Predicting the Mean Annual Travel Distance of Streambed Particles: A Meta-analysis of Published Data

The cumulative effect of sediment inputs on channel morphology and downstream aquatic resources depends on the rate of downstream particle transport. The use of physically-based models to predict the downstream migration of bedload waves is limited by the need for detailed channel and discharge data. The goal of our study is to determine whether simpler procedures can provide first-order estimates of sediment transport distances and hence the likely magnitude and timing of downstream effects. The specific objectives were to: (1) collect and review existing data on bedload and bedform travel distances; and (2) determine whether mean annual travel distance could be related to easily-measured variables such as stream type, stream gradient, stream width, basin area, size class of the particle(s) being observed, or bed material D₅₀. Our literature review yielded 55 references with 84 observations of travel distance from individual tracer studies or the downstream migration of bedforms and bedload waves. In many cases we had to extrapolate the mean annual travel distance from shorter-term data. Hence the large variability in annual travel distances may partly due to the uncertainty in these extrapolations as well as differences in methodology and study design. Initial results indicate that mean annual travel distance decreases with stream gradient (R²=0.29), increases with basin area (R²=0.29), and decreases with increasing bed material D₅₀ (R²=0.26). In contrast to Beechie (2001), stream width was a relatively poor predictor of mean annual travel distance. Strong correlations between the independent variables meant that multiple regression could explain only 39% of the variability in mean annual transport distances. Efforts to stratify the data showed that mean annual travel distances tended to be larger for streams with finer bed material, braided streams, and migrating bedforms. Coarse-bedded pool-riffle streams, step-pool streams, and ephemeral streams tended to have shorter mean annual travel distances than predicted by the global regressions. These patterns in bedload transport distances suggest that site- or stream-specific correction factors could be developed to improve the predictions derived from the complete data set. Additional analyses are being conducted to better understand the causes of the observed variability and determine the usefulness of this approach for predicting cumulative watershed effects.