A comparison of three methods for estimating substrate coarseness in rivers

Three methods were used to estimate coarseness of substrate within experimental stations in four Newfoundland rivers: (1) a categorical rating was derived by estimating proportions of types of all the substrate within an area; (2) a similar type of rating was derived, but from substrate types within 30 cm sections delineated on a lead rope stretched across the station; and (3) an index of coarseness of the substrate was measured from a depth sampler constructed of 11 sliding rods supported within a 1 m long frame. The rod, or depth, sampler was considered to provide the least subjective method and could therefore provide an index to validate the other methods. Compared with the standard deviations of depths for the rod sampler, the categorical ratings from the general estimate had an r_s= 0.85, and the 30 cm sampler an r_s= 0.86. It was deduced that the two methods of visually estimating substrate types in shallow clear rivers are reasonably adequate methods for describing coarseness, but that the depth sampler would be the more precise under certain conditions.