Current Research


Trophic positions of fishes in relation to hydrological connectivity gradients in the Upper Mississippi River

Katherine Roach, James Thorp and Mike Delong

Structurally complex rivers consist of the main channel and various slackwater habitats which differ in hydrological connectivity within the riverscape. Differences in current velocity and other factors alter abiotic and biotic conditions along lateral gradients from main channel to backwater habitats. Fish assemblages using habitat patches along lateral gradients also change in response to behavioral and life history adaptations and dynamic habitat conditions. This study focused on the importance of lateral connectivity gradients in structuring aquatic food webs in the Upper Mississippi River during spring and summer sampling periods. We used stable isotope analysis to determine patterns of trophic position and food chain length in four feeding guilds of fishes. Over 1000 tissue samples were taken in four lateral sites (main channel, secondary and tertiary channels, and backwaters) in each of two reaches of two river “pools” during June and August of 2004. Trophic position differed significantly among feeding guilds and across lateral connectivity gradients. Site along the lateral connectivity gradient affected mean trophic position, variability in trophic position, and food chain length in both seasons. Food chain length for the two dominant piscivore species (Micropterus salmoides and M. dolomieu) peaked at intermediate points on the connectivity gradient in both seasons. Temporal changes in current velocity occurring from the flood pulse seem important in structuring food webs along lateral connectivity gradients. Patches with intermediate connectivity may allow a greater diversity of competitors to feed on a variety of trophic levels.

A model of carbon sources supporting fishes from three temperate and three tropical floodplain rivers

Katherine Roach and Kirk Winemiller

Identification of the basal production sources that support the food web is essential for understanding how environmental changes will influence the biogeochemistry of floodplain rivers. The Riverine Productivity Model hypothesized that in floodplain rivers, the majority of consumers would be supported by algal carbon because it is more nutritious and labile than allochthonous carbon sources. Indeed, the majority of research utilizing stable isotope analysis indicates that C4 macrophytes are a relatively unimportant carbon source in many streams and floodplain rivers and that the majority of floodplain river food webs are supported by autochthonous carbon sources. However, most studies examining basal production sources have been conducted during the low-water period in rivers with low levels of suspended sediments. 

Other recent studies have found that after periods of higher flows, consumers can be supported primarily by terrestrial organic carbon sources (i.e., Zeug and Winemiller, 2008, Ecology 89: 1733-1743; Huryn et al., 2001, Freshwat Biol 46: 213-226). These studies suggest that terrestrial-based carbon sources support the food web when physical factors such as discharge, suspended sediment load, and nutrients are limiting the availability of algae. This study will examine how discharge, turbidity, and nutrients limit algal production and influence the proportion of terrestrial-based versus algal-based carbon sources supporting the food web during the low-flow and high-flow period in six floodplain rivers: the Brazos River, the Neches River, and the Guadalupe River in Texas, the Tambopata River in Peru, and the Rupununi River and Pirara River in Guyana.  These rivers are very different in terms suspended sediments, nutrients, and primary productivity, and thus are ideal systems for developing a model to understand the biogeochemistry of floodplain rivers. In each river, during both the low- and high-water period I will measure turbidity, nutrients (SRP, NH4, and NO3/NO2), and discharge, and will estimate net ecosystem production of the benthos and water column by using light and dark chambers to measure fluxes of dissolved oxygen. Finally, I will use stable isotope analysis to estimate the relative proportion of autochthonous versus allochthonous carbon sources supporting the fish food web. 

Back to Katie's main page