Roughness lengths are used in wave-current bottom boundary layer models to parameterize drag associated with grain roughness, the effect of saltating grains during sediment transport, and small-scale bottom topography (ripples and biogenic features). We made field measurements of flow parameters and recorded sonar images of ripples at the boundary of a sorted-bedform at ~12-m depth on the inner shelf for a range of wave and current conditions over two months. We compared estimates of apparent bottom roughness inferred from the flow measurements with bottom roughness calculated using ripple geometry and the Madsen (1994) one-dimensional (vertical) wave-current bottom boundary layer model. One result of these comparisons was that the model over predicted roughness of flow from the dormant large ripples when waves were small. We developed a correction to the ripple-roughness model that incorporates an apparent ripple wavelength related to the combined wave-current flow direction. This correction provides a slight improvement for low-wave conditions, but does not address several other differences between observations and the modeled roughness.