominant
Decentralization and the Capacity to Manage Biodiversity. piscivore on pelagic food web structure. Limnol. Oceanogr.
World Resources Institute, Washington, DC 12 pp. 42, 722 – 729.
Mitsch, W.J., Jørgensen, S.E. (Eds.), 1989. Ecological Engi- Postel, S., Carpenter, S., 1997. Freshwater ecosystem services.
neering: an Introduction to Ecotechnology. Wiley, New In: Daily, G.C. (Ed.), Nature’s Services. Island Press,
York 472 pp. Washington, DC, pp. 195 – 214.
Montgomery, D.R., Buffington, J.M., Peterson, N.P., Schuett- Powell, G.V.N., Fourqurean, J.W., Kenworthy, W.J., Zieman,
Hames, D., Quinn, T.P., 1996. Stream-bed scour, egg J.C., 1991. Bird colonies cause seagrass enrichment in a
burial depths, and the influence of salmonid spawning on subtropical estuary: observational and experimental evi-
bed surface mobility and embryo survival. Can. J. Fish. dence. Estuar. Coast. Shelf Sci. 32, 567 – 579.
Aquat. Sci. 53, 1061 –1070. Primavera, J.H., 1991. Intensive prawn farming in the Philip-
Mooney, H.A., Lubchenco, J., Dirzo, R., Sala, O.E., 1995. pines: ecological, social and economic implications. Ambio
Biodiversity and ecosystem functioning—basic principles. 20, 28 – 33.
In: Heywood, V.H., Watson, R.T. (Eds.), Global Biodiver- Reese, E.S., 1995. The use of indicator species to detect change
sity Assessment. UNEP, Cambridge University Press, on coral reefs: butterfly fishes of the family Chaetodontidae
Cambridge, pp. 279 –325. as indicators for Indo-Pacific coral reefs. A coral reef
Moyle, P.B., Moyle, P.R., 1995. Endangered fishes and eco- symposium on practical, reliable, low cost monitoring
nomics: intergenerational obligations. Environ. Biol. Fish. methods for assessing the biota and habitat conditions of
43, 29 – 37. coral reefs. ONLINE. Available: http/www.epa.gov/
Myers, N., 1997. Mass extinction and evolution. Science 278, owowwtr1/oceans/coral/symposium.html
597 – 598. Rosenberry, B. (Ed.), 1996. World Shrimp Farming. Shrimp
Naeem, S., Thompson, L.J., Lawler, S.P., Lawton, J.H., News International, San Diego. An annual report pub-
Woodfin, R.M., 1994. Declining biodiversity can alter the lished by Shrimp News International, 164 pp.
performance of ecosystems. Nature 368, 734–737. Rudstam, L.G., Aneer, G., Hildén, M., 1994. Top-down con-
National Environmental Protection Board, 1989. Climate and trol in the pelagic Baltic ecosystem. Dana 10, 105 – 129.
the natural environment. Monitor 1989. Ingvar Bingman Ryman, N., Utter, F. (Eds.), 1986. Population Genetics and
National Environmental Protection Board INFORMS, Fishery Management. University of Washington Press,
SMHI106 Helsingborg. (in Swedish). Seattle, 420 pp.
Nelson, J.S. (Ed.), 1994. Fishes of the World. Wiley, New Sasekumar, A., Chong, V.C., 1987. Mangroves and prawns:
York. further perspectives. In: Sasekumar, A., Phang, S.M.,
268 C.M. Holmlund, M. Hammer / Ecological Economics 29 (1999) 253–268

Chong, E.L. (Eds.), Towards Conserving Malyasia’s Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M.,
Marine Heritage, pp. 10–15 Proceedings of the 10th An- Siemann, E., 1997. The influence of functional diversity
nual Seminar of the Malayasian Society of Marine and composition on ecosystem processes. Science 277,
Sciences. 1300 – 1302.
Schindler, D.W., 1990. Experimental perturbations of whole UNEP, 1998. Report of the workshop on the ecosystem
lakes as tests of hypotheses concerning ecosystem structure approach: convention on biological diversity, Lilongwe,
and function. Oikos 57, 25–41. Malawi, 26 – 28 January. UNEP/CBD/COP/4/INF.9.
Schindler, D.E., 1992. Nutrient regeneration by Sockeye Vanni, M.J., Layne, C.D., 1997. Nutrient recycling and her-
salmon (Oncorhynchus nerka) fry and subsequent effects on bivory as mechanisms in the ‘top-down’ effect of fish on
zooplankton and phytoplankton. Can. J. Fish. Aquat. Sci. algae in lakes. Ecology 78, 21 – 40.
49, 2498 – 2506. Vanni, M.J., Luecke, C., Kitchell, J.F., Allen, Y., Temte, J.,
Schindler, D.E., Carpenter, S.R., Cole, J.J., Kitchell, J.F., Magnusson, J.J., 1990. Effects on lower trophic levels of
Pace, M.L., 1997. Influence of food web structure on massive fish mortality. Nature 344, 333 – 335.
carbon exchange between lakes and the atmosphere. Sci- Vitousek, P.M., 1990. Biological invasions and ecosystem pro-
ence 277, 248 – 251. cesses: towards an integration of population biology and
Schramm, H.L.J., Mudrak, V.A., 1994. Use of sport fish ecosystem studies. Oikos 57, 7 – 13.
restoration funds for put-and-take trout stocking: benefi- Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H., Matson, P.A.,
cial uses of put-and-take trout stocking. Fisheries 19, 6–7. 1986. Human appropriation of the products of photosyn-
Schultze, E.-D., Mooney, H.A. (Eds.), 1994. Biodiversity and thesis. BioScience 36, 368 – 373.
Ecosystem Function. Springer, Berlin. Walker, B.H., 1993. Rangeland ecology: understanding and
Service, R.F., 1997. Microbiologists explore life’s rich, hidden managing change. Ambio 22, 80 – 87.
kingdoms. Science 275, 1740–1742. Waples, R.S., 1991. Genetic interactions between hatchery and
Shapiro, J., Wright, D.I., 1984. Lake restoration by biomanip- wild salmonids: lessons from the Pacific Northwest. Can. J.
ulation: Round Lake, Minnesota, the first two years. Fish. Aquat. Sci. 48, 124 – 133.
Freshw. Biol. 14, 371–383. Welcomme, R.L., 1988. International Introductions of Inland
Sheffer, M., 1990. Multiplicity of stable states in freshwater Aquatic Species. FAO, Rome FAO Fisheries Technical
systems. Hydrobiologica 200/201, 475–486. Paper, no. 294.
Sparholt, H., 1994. Fish species interactions in the Baltic Sea. Wilson, J.A., Acheson, J.M., Metcalfe, M., Kleban, P., 1994.
Dana 10, 131 – 162. Chaos, complexity and community management of
Spencer, C.N., McClelland, B.R., Stanford, J.A., 1991. Shrimp fisheries. Mar. Pol. 18, 291 – 305.
stocking, salmon collapse, and eagle displacement. Bio- Woodin, S.A., 1978. Refuges, disturbance, and community
Science 41, 14 – 21. structure: a marine soft-bottom example. Ecology 59, 274 –
Steneck, R.S., 1998. Human influences on coastal ecosystems: 284.
does overfishing create trophic cascades? Tree 13, 429–430. World Resources Institute, 1996. World Resources: a Guide to
Sutherland, J.P., 1974. Multiple stable points in natural com- the Global Environment: the Urban Environment 1996 –
munities. Am. Nat. 108, 859–873. 97. Oxford University Press, Oxford, 365 pp.