The precision of sampling grain‐size percentiles using the Wolman method

The precision of sampling different grain‐size percentiles in comparison to each other is of significance for choosing which percentile is appropriate to characterize riverbed surface sediments. The traditional view, based on normal grain‐size distributions derived by the Wolman grid‐by‐number method, has identified the median percentile as having the highest sampling precision. However, grain‐size distributions are highly variable and often fail to follow a theoretical distribution. This paper presents both theoretical and empirical techniques to determine the precision of sampling different percentiles. These methods are then applied to a range of natural distributions found in UK rivers. It is found that grains between the 66th and 91st percentiles are the most precisely sampled, with maximum precision obtained with the 73rd percentile.