Top-down, bottom-up, or side to side? Within-trophic-level interactions modify trophic dynamics of a salt marsh herbivore

Many factors can influence the top-down and bottom-up dynamics of phytophagous insects. Although interactions between herbivore species have been frequently shown to be ecologically important, the effects of such horizontal trophic interactions on the relative roles of top-down and bottom-Lip forces have gone largely unstudied. In this paper we report on the results of a factorial field experiment in which we examined the effects of within-trophic-level interactions on the top-down and bottom-up dynamics of a salt marsh planthopper. We manipulated the bottom-up effects of plant quality by increasing soil salinity, and manipulated top-down effects by decreasing the intensity of parasitoid attack with yellow sticky traps that removed hymenopteran parasitoids. We applied these treatments to plots in two patches of the host plant, one with low densities of lepidopteran stem borer larvae, and one with high densities of stem borers. We maintained the treatments and monitored planthopper density for ten months, from March through December 1999. Increased salinity significantly increased planthopper density within one month of the first application of salt. The rapid response of the planthopper to salt treatments suggested a chemical mechanism, perhaps mobilization of bound nitrogen. Yellow sticky traps, although significantly reducing parasitism of planthopper eggs, had little impact on hopper (density. The density of lepidopteran stem borers, however, had an even greater impact on planthopper density than did salt treatments, with high stem borer plots Supporting Much lower densities of hoppers. Stem borer density also reduced the response of the planthopper to other treatments, especially salt supplementation. The results of this study show that the impact of within-trophic-level interactions can significantly chan.-c herbivore trophic dynamics and can be even more important than either top-down or bottom-up effects in determining herbivore density.