The identification and geometric definition of individual cascade and step-pool bedforms are investigated in a steep, coarse-grained, mountain stream, Mosquito Creek, by testing four analytical techniques: visual identification, zero-crossing, bedform differencing, and power spectral analysis. The test is the first use of these techniques in a headwater stream, and the analysis of two bed profiles showed that visual identification was able to (i) identify, (ii) determine the geometry of, and (iii) classify the type of individual bedforms better than the other methods. The other techniques were not able to differentiate step-pools from cascades, and the large range of grain sizes and bedform heights hampered their ability to consistently identify stepped bedforms. The step-pool (pronounced, channel-spanning steps that alternate with channel-spanning pools) and cascade (multi-tiered, partially channel-spanning structures) morphology in Mosquito Creek has formed in the last 20 years as fluvial action has restructured its previously engineered, revetment-lined, planar bed. The channel bed exhibits a morphologic regularity that power spectral analysis captured as periodic fluctuations in the bed profiles, with mean wavelengths slightly greater than those identified by the other methods. Further, the active reorganization of revetment has formed stepped structures with geometries similar (i.e., height to wavelength ratios) to stepped features found in natural mountain streams. Channel slope partially controlled bedform geometry (wavelength and height), and bedform height weakly controlled individual step spacing, but there was no relation between wavelength and grain size (D90).