The effects of grain size heterogeneity on sediment transport modeling

Ross Steven Wickham

Sediment is a major pollutant of U.S. waterways, affecting both people and the environment in numerous ways. Increases or decreases in the sediment supply of a waterway may damage infrastructure or degrade habitat quality, so it is important to accurately predict sediment transport. Computational modeling of sediment transport has become increasingly more advanced in recent decades. However, numerical model predictions are only valid if the natural environment is appropriately represented. The heterogeneous surface of gravel bedded streams presents a source of uncertainty in numerical model representation, and is the focus of this thesis. The analyses presented in this study may be divided into three main components: statistical analysis, hydraulic modeling, and disturbance predictions. First, various statistical tests are applied to grain size samples to establish statistically similar groups. A multi-sample, non-parametric statistical test is identified as most appropriate with respect to grain size analysis. Properties of the test established grain size groups that are translated to roughness values with the least amount of redundancy. In the second analysis component, the heterogeneity of both roughness and grain size are analyzed with a hydraulic model capable of simulating sediment transport. Findings show that roughness heterogeneity alone does not produce a difference in sediment transport predictions, but is important when considering flow properties such as velocity. The effects of grain size heterogeneity have significant impacts on bedload transport predictions. Lastly, impacts to sediment yield and bedload transport due to biomass removal following timber harvesting in forested watersheds are assessed. Hillslope predictions show an increase in sediment yield of between 6 to 65%, which would result in a subsequent bedload transport increase of 1 to 6%. Mean bed material diameter is also predicted to decrease by up to 4 mm. Results of the study highlight the importance of appropriate representation of grain size heterogeneity in computational models. Simulations of uniform and heterogeneous surface types showed significant differences in predicted flow and sediment transport properties. The uncertainty associated with sediment transport models will be reduced if the heterogeneity of the stream surface is considered, providing better estimates for flood control, habitat quality, and other purposes.

The effects of grain size heterogeneity on sediment transport modeling

Files and links (1)

Abstract

Metrics

Details