Trophic transfer efficiency in the Lake Superior food web: assessing the impacts of non-native species

Ecosystem-based management relies on understanding how perturbations influence ecosystem structure and function (e.g., invasive species, exploitation, abiotic changes). However, data on unimpacted systems are scarce, therefore, we often rely on impacted systems to make inferences about 'natural states.' Among the Laurentian Great Lakes, Lake Superior provides a unique case study to address non-native species impacts because the food web is dominated by native species. Additionally, Lake Superior is both vertically (benthic versus pelagic) and horizontally (nearshore versus offshore) structured by depth, providing an opportunity to compare the function of these sub-food webs. We developed an updated Lake Superior EcoPath model using data from the 2005/2006 lake-wide multi-agency surveys covering multiple trophic levels. We then compared trophic transfer efficiency (TTE) to previously published EcoPath models. Finally, we compared ecosystem function of the 2005/2006 ecosystem to that with non-native linkages removed and compared native versus non-native species-specific approximations of TTE and trophic flow. Lake Superior was relatively efficient (TTE = 0.14) compared to systems reported in a global review (average TTE = 0.09) and the microbial loop was highly efficient (TTE > 0.20). Non-native species represented a very small proportion (<0.01%) of total biomass and were generally more efficient and had higher trophic flow compared to native species. Our results provide valuable insight into the importance of the microbial loop and represent a baseline estimate of non-native species impacts on Lake Superior. Finally, this work is a starting point for further model development to predict future changes in the Lake Superior ecosystem.

Seasonal changes in partial, reverse diel vertical migrations of cisco Coregonus artedi

The objectives of this study were to (1) document changes in partial, reverse diel vertical migrations (DVM) patterns of cisco Coregonus artedi in Ten Mile Lake, MN, U.S.A., throughout the year and (2) evaluate the mechanisms that may cause shifts in migration behaviour. Results indicated that C. artedi vertical distributions remained deep in the water column during the day and night of the spring and autumn, which was related to a low risk, low reward strategy. During summer, a partial migration occurred where a portion of the population remained deeper according to the low risk, low reward strategy, while the other portion performed a more extensive high risk, high reward reverse DVM. In winter, C. artedi did not migrate because there were only low risk, low reward conditions present at all depths. The extensive partial, reverse DVM during summer probably increased the growth potential of C. artedi, helping individuals survive in a lake with low zooplankton prey resources.

Foraging, bioenergetic and predation constraints on diel vertical migration: field observations and modelling of reverse migration by young-of-the-year herring Clupea harengus

Diel vertical migration (DVM) of young-of-the-year (YOY) herring Clupea harengus and one of their major predators, pikeperch Sander lucioperca, was examined using bottom-mounted hydroacoustics in Himmerfjärden, a brackish bay of the Baltic Sea, in summer. In contrast to previous studies on DVM of C. harengus aggregated across size and age classes, YOY C. harengus showed a reverse DVM trajectory, deeper at night and, on average, shallower during the day. This pattern was observed consistently on five acoustic sampling occasions in 3 years and was corroborated by two out of three trawl surveys. Large acoustic targets (target strength >-33 dB, probably piscivorous S. lucioperca >45 cm) showed a classic DVM trajectory, shallow at night and deeper during the day. Variability in YOY C. harengus vertical distribution peaked at dawn and dusk, and their vertical distribution at midday was distinctly bimodal. This reverse DVM pattern was consistent with bioenergetic model predictions for YOY C. harengus which have rapid gut evacuation rates and do not feed at night. Reverse DVM also resulted in low spatial overlap with predators.

A simulation of food-web interactions leading to rainbow smelt Osmerus mordax dominance in Sparkling Lake, Wisconsin

A process-based simulation model was used to examine the nature and intensity of food-web interactions that allow Osmerus mordax to dominate invaded lakes. The model simulates food-web interactions among linked populations of O. mordax, Coregonus artedi and Sander vitreus. Simulations indicated that O. mordax dominate where: (1) adult O. mordax prey on young-of-the-year (YOY) C. artedi, (2) YOY O. mordax negatively affect YOY S. vitreus through competition and (3) adult S. vitreus experience moderate fishing mortality. Osmerus mordax dominated simulations across a broad range of variable values that regulated competition and predation, and displayed threshold responses to increasing angler harvest. Consequently, angler harvest should be carefully managed in lakes susceptible to O. mordax invasions because the alternative could lead to fishery collapse.

Recent declines in mercury concentration in a freshwater fishery: isolating the effects of de-acidification and decreased atmospheric mercury deposition in Little Rock Lake

The atmospheric deposition of H+, SO4, and Hg to Little Rock Lake in northern Wisconsin has declined substantially during the past decade. Parallel decreases have been observed in the surface waters of the lake. Here we extend the observations to the fish community and we present evidence of a contemporaneous decline in levels of Hg in fish tissue. By comparing data from two separated basins of the lake, we then make an initial effort to isolate and quantify the relative importance of de-acidification and reduced Hg deposition on mercury contamination in fish. Statistical modeling indicates that fish Hg in both basins decreased by roughly 30% between 1994 and 2000 (-5%/y) due to decreased atmospheric Hg loading. De-acidification could account for an additional 5% decrease in one basin (-0.8%/y) and a further 30% decrease in the other basin (-5%/y), since the basins de-acidified at very different rates. These results are consistent with the hypothesis that depositional inputs of SO4 and Hg(II) co-mediate the biosynthesis of methyl mercury and thereby co-limit bioaccumulation. And they suggest that modest changes in acid rain or mercury deposition can significantly affect mercury bioaccumulation over short-time scales.