"Proceedings, Federal Interagency Workshop,
"Sediment Technology for the 21'st Century,"
St. Petersburg, FL, February 17-19, 1998"

Sampling Bedload Transport in Coarse-Grained Mountain Channels using Portable Samplers

By Sandra E. Ryan

Abstract

Measuring bedload transport in coarse-grained channels can be particularly difficult because flows necessary for transporting larger particles are usually deep, turbid, and turbulent, making difficult the direct physical measurement or visual observation of particle motion. Bedload movement, though rarely observed, is measured using various traps, tracers, or samplers. Difficulties in measuring bedload are compounded by erratic transport patterns, even under stable conditions of flow (Emmett, 1980; Gomez and Church, 1989). Developing representative sampling procedures in steep mountain channels is particularly problematic due to continually fluctuating flows, the presence of large roughness elements, and uneven bed topography. Yet, the need to understand channel processes requires that some effort to gauge bedload movement be undertaken, recognizing the limits of our present capability to measure the processes.

Since 1992, USDA Forest Service research crews from the central Rocky Mountain region have made thousands of bedload measurements in steep, coarse-grained channels in Colorado and Wyoming. These steams are small in comparison to some lowland systems, but they are typical of the types of channels most Forest Service managers contend with on a daily basis. The primary instrumentation has been a Helley-Smith bedload sampler (Helley and Smith, 1971). The Helley-Smith sampler traps bedload moving along the bed of the channel in a metal container with an attached catch bag. The original version of the Helley-Smith sampler is constructed of 1/4" thick cast aluminum, with a 3 x 3 inch (inner dimension) intake, and an expansion ratio (exit area/entrance area) of 3.2. Flow decelerates as it moves into the expanded portion of the sampler, allowing material to collect in the catch bag. Since originally developed, a lighter, less expensive version has been manufactured commercially. This "GBC-type" (or sheetmetal) Helley-Smith is identical to the original, except for wall thickness and the material from which it is constructed (16 gauge stainless steel instead of 1/4" cast aluminum). Both types have been used in these studies, with the same device used consistently at a site because there is some concern that the collection rate of the modified sampler differs from the original. Though some versions of the sampler are cable mounted and suspended from trucks or bridges, we have used a wading version of the sampler from platforms spanning the width of the channel at all sites.

A bedload sample usually consists of all the material collected at a number of verticals (or channel positions) within a cross-section. Typically, 10-20 verticals are measured each site visit, the number of which depends largely upon the width of the channel (Ryan and Troendle, 1997). One key aspect of this effort has been to obtain a large number of samples over a wide range of flows (base flow to greater than bankfull) at several sites; there have been over 200 visits to some sites over 5 runoff seasons (Figure 1). This data saturation helps better define mean transport rates and develop confidence limits for functions fitted to bedload and flow data. Repetitive measurements at several sites help confirm that the observed transport patterns are widespread in these coarse-bed channels.

There are several recognized limitations to bedload data collected by portable samplers, such as the Helley-Smith. Their main weakness is that the confidence in measurements can be relatively low, at least in comparison to other sampling methods, such as instream traps. The reason for low confidence in estimates of transport is that the spatial and temporal variability may not be adequately assessed by these samplers and methods. There are several different "philosophies" on how the samplers should be used to obtain the most reliable measurements for mean transport. For instance, there is little consensus on the correct number of verticals needed or the number of times an individual vertical should be measured. Additional weaknesses result from difficulties in designing a suitable device for measuring bedload for a wide range of conditions. This is problematic because introducing any device to the stream alters local flow hydraulics, which, in turn, may affect local transport patterns. There is some indication that slight differences in sampler design can produce significant differences in bedload collection rates (e.g., Ryan, in progress). Also, there is a limit to the size of material that may be sampled with a relatively small nozzle (3 x 3" in our case) so estimates of flow competence based solely on data from portable samplers are suspect. Finally, sampling efforts using portable samplers are highly labor intensive and, therefore, can be quite expensive.

Still, there is some evidence that data from portable samplers does provide a reasonable estimate of actual bedload transport rates and patterns. For instance, annual accumulations predicted by transport functions integrated over historical gage records predicted measured accumulations in a weir pond (Figure 2), often within a factor of 2 (Troendle et al., 1996; Ryan, in progress). Additionally, comparisons between bedload samples trapped with portable samplers and grain sizes extracted from the ponds indicate that 85% or more (by weight) of the grain sizes from the weir ponds fit into a sampler with a 3 x 3" opening (Wilcox, et al., 1996). This indicates that even small samplers are capable of trapping a majority of grain sizes moved (consistently) as bedload in these coarse-grained systems. An additional advantage of portable samplers is that they may be used at a number of sites and are easily transported to remote areas. The sampling scheme can also be adjusted to meet the objectives of an individual study, whether the interest is in mean transport rates or variation in transport patterns within a cross-section. Finally, databases on bedload transport may be developed within a year or two, depending on flow levels reached during the sampling periods.

figure 1

Figure 1. An example of bedload transport data measured at one site on the Fraser Experimental Forest between 1992 and 1997 (n ~ 200). Two fits, including power and piecewise linear regression functions, are shown.
figure 2
Figure 2. Comparison of predicted sediment yield derived from bedload transport functions for 2 samplers and measured annual accumulations in a weir pond on East St. Louis Creek, Fraser Experimental Forest (n = 31).

References

Emmett, W.W. 1980. A field calibration of the sediment-trapping characteristics of the Helley-Smith bedload sampler. USGS Professional Paper 1139. 44 pp.

Gomez, B. and Church, M. 1989. An assessment of bed load transport formulae for gravel bed rivers. Water Resources Research 25:1161-1186.

Helley, E.J. and Smith, W. 1971. Development and calibration of a pressure difference bedload sampler. USGS Water Resources Division Open-file report. 18 p.

Ryan, S.E. and Troendle, C.A. 1997. Measuring bedload in coarse-grained mountain channels: procedures, problems, and recommendations. In: Water Resources Education, Training, and Practice: Opportunities for the Next Century. AWRA Symposium, June 29- July 3, Keystone, CO. pp. 949-958.

Troendle, C.A., Nankervis, J.M. and Ryan, S.E. 1996. Sediment transport from small, steep-gradient watersheds in Colorado and Wyoming. In: Sedimentation Technologies for Management of Natural Resources in the 21st Century. Sixth Federal Interagency Sedimentation Conference, March 10-14, 1996, Las Vegas, NV. pp IX-39 -IX45.

Wilcox, M.S., Troendle, C.A., and Nankervis, J.M. 1996. Bedload transport in gravel bed streams in Wyoming. In: Sedimentation Technologies for Management of Natural Resources in the 21st Century. Sixth Federal Interagency Sedimentation Conference, March 10-14, 1996, Las Vegas, NV. pp VI-28-VI33.


upWorkshop Contributions


[an error occurred while processing this directive]
[an error occurred while processing this directive]