Seite - 161 - in Book of Full Papers - Symposium Hydro Engineering

of slab. The uplift force was a product of both the uplift pressure and the area over which it was applied, and the uplift pressure likely involved transmission of stagnation pressure through the slab to the foundation. The resistance to uplift was provided by a combination of the weight of the slab, the weight of the water above the slab, the uplift resistance provided by the foundation anchor system, the structural interaction of slab panels with adjacent slab panels, and any bond between the concrete slab and the foundation. Once the upstream end of the slab section lifted, creating an offset into the flow, the pressure under the slab rapidly increased due to greater stagnation pressures and produced the sudden failure of the slab. The initial chute slab failure section could have been small. Once even a small section of slab was removed, the high velocity flow would have quickly attacked the underlying erodible foundation material and would have begun to remove additional sections of the chute slab. Additionally, the edges of the remaining downstream slab adjacent to the removed section would have presented a significant offset into the flow, which would generate high stagnation pressures that could jack out all but the most resistant slab sections. Subsequent downstream slab sections were removed until the flow encountered a slab section with enough resistance to counter these high uplift forces. The excessive flow into the foundation was mainly due to the high velocity spillway flow injecting water into and through slab surface features, such as open joints, unsealed cracks over the herringbone drains, spalled concrete at either a joint or drain location in either a new or previously repaired area, or some combination of these features that produced offsets into the flow. Localized slab deterioration and repairs existed in the area of the initial slab failure prior to February 7, and these localized areas would have been vulnerable to damage during high velocity spillway flow. Flows into the foundation would generally increase as flow velocities near the chute surface increased. The chute failure occurred very shortly after the spillway gates were opened more to increase discharge down the spillway chute, which led to higher surface flow velocities, and likely higher injection flows which exceeded drainage capacity and resulted in increased uplift pressures. Over time, there was likely also some shallow underslab erosion of fines from the weathered rock and some loss of underdrain system effectiveness, which contributed to increased slab uplift forces. In evaluating the physical factors, it is important to consider why the chute failure happened in 2017 at a spillway discharge of about 1,490 m3/s (52,500 ft3/s), whereas the service spillway chute had not failed previously during higher discharges, most recently a discharge in excess of 1,420 m3/s (70,000 ft3/s) in 2006, and historically discharges as large as about 2,270 m3/s (160,000 ft3/s) in 1997. In other words, what changed from 2006 until 2017 such that the chute slab failure happened in 2017, rather than 2006 or earlier? 161