Predation and Food Webs - Freshwater Ecology: Concepts and Environmental Applications - page 407

Little is known about the trophic cascade in wetlands, though the food
webs should not differ drastically in structure from other benthic food
webs; therefore, there is no reason to believe that the interplay between
top-down and bottom-up controls is not important. For example, preda-
ceous hydrophilid beetle larvae can control chironomid midge larvae in a
seasonal wetland. The midge larvae graze periphyton. When predation was
not present, the midge larvae outstripped their food supply (Batzer and
Resh, 1991).
As with wetlands, the trophic cascade in groundwaters is poorly docu-
mented. Grazer control of bacterial production has been demonstrated in sev-
eral groundwater habitats. The isopod Caecidotea tridentata stimulated bac-
terial production in a limestone aquifer in which carbon additions did not
(Edler and Dodds, 1996). A case of bottom-up control was demonstrated in
which organic contamination of an aquifer stimulated bacterial and proto-
zoan production relative to nearby uncontaminated groundwater (Madsen
et al., 1991). As more groundwater food webs are described, numerous cases
of top-down and bottom-up control are likely to be documented.
THEORETICAL COMMUNITY ECOLOGY AND AQUATIC FOOD WEBS
Lake food webs are well characterized and have been analyzed by the-
oretical community ecologists in attempts to describe general patterns
(Pimm, 1982). General questions that have been asked include the follow-
ing: What limits the length of food chains? Are complex systems more or
less stable? How interconnected are large food webs? Does aggregation of
species into trophic groups alter the results obtained from analyses of food
webs? and How variable are food webs over space and time?
It is not clear what limits the lengths of aquatic food chains. The length
of the food chain is of particular importance when determining biomagni-
fication of lipid-soluble pollutants. General analyses suggest that the amount
of primary production at the base of the food chain is not a good indica-
tor of food chain length (Briand and Cohen, 1990). This is not necessarily
consistent with the idea that energy is lost at each tropic level. However,
at least one study suggests that trophic transfer of energy in the Okefeno-
kee swamp may be very efficient (Patten, 1993), so our concepts of food
web efficiency and how it should link to the number of trophic levels may
need to accommodate situations with high energy transfer efficiencies.
Groundwaters in which the number of large animals is physically limited
by their ability to move through aquifers typically have short food chains.
Marine pelagic food chains are considerably longer than most pelagic
freshwater chains. It has been suggested that stream food webs tend to be
short but wider relative to those of lakes (i.e., with more species as primary
consumers). Large rivers and lakes tend to have similar food web structure
(Briand, 1985). Movement of energy in freshwater ecosystems will be dis-
cussed in more detail in Chapter 22.
It is also not clear if complex food webs are more or less stable. Early
ecologists viewed simple systems as less stable (MacArthur, 1955; Elton,
1958). Mathematical analyses of simple linear models of randomly assem-
bled “communities” then suggested decreased stability with more interact-