Joint Federal Interagency Sedimentation Conferenet and Federal Interagency Hydrologic Modeling Conference 2015

Full Program »

Inaccuracies in Sediment Budgets Arising from Estimations of Tributary Sediment Inputs: An Example from a Monitoring Network on the Southern Colorado Plateau

Technical Paper
View File

Sediment budgets are an important tool for understanding how riverine ecosystems respond to perturbations. Changes in the quantity and grain-size distribution of sediment within river systems affect the channel morphology and related habitat resources. It is therefore important for resource managers to know if a channel reach is in a state of sediment accumulation, deficit or stasis. Many studies have estimated sediment loads from ungaged tributaries using regional sediment-yield equations or other similar techniques. While these approaches may be valid in regions where rainfall and geology are uniform over large areas, use of sediment-yield equations may lead to poor estimations of sediment loads in semi-arid climates, where rainfall events, contributing geology, and vegetation have large spatial variability.

Previous estimates of the annual sediment load from the ungaged tributaries to the Colorado River downstream from Glen Canyon Dam vary by an order of magnitude; this range in sediment loads has resulted in different researchers reaching opposite conclusions on the sign (accumulation or deficit) of the sediment budget in the Colorado River. To better estimate the supply of fine sediment (sand, silt, and clay) from these tributaries to the Colorado River, eight gages have been established on previously ungaged lesser tributaries in Glen, Marble, and Grand canyons. The remote locations of these streams and short duration of floods make it prohibitively expensive, if not impossible, to directly measure streamflow or to use conventional depth-integrating suspended-sediment samplers. Discharges are therefore calculated using a stage-discharge relation developed from a series of modeled flows and a stage record measured by a downward looking sonic ranging sensor. Flows are modeled using surveyed high-water marks, surveyed channel topography, and Z0 bed roughness constrained by pebble counts. Suspended-sediment measurements are made with passive US U-59 samplers and, at some tributary gages, stage-triggered pump samplers. During floods with a sufficient number of suspended-sediment samples, loads are calculated by interpolating sediment concentrations between the physical samples. When few or no physical samples are collected for a given flood event, regression relations between discharge and sediment concentrations are used if the relations are statistically significant. For gages with no significant relation between discharge and sediment concentrations, mean sediment concentrations – averaged over the period of record – are used. Using these methods, suspended-silt-and-clay and suspended-sand loads transported past each of the lesser-tributary gages are calculated.

Results from this sediment-monitoring network show that previous estimates of annual sediment load from the tributaries were too high, and that the sediment budget for the Colorado River below Glen Canyon Dam is in greater deficit than previously concluded by most researchers. In addition, we found that floods of the same magnitude may have different source areas, resulting in large differences in sediment loads between equal magnitude flows. Because sediment loads do not necessarily correlate with drainage size, and cumulative sediment loads may vary by two orders of magnitude on an annual basis, using techniques such as sediment-yield equations to estimate sediment loads from ungaged tributaries may lead to large errors in sediment budgets.


Ronald Griffiths    
USGS Grand Canyon Monitoring and Research Center

David Topping    
USGS Grand Canyon Monitoring and Research Center


Powered by OpenConf®
Copyright ©2002-2013 Zakon Group LLC