Bedload transport has long been known for its complexity. Despite decades of research, significant gaps of understanding exist in the ability to assess and predict bedload movement. This work introduces a comprehensive bedload database that is a compilation of field samples collected over the past 40 years; compares prediction formulae using a subset of the database; evaluates the influence of the armor layer on stream response to sediment input based on a hypothesis linked to one of the tested formulae, presents a mathematically manipulation of the empirical Pagosa Good/Fair formula for bedload transport into a format similar to the semi-empirical Parker Surface-Based 1990 formula; and addresses the complications of bedload transport by collecting bedload samples on a stream in Central Utah. A comprehensive review of available bedload data resulted in a publicly available database with more than 8,000 individual bedload samples on gravel bed streams. Each measurement included extensive and detailed information regarding channel, site, and hydraulic characteristics. A subset of this database was used to compare four calibrated (a single calibration point of a measured bedload transport rate near bankfull discharge is used to improve formula prediction accuracy) and two un-calibrated bedload prediction formulae. The four calibrated formulae include three semi-empirical (a theoretical treatment adjusted to fit bedload measurements) formulae and one empirical (solely based on regression of bedload measurements) formula; the two un-calibrated formulae are both semi-empirical. Of the formulae compared, the empirical Pagosa Good/Fair formula (a calibrated formula) provided the most accurate prediction results with an overall root mean square error of 6.4%, an improvement of several orders of magnitude over the un-calibrated formulae. The Pagosa Good/Fair formula is cast in a form similar to the Parker 1990 formula, suggesting that criticisms stating that the empirical Pagosa method lacks a theoretical basis are unfounded. The hypothesis of equal mobility that states the gradation of the average annual gravel bedload yield for a given stream matches the particle size distribution of the subsurface material is evaluated with relation to the armor layer. Equal mobility is found to correlate to armor layer such that lower armor ratios indicate a greater tendency to uphold the equal mobility hypothesis and increasing armor ratio values tending to move toward supply limited conditions. This correlation provides an upper limit for lightly armored streams. Bedload sampling efforts described in this work compare the Helley-Smith sampler with the net trap sampler and duplicate previous observations that bedload transport collected using net traps increase more rapidly with discharge than for data collected using Helley-Smith samplers. An alternative, relatively low-cost method for collecting bedload during relatively high discharges on highly urbanized streams is also proposed.



College and Department

Ira A. Fulton College of Engineering and Technology; Civil and Environmental Engineering



Date Submitted


Document Type





bedload, sampling, database, bedload prediction, armor layer