River food webs: an integrative approach to bottom‐up flow webs, top‐down impact webs, and trophic position

The majority of food web studies are based on connectivity, top‐down impacts, bottom‐up flows, or trophic position ( TP ), and ecologists have argued for decades which is best. Rarely have any two been considered simultaneously. The present study uses a procedure that integrates the last three approaches based on taxon‐specific secondary production and gut analyses. Ingestion flows are quantified to create a flow web and the same data are used to quantify TP for all taxa. An individual predator's impacts also are estimated using the ratio of its ingestion ( I ) of each prey to prey production ( P ) to create an I / P web. This procedure was applied to 41 invertebrate taxa inhabiting submerged woody habitat in a southeastern U.S. river. A complex flow web starting with five basal food resources had 462 flows >1 mg·m −2 ·yr −1 , providing far more information than a connectivity web. Total flows from basal resources to primary consumers/omnivores were dominated by allochthonous amorphous detritus and ranged from 1 to >50,000 mg·m −2 ·yr −1 . Most predator–prey flows were much lower (<50 mg·m −2 ·yr −1 ), but some were >1,000  mg·m −2 ·yr −1 . The I / P web showed that 83% of individual predator impacts were weak (<10%), whereas total predator impacts were often strong (e.g., 35% of prey sustained an impact >90%). Quantitative estimates of TP ranged from 2 to 3.7, contrasting sharply with seven integer‐based trophic levels based on longest feeding chain. Traditional omnivores ( TP  = 2.4–2.9) played an important role by consuming more prey and exerting higher impacts on primary consumers than strict predators ( TP ≥ 3). This study illustrates how simultaneous quantification of flow pathways, predator impacts, and TP together provide an integrated characterization of natural food webs.