7930 - Laserfiche WebLink

Home Browse Search

Sediment Rating Equations: An Evaluation for Streams in the Idaho Batholith Gary L. Ketcheson INTRODUCTION Sediment transport rates in streams and rivers have been measured for various reasons for many years. The U.S. Army Corps of Engineers began sediment sampling as early as 1838 in relation to navigation work on the Mississippi River (Livesey 1975). Most of the early sampling was done by groups such as the Corps, the Bureau of Reclama- tion, and the Soil Conservation Service (SCS) in response to the Flood Control Acts of 1928 and 1936. The purpose of the sampling was to predict sediment yields for planning flood prevention and control, water storage facilities, and large river basin studies. The sediment rating curve method of sedi- ment prediction was developed in the 1930's. It was used extensively by the Corps of Engineers (Livesey 1975), the Bureau of Reclamation (Strand 1975), and the Soil Conservation Service (Holeman 1975) for large river analyses during the 1930's and 1940's. In the 1940's and 1950'x, emphasis in the Corps began to change from large river analyses to smaller watersheds. At the same time, analyses were being completed on an ever-increasing number of sites with shorter times from planning to implementation. This caused a shift from the rating equation method to the use of predictive equations correlating sedi- ment yield from reservoir surveys to physical water- shed characteristics and management activities. The SCS has moved away from the rating equation method due to shorter analysis times and to the variability of the sediment-discharge relationship (Holeman 1975). Land management agencies have become increas- ingly involved in sediment measurements over the last 30 years. Sediment in streams was identified as a pollutant in the Federal Water Pollution Control Act Amendments of 1972 (PL 92-500). This and the passage of the National Forest Management Act of 1976 (NFMA) created a need to quantify sediment production, especially from National Forest lands. The development of Forest land management required quantified estimates of sediment yields. State water quality laws also became more specific in relation to sediment. Suspended sediment measurement techniques have been well described and tested since the 1940's. Effective bedload measurement methods have only recently been developed. For this reason discussions regarding sediment rating equations in the literature have been restricted to suspended sediment rating equations. Also, much of the sediment data collected in the past are from large river systems where bed- load sediment accounts for only 5 to 10 percent of the total sediment load (Anderson 1975; Emmett 1975). Rannie (1977) notes that even though the sus- pended sediment rating equation has received nearly universal acceptance as a hydrologic and geomorpho- logic tool, the curves themselves have received little attention. Rating curves are usually reported for individual studies with no general standards for comparing them with curves developed elsewhere. Rannie used watershed characteristics to predict rat- ing equation coefficients. This follows from the assumption that suspended sediment is derived mainly from processes that operate throughout the watershed and that are imposed upon the channel, versus bedload transport, which is more dependent on channel hydraulics. There has been limited application of sediment rat- ing curves for small streams in the Northern Rocky Mountains. Much of the work is from third-order streams underlain by highly metamorphosed sedimentary rocks and glaciated granitics. Bedload makes up only 5 percent of the total sediment load from these watersheds (Rosgen 1980). Little atten- tion has been given to the reliability of the sediment rating curves or to the statistics associated with the regressions. This paper discusses sediment rating curves for streams in the Idaho batholith region. The granitic bedrock in this region is highly weathered, and the resulting soils are shallow, coarse textured, and non- cohesive, which results in high erosion rates. The annual sediment yield contains 60 to 70 percent bed- load. This paper will discuss the statistical reliability of the rating equations, their usefulness in monitor- ing watershed response to management, and their effectiveness in determining sediment yields, and will compare different techniques for estimating sediment yields. Suspended sediment rating equations take the form: Qs = aQb 11? where: Qs = sediment concentration Q = stream discharge a and b = coefficients.