Rivers and their flooded alluvial plains integrate physical, biological, and human processes at the scale of continents. Despite their ecological and economic values, these complex ecosystems are poorly understood and highly modified by humans. A primary problem is that most research in fluvial ecosystems has been conducted in small streams and then scaled up to rivers. Furthermore, the point where a stream transitions into a river is not well understood. Although many conceptual models exist, the role that large river–floodplain complexes play within these frameworks is lacking. These models focus on flooding as a temporary reset to river ecosystems, but floodplains and rivers may continue to interact long after floodwaters recede. We revisit the concept of the riverscape, a unique mosaic of perennially interacting wetland and channel habitats that have unique ecological properties during both non-flood and flooding periods relative to the small tributary streams within the riverscape network. This strong bidirectional interaction within low-lying alluvial plains may define large rivers. To determine whether a riverscape is indeed a useful unit of study for river ecology, conservation, and restoration, baseline conditions with measurable, comparable metrics, such as primary and secondary production need to be established. Responses of these metrics to multiple stressors and restoration such as levee setbacks, wetland mitigation, and dam removals will inform both basic models of riverscape function and future management actions. Because humans currently affect nearly all aspects of the environmental structure and function of riverscapes, human perceptions of riverscape value and threat need to be considered as a fundamental component of riverscape ecology.