SEDIMENT TRANSPORT IN A COBBLE- AND BOULDER-BED URBAN HEADWATER CHANNEL, PRELIMINARY RESULTS FROM WORCESTER, MA

The 11 km² watershed of Beaver Brook, a small headwater tributary of the Blackstone River, is located entirely within the city limits of Worcester, Massachusetts, population 183,000. Local “common wisdom” regarding sediment transport of this stream is that the coarse channel sediment (cobbles and boulders) is winnowed from the underlying glacial till and now forms an immobile lag. Furthermore, a relatively large amount of sand is said to be sourced from roads and construction sites and purportedly washes downstream rapidly. Undergraduate students at the College of the Holy Cross in Worcester have begun research projects aimed at testing these anecdotal descriptions of sediment movement and supply. This project will address the following questions: 1) Is the channel sediment mobile in bankfull flow events; and 2) have land-use and/or infrastructure changes in the Beaver Brook watershed contributed to either the mobility or immobility of the bedload? We have begun to address the first question based on field observations and excess shear stress calculations.

In contrast to the lag hypothesis, field observations reveal that the boulders and cobbles in several steep (1.8-7.0%) reaches of Beaver Brook are arranged in steps and bars, indicating that the coarse bed material has been mobile in the past. Furthermore, even very large boulders (> 510 mm) lack abrasion features that might be expected on immobile clasts. We made preliminary dimensionless shear stress (t*) calculations based on channel surface grain size distribution, reach-averaged slopes derived from topographic maps, and estimates of t*_c50 from the literature. Results suggest flow depths < 0.7 m (less than the bankfull channel depth) are sufficient to mobilize the D50 grain size in several reaches (D₅₀ = 78 - 145 mm). If one does not assume equimobility, flows in some reaches would need to be significantly greater in depth (up to 2.6 m) to move the D₈₄ grain size, well above the bankfull channel depth. The implication of these preliminary results is that coarse (though perhaps not the largest) channel sediment is mobilized during large flows. Additional survey work should provide better parameterization for the shear stress calculations. In addition, we plan to conduct tagged clast surveys in a range of sediment sizes (sand to boulders) to test these model results.