Simulation of Sediment Delivery Ratio and Quantification of Sediment Yield of a Watershed


Current understanding of landform development allows us to identify general causative factors and reasons for occurrences of erosion and deposition of sediment in our environment. General locations for erosion and deposition can be identified based on topographic and geologic conditions. Likewise, the potentials for erosion and deposition can be estimated from rainfall energy and flow capacity. For example, sheet and rill erosion generally occur in fields or tilled lands; channel erosion including streambed degradation, streambank scour, and floodplain scour occurs in the drainageways; and deposition typically happens at the base of steep slopes in the river valleys. Nevertheless, it is difficult to measure the individual effects of these causative factors, and comprehend the complex relationship among them [1]. Also, complexity in the behavior of detached sediment arises from uncertainties in rainfall energy and its impact on surface water runoff, consequent influences of roughness and gradient characteristics in transport and storage of sediment, and changes in transport capacity according to residence time and intermittent flow [2].

Study Area

The ratio between sediment production and sediment yield at the watershed outlet can be expressed as Sediment Delivery Ratio (SDR) [3], which has been shown to take a wide range of values depending on different geomorphological, hydrological, and social factors including landscape relief and slope characteristics; drainage patterns and channel conditions; and landuse, management practices, soil texture, vegetation, location and extent of sediment sources [2].    

We can make inferences about the rate at which sediment is transported to the basin outlet based on the annual gross erosion rates at the field level and annual sediment yield at the watershed outlet. Together with analysis of topographic attributes such as flow lengths, gradients, curvature, and contributing area, we can capture some of the complexities in the behaviors of detached sediment without incorporating specific physical processes such as sediment erosion, transport or storage mechanisms. Thus, proposed is a method to estimate SDR capturing the key topographic parameters, flow lengths and gradient of the overland flow and in-channel flows for different stream orders, and soil loss information.                 


With widely available high-resolution Digital Elevation Model (DEM) and advanced spatial data analysis methods, the topographic parameters can be obtained within a geographical information system (GIS) environment for assessment of sediment delivery rates of soil eroded from uplands to watershed outlet. Within the context of raster cell environment of DEM, a distributed model for SDR can be obtained where sediment delivery rate of each raster cell is estimated. Accordingly, Sediment Yield (SY) at each raster cell can be calculated from soil loss estimates obtained through soil loss maps put together following Universal Soil Loss Equation (USLE) method and the estimated SDR at each cell.


The unknown parameters can be obtained through optimization of estimated SY of the basin against observed Total Suspended Solid (TSS) at the gages located at the basin outlet. However, there can be multiple optima as some of the parameters are related to one another. Also, the unknown parameters and inaccuracies in soil loss estimates can introduce uncertainties in the model. Hence, in order to quantify the degree of belief associated with parameter sets, multiple sets of parameters are tested using Monte Carlo simulation by randomly generating the parameters within reasonable ranges of the values for each set. The outputs of the simulation quantify the uncertainties associated with model parameters and sensitivity of the output to the input parameters.

The motivation of our analysis lies in the extensive information of field-scale erosion estimates available within the general approach of the Universal Soil Loss Equation (USLE); together with the comprehensive local soil mapping that has produced county soil maps for nearly all of the USA. Soil loss information without concrete understanding of sediment delivery to the watershed outlet cannot provide reliable watershed sediment yield. The central research question is: can high-resolution topography provide SDR that can utilize the soil loss information from USLE to provide reliable estimate of watershed sediment yield? Our analysis is built upon topographic algorithm that captures the fundamental hydrologic process connecting sediment source to outlet; however it does not represent specific sediment erosion, transport, or storage mechanism. Thus, we explore the applicability of spatial analysis routines as an automated tool that accounts for the effects of topography on sediment delivery.

[1] R. F. Piest and C. R. Miller. A. Sedimentation Engineering (Chapter IV: Sediment Sources and Sediment Yields). American Society of Civil Engi- neers, Reston, VA, 2006.

[2] D. E. Walling. Sediment delivery problem. Journal of Hydrology (Elsevier Science Publishers B.V.), pages 209–237, 1983.

[3] Louis M. Glymph. Studies of sediment yield from watersheds. International Association for Hydrological Science Publication, pages 173–191, 1954.


Sediment delivery ratio, Digital Elevation Model (DEM), Sediment yield, Geomorphic processes

Article Copyright:

Creative Commons License This article is licensed under a Creative Commons Attribution-Share Alike 3.0 United States License

Article Disclaimer:

The views expressed in this article are those of the author(s) and do not reflect the official policy or position of Johns Hopkins University or the Johns Hopkins University Global Water Program.