Field surveys in the Willapa River basin, Washington State, indicate that the drainage area–channel slope threshold describing the distribution of bedrock and alluvial channels is influenced by the underlying lithology and that local variations in sediment supply can overwhelm basinwide trends. Field data from 90 short-reach surveys indicate that about one-eighth of the surveyed reaches do not conform to a threshold defined by data from free-formed alluvial and bedrock reaches due to the effects of logjams or local sediment sources or sinks. Mapping of channel type distributions in 18 extended reconnaissance surveys of >100 channel widths in channel length show that ∼75% of the channel network was alluvial, but that the proportion of forced alluvial channels varies from 0% to 84%. Using the drainage area–slope thresholds defined by bedrock and alluvial data from the short-reach surveys, only 40% of the total channel length mapped in the longer reconnaissance surveys was correctly classified from a 10 m grid digital elevation model. Of the misclassified reaches, 80% of the alluvial channels predicted to be bedrock had forced alluvial morphologies, while almost half of the bedrock channels predicted to be alluvial were forced by low sediment supply, typically due to their location immediately downstream of large channel-spanning logjams. Poor representation of reach-scale slope in the digital topography and/or a stochastic influence of sediment wave propagation likely account for the remaining misclassified channels, which together compose 7% of the total surveyed channel length. Although variations in sediment supply can locally overwhelm the channel type predicted by the threshold model, the effect of logjams masks any influence of propagating sediment waves on the distribution of bedrock and alluvial channels in the Willapa River basin.