Findings:

the time to peak is about two times larger for an undeveloped than for a developed watershed of the same characteristic dimension.

the shape factor for an undeveloped watershed is one and one-half times larger that for a developed watershed of the same characteristic length.  

the dimensionless slope on the undeveloped watersheds studied is larger than on the developed watersheds.

the dimensionless leinhard hydrograph at N=2.25 approximates the NRCS DUH, thus for this paper the NRCS-DUH is considered a particular case of the Leinhard hydrograph.

the dimensionless leinhard hydrograph at N=3.35 (The average value) is nearly identical to the NRCS DUH until a dimensionless time ratio of 1.7.  At dimensionless time ratios larger than this value the NRCS DUH has larger values of dimensionless discharge -- the tail of the distribution decays slower in the dimensionless time.  

The value of shape parameter in the regional regression model is nearly a constant, implying that a constant shape hydrograph is appropriate for the watersheds studied.  This result is the same regardless of watershed development classification.

Taking N=2.25 as the approximate NRCS DUH, the results imply that the constant shape hydrograph for these watersheds should have a slighlty shorter decay in dimensionless space (narrower hydrograph about the peak) as indicated by the mean value of N=3.35.

Interpretation: