Please use this identifier to cite or link to this item:
|Title:||Geotechnical aspects of rock erosion in emergency spillway channels. Report 4, Geologic and hydrodynamic controls on the mechanics of knickpoint migration|
|Authors:||Repair, Evaluation, Maintenance, and Rehabilitation Research Program.|
May, James H.
|Publisher:||Geotechnical Laboratory (U.S.)|
Engineer Research and Development Center (U.S.)
Abstract: During the last decade, occurrences of emergency spillway discharges at Soil Conservation Service, Corps of Engineers, and private reservoirs have increased and, in certain circumstances, resulted in erosional damages to the spillways. Rapid headward erosion in unlined emergency spillways at Corps of Engineers reservoirs including Grapevine, Saylorville, and Black Butte caused the Corps to take a serious look at the available methods used to predict erosion damage in unlined emergency spillways. The catastrophic loss of private and Soil Conservation Service reservoirs because of knickpoint migration in emergency spillway channels was additional evidence that the mechanics of knickpoint erosion were not clearly understood. Severe erosion was documented at several Corps spillways where the flow was less than one-tenth of the designed capacity. Preliminary research revealed that the severe cases of knickpoint erosion were usually caused by the mass failure of large blocks of material and not by the tractive force scour of individual grains. The details of the mechanisms which actually caused the mass failures were largely unknown. Without an understanding of these mechanisms it would not be possible to develop scale hydraulic models or computer models. The purpose of this research was to study knickpoint erosion phenomena with respect to the combined effects of the geologic and hydrodynamic controls. In order to study the mechanisms working at the knickpoint, several obstacles had to be overcome. First a material had to be developed which would erode like rock but would keep the eroding water clear so that the failure mechanisms could be observed. Sodium silicate and gelatin-cemented gravel in combination with Plexiglass were used to simulate knickpoints in layered rock. Next, a hydraulic flume had to be modified to accommodate layered samples. The designed drop structure, which is constructed in streams or channels to dissipate erosive energy, was used as an analog to study the knickpoint phenomena. The research revealed that the potential for headward knickpoint erosion is controlled by the geometry of the knickpoint, the velocity of eroding water, and the pressure underneath the nappe. The geometry of the knickpoint is in turn controlled by the site-specific geology. The erosion rate was found to increase significantly when the thickness of the erodible lower layer in a two-layer model exceeded the diameter of the back roller at the toe of the knickpoint. It was found that headward erosion in the flume could be controlled or completely stopped by controlling the pressure underneath the nappe. Headward erosion was orders of magnitude greater when the area underneath the nappe was not vented to the atmosphere. The development of an unvented condition was found to cause rapid headward erosion at flow conditions well below the maximum discharge.
|Appears in Collections:||Technical Report|