Background: Large-river decision-makers are charged with maintaining diverse ecosystem services through unprecedented social-ecological transformations as climate change and other global stressors intensify. The interconnected, dendritic habitats of rivers, which often demarcate jurisdictional boundaries, generate complex management challenges. Here, we explore how the Resist–Accept–Direct (RAD) framework may enhance large-river management by promoting coordinated and deliberate responses to social-ecological trajectories of change. The RAD framework identifies the full decision space of potential management approaches, wherein managers may resist change to maintain historical conditions, accept change toward different conditions, or direct change to a specified future with novel conditions. In the Upper Mississippi River System, managers are facing social-ecological transformations from more frequent and extreme high-water events. We illustrate how RAD-informed basin-, reach-, and site-scale decisions could: (1) provide cross-spatial scale framing; (2) open the entire decision space of potential management approaches; and (3) enhance coordinated inter-jurisdictional management in response to the trajectory of the Upper Mississippi River hydrograph. Results: The RAD framework helps identify plausible long-term trajectories in different reaches (or subbasins) of the river and how the associated social-ecological transformations could be managed by altering site-scale conditions. Strategic reach-scale objectives may reprioritize how, where, and when site conditions could be altered to contribute to the basin goal, given the basin’s plausible trajectories of change (e.g., by coordinating action across sites to alter habitat connectivity, diversity, and redundancy in the river mosaic). Conclusions: When faced with long-term systemic transformations (e.g., > 50 years), the RAD framework helps explicitly consider whether or when the basin vision or goals may no longer be achievable, and direct options may open yet unconsidered potential for the basin. Embedding the RAD framework in hierarchical decision-making clarifies that the selection of actions in space and time should be derived from basin-wide goals and reach-scale objectives to ensure that site-scale actions contribute effectively to the larger river habitat mosaic. Embedding the RAD framework in large-river decisions can provide the necessary conduit to link flexibility and innovation at the site scale with stability at larger scales for adaptive governance of changing social-ecological systems.

In 2015, the Illinois Department of Natural Resources changed size and harvest limits for catfishes (Blue Catfish Ictalurus furcatus, Flathead Catfish Pylodictis olivaris, and Channel Catfish I. punctatus) in the lower Ohio River to match those of neighboring states. The new regulations imposed a daily limit of one 88.9-cm (35-in) or larger Blue or Flathead Catfish per fisher, one 71.1-cm (28-in) or larger Channel Catfish per fisher, and a 33.0-cm (13-in) minimum length limit for all species, with no harvest restrictions between these limit lengths (permissive slot). When these management regulations were implemented, potential efficacy was unknown. The goal of this study was to evaluate the potential effects of these regulations using the limited data for baseline demographic parameters. Bayesian modeling was used to estimate growth and mortality because of the ability to estimate derived functions of parameters and associated uncertainty (as credible intervals). Yield-per-recruit (YPR) modeling was used to estimate and evaluate the fishing mortality rates (F) that would induce growth overfishing (F_MAX) and the yield at F_MAX (YPR_MAX) for three management scenarios: no harvest restrictions (former regulation), a 33.0-cm minimum length limit, and the enacted permissive slot limits. Model results suggested that catfish stocks are unlikely to benefit from current regulations given the near-complete overlap of YPR_MAX credible intervals for all estimable scenarios and the small significant differences (1–3%) based on F_MAX between the most permissive and most restrictive scenarios. Despite the existence of significant differences in F_MAX, they are likely biologically irrelevant based on the life histories of these species. Although modeling results suggested that the implemented harvest regulations will not likely affect catfish population demographics or fishery yield in the lower Ohio River, a post hoc evaluation examining the accuracy of these predictions is needed and would improve the accuracy and precision of parameter estimates.