Discussion in 'Oroville Dam' started by Mick West, Feb 8, 2017.
An interpolated view of appx. elevations for front of the Emergency ogee weir ...
is a good representation of the bedrock on the Emergency weir side of the main spillway ... there is some weathering, not unexpected with close to 50 years exposure - but the basic blue-green bedrock is apparent ...
The image is a composite and contrasts the spillway during the process of failure with the spillway once the slab failure has been recognized by dam operating staff and the flow down the spillway curtailed for inspection of the surface.
Both images show a group of people standing at the same vantage point viewing the spillway. The slabs in the right hand image are believed to still be present in the spillway but have likely shifted under the force of the turbulent flow and are in the process of being destroyed and / or shifted downstream.
This image seeks to situate the image of spillway damage together with an enhanced image of two water boils occurring on the surface of the spillway. The images appear to have been taken from the same position. The focal length used, and the framing, differs slightly. This difference has been compensated for in Photoshop. The right hand image has been enhanced to make the water boils more visible.
The top black line serves to provide a base alignment for the two images. If extended fully to the left this line would align along the base of the electrical towers.
Below that, on image left and image right, there is a red circle. Both images both show a small group of observers standing on a knoll on the left side of the spillway. It is assumed there is a known vantage point at this location and the observers stand in the same location. The surrounding vegetation can be observed to be in the same position in each image.
The black line below that is aligned with the upper margin of the damaged area of the spillway shown in the left image. It can be seen that the area of white turbulence in the right hand image is occurring in the immediate area of the void to the centre left of the spillway. I do not believe the void has yet been fully created or the spillway slabs totally removed. I suspect the spillway is in the process of failing and the slabs have been displaced such that the spillway flow is entering beneath them, and commencing the under-burden scour that will ultimately result in slab failure. The significant water plume is likely due to the flowing water hitting the area of competent blue-grey bedrock visible the left hand image.
The black line below that is aligned with the bottom margin of the damaged area of the spillway in the left image. In the right image, the water has taken on a distinct orange-red coloration. This suggests that some portion of the water jetting beneath the spillway slab has been diverted to the right side of the spillway and is commencing to scour out weathered rock, or loose soils, and create an erosion channel that will undermine the right spillway wall. It can be seen that large flows of water have already exited the spillway and have commenced to scour down the right bank outside the spillway wall.
There is considerable water spray and mist on the right hand side. I suspect this is due to a fountain effect. Prior posts have identified a series of what are believed to be ventilation fixtures, or inspection ports, on each side of the spillway. These fixtures are the black dots located beneath each of the yellow arrows in the left side image. These fixtures are believed to be some aspect of the under-slab drainage system but their exact purpose is not yet known.
I believe these fixtures function as air vents for the drainage system and prevent a vacuum siphon and that they also serve as inspection ports and drain clean out ports. The lids are only loosely secured. I believe the fountain plume shown in the right hand image is due to the drainage system on that side becoming obstructed and this has resulted in pressurized water back flowing through the drainage system on that side and fountaining up through the ventilation / inspection port fittings.
I had wondered about that as well, but it seems like then they would have to increase flow if seepage increased or lose suction, and it would be difficult to monitor and manage. Not to mention the obvious huge issues of it causing flooding under the spillway if backpressure from a clog happens.
All that water is coming from somewhere, the suggested obvious source of head pressure from leaking and seepage under the spillway then makes me think the amount is either far to much or the acceptable amount is is extremely unintuitive and the underside of the spillway can, or is at least supposed to, handle a massive flow of water and leaking. Which is the question we had a few days ago, 'whats with all that water from the drains?'.
I would think if the drains were broken and spilling for years we should see increased plant growth in that area and some soil saturation. that is sort of bothering me with the concept they were broken or bypassing into the soil before this rain season. I looked a bit, there is some obvious groundwater seepage from the reservoir in the areas that washed out on the emergency spillway, they are the only green grass on that slope. But nothing obvious near the main spillway to indicate an extra amount of water under the ground.
Holy cow! I would have never thought of that.
However, I think it's clearer when you see the whole graph.
You can then see that the combined flow eventually reaches 646k cfs at elev. 917.00. In order to overtop the dam, the flow must reach 750k cfs.
Found this site via LinkedIn, and am thrilled as it is surrounded by some uber smart people!
Anyhow, not sure if this is the right place to post this (pls correct me if not) but noticed folks were clamoring for more detailed pics. So wanted to share this pic I took just 2 months ago inside the sluice of the spillway, looking towards the gates. In it you can see the beginning of the emergency spillway, along with what seems to be very weathered rock below it. Hoping this might be helpful to you geology folks....
