Fig. 1 Conceptual illustration of two hypotheses underlying trophic cascade strengths: the productivity hypothesis (a) and the energy transfer hypothesis (b) . R (horizontal green bars), H (horizontal brown bars), and C (horizontal red bars) represent the biomass of plants, herbivores, and predators, respectively. The upper arrows between adjacent trophic levels represent energy transfer, where Eher (green arrows) andEpre (brown bars) represent herbivore and predator efficiency, respectively, with the thickness of arrows representing the relative magnitude. The bottom dark-green upper arrows (P ) refer to the primary productivity. The dashed arrows indicate the indirect effects of the apex predator on plant biomass (i.e., trophic cascades) with their thickness representing the relative strength. The productivity hypothesis posits that a higher productivity can increase the biomass of apex predators and thereby enhance trophic cascades (Fig. 1a, Hypothesis I). The energy transfer hypothesis posits that a higher efficiency of energy transfer promotes the biomass accumulation of apex predators and subsequently strengthens trophic cascades (Fig. 1b, Hypothesis II). The dashed line with a smaller slope in Fig 1a (resp. Fig. 1b) indicates weakened production effects (resp. weakened effects of energy transfer) under a lower energy transfer efficiency (resp. lower production).
Fig. 2 Responses of standing biomass at different trophic levels to experimental manipulations: (a) predator E.modestus , (b, d) zooplankton herbivores, (c, e) phytoplankton. Different capital characters indicate significant differences (P< 0.05) among treatments varying in zooplankton community composition, and different lowercases indicate significant differences among treatments varying in nutrient supply. The bars and error bars show the averages and standard errors across three replicates, respectively.
Fig. 3 Responses of energy transfer efficiencies across trophic levels to experimental manipulations: (a-b) Herbivore efficiency, (b) predator efficiency, (d) FCE. The red arrows in the illustration schemes represent the focal energy transfer processes. Different capital characters indicate significant differences (P < 0.05) among treatments varying in zooplankton community composition, and different lowercases indicate significant differences among treatments varying in nutrient supply. The bars and error bars show the averages and standard errors across three replicates, respectively.
Fig. 4Responses of trophic cascade strength to experimental manipulations. Different capital characters indicate significant differences (P< 0.05) among treatments varying in zooplankton community composition, and different lowercases indicate significant differences among treatments varying in nutrient supply. The bars and error bars show the averages and standard errors across three replicates, respectively. The horizontal grey lines correspond to no cascading effect, i.e., no change in phytoplankton biomass following the removal of the top predator.
Fig. 5 Relationships between trophic cascade strength with primary production (a, c ) and predator efficiency (b, d ), derived from experimental data (a, b ) and theoretical models (c, d ). In a and b , each symbol represents one mesocosm, where circles (SD) represent zooplankton communities consisting of S. dorrii and D. brachyurum,triangles (DD) represent zooplankton communities consisting of D. magna and D. brachyurum ; and squares (SDD) represent zooplankton communities containing all the three species. The dashed black line in (a ) represents a non-significant relationship (P> 0.05), whereas the solid black line in (b ) indicates a significant relationship (P < 0.05). Shaded areas are the error bands and denote 95% confidence intervals. Inc and d , each point represents one simulated food webs with randomized assimilation efficiencies, i.e.,\(e_{H}\sim U\left[0.3,0.6\right]\) and\(e_{C}\sim U\left[0.5,0.9\right]\), where trophic interactions follow the type II functional response. Other parameters are specified in Methods. The horizontal grey lines correspond to no cascading effect.