Below is another perspective from Robert Bea, UC Berkeley, commenting on a known trouble spot in the main spillway (see 2013 image below).
However, I believe there is more to it than just the surface concrete slab crack repairs (i.e. detected defects & enacted repairs).
Engineering Professor (emeritus) Robert Bea gives a form of a simplified answer on associating repeated defect repairs & the breach. The discussion should have delved deeper. (note: I realize the prof was asked to review reports for the interview - the prof did not have any input to these original 14 reports generated by the Division of Safety of Dams- part of DWR).
How? The recurrence of these cracks should have triggered a more thorough level of expertise in analysis - such as bringing in expertise in an investigation on other mechanisms at play: such as slumping or even "voiding" underneath the slab(s), from excessive scour erosion or subsurface material integrity issues, including a failure of one of the "herring bone" angled drain pipes. Any indicator that may be causing unseen stresses on the concrete.
Why? Because repeated cracking means (1) the repairs were insufficient to stop the cracking and (2) there is likely another causal mechanism at play.
(note: this of course is giving the benefit of the doubt on competency of the expertise applied in the "crack repairs").
NOTE (error correction): Professor gives an incorrect length comment of the spillway of 1,730ft (spillway length is stated to be at/near 3,055ft).
=== Article excerpt
http://www.govtech.com/em/disaster/Alarms-raised-years-ago-about-risks-of-Oroville-Dams-spillways.html
excerpt:
Alarms Raised Years Ago About Risks of Oroville Dam's Spillways
The cratering of the main spillway — which spiraled into the current crisis in Butte County — occurred in a spot where cracks and other defects had been found repeatedly since 2009, said Robert Bea, a professor emeritus and engineering expert at UC Berkeley.
But the defects do not appear to have been adequately repaired or resolved by the state Department of Water Resources, which runs the dam, and the faulty work probably resulted in the fissure that opened up last week on the 1,730-foot-long spillway, Bea said.
“My God, we had evidence that there was trouble going back to 2008, 2009,” said Bea, who at The Chronicle’s request reviewed 14 dam inspections from 2008 to 2016 conducted by the Division of Safety of Dams, which is part of the Department of Water Resources.
“Yes, they had detected the defects (in the main spillway) and yes, they had put into gear remedial measures,” Bea said. “Were those repairs sufficient? No. The result was a breach.”
=====
Image of "repeated trouble spot": This was in 2013. This trouble spot is at the location of the "access road" (2,039ft) just at/near where the failure occurred.
Image of normal sealing and repairs at the bottom of the Main Spillway. Notice the "water seepage" from the cracks under repair. This identifies that water flows upward from below the slabs. When the spillway has high water pressure forces from heavy volume waterflow in spillway operation, the flow reverses to under the spillway from high pressure. Thus the need for the extensive drain piping designed into & underneath the concrete slabs. The higher the pressure, from more spillway cfs, the more sub waterflow under the slabs. Increased drainage, under pressure, reveals a "jetting" of water in the drain outlets in the sidewalls. Thus, as dam expert Scott Cahill pointed out -a Good Sealed spillway should have very little waterflow coming from the sidewall drains.