Great find!! I meant maximum controlled release. An ogee crest spillway at ~1770 ft long can indeed discharge a BUNCH of water - and kiss that drainage slope just downstream of the emergency spillway good-bye. Thanks for posting that design description.
Thanks for pointing this out. I stand corrected. Good Job!
There appears to be a small road approaching that area from the right (the dam side). I wonder if it was an area of concern that they thought they'd need to look at often.
That road has been there since construction. There's some kind of solar powered sensor there, possible a weather station or seismograph.
Lots of comments and questions about anchors in this thread (what are they? were they corrosion protected? how many of them were there?, etc)...in just this one picture below (http://pixel-ca-dwr.photoshelter.co...0kN9PORvuykE/KG-oro-spillway-damage-10060-jpg), I count the remains of 22 anchors visible, most of which are still very much permanently embedded into solid bedrock. Of those 22 anchors, I only see 2 of which show any signs of corrosion at all (both are brown colored...likely rusted).
1) What's a little disconcerting to me is how many of the anchors look to have no damage at all to them, despite supposedly being embedded into the concrete poured above them and supposedly then having that concrete forceably stripped off of them by unreal forces. Someone with knowledge of concrete slab failure will have to enlighten us on how this is possible.
2) On-center spacing laterally appears to be fairly consistent, but the spacing up and down the slope seems to be sporadic. Would this be due to entire lines of anchors being pulled from their locations as the concrete failed in certain places while concrete failure did not pull anchors in other locations, or would this spacing have just been randomness in how it was originally constructed?
3) In browsing this photo, I believe I found the broken drain pipe for the next drain downstream. I circled this area in blue and enlarged it. The lateral location, the orientation, the shape and size of the object, and the position up and down the slope all support the idea that this might be a drain pipe that has turned the corner out beyond the far edge of the slab and is now heading downstream for the next drain exit. The next question this brings up...what is this pipe made of? These days, it's PVC. Back then? This picture and pictures elsewhere of the 90 degree elbow still attached to the sidewall indicate heavy material corrosion (looks like rust, especially for the 90 degree elbow). Would these pipes be heavily rusted, and therefore have contributed to underslab erosion due to material failure?
The dscussion of historical photos of the damage area is split to:
If you look carefully at the detail in the upper right hand corner, it indicates that the anchors are at 10' centers each way.
Ward’s Partial Failure of the Oroville Emergency Spillway Model
2 million cfs flow would flood the primary Highway 70 evacuation route within 30 minutes! (Last week’s evacuation of 188,000 took > 70 hours.)
Simulation Shows Oroville Dam Spillway Failure http://www.capradio.org/90618
See Ward's models at https://websites.pmc.ucsc.edu/~ward/
Partial Failure – “3D” – of the Oroville Emergency Spillway
Partial Failure Oroville Emergency Spillway with Street Map
Partial Failure Oroville Emergency Spillway - 9 hours
Complete Failure Oroville Dam https://websites.pmc.ucsc.edu/~ward/oroville1.mov
Here is an interesting 3d dwg of the area.
Thanks, I'd been looking for that but somehow didn't spot it!
I've never worked on dams or spillways or other things that transport water, but have designed anchors into rock that hold concrete to it.
With an unstoppable force (which I'd consider 100k cfs of water going down a significant slope to essentially be), there are 3 possible modes of failure. The anchor can pull out of the rock, the anchor can pull out of the concrete, or the anchor itself can fail.
Given the degree of damage, if the anchor pulls out of the rock we likely can't tell that it was ever there. If the anchor is in solid rock, and the concrete is removed, we can still see it sticking out. If the anchor itself fails, I'm not sure if we'd be able to see it - it's a fairly small item compared to the scale of the photos, and not sticking out much (if at all).
I have much less experience with anchors than rebar, but I know that for rebar the last thing you want to have happen is pull out. You want to make absolutely sure that the strength of the rebar is developed, so that you can yield the rebar and achieve ductility. In the case of rock anchors, I'm not sure how important ductility is - once you've yielded your anchors, your slab has moved and the water is most likely applying far more force to the slab, so it's going to fail in short order. However, this was designed 50+ years ago and there's been significant change in design techniques since then.
However, with the extensive damage to the concrete, it's possible that the concrete is stripped off of the anchor giving the appearance of it having pulled out when in fact the anchor didn't fail. That's my best guess as to what happened (but it's just a guess).
I suspect the reason for sporadic spacing is that many were anchored in weathered rock that was eroded away (I don't know whether or not it was weathered when they were installed). If the rock is gone, clearly the anchors will be too.
Regarding the drain pipe, that's a good catch. It's interesting that we found documents indicating initially they designed 4" drains and then stepped up to 6", but these appear far larger than that. I'm also curious as to what type of pipes they used at the time.
