Experimental evidence for enhanced top-down control of freshwater macrophytes with nutrient enrichment

Abstract

The abundance of primary producers is controlled by bottom-up and top-down forces. Despite the fact that there is consensus that the abundance of freshwater macrophytes is strongly influenced by the availability of resources for plant growth, the importance of top-down control by vertebrate consumers is debated, because field studies yield contrasting results. We hypothesized that these bottom-up and top-down forces may interact, and that consumer impact on macrophyte abundance depends on the nutrient status of the water body. To test this hypothesis, experimental ponds with submerged vegetation containing a mixture of species were subjected to a fertilization treatment and we introduced consumers (mallard ducks, for 8 days) on half of the ponds in a full factorial design. Over the whole 66-day experiment fertilized ponds became dominated by Elodea nuttallii and ponds without extra nutrients by Chara globularis. Nutrient addition significantly increased plant N and P concentrations. There was a strong interactive effect of duck presence and pond nutrient status: macrophyte biomass was reduced (by 50%) after the presence of the ducks on fertilized ponds, but not in the unfertilized ponds. We conclude that nutrient availability interacts with top-down control of submerged vegetation. This may be explained by higher plant palatability at higher nutrient levels, either by a higher plant nutrient concentration or by a shift towards dominance of more palatable plant species, resulting in higher consumer pressure. Including nutrient availability may offer a framework to explain part of the contrasting field observations of consumer control of macrophyte abundance.

Fig. 1

Macrophyte biomass in a unfertilized and b fertilized ponds before the ducks were introduced (15 June), immediately after the ducks had been present (11 July) and 6 weeks after ducks had been present (20 August). Ducks stayed on the ponds from 2 to 11 July (grey bar). Asterisk indicates significantly different biomasses between ponds with ducks and ponds without ducks in unfertilized and fertilized treatments (P < 0.05); NS not significant. Data are means + SE (n = 5). Results of the repeated-measures ANOVA and paired t-tests are presented in Tables 1 and 2

Fig. 2a–d

Plant nutrient concentrations in experimental ponds under different nutrient treatments and duck presence in Chara and Elodea plants. Data are means + SE (n = 5) and represent the values on 11 July, immediately after the ducks were removed from the ponds. Different letters indicate significantly different plant nutrient concentrations between nutrient treatments and plant species. Differences among duck treatments were not significant. For N concentrations there was a significant interaction between nutrient treatment and plant species, for the P concentrations only the main effects of nutrient treatment (indicated by asterisks) and plant species (indicated by capital letters) were significant. See Table 3 and “Results” for results of the statistical analyses. ***P < 0.001

Fig. 3a–d

Nutrient and light availability in the experimental ponds. Data were collected before (2 July), immediately after (11 July) and 6 weeks (20 August) after the presence of ducks on the ponds. a NO₃, b NH₄, c PO₄ and d light availability at the bottom of the ponds relative to ambient light. Data are means + SE (n = 5). Data were tested with repeated-measures ANOVAs; see Table 1 for results. Different letters indicate statistically different treatments for each parameter tested per date (two-way ANOVA, followed by post hoc Tukey test if a significant interaction between nutrient and duck treatment was present; P < 0.016, as a result of Bonferroni correction for three dates tested per parameter). Capital letters indicate a significant main effect of nutrient treatment; small letters indicate significant differences among nutrient and duck treatments as a result of a significant interaction between nutrient and duck treatments. Dashed line Unfertilized, solid line fertilized, open symbols no ducks, filled symbols with ducks

Fig. 4

a Epiphyton biomass on Elodea (mg dry weight mg⁻¹ Elodea dry weight) at the end of July (n = 5). Different letters indicate statistically different amounts of epiphyton on Elodea (Kruskall-Wallis test) across the nutrient and duck treatments. b Snail presence expressed as the number of floating Lymnea snails on the water surface in each pond at the end of July. Different letters indicate statistically different numbers of snails between the nutrient treatments; there were no differences between the duck treatments (two-way ANOVA). See “Results” for statistical tests