Feedback between bottom-up and top-down control of stream biofilm mediated through eutrophication effects on grazer growth

Abstract

Algal biofilms in streams are simultaneously controlled by light and nutrient availability (bottom-up control) and by grazing activity (top-down control). In addition to promoting algal growth, light and nutrients also determine the nutritional quality of algae for grazers. While short-term experiments have shown that grazers increase consumption rates of nutrient-poor algae due to compensatory feeding, nutrient limitation in the long run can constrain grazer growth and hence limit the strength of grazing activity. In this study, we tested the effects of light and phosphorus availability on grazer growth and thus on the long-term control of algal biomass. At the end of the experiment, algal biomass was significantly affected by light, phosphorus and grazing, but the interactive effects of the three factors significantly changed over time. At both high light and phosphorus supply, grazing did not initially reduce algal biomass, but the effect of grazing became stronger in the final three weeks of the experiment. Snail growth was enhanced by light, rather than phosphorus, suggesting that algal quantity rather than quality was the main limiting factor for grazer growth. Our results highlight the role of feedback effects and the importance of long-term experiments in the study of foodweb interactions.

Figure 1

Interactive effects of phosphorus, light and grazing on periphyton biomass over time. Graphs depict periphyton dry mass over time at the beginning and during the grazing phase, at high phosphorus (P+) and low phosphorus supply (P−), high light (L+) and low light availability (L−), with grazers (black bars, G+) and without grazers (grey bars, G−). Phosphorus and light significantly increased periphyton dry mass at the beginning of the grazing phase (two-way ANOVA, Table 1), while time significantly interacted with grazing and phosphorus × light to determine periphyton dry mass during the grazing phase (linear mixed effects model, Table 2).

Figure 2

Interactive effects of phosphorus, light and grazing on periphyton C:P stoichiometry. Graphs depict periphyton molar C:P ratio at the beginning (10 June, a) and end (7 August, b) of the grazing phase at high phosphorus (P+) and low phosphorus supply (P−), high light (L+) and low light availability (L−) and in the presence and absence of grazers (G+ and G−, respectively). Values are mean ± SD of n = 6 (a) and n = 3 (b) replicate flumes. Phosphorus addition significantly decreased C:P ratios both at the beginning (two-way ANOVA; F_1,20 = 457.01, p < 0.001) and at the end of the grazing phase (three-way ANOVA; F_1,16 = 104.64, p < 0.001), whereas light and grazing had no effect. Different letters in each panel indicate significant differences between treatments (Tukey’s HSD post-hoc test).

Figure 3

Interactive effects of phosphorus, light and grazing on periphyton taxonomic composition. Graphs depict periphyton relative abundance of diatoms and chlorophytes, measured as percentage contribution to total chlorophyll a at the beginning (10 June, a) and end (7 August, b) of the grazing phase at high phosphorus (P+) and low phosphorus supply (P−), high light (L+) and low light availability (L−) and in the presence and absence of grazers (G+ and G−, respectively). Values are mean ±SD of n = 6 (a) and n = 3 (b) replicate flumes. Diatom proportion was significantly increased by the interactive effects of P addition and light at the beginning of the grazing phase (two-way ANOVA; F_1,20 = 14.40, p < 0.001), whereas it was significantly decreased by the interactive effects of low P supply and grazing at the end of the grazing phase (three-way ANOVA; F_1,16 = 10.50, p = 0.005). Different letters in each panel indicate significant differences between treatments (Tukey’s HSD post-hoc test).

Figure 4

Interactive effects of phosphorus and light on snail growth. Graphs depict the shell length (a) and the soft body mass (b) of Ancylus fluviatilis and at the end of the grazing phase (7 August), at high phosphorus (P+) and low phosphorus supply (P−) and at high light (L+) and low light availability (L−). Values are mean ± SD of n = 3 replicate flumes. Shell length was significantly increased by high light (two-way ANOVA; F_1,8 = 52.69, p < 0.001), whereas P had no effect. Soft body mass was significantly increased by both high light (two-way ANOVA; F_1,8 = 31.51, p < 0.001) and low P supply (two-way ANOVA; F_1,8 = 20.63, p = 0.002). Different letters indicate significant differences between treatments (Tukey’s HSD post-hoc test).