310
Daniel E. Schindler, Brian R. Herwig, and Stephen R. Carpenter
system experimentation (Northcote 1988). Unlike
species removal experiments, stocking can be
taxon-specific. Some knowledge of the natural history of the organism (e.g., spawning habitat, thermal requirements) should allow evaluation of
whether the experimental system is habitable. The
cheapest method to add new species is to introduce
a small spawning stock and let nature take its
course. This is especially effective when stocking
piscivorous fishes into systems that lack other large
predators (e.g., Carpenter and Kitchell 1993a; Brabrand and Faafeng 1993). In ecosystems with public access, cooperation with regulatory agencies to
restrict exploitation may be needed to ensure successful introductions of game fishes (Kitchell 1992;
Benndorf et al. 1988). Intensive stocking of juveniles can also establish the target species at the desired population density. This method can be very
expensive because the cost of fish increases substantially with age (Madenjian et al. 1991). Manipulations to alter food web structure in systems dominated by omnivorous fishes with high reproductive
capacity (e.g., carp, Cyprinus carpio; gizzard shad,
Dorosoma cepedianum; roach, Rutilis rutilis), may
require a combination of piscivore stocking and
physical removal of resident species to alter community composition (e.g., Shapiro and Wright
1984; Kasprzak et al. 1993; Meijer et al. 1994;
Hansson et al. 1998). Prey fish introductions to
lakes with large predator popUlations are unlikely
to be successful without removal of some of the
predators.
Detailed descriptions of fish stocking techniques
are found in Nielsen and Johnson (1983). General
rules of thumb for minimizing stress to stocked
fishes are: (1) stock when water temperature is cold,
(2) prevent thermal shock by minimizing temperature differences between collection site, holding
water, and stocking site, (3) keep holding water
well aerated (> 5 mg liter - 1 02), (4) minimize handling stress during capture and holding, and
(5) treat fish with a general external antibiotic or
parasiticidal solution prior to stocking.
Ecosystem-scale stocking of invertebrates has
not been done as often as fish stocking. Notable
exceptions to this are the stocking of Mysis relicta
throughout western North America and Scandinavia (Lasenby et al. 1986), amphipods (Hill et al.
1990) and calanoid copepods in alpine lakes
(McNaught et al. 1999). In these experiments, relatively small numbers of reproductively mature individuals were collected with trawls or large nets,
transported in cold water, and stocked into experimental ecosystems.
Inadvertent introductions of species to aquatic
systems also represent potential learning opportunities at the ecosystem scale. For example, the rainbow smelt (Osmerus mordax) (Hrabik et al. 1998)
and the rusty crayfish (Orconectes rusticus) (Capelli 1982) have invaded and greatly altered ecosystem structure in lakes throughout the upper Midwest in the United States. When coupled with
small-scale experiments or process studies, invasions often offer insight into the mechanisms by
which exotics alter ecosystem structure and function. However, studies of inadvertent introductions
are not as powerful as direct ecosystem manipulations because species may be present for several
years before being detected, and preinvasion and
reference data may not be available for use in statistical analyses of the effects of invasion. Studies
of inadvertent introductions and the effects of other
human activities are most valuable when treated as
case studies rather than as controlled experiments.
Case studies at the ecosystem scale represent important opportunities for learning in ecosystem
ecology and conservation (Schrader-Frechette and
McCoy 1993), but are not an alternative to direct
ecosystem manipulation.
Habitat Manipulations
Macrophyte Restoration and Removal
in Lakes
Macrophyte manipulations have been an important
arena for ecosystem experiments in systems where
exotic plants have proliferated and often displaced
native species, and others where native plants have
been lost due to eutrophication and human development. Macrophyte manipulations are often performed to determine their effects on other processes
such as eutrophication (Moss et al. 1996) and fish
production (Carpenter et al. 1997). In systems with
extensive macrophytes, plant removal is generally
accomplished through water-level drawdown,
dredging, mechanical cutting, raking and derooting,
or application of herbicides (Cooke et al. 1993).
