Will massive poured concrete structures like dams degrade over millennia or not? What is the estimated structural lifetime of a working dam?
Composition and dimensions[edit]
IIRC, I have heard that the hydro dams like on the Grand River in OK will silt in before the life of the dam is reached. Supposedly happening faster than expected.
So, on a river carrying as much silt as the one in China, I think that just adds to the bad idea of doing it in the first place.
No cites, just stuff I remember from somewhere. Maybe my imagination.
Are you talking about millennia without any human involvement? Or millennia with proper maintenance?
When you are dealing with any large concrete structure, the surface of the concrete is prone to be damaged. There can be cracking, pitting, and delamination. If the surface damage isn’t repaired, the damage goes deeper. It will eventually hit rebar, which is then exposed to the air/moisture, and rusts. Then, you have other things to deal with.
If the concrete is well maintained, most concrete dam failures would be due to design problems. The dam could be not sized appropriately, and could be overtopped by a storm. It could be improperly designed for an earthquake (or there can be an earthquake that is so major that it couldn’t have been reasonable planned for). Often when there is a seepage problem, the place where the concrete touches the bedrock has a bad contact. That can be repaired, as well, but it is better to take care of it in the original design.
There are really too many variables to say how long a dam is expected to survive. I have personally dealt with 250 year old dams that had minimal maintenance on them. The original design was probably done by a farmer or miller. The original construction was done by him and his sons. We would never build such a dam today, but they are still standing up - albeit is pretty poor condition. I wouldn’t want to live downstream of an unmaintained dam of any design.
Rivers can certainly silt in, but we can also dredge out the silt. Again, it all depends on whether there is human intervention. It will probably be much easier for the Chinese to take care of the silt problems, because dredging can be difficult to get through a permitting process in the US because of environmental and wildlife concerns.
Then you have a situation like the Mosul dam (this was the one briefly taken over by the Islamic State):
The Masterequivocates. Although the answer seems to be that they’re now designed for 100 years if they’re small and longer if they’re bigger.
How much longer? Well, Zidbits took a shot at an answer for the Hoover Dam. But they took a zombie apocalypse approach, possibly because they kept being told that with proper maintenance and care, it could keep going forever, theoretically.
So they asked how long it would take with no human intervention. The first answer was that the turbines would stop producing electricity in a few weeks to a few months, due to fouling by quagga mussels. Fouling would eventually clog the outflow entirely, causing Lake Mead to rise until water found a way around or over the dam.
About the dam, itself:
I’m guessing that if the Three Gorges Dam got similar engineering, concrete, and maintenance, the results would be similar.
Assuming no human intervention, water levels rising after all drainage ports clog up, and the water can not cut a path around the dam, then water is going to flow over the top and the dam would be lucky to last for 100 more years.
As to Hoover … there are diversion tunnels which are sufficient to offset the full emergency flow of the Colorado. These are built higher than the normal intakes, but well below the level that would lead to overtopping. AFAIK, there are no valves, they just rely on the fact their intake is as high as the designers ever wanted the water level to get. Much like the overflow drain in a bathroom sink or bathtub.
So until those tunnels, which are wicked steep and empty into the lower gorge well above the river level, somehow silt or clog up, they’ll serve to protect the dam from overtopping.
Based on the silting problems of the Aswan dam (which is actually a rammed earth dam?) the Three Gorges was, from what I read, designed to avoid the silting problem by allowing water flow to come from the bottom of the dam as necessary to flush out accumulating silt. This is not perfect, but it will allow the back of the dam to avoid being silted in - which means that the rest could simply be dredged and dumped at the back of the dam where it can be bypassed downstream.
The Pantheon in Rome is about 2000 years old, a massive concrete dome structure that despite minimal maintenance at times shows no sign of failure. Of course, it does not have rebar. the problem with rusting rebar is it not only no longer has structural strength, but rusting caused it to swell and crack the surrounding concrete accelerating the decay.
Good to know. That sounds like it would extend the life of the dam significantly, post zombie apocolypse.
I read the article, which sounded like it was describing the breakup of the main spillway of the Glen Canyon Dam, located upstream of the Hoover Dam, in 1983 (map). It sounded like the spillway wasn’t designed to handle the velocity of water that was dropped onto it. This seemed to be a management issue complicated by the fact that:
(More than you want to know about dam management below the -----.)
But I googled around and the DOI website says that, yes, the Glen Canyon Dam does have diverting tunnels and that they were badly damaged in 1983, but have been redesigned since then. The designers had not been expecting cavitation (the creation of bubbles, which send out destructive shock waves when they collapse). I suspect that the design of other diversion tunnels have been checked for cavitation problems. At least I hope so. Thanks for the link.
The DOI webpage reminded me that big dams were considered the wave of the future when they were installed. Building of the Glen Canyon Dam began in 1956, 21 years after the dedication of the Hoover Dam and 14 years after the startup of the Grand Coolee.
It began collecting water in 1963, began producing electricity in 1964, and was considered complete in 1966. The lake it created, Lake Powel, was not filled until 1980, 17 years after collection began. So The Damaging Event happened only three years after the dam was operating with a filled lake behind it. The first time they had to use the diversion tunnels, they found a bug.
If it hasn’t happened again since 1983, maybe we can hope that the multi-jurisdictional, multi-dam system has management procedures in place to prevent it happening again. Or maybe we’ve just been lucky so far.
[Skip if you’re not interested in dam management.] Dams can do three things, mainly, and when you’re managing them you have to choose your priorities. If you want flood control, you want to keep your dam as empty as possible. If you want a reservoir, you want to keep your dam as full as possible. (The other thing a dam can do is produce hydroelectric power, in which case you’d want it full-ish, but not fuller than your turbines and spillway can handle.)
The fact Glen Canyon is 25-ish years newer than Hoover, smaller, and still had design / fabrication problems with the diversion tunnels leads me to wonder whether the Hoover tunnels would suffer a similar failure mode if put to the ultimate test.
They have been used / tested. Once. AFAIK there was no significant damage, but I have no data on how severe the conditions were or how much minor damage there was. And like any degenerative process, the failure scenario grows exponentially. They may have experienced only very minor spalling which would have turned into total disaster a week later.
Then again, the way a diversion tunnel fails is by eroding bigger and draining too much too fast. The only thing that really suffers is the downstream environs. And for a post-human apocalypse scenario, well who cares about the downstream environs. There’s not much immediately below Hover to care about now, much less then. We’re just keeping score on whether the dam proper is still standing.