8
The Eutrophication Problem in Temperate Lakes: Practical Aspects and Theoratical ...
1.3 Stability and Persistence of Grazer-Controlled
Systems
Community Structure and the HSS Hypothesis. While the stability of grazercontrolled plankton communities has direct practical consequences in
water resource management, such questions also have a deep relationship
to several central problems in ecological theory. The main ideas behind the
biomanipulation concept can be traced back to a theory by Hairston et al.
(1960), often referred to as the HSS hypothesis, proposing that populations
in simple linear food chains appear to be alternately controlled by competition or predation. In food chains with an even number of trophic levels,
herbivores would deplete plants and produce barren habitats, while an odd
number of trophic levels would release plants from herbivory and produce
green habitats where plants become limited by nutrients or other resources.
Oksanen et al. (1981) expanded on this view by a conceptualization
where the number of trophic levels is dependent on the richness of the
system, and where enrichment sufficient to introduce an additional top
predator level would imply a shift from predatory to competitive control,
and vice versa, in all the supporting trophic levels. In an alternate model of
trophic control, Menge and Sutherland (1976,1987) argued that the prevalence of omnivory in food webs should lead to increasing control by predation, and decreasing control by resource limitation, for populations at
lower trophic levels. The removal of a completely omnivorous top predator
would, in the model of Menge and Sutherland (1976, 1987), not be expected
to have the same dramatic effects on lower trophic levels as predicted by
the HSS model.
McQueen et al. (1986, 1989) challenged the HSS model from a more
empirical point of view by correlation analysis of regional and experimental data on total phosphorus and biomasses of algae, zooplankton and fish.
Borrowing from the theory of structured programming (e.g., Dijkstra
1976), they coined the phrases bottom-up and top-down control, denoting
populations that are controlled by either nutrient supply or predation.
From the observed pattern in regression slopes, they inferred a tendency
for gradually increasing top-down control and decreasing bottom-up control with increasing trophic level, instead of strict alternations between
competitive and predatory control on adjacent trophic levels.
The results of McQueen et al. (1986) initiated a heated debate for some
years with sharp frontiers between those supporting fish or phosphorus as
the major determinants of phytoplankton biomass in lakes. This argument
seems to be settling down now with most participants admitting that there
cannot be a single unequivocal answer to such a question, and acknowledging that in a network of interacting components, "effects" may propagate in one direction or another, but not all control comes from either the
top or the bottom. Sterner (1989) summarizes the "fish or phosphorus"
The Eutrophication Problem in Temperate Lakes: Practical Aspects and Theoratical ...
1.3 Stability and Persistence of Grazer-Controlled
Systems
Community Structure and the HSS Hypothesis. While the stability of grazercontrolled plankton communities has direct practical consequences in
water resource management, such questions also have a deep relationship
to several central problems in ecological theory. The main ideas behind the
biomanipulation concept can be traced back to a theory by Hairston et al.
(1960), often referred to as the HSS hypothesis, proposing that populations
in simple linear food chains appear to be alternately controlled by competition or predation. In food chains with an even number of trophic levels,
herbivores would deplete plants and produce barren habitats, while an odd
number of trophic levels would release plants from herbivory and produce
green habitats where plants become limited by nutrients or other resources.
Oksanen et al. (1981) expanded on this view by a conceptualization
where the number of trophic levels is dependent on the richness of the
system, and where enrichment sufficient to introduce an additional top
predator level would imply a shift from predatory to competitive control,
and vice versa, in all the supporting trophic levels. In an alternate model of
trophic control, Menge and Sutherland (1976,1987) argued that the prevalence of omnivory in food webs should lead to increasing control by predation, and decreasing control by resource limitation, for populations at
lower trophic levels. The removal of a completely omnivorous top predator
would, in the model of Menge and Sutherland (1976, 1987), not be expected
to have the same dramatic effects on lower trophic levels as predicted by
the HSS model.
McQueen et al. (1986, 1989) challenged the HSS model from a more
empirical point of view by correlation analysis of regional and experimental data on total phosphorus and biomasses of algae, zooplankton and fish.
Borrowing from the theory of structured programming (e.g., Dijkstra
1976), they coined the phrases bottom-up and top-down control, denoting
populations that are controlled by either nutrient supply or predation.
From the observed pattern in regression slopes, they inferred a tendency
for gradually increasing top-down control and decreasing bottom-up control with increasing trophic level, instead of strict alternations between
competitive and predatory control on adjacent trophic levels.
The results of McQueen et al. (1986) initiated a heated debate for some
years with sharp frontiers between those supporting fish or phosphorus as
the major determinants of phytoplankton biomass in lakes. This argument
seems to be settling down now with most participants admitting that there
cannot be a single unequivocal answer to such a question, and acknowledging that in a network of interacting components, "effects" may propagate in one direction or another, but not all control comes from either the
top or the bottom. Sterner (1989) summarizes the "fish or phosphorus"