Stream flow mediates biomass, associations, and nutrient cycling of dominant animal functional groups

Abstract

Animals can be important mediators of resource heterogeneity and fluxes. While the properties of ecosystems generated by animals often result from interactions among multiple taxonomic and functional groups and environmental factors, most studies reduce these processes by examining processes performed by a single animal group under relatively static environmental conditions. Thus, our understanding of how animal-mediated ecosystem processes vary with abiotic and biotic context is limited. I propose a conceptual framework and present empirical evidence for animal-mediated nutrient cycling that considers potential effects of spatially overlapping animal groups within dynamic ecosystems to address this issue. First, I evaluate this framework by testing if biogeochemical hotspots generated by stable aggregations of mussels attract fishes. I quantified how different fish assemblage biomass was distributed between mussel bed reaches and reaches without mussel under different hydrologic conditions. I compared fish and mussel biomass at mussel beds to test whether differences in animal biomass mediate their contributions to nutrient cycling through nitrogen (N) and phosphorous (P) excretion. Hydrology influenced fish biomass distribution relative to stable mussel beds, with fish aggregating on mussel beds during low flows. Mussel biomass was consistently 10-fold higher than fish biomass, resulting in large differences among mussel and fish assemblage excretion rates, regardless of hydrologic conditions. Second, I evaluated the potential for spatial overlap of fish and mussels at fine spatial scales by conducting a 12-week field experiment to test if fish distribution was influenced by the presence of subsidies associated with live mussels or biogenic habitat of shells. I used underwater video footage to quantify fish occurrences at 50 0.25-m² enclosures stocked with either live mussels (two-species assemblages), sham mussels (shells filled with sand), or sediment only. Contrary to my predictions, I found that live mussels did not increase trophic subsidies to fish during the experiment, which may be attributed to unusually high stream flows homogenizing treatment effects. I also found that mussel shells (live and sham) influenced the distribution of fishes within the experimental reach and may provide habitat for fishes at fine spatial scales. Third, I investigated the potential for extreme low flow events to alter stream animals’ nutrient contributions through shifts in species composition and biomass. I tested how biomass and nutrient cycling rates of an intermittent prairie stream community changed during a drought. I quantified the biomass and contributions to nutrient cycling for assemblages comprising fishes, crayfish, and tadpoles in 12 isolated pools during the harshest drought on record for Kings Creek, KS. I found that assemblage biomass declined with decreasing pool size and assemblage composition shifted toward species with more drought resistant traits. Assemblage N excretion rates declined as pool biomass was reduced by mortality, emigration, or metamorphosis. P excretion rates were reduced initially but increased as species with high P excretion rates maintained similar proportional biomass and non-native fish biomass increased, consequently reducing assemblage excretion N:P. I conclude that taxonomically and functionally diverse animal groups coexisting in dynamic ecosystems generate the potential for periodic overlap of animal groups through facilitation or abiotic forcing. These studies demonstrate the context dependency of processes performed by animals, but also illustrate the general role of stoichiometric traits, biomass and density of organisms, and ecosystem size in governing animal processes.