Where would a discussion of what caused this go? The following link is informative and about that exact issue.
The lateral flow on the upstream side of the spillway has deposited a considerable amount of detritus at its entry into the river channel, causing obstruction to the possible outflow from the hydro works; this will need to be removed before the hydro station can be brought back on line.
PS For parameters on his Partial Failure – “3D” – of the Oroville Emergency Spillway model, Steven Ward (personal communication) states:
Steven Ward posted a movie on the Banqiao Dam Disaster
That emergency spillway scenario does not seem likely, or even possible. as it would require removal of both spillways all the way to the hill past the parking lot
The parking "spillway" is only a meter or so down to rock.
So really his simulation is pretty much a "worst case spillway collapse"
Nice work on identifying the anchors.
1) It's a question of what's the weakest link. Cavitation was likely not anticipated by the designers and is known to have a jackhammer effect on concrete, but it wouldnt have much effect on the anchors. It is a possible cause. Once a hole opened up, you can imagine the flow acted as a hydraulic ram between the concrete and rock below, peeling away at the panels. With 10' anchor spacing, this could easily crack the slab, and the internal slab reinforcing (wire mesh?) would be no match.
3) Based on my mental 3-D model of that pic, that is to low in elevation to be a pipe. My guess is the pipes are steel based on what I have seen. PVC is too brittle, and I don't thing they had HDPE pipes then. Steel would match this vintage. Yes, it does look like corrosion.
One concern I have, and perhaps it is a bit premature, is that I think they will have to rely on the emergency spillway as the primary "safeguard" for at least one rainy season, as the primary spillway is repaired or perhaps replaced (this is likely to be a significant project). I think the design bases for this scenario need to be looked at very closely, along with the apparent pattern of more "extreme" weather events (ie, the concept of 100 year floods now happening every 10 years (exact data/forecasts/probabilities to be provided by climatoligists)).
How good is that "rock"? From the erosion following the emergency spillway overflow, it seemed very broken with numerous cracks and faults. What confidence can we have that that "rock" is reliable? Have there been any deep core samples taken?
Plus the changing pattern of rain vs. snow. We're seeing more significant rain events as a result of warming as opposed to snow ones. All of these dams were built--and have been operated--on the assumption the big pulse of water would be from snowmelt. That may not always be the case anymore.
Once they get the power plant operating again, they can drain it down a long way in order to have significant capacity available in case of another storm. Obviously this will mean they don't have that water available for agriculture etc, but I bet they'll do everything they can to avoid any more use of the emergency spillway. They can only draw it down so far with the spillway, but it can go way lower with the power plant.
My understanding is that the worst case flood scenario is a long heavy snowstorm depositing lots of snow down to relatively low elevations, followed by a tropical storm with heavy rain up to high elevations. That's essentially what happened in 1862.
Good points. Can they draw it down low enough so that they can get though a rainy season without any expectation of using/needing any spillway? Perhaps someone has already crunched the numbers? And they won't be generating a lot of power with levels down that low, but I agree that is a secondary concern...
It would certainly erode in places, as seen by the current erosion. But it's far more likely in the short term a notch would form and deepen rather than 600m off hillside neatly shearing off to 25m.
http://www.water.ca.gov/orovillerel...nse_2005/Vol_IV_App B_Project Description.pdf
Routine maintenance states they are to:
Does anyone know when they last made repairs for erosion on the underdrains?
I think it's quite possible they will use the main spillway again if they get a large rainstorm, even after the power station is up and running - however it would probably take two in a row.
I suspect they'll spend quite a bit more money than they otherwise would in order to get the repair done during a dry season. Add in use of the power plant at 100% 24/7 and I suspect they'll be ok without spillway use. Obviously there's some luck involved.
I think part of why they had problems this time is that, with the long drought we now seem to be getting over, they were retaining as much water as they could.
There is an evacuation going on now in the nearby town of Maxwell.
This is unrelated to the Oroville situation (other than it being from lots of rain). Maxwell gets its water from the hills to the West. However due to the proximity people might get confused.
Question for the civil engineers - can this type of failure be modeled with open channel flow simulation software (if there is such software out there)? The objective would be to calculate dynamic and static forces on the concrete weir, all for a given range of V notch sizes/flows? Is there any way that the end of the weir could could be pushed aside, if one end were to be bypassed and a V notch formed? Or is the weir too massive to be pushed aside or pushed forward by any foreseeable flow?
Well the weir is quite capable, as we have seen, of holding back the entire top of the lake. I don't think water rushing past one end is going to add that much additional force. Plus it's notched into the bedrock. Expansion of the notch would seem to be more of an issue. All very unquantified though.
