Inferring Food Web Structure to Identify Seasonal and Longitudinal Patterns in Ogeechee River Invertebrate Communities

Understanding how the structure and function of aquatic communities vary across space and time is essential for proper management of freshwater ecosystems. Current management relies on rapid biomonitoring using metrics of community structure, but metrics that incorporate ecosystem processes and functions are only just beginning to see use in assessment and management. Food webs inferred from known species interactions have been proposed as a method of incorporating function into bioassessment without expending extra effort or sacrificing the cost-effectiveness of current monitoring schemes. To apply food webs in biomonitoring, it is first necessary to understand how communities and food webs vary across temporal and longitudinal gradients so that changes can be interpreted correctly. I conducted a study to establish such a baseline for the Ogeechee River, a large coastal plain blackwater river in the southeastern United States. Samples of invertebrate assemblages were taken quarterly for three years from 6 sites along the main course of the Ogeechee River. I then investigated the spatial and seasonal structure of communities and food webs. Food webs were inferred using a database of invertebrate diets compiled from the literature. Communities varied across sites and seasons, with lower diversity and more lentic and estuarine taxa at downstream sites, and lower diversity during the flood pulse in the winter and spring due to high numbers of certain dominant taxa. Food web structure changed little over time and space, with some diversity dependent increases in links, linkage density and connectance at upstream sites, and some diversity independent decreases in prey to predator ratios during the winter. Results suggest that community structure was both more informative and reliable than food web structure for detecting changes in river communities. Better documentation of species interactions in the literature could improve inferred food webs so that they could work as a supplemental tool for interpreting the results of biomonitoring and for making management decisions