GSA Annual Meeting in Denver, Colorado, USA - 2016

Paper No. 22-7
Presentation Time: 9:50 AM


BUNTE, Kristin, SWINGLE, Kurt W., ETTEMA, Robert, ABT, Steven R. and CENDERELLI, Daniel A., Dept. of Civil and Environmental Engineering, Colorado State University, Engineering Research Center, 1320 Campus Delivery, Fort Collins, CO 80523,

Mountain streams encompass a variety of morphologies, as indicated within the Montgomery-Buffington (1997) stream classification. The question arises whether gravel transport relations and longterm mean annual gravel loads differ systematically among stream types.

This study compiled a worldwide set of 55 gravel transport relations measured in mountain streams using samplers considered most suited for coarse beds: bedload traps, vortex, baskets, and pit-type samplers. Power functions in the form of QB = a Q b were fitted to the sampled transport rates QB (g/s) and the discharge Q (m3/s) at the time of sampling in each data set. The a-coefficients and b-exponents are empirically determined.

Individual stream types showed characteristic combinations of a-coefficients and b-exponents. Steep plane-bed streams had high b-exponents (5 – 18), while their a-coefficients were low (1 – 1E-12). Less steep plane-bed streams with occasional forced pool-riffle sequences had lower a-coefficients (0.0001 – 1E-13) and lower b-exponents (6 – 11). Step-pool streams have notably lower b-exponents (3 – 9) but higher a-coefficients (0.001 – 1,000), while steep mountain torrents (gravel-cobble, high-flow bed with incised step-pool low-flow bed) have even lower b-exponents (2.5 – 7) but higher a-coefficients (100 – 100,000).

Longterm annual load was computed for all 55 gravel transport relations (and for many other combinations of b and a) using a longterm record of mean daily flows from a Rocky Mountain stream with snowmelt regime (this assumes all 55 streams are of similar size). All annual loads were ranked into five groups from very low, low, moderate, high and very high. Load calculations were repeated with another snowmelt flow record, and results were similar: Computed annual loads were high to very high in steep plane-bed streams, but low to moderate in the less steep plane-bed streams. Step-pool streams had a wide range of annual loads from low to very high, while steep torrents have moderate to very high annual loads.

These findings facilitate rough estimates of the relative magnitude of longterm gravel loads for various stream types. Additional assessments of sediment supply (hillslope-channel connectivity) and downstream conveyance potential (e.g., beaver dams) from aerial photography help to narrow down the variability.