Figure legends
Figure 1. Conceptual diagram of the river trophic network model. The red dashed arrows represent flux leaving the system (i.e., fish respiration), and the black arrows represent flux transferring from one trophic compartment to another through the food web. Plants, invertebrates, algae, and detritus were considered basal resources for the fish community, so they have no input flux and respiration being represented here. Trophic interaction through the food web was defined by the use of gut contents. Illustration credit: Margenny Barrios (Departamento de Ecología y Gestión Ambiental CURE, Universidad de la República, Uruguay).
Figure 2.Long-term trends in diversity and energy flux of fish trophic guilds. The mean (± s.e., n = 179) of rarefied species richness (a) , abundance(b) , and energy flux (c) of whole-community (black points) and individual trophic guilds (colored lines) along 17 years.(d) mean of changes in the energy flux of each single trophic guild from year 1 to 17 year. Error bars show 95% confidence intervals.(e) relative contributions of fish trophic guilds to the whole food web energy flux (n = 179).
Figure 3. Response of energy flux (whole-community and individual trophic guilds) to changes in species richness (n = 179).Relationship between species richness and community energy fluxes. Linear mixed effects models for (a) whole-communities, (b) top-carnivores, (c) mesocarnivore, (d) omnivore, and (e) detritivore. Black dashed lines denote overall model fits and coloured lines indicate different river sites (Nuevo Berbín, Fray Bentos, and Las cañas).
Figure 5. Direct and indirect effects of HFP, climate and water properties on the diversity and energy flux of fish trophic guilds.Direct and indirect pathways by which HFP, climate and water properties affect diversity and energy flux of fish trophic guilds. Specifically, the structural equation modelling was used to disentangle the direct and indirect diversity-mediated effects of HFP, climate, and water properties on energy flux of four fish trophic guilds, including (a-c) top-carnivores, (d-f) mesocarnivores, (g-i) omnivores, and (j-l) detritivores. Solid blue and red arrows represent significant positive and negative pathways, respectively (P  ≤ 0.05), while solid grey arrows non-significant pathways (P  ≥ 0.05). The thickness of the significant pathways (arrows) represents the magnitude of the standardized regression coefficient. Bar graphs illustrates the standardized effect size from SEMs (a, d, g, j) for both species’ richness and energy flux of (b-c) top-carnivores, (e-f) mesocarnivores, (h-i) omnivores, and (k-l) detritivores. Asterisk adjacent to bar represents the significance of the effects: *P  < 0.05, **P  < 0.01, ***P  < 0.001. HFP = human footprint, MMP = mean monthly precipitation and WD = water discharge. Indirect effects of drivers on energy flux are calculated by multiplying the path coefficient for the effect of drivers on species richness with the path coefficient for the effect of species richness on energy flux (see Table S5).