In my opinion, a competent engineer or manager would not rely on the emergency spillway for reservoir control. This structure exists primarily for dam safety based on the designs we have seen documented previously. The lack of flow management and its high spill-over height (901') make it poorly suited for flow control and reservoir management use.
Were I charged to resolve this problem, I would determine how far back towards the flood control structure gates I needed to cut back the existing structure to reach a stable transition point and then begin construction down the slope towards the river. If I were unable to complete the entire spillway in the given time, I would include a sacrificial edge structure at the terminus of the new work.
I don't believe it would be unreasonable to expect that the remaining lower section is quite stable given the levels of erosion and hydraulic wear it has been subjected to during this event. Also, it would appear that the area to the SouthEast of the flow dispersal section has been worn down to stable bedrock based on the lack of additional deposits in front of the washout.
The "sacrificial" section mentioned above could be constructed to direct the flow into this existing - though recently created - channel for the 2017/18 season. I would then complete the remainder of the lower section replacement during the 2018 summer season and also work any additional replacements or modifications on the upper spillway channel that has thus far not been impacted by this event.
Given the strong possibility that the drain system in conjunction with flow related damage created the failure, I would not be comfortable retaining any of the existing drainage system within the spillway system - I would choose to replace it all with a system meeting current standards. This drainage improvement, along with more modern water incursion prevention at the control joints would likely provide a significant improvement in the amount of water moving below the spillway structure.
Thanks for your detailed response. Given that you are not comfortable (nor am I) with the existing drainage system (full extent of the spillway?), would you replace the slab right up to the gates? I can't see how any of the under-slab drainage system can be replaced, or even inspected, without removing the slab?
In case anyone would like some additional background on how the outlet and spillway was designed, I found this report on the "Hydraulic Model Studies of the Flood Control Outlet and Spillway for Oroville Dam" insightful (particularly, the trial and error aspects of it). The short of it is that they built 1:48 and 1:72 scale models to test and finalize aspects of the outlet and spillway design before the actual construction was completed.
CRM, thanks for that information.
The info they post is missing what seems like an important value to me: inflow minus main spillway max outflow. If you can release 250k indefinitely (which is what they thought they could do, knowing it would flood downstream) then you can deal with a long period of 250k inflow. If you start off reasonably near full, and can only release 250k, you can't deal with 500k inflow for very long.
retired mech eng, given the extent of the damage I'd be surprised if they don't replace the entire spillway (but I've never designed one). It wouldn't be a big surprise to me if they take over part of the existing emergency spillway area to create a new spillway, so they can have a (somewhat) functional spillway during construction.
Note Section B-B in the main spillway blueprint above...
This spec calls for "Backfill Concrete" up to cut grade for any area where the 'Est. "Acceptable Rock" Line' does not meet the elevation requirement. Also...'Note: Details are typical for case where "accept. rock" is below cut grade'.
Has anyone here seen any evidence whatsoever in any picture they've looked at so far that there was Backfill Concrete used at any point in the construction of the main spillway?
There were accounts earlier in this thread that detailed the great effort undertaken to make sure that the emergency spillway weir sat on a suitable foundation involving backfilled concrete, and other accounts detailing where 90% of the main spillway sat on suitable bedrock, but that the other 10% needed different anchors, etc, because of the unexpectedly poor rock conditions that were encountered (I can take the time to go back thru the first 25 pages of this thread and find the exact posts to quote if someone wants me to, but that would take a while).
The question is...was the 10% of the spillway built on unsuitable foundation built to spec? We seem to be mostly in agreement that the visual aspects of the foundation underneath the failed section of the spillway appear to look really poor to the naked (and for most of us, untrained) eye. So, per the specs, should the foundation in that failed area (and any other "bad" areas) have been prepared differently, thus giving us a different outcome last week when the waters rose?
Are there any Sacramento River flow models in public domain? It would be interesting to see what the design plan is for keeping the Capital dry, during a warm water valley rain storm event, with primary Dam's releasing max normal flows, and Sac river weirs fully opened.
The news about flooding in Maxwell, is an example. The area around them was hit with significant rain fall (doppler showed up to 5 inches last night for 24H count, now at 2 inches) in the last 36 hours which isn't in a controlled flood basin. Its hill side run off becoming new flow to the river, joining the Dam flows in route.
In the latest Shasta comments, they talked about throttling down releases while storms passed, I believe for river run off to have volume available. Everyones models appears to have the heavy precipitation in the higher elevations. What is the system capacity for heavy rains "In the valley?". My point is, the Dam's could find themselves restricting flows do to valley flows, which brings the Oroville E spill way repairs back into play.
Separate names with a comma.