Skip to main content
ARS Home » Southeast Area » Oxford, Mississippi » National Sedimentation Laboratory » Watershed Physical Processes Research » Research » Publications at this Location » Publication #333944

Title: Sampling interval analysis and CDF generation for grain-scale gravel bed topography

Author
item Wren, Daniel
item Rigby Jr, James
item Langendoen, Eddy
item Kuhnle, Roger

Submitted to: Journal of Hydraulic Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/26/2018
Publication Date: 7/24/2018
Citation: Wren, D.G., Rigby Jr, J.R., Langendoen, E.J., Kuhnle, R.A. 2018. Sampling interval analysis and CDF generation for grain-scale gravel bed topography. Journal of Hydraulic Engineering. 144(10). https://doi.org/10.1061/(ASCE)HY.1943-7900.0001522.
DOI: https://doi.org/10.1061/(ASCE)HY.1943-7900.0001522

Interpretive Summary: In gravel-bed rivers and streams, the distribution of elevation with the top layer of particles has been shown to be critical information for predicting sediment transport. The goals of this study were to find efficient ways to measure the elevations in a rough bed. There is little guidance for choosing an appropriate number of points to measure for bed elevation. Additionally, there is a lack of guidance for explicitly confirming that the true distribution of elevation is represented by the measurements. Gravel and other rough-bed flows represent an active area of research that is being driven by the need for improved ability to predict fine-sediment transport over and through coarse beds. Detailed measurements of elevation for a bed made of gravel in a laboratory flume were used to find how the data were distributed. Statistical methods were then used to examine the distributions and calculate the error introduced by using different numbers of data points to estimate descriptive statistics. It was found that approximately 1000-5000 single-point elevation measurements from random locations in a flattened gravel bed were needed to quantify the distribution of elevations for the gravel bed. The relative error in estimating the standard deviation of the entire population of elevations was not sensitive to the number and spacing of elevation measurements when the spacing exceeded the median size of particles in the gravel bed.

Technical Abstract: In river hydraulics, there is a continuing need for characterizing bed elevations to arrive at quantitative roughness measures that can be used in predicting flow depth and for improved prediction of fine-sediment transport over and through coarse beds. Recently published prediction methods require a method for estimating the cumulative distribution function (CDF) of bed elevations. Representing the distribution of elevations for rough beds requires a correct choice for the number and spacing of measurement locations. Laboratory experiments over a screeded flat gravel bed with D50=35 mm were used to determine the number and spatial coverage of measurements needed to define the elevation field for a range of uncertainty levels. Approximately 1000-5000 randomly collected single elevation measurements were needed to quantify the distribution of elevations for the gravel bed. Relative error in estimating the standard deviation of elevations was insensitive to the number and spacing of elevation measurements when the spacing exceeded D50 (35 mm). The standard deviation of bed elevations was found to be proportional to the median bed material size for several different gravel beds. A method is presented for generating a CDF for bed elevations using the elevation-distribution standard deviation with a randomly sampled distribution function.