LIMNOLOGY AND July 1999 Number 5 OCEANOGRAPHY Volume 44 L,IIIIIO/ Oc~o~~~~g~ , 44(5). 19%. 11x7-1197 C 1999. b) the 4mrl1c.in Soc~cty oi Llmnology ,ind Ocranogt,~phy. inc Trophic interactions in open systems: Effects of predators and nutrients on stream food chains Graham E. Forrester Department of Organismic Biology, Ecology and Evolution, University of California, Los Angeles, California 90095-1606 Tom L. Dudley Department of Integrative Biology, University of California, Berkeley, California 94720-3140 Nancy B. Grirnrn Department of Zoology, Arizona State University, Tempe, Arizona 85287-1501 Abstract Theory and empirical work on food chains has focused primarily on closed, equilibrial environments. We tested the combined effects of secondary consumers (fish) and limiting nutrients (nitrogen) on intermediate trophic levels in an open stream environment, where flow redistributes organisms and materials among patches of streambed habitat. Fish reduced the biomass of the dominant herbivore (baetid mayflies) within habitat patches both by direct predation and by causing increased emigration from the patches. The resulting decrease in herbivory caused an increase in the growth and biomass of primary producers (algae) in areas containing fish. Independent of the effect of fish, algal growth and biomass was increased by augmenting the nutrient supply to patches. Nutrient-enriched areas (with high algal biomass) also supported greater populations of herbivores because they either grew faster in these areas or emigrated less frequently from them. Controlling influences on trophic structure came from both the top and the bottom of the food chain, and these influences were not conditional upon one another. Trophic structure in this system was determined by a mix of behavioral and trophic interactions between the major taxa, most of which were specific to open systems and not predicted by conventional theory. Ecologists have focused much attention on predicting the al. 1987; McQueen et al. 1989; Power 1992; Menge 1992; relative biomass of organisms at different trophic levels in Strong 1992; Polis and Winemiller 1996). Although there is systems where food webs can be represented as simple general agreement that both top-down and bottom-up forces chains (McNaughton et al. 1989; Brett and Goldman 1996). are likely to operate in most ecosystems (Carpenter et al. A central, and controversial, issue in this literature is defin- 1987; McQueen et al. 1989; Power 1992; Menge 1992; ing the relative effects of control from the top of the food Strong 1992; Polis and Winemiller 1996; Hunter et al. 1997), chain by predators, and control from the bottom by nutrients, surprisingly few studies have tested the simultaneous influ- or other resources limiting primary production (Carpenter et ences of predators and nutrients so that the nature of their effects can be compared directly. Of the tests that have been conducted, most were done in the pelagic zone of lakes Acknowledgments (Brett and Goldman 1997) and very few have been done in Many thanks to Rosa Gomez, Sam Neil, Linda O'Bryan, Kevin other habitats (Hartvigsen et al. 1995; Osenberg and Mittel- Petrone, Elizabeth kchards, Laura Ross, David Simonelle, Claudia bach 1996; Stiling and Rossi 1997). Torres-Navarro, Amy Weibel, and Jill Welter for help with field Pelagic communities in lakes are relatively self-contained work and analyses. G.E thanks Stuart Fisher and the members of and can reasonably be considered as closed equilibrial sys- the Arizona State University stream lab for their hospitality while tems at the time scales over which most food-web studies this research was being conducted. The manuscript benefited from insightful comments by two anonymous reviewers. We are grateful are performed (Brett and Goldman 1997). Results of empir- for financial support for this research in the form of grants from the ical food-chain manipulations can thus be compared to the National Science Foundation Ecology Program (BSR-9016445 to predictions of conventional food-chain theory that forecasts G.E and Peter Sale) and Ecosystem Studies Program (BSR-9008114 outcomes only for closed systems that have reached a stable to Stuart Fisher, T.D., and N.G.). equilibrium (Hairston et al. 1960; Fretwell 1977; Oksanen