Snowmelt-dominated basins in northern latitudes provide critical habitat for salmonids. As such, these systems may be especially vulnerable to climate change because of potential shifts in the frequency, magnitude, and timing of flows that can scour incubating embryos. A general framework is presented to examine this issue, using a series of physical models that link climate change, streamflow, and channel morphology to predict the magnitude and spatial distribution of streambed scour and consequent risk to salmonid embryos at basin scales. The approach is demonstrated for a mountain catchment in the Northern Rocky Mountains, USA. Results show that risk of critical scour varies as a function of species and life history and is modulated by local variations in lithology and channel confinement. Embryos of smaller-bodied fall spawners may be at greater risk because of shallow egg burial depths and increased rain-on-snow events during their incubation period. Scour risk for all species is reduced when changes in channel morphology (width, depth, and grain size) keep pace with climate-driven changes in streamflow. Although climate change is predicted to increase scour magnitude, the frequency of scouring events relative to typical salmonid life cycles is relatively low, indicating that individual year classes may be impacted by critical scour, but extirpation of entire populations is not expected. Furthermore, refugia are predicted to occur in unconfined portions of the stream network, where scouring shear stresses are limited to bankfull stage because overbank flows spread across alluvial floodplains; conversely, confined valleys will likely exacerbate climate-driven changes in flow and scour. Our approach can be used to prioritize management strategies according to relative risk to different species or spatial distributions of risk and can be used to predict temporal shifts in the spatial distribution of suitable spawning habitats. A critical unknown issue is whether biological adaptation can keep pace with rates of climate change and channel response.