The round goby (Neogobius melanostomus) is among the fastest‐spreading introduced aquatic species in North America and is radiating inland from the Great Lakes into freshwater ecosystems across the landscape. Predicting and managing the impacts of round gobies requires information on the factors influencing their distribution in habitats along the invasion front, yet this information is not available for many recently invaded ecosystems. We evaluated the seasonal habitat use and biomass of round gobies in an inland temperate lake to define the spatiotemporal scope of biological interactions at the leading edge of the round goby invasion. Using novel statistical approaches, we combined hierarchical models that control for imperfect species detection with flexible smooth terms to describe non‐linear relationships between round goby abundance and environmental gradients. Subsequently, we generated accurate detection‐corrected estimates of the standing stock biomass of round gobies. Our results show seasonally differentiated habitat niches, where suitable round goby habitat in summer months is restricted to shallow depths (<18.4 m) with a mixture of vegetative and mussel cover. We found high round goby biomass of 122 kg/ha in occupied habitats during the summer, with a total lake‐wide biomass of 766,000 kg. In winter, round gobies migrate to deep offshore habitats and disperse, dramatically altering their scope for biological interactions with resident aquatic species across summer and winter seasons. The results of this study indicate that the scope of biological interactions in inland lakes may be seasonally variable, with potential for high round goby biomass in shallow lakes or at the periphery of deep lakes in the summer months. Such shallow‐water habitats may therefore present higher risk of ecological impacts from round gobies in invaded lentic ecosystems. As round gobies expand inland, consideration of seasonal habitat use will be an important factor in predicting the impacts of this pervasive invader.

Worsening marine hypoxia has had severe negative consequences for fish communities across the globe. While individual­ and population­level impacts of deoxygenation have been identified, it is unknown how they interact to drive changes in food webs. To address this, we incorporated several major impacts of hypoxia, including declines in benthic re­ sources, habitat shifts, increasing mortality, and changes to rates of feeding, assimilation, and reproductive efficiency, into an existing size spectrum food web modeling framework. We used this structure to ask the following questions: which of these direct effects are most critical to capturing population and community dynamics in a representative hypoxic system, how do they interact to result in community responses to deoxygenation, and what are the potential consequences of these effects in the context of accelerating deoxygenation? We tested the effect of different combinations of oxygen­dependent processes, driven by observed oxygen levels, on the food web model’s ability to explain time series of observed somatic growth, diets, biomass, and fishery yields of commercially relevant species in the Baltic Sea. Model results suggest that food availability is most critical to capturing observed dynamics. There is also some evidence for oxygen­dependent habitat use and physological rates as drivers of observed dynamics. Deoxygenation results in declining growth both of benthic and benthopelagic fish species, as the latter are unable to compensate for the loss of benthic resources by consuming more pelagic fish and resources. Analysis of scenarios of ideal, declining, and degraded oxygen conditions show that deoxygenation results in a decline in somatic growth of predators, an altered habitat occupancy resulting in changing species interactions, and a shift in energy flow to benthopelagic predators from benthic to pelagic resources. This may have important implications for management as oxygen declines or improves.