George was feeling queasy. He and a dam inspection and survey team were flying in a Huey helicopter. They were performing the annual dam safety inspection for a large hydroelectric reservoir. The work involved flying low over about 50 dams and dykes to inspect rip-rap condition and look for signs of slope slippage and seepage. The helicopter was flying north, into a severe windstorm. During takeoff, the pilot mentioned that if the wind was just a few knots higher, they could not fly.
The team quickly realized that they were witnessing a test of the rip-rap. With the increasing wind and heavy downdrafts, the pilot was having more difficulty controlling the helicopter. He had to fly higher than usual above the dykes, making the survey more challenging.
After crossing a series of hills, the helicopter descended into the last valley, which contained the most northerly dykes. The team was astonished to see what appeared to be waves washing over the dykes. Circling these dykes as low as was deemed safe, they could see that waves were reaching close to the crest of the dykes and that spray was contin-ually being swept over. This was a concern because the reservoir was below the fully supply level and the wind was lower than the design wind speed (the maximum wind speed the structure is designed to withstand).
The pilot flew to where pickup trucks had been parked the previous day for the inspection team. The team used the trucks to travel to the dykes. When the team arrived, the high winds made it almost impossible to stand on the crest of the dykes. All the team members could do was observe and take notes and photos. They could not measure the wind speed on the dyke crests.
When they returned to the helicopter, the team asked the pilot to land on one dyke crest and measure the wind speed. They obtained a reading of 55 kilometers per hour, with gusts to 65 kilometers per hour.
The dykes and dams around the reservoir, commissioned about 35 years previously, had been designed to the highest standards. However, because of the large number of dykes, it had not been possible to calculate the fetch, significant wave height, wave run-up, and rip-rap size for each dyke.
With the advent of computers, a few years earlier the inspection team had recommended a re-survey and computation of the required dam crest elevation and rip-rap size. Two engineering students were hired to check the crest elevation, assess the rip-rap size and elevation, measure the level of the highest driftwood found in the rip-rap, determine the fetch from topographic maps, and take a panoramic video of the reservoir from the center of each dyke. This data was entered into a spreadsheet designed to calculate the required dam crest elevation and rip-rap size and compare that to actual and theoretical values. The results were reassuring, indicating that only a few dykes had suspect crests. However, the spray washing over the dykes signaled that the crest height was suspect.
George used the data from the recent inspection to calculate the dyke crest level with the 55 kilometers-per-hour wind speed and also included reservoir level and data obtained from recording gages around the reservoir. He then increased the wind speed to the design 105 kilometers per hour, raised the reservoir by 0.9 meter to the full supply level, and increased set-up in proportion to the square root of the wind speed. George obtained a crest level some 2 meters higher, a definite indication that the crest was too low.
When George returned the next day to continue the inspection, there could not have been a greater contrast. The weather was clear and calm, with the water well below crest level. George reflected that it would be impossible to imagine waves reaching crest level without the photographic evidence to prove it.
As a result of this inspection and analysis, the dam owner increased the dykes’ crest level by 1 meter, then placed a rip-rap berm on the crest to add another meter.
The owner of this dam had performed meticulous inspections and continual maintenance of the dykes. There had never been any mention of significant spray washing over the dykes. However, the furthest north dykes required 3.5 hours of travel over rough gravel roads, so inspections during inclement weather were never attempted. Also, the vast multi-year reservoir only reached full supply level on rare occasions. It is not surprising that spray wash-overs were never reported. However, a 2-meter increase in crest height is very significant and should have been detected during the design. A review of the design was undertaken on several occasions and was always found to be correct. If not the design, what about the methodology?
This is where the story becomes murky. All wave height and run-up equations are based on measurements on a straight front, facing out into the ocean, with no islands. A reservoir is very different, with islands and adjacent shorelines providing an opportunity for wave reflection. It is a well-known fact that a V-shaped shoreline significantly enhances wave height. Attempts have been made in Norway to use this phenomenon to generate water power at the head of fjords, only to have the power plant destroyed by the proverbial giant wave after about two years of operation.
Based on these incidents, it would be prudent to review dam crest elevations where there is a large fetch and the possibility of wave reflection from adjacent shorelines. Another clear indication of insufficient freeboard is the level of driftwood and debris lodged in the rip-rap. If it is near the rip-rap crest or blown over the dyke, a review should be undertaken.
By James L. Gordon, B.Sc., hydropower consultant. Mr. Gordon may be contacted at (1) 514-695-2884; E-mail: [email protected].