Repository logo

Geotechnical aspects of rock erosion in emergency spillway channels. Report 5, Summary of results, conclusions, and recommendations

Thumbnail Image




Cameron, Christopher P.
Patrick, David M.
May, James H.
Palmerton, John B.
McAneny, Colin C.
Hatheway, Allen W.
Bartholomew, Craig O.
Mathewson, Christopher C.
Cato, Kerry Don, 1959-

Journal Title

Journal ISSN

Volume Title


Geotechnical Laboratory (U.S.)
Engineer Research and Development Center (U.S.)



Technical report
Abstract: This document is a final report of a series of REMR reports which summarizes 5 years of research on erosion in unlined emergency spillway channels. Experience has shown that severe erosion of rock and soils flooring unlined emergency spillway channels may cause undermining or failure of spillway structures and catastrophic release of reservoirs waters. Significant erosion-induced damage is well documented in spillway channels at projects built and managed by the US Army Corps of Engineers (USACE) and other Federal Agencies, and one large privately owned dam lost impoundment by spillway failure. An observational data base was developed to document cases of spillway erosion using data from site visits, Periodic Inspection Reports, videos of spillway flow, and the literature, The data base showed that severe erosion occurred at discharges which were less than 10 percent of Project Maximum Floods, and at velocities which were greater than those recommended by current guidelines; spillway channel erosion was driven by processes similar to knickpoint migration (headcutting) in natural stream channels; and the occurrence of stratigraphic and structural discontinuities in the spillway foundation were important factors in controlling the occurrence and extent of erosion. Laboratory flume studies using simulated earth materials were conducted to investigate knickpoint migration at a waterfall. The waterfall consisted of an erosion resistant layer overlying a material of low erosion resistance. The geometric and hydraulic conditions at the waterfall were mathematically defined using a vented, erosion-dissipating drop structure as an analog. The flume studies showed that erosion and headcutting were maximized when the waterfall became unvented, the ratio of stage height-to-height of waterfall was 1:8 or less, the ratio of thickness of the erosion resistant layer to height of backroller was greater than 1:5, and there were structural discontinuities in the resistant layer. Furthermore, the model studies and computer simulations showed that erosion did not accompany peak discharge but rather it occurs on the lower portions of the rising and falling limbs of the hydrograph. These findings support the observation that severe erosion may occur at discharges significantly lower than Project Maximum Flood or Spillway Design Flood. The evaluation of spillways channels having experienced erosion or suspected to be susceptible to erosion requires preparation of detailed engineering geologic maps and cross sections showing distribution of rock and rock mass properties, particularly lithostratigraphic and structural discontinuities; data from borings including geophysical logs, and information relative to the flood history of the facility (including hydrographs). Spillway erosion is a site-specific problem, The extent of erosion in a spillway may be described and compared with other sites by horizontal and volumetric erosion rankings which, respectively, show the relative amount of headcutting toward the spillway crest and the relative amount of material eroded from the spillway channel. Erosion susceptibility, similar to spillway evaluation, must emphasize rock-mass ratings or classification systems (e.g. rippability) which, when combined with lithostratigraphic discontinuity and hydraulic data, may provide indices indicative of conditions conducive to severe erosion. Remedial and preventive measures at sites having experienced erosion should be designed and emplaced to secure and protect rocks at the top and face of the knickpoint; gabions, rockbolts, and standard cement-based techniques are the most common used. The following recommendations are derivatives from these research activities: the CE should maintain close liaison and cooperation with the SCS in terms of site visits and the sharing of research information; sites which have experienced spillway flow should be documented, preferably by video, and described geotechnically, as well, and hydraulically; and future research should address erosion prediction, knickpoint pressure differentials and venting by flume and prototype testing, numeric modeling and computer simulation of knickpoint erosion, risk analyses (at conditions less than Project Maximum Flood or Spillway Design Flood), and the downstream sedimentation effects of severe erosion in spillway channels.


Erosion, Headcutting, Hydraulic structures, Knickpoint, Spillways, Rock, Rock mass



chat loading...