There's a data feed from sensors at the dam.[1] "Level" is probably good data. "Out" flow rate may not be meaningful; the emergency spillway probably lacks a flow meter. Here's a plot of "Level".[2] When it's below 900, the auxiliary spillway should stop flowing. As of 2330, it's at 900 feet almost exactly.
[1] http://rdcfeeds.redding.com/lakelevels/oro.cfm [2] http://cdec.water.ca.gov/jspplot/jspPlotServlet.jsp?sensor_n...
They had reduced the main spillway flow prior to water starting to flow out the emergency spillway, but they increased it after that happened.
edit: it seems that the water level is below the emergency spillway now, but the main spillway is still open, probably at least partly out of concern about coming rain.
https://www.youtube.com/watch?v=EQSUJGzjmAI&t=0s
The video clearly shows the over-topping of the emergency spillway and the flooding/over-topping of the recreational area.
It also shows the main spillway and the damage is obvious, particularly at the 22 minute mark.
It's actually surprising how bad the damage is to the main spillway, it's not just a 'sink-hole' as was earlier described, it appears totally destroyed across a large portion of the middle part causing severe erosion of the land underneath and water spilling to the left side as well as rebounding up and over to the remaining lower section.
This video shows the damage to the main spillway even more clearly:
edit: it should be noted that the main spillway damage in itself doesn't threaten the hill and the lake, the issue is fragments or rocks and concrete sent downstream, that might partially clog the river or destroy some infrastructure.
I imagine the most straightforward way to patch a hole, at least for a short period, is to use the water pressure to your advantage and inject objects to plug the hole from the dam side. Perhaps a thick metal plate lowered onto the main breach, and then malleable plastics/rubbers to slow seepage around the plate. Could probably push a hotfix out to master in a few days, for far less than 100 million dollars.
Disclaimer: I haven't found any details on the design of the emergency spillway and the suspected damage, so this is speculation.
You can see the extreme damage in this photo: http://fm.cnbc.com/applications/cnbc.com/resources/img/edito...
The current solution is to drain the reservoir of water as fast as possible--this removes pressure on the hillside, and reduces the scope for damage should the thing actually go.
As for actually patching it, the solution is dumping lots of grout. But that only helps if the slope is actually stable and capable of supporting the embankment and dam. If the slope is washing away, all of your patching is just adding more weight to a structure that can't support its own weight anymore.
While looking at old examples of dam failures, I came across https://en.wikipedia.org/wiki/Sweetwater_Dam#1916_failure ... this seems awfully similar to what's happening at Oroville right now.
Anyway, the Sweetwater Dam article led me to an entertaining read about Charles Hatfield, the professional rainmaker: https://en.wikipedia.org/wiki/Charles_Hatfield
Now this guy was a great salesman. It leads you to wonder if his apparently high success ratio was a result of pure luck, if there's some truth to his "evaporative mix," or as suggested in the article, he had great meteorological prediction skills.
If weather patterns are not independent one year to the next, then just the fact that only statistically significant drought-ridden cities call for his help probably suggests that all of these cities are "due" for rain regardless.
https://apps.axibase.com/chartlab/dee79515
The overflow lasted for 45 hours.
The water level appears to be dropping quickly at the moment:
See pic: http://s4.reutersmedia.net/resources/r/?m=02&d=20170213&t=2&...
I believe they're worried about the erosion tearing apart the hillside and thus destabilizing the entire hill; at which point the whole hill would just collapse.
If it fails, it will be a double disaster, many people will lose property, and California will lose some or all of the water in its largest reservoir. The challenge is going to be keeping ahead of it through to the next storm.
There's definitely worries about the erosion on the main spillway moving back upstream towards the dam, but that is something folks are going to have to look at and consider over the remaining duration of the water season, unlikely to be a danger tonight.
I wonder, how do you test for something like this? I mean, you can certainly plan, and I imagine that the emergency spillway was examined at the time it was planned, but no plan survives contact with the enemy, and the best thing to do is to test.
But, to test the emergency spillway you need to fill the dam to the point where it's overflowing, and I believe it's been a few years since we've gotten close to that. Certainly not any time in the last year.
Besides, if it was tested in the past, it may be that the drought conditions--followed by all the rain--changed the soil in such a way as to enable this.
I need to make a note to read the report when it comes out; I'm sure it'll be interesting.
It was supposedly designed to handle 350,000 cubic feet per second, but began to fail at around 6,000 to 12,000 cubic feet per second.
Same principle seems to apply to spillways.
Under a stream of water moving 65,000 CFS, ripping out chunks of solid cement, probably not so much.
Or I also wonder if they could drill into the bottom of the Resavoir and let the water drain into the ground?
The main spillway has been running at 100000 cubic feet per second. To carry that in 3 foot pipes (7 square foot cross section), you'd need hundreds or thousands of them (so a few just wouldn't do much of anything).
Same with trying to bypass the water to somewhere else. To make a difference the bypass needs to be at a similar scale to the spillway, carrying tens of thousands of cubic feet per second.
http://www.hcn.org/external_files/40years/blog/NearBreachGle...
It's a fascinating read.
I'm sure the California folks right now are studying up notes from the Glen Canyon incident.
40 million cubic meters in 24 hours means just 15 2m diameter pipes running at 10 m/s.
I would imagine the same amount of water flowing over a longer period time would cause less erosion even if the pipes wouldn't reach all the way down...
Edit: You could even do it with 6000 20 cm diameter pipes, probably plastic would do.
Good idea, but this is happening in CA, where pipelines don't sprout so quickly.
http://cdec.water.ca.gov/jspplot/jspPlotServlet.jsp?sensor_n...
1: http://cdec.water.ca.gov/jspplot/jspPlotServlet.jsp?sensor_n...
2: http://www.abc10.com/weather/sierra-snowmelt-peaking-earlier...
http://cdec.water.ca.gov/histPlot/DataPlotter.jsp?staid=ORO&...
Live coverage.
They say situation is better now that drain is sufficient.
https://www.youtube.com/watch?v=hSaAE6MbKE0
Not that you can see much as it's dark...
Fingers crossed that things stay drier and cooler for the sake of all these people
> Sunday night gridlock as Marysville residents evacuate
> After having to evacuate several times before, Agrifoglio said, she was used it.
> She’s confident her husband will be able to evacuate if he needs to, but Tommy would have to wait out a flood.
> Erin English of Linda said she got a robo-call a few minutes ago telling her to evacuate and get to higher ground.
"The Wire: to evacuate":
Thankfully, it appears the actual body of the dam is unlikely to fail, so hopefully even if the emergency spillway fails, it won't be a catastrophic, immediate draining of the whole lake.
Between 130,000 and 180,000 people are being evacuated upstream.
(I just realized the Oroville Dam is for a reservoir rather than power generation, but the risks are similar.)
