Home Archived January 13, 2017

Potential Drift Accumulation at Bridges

Management of Drift

Management of drift in the channel has typically focused on drift removal, bank clearing, and channel modifications. Removal of accumulated drift at bridges and in upstream channels is common (Brice and others, 1978a; Lagasse and others, 1991). At some sites, trash racks intercept and collect transported drift. The complex problems of stabilizing channels to reduce drift generation have been addressed in few locations (Gippel, 1989; Gippel and others, 1992).

Many wide streams in the United States were impeded by large drift accumulations at the time of European settlement (Young, 1837; Sedell and Frogatt, 1984; Triska, 1984; Wallace and Benke, 1984; Harmon and others, 1986; Sedell and others, 1988; Orme, 1990). Drift was abundant in streams of all sizes. Beaver created drift by felling trees and creating ponds that killed trees, but their dams probably stabilized more drift than they mobilized (Naiman and others, 1986).

The relatively unimpeded channels now present are the product of intensive ongoing drift removal (Nunnally and Keller, 1979; Shields and Nunnally, 1984). Beginning shortly after European settlement, drift jams were removed to allow navigation and promote drainage. Logging of virgin timber was accompanied by efforts to increase the capacity of channels to transport logs (Sedell and Duval, 1985; Perham, 1988). Clearing and snagging of channels has been a common practice from the time of settlement to the present (1996).

Drift removal has mixed effects on flood conveyance (Gippel and others, 1992; Smith and Shields, 1990; Young, 1991; Shields and Gippel, 1995). A detailed study of partial drift removal in the South Fork Obion River, Tennessee, reported that "...flood control benefits of LWD [large woody debris] removal may be extremely limited in channels similar to the one studied." (Smith and others, 1992). Not all claims of greater flow velocity after drift removal are justified. When increased velocity does result, the flood wave may move downstream faster and increase flood stages downstream from the cleared reach. In large rivers, the effect of drift removal on bank-full conveyance may be negligible. The practice of clearing and snagging, which combines drift removal and bank clearing, is typically intended to increase conveyance, but may cause channel instability and produce a wider and shallower channel with lower conveyance (Thorne, 1990).

Channel modifications to prevent drift accumulation include channel straightening, channel enlargement, drift removal, and the elimination of islands, sloughs, and side channels. These modifications, which resemble those historically used to enable water transportation of logs, are generally successful in promoting drift transport (Sedell and Duval, 1985). Where channel modifications cause channel widening, the abundant drift generated is delivered downstream (Diehl, 1994).

Drift removal, once considered beneficial to fish, is now regarded as detrimental to stream ecology (Gippel, 1989; Gippel and others, 1992). Large woody debris provides a substrate for aquatic invertebrates, stores sediment, and through its effects on channel morphology creates invertebrate and fish habitat (Zimmerman and others, 1967; Megahan, 1982; Wallace and Benke, 1984; Benke and others, 1985; Heede, 1985; Shine, 1985; Bisson and others, 1987; Klein and others, 1987; MacDonald and Keller, 1987; Cherry and Beschta, 1989; Smock and others, 1989; Benke and Wallace, 1990; Carlson and others, 1990; Sedell and others, 1990). Woody debris in road ditches has effects similar to those in low-order channels, trapping sediment and providing invertebrate habitat (Duncan and others, 1987; Hammer, 1989).

New forms of channel management are designed to remove as little drift as possible while still achieving management goals. Forest is allowed to remain in the stream corridor to provide a source of large woody debris sufficient to sustain a large debris inventory in the stream over the long term (Froehlich, 1973; Dykstra and Froehlich, 1976; Bisson and others, 1987; Bilby and Wasserman, 1989). In some areas of the United States, managers have begun to place large woody debris in streams from which it was formerly removed (Gippel and others, 1992). As a result of such changes in management, logs from large mature trees may become more abundant in rivers in the next 50 to 200 years (Bilby and Likens, 1980; Likens and Bilby, 1982; Hogan, 1987; Andrus and others, 1988; Hauer, 1989; Gippel and others, 1992).

Trash racks and holding booms are well-known devices for collecting drift (Perham, 1987; Perham, 1988). Trash racks can collect all the large drift transported in the channel (figure 13). Flow transporting drift distributes it to unblocked portions of the rack, which then become clogged. Booms seem most effective where currents are slow, but net booms are effective even in faster flow.

171K GIF version of this photograph

Figure 13. Trash rack and accumulated drift in Georges Creek near New Columbia, Illinois.

The most serious problem in using trash racks or holding booms to protect highway bridges from drift delivery is the large amount of drift that can be transported through the relatively small channel cross section. In one study, the estimated amount of floating debris transported from a 260-square-kilometer (100-square-mile) basin in the design flood was 54,000 cubic meters (1,925,000 cubic feet) (Martin, 1989). This is much more than enough to completely block the channel. Site studies conducted as part of this project and scour-potential studies include observations of many drift accumulations that formed on bridges with a much lower trapping efficiency than a trash rack, yet blocked most of the channel (figure 14). The existence of such accumulations indicates that a single flood in some rivers may transport more than enough drift to block the channel completely.

. . . 152K GIF version . . . . . . . . . . . . . . 88K GIF version . . . . . . . . . . 89K GIF version

Figure 14. Nearly complete channel blockage of the Harpeth River at Interstate 40, Davidson County, Tennessee.

If the trash rack clogs completely, blockage of the channel will produce significant backwater upstream. Unless the trash rack is designed to function as a dam, scour may be severe downstream from the rack or where flow re-enters the channel. If the trash rack is just upstream from the bridge, it could increase contraction and local scour.

A second, related problem is the cost of drift removal. A trash rack collects much more drift than a bridge at the same location would trap. This drift must be removed to maintain the function of the rack, so the cost of removal should be considered as part of the cost of the trash rack. The chance of drift remaining in place until the next flood is higher where removal is difficult.

previous || table of contents || next

U.S. Department of the Interior, U.S. Geological Survey
Last update: Tuesday, 08-Jan-2013 14:46:21 EST
URL: http://tn.water.usgs.gov /publications/FHWA-RD-97-028/manage.htm
USGS Privacy Policy || Disclaimer