Landscape-scale hydrologic characteristics differentiate patterns of carbon flow in large-river food webs

Abstract

Efforts to conserve, restore, or otherwise manage large rivers and the services they provide are hindered by limited understanding of the functional dynamics of these systems. This shortcoming is especially evident with regard to trophic structure and energy flow. We used natural abundances of carbon and nitrogen isotopes to examine patterns of material flow in ten large-river food webs characterized by different landscape-scale hydrologic characteristics (low-gradient river, high-gradient river, river stretches downstream of reservoirs, and reservoirs), and tested predictions from three ecosystem concepts commonly applied to large-rivers: The River Continuum Concept, The Flood Pulse Concept and the Riverine Productivity Model. Carbon derived from aquatic C3 plants and phytoplankton were the dominant energy sources supporting secondary consumers across the ten largeriver food webs examined, but relative contributions differed significantly among landscape types. For low-gradient river food webs, aquatic C3 plants were the principal carbon source, contributing as much as 80% of carbon assimilated by top consumers, with phytoplankton secondarily important. The estimated relative importance of phytoplankton was greatest for food webs of reservoirs and river stretches downriver from impoundments, although aquatic C3 plants contributed similar amounts in both landscape types. Highest 99th percentile source contribution estimates for C4 plants and filamentous algae (both approximately 40%) were observed for high-gradient river food webs. Our results for low-gradient rivers supported predictions of the Flood Pulse Concept, whereas results for the three other landscape types supported the Riverine Productivity Model to varying degrees. Incorporation of landscape-scale hydrologic or geomorphic characteristics, such as river slope or floodplain width, may promote integration of fluvial ecosystem concepts. Expanding these models to include hydrologically impacted landscapes should lead to a more holistic understanding of ecosystem processes in large-river systems.