Daniel E. Schindler, Brian R. Herwig, and Stephen R. Carpenter
system experimentation (Northcote 1988). Unlike
species removal experiments, stocking can be
taxon-specific. Some knowledge of the natural history of the organism (e.g., spawning habitat, thermal requirements) should allow evaluation of
whether the experimental system is habitable. The
cheapest method to add new species is to introduce
a small spawning stock and let nature take its
course. This is especially effective when stocking
piscivorous fishes into systems that lack other large
predators (e.g., Carpenter and Kitchell 1993a; Brabrand and Faafeng 1993). In ecosystems with public access, cooperation with regulatory agencies to
restrict exploitation may be needed to ensure successful introductions of game fishes (Kitchell 1992;
Benndorf et al. 1988). Intensive stocking of juveniles can also establish the target species at the desired population density. This method can be very
expensive because the cost of fish increases substantially with age (Madenjian et al. 1991). Manipulations to alter food web structure in systems dominated by omnivorous fishes with high reproductive
capacity (e.g., carp, Cyprinus carpio; gizzard shad,
Dorosoma cepedianum; roach, Rutilis rutilis), may
require a combination of piscivore stocking and
physical removal of resident species to alter community composition (e.g., Shapiro and Wright
1984; Kasprzak et al. 1993; Meijer et al. 1994;
Hansson et al. 1998). Prey fish introductions to
lakes with large predator popUlations are unlikely
to be successful without removal of some of the
predators.
Detailed descriptions of fish stocking techniques
are found in Nielsen and Johnson (1983). General
rules of thumb for minimizing stress to stocked
fishes are: (1) stock when water temperature is cold,
(2) prevent thermal shock by minimizing temperature differences between collection site, holding
water, and stocking site, (3) keep holding water
well aerated (> 5 mg liter - 1 02), (4) minimize handling stress during capture and holding, and
(5) treat fish with a general external antibiotic or
parasiticidal solution prior to stocking.
Ecosystem-scale stocking of invertebrates has
not been done as often as fish stocking. Notable
exceptions to this are the stocking of Mysis relicta
throughout western North America and Scandinavia (Lasenby et al. 1986), amphipods (Hill et al.
1990) and calanoid copepods in alpine lakes
(McNaught et al. 1999). In these experiments, relatively small numbers of reproductively mature individuals were collected with trawls or large nets,
transported in cold water, and stocked into experimental ecosystems.
Inadvertent introductions of species to aquatic
systems also represent potential learning opportunities at the ecosystem scale. For example, the rainbow smelt (Osmerus mordax) (Hrabik et al. 1998)
and the rusty crayfish (Orconectes rusticus) (Capelli 1982) have invaded and greatly altered ecosystem structure in lakes throughout the upper Midwest in the United States. When coupled with
small-scale experiments or process studies, invasions often offer insight into the mechanisms by
which exotics alter ecosystem structure and function. However, studies of inadvertent introductions
are not as powerful as direct ecosystem manipulations because species may be present for several
years before being detected, and preinvasion and
reference data may not be available for use in statistical analyses of the effects of invasion. Studies
of inadvertent introductions and the effects of other
human activities are most valuable when treated as
case studies rather than as controlled experiments.
Case studies at the ecosystem scale represent important opportunities for learning in ecosystem
ecology and conservation (Schrader-Frechette and
McCoy 1993), but are not an alternative to direct
ecosystem manipulation.
Habitat Manipulations
Macrophyte Restoration and Removal
in Lakes
Macrophyte manipulations have been an important
arena for ecosystem experiments in systems where
exotic plants have proliferated and often displaced
native species, and others where native plants have
been lost due to eutrophication and human development. Macrophyte manipulations are often performed to determine their effects on other processes
such as eutrophication (Moss et al. 1996) and fish
production (Carpenter et al. 1997). In systems with
extensive macrophytes, plant removal is generally
accomplished through water-level drawdown,
dredging, mechanical cutting, raking and derooting,
or application of herbicides (Cooke et al. 1993).