15 June 2010
Attabad – remarkable stability on the spillway – and some thoughts on what to do next
Posted by Dave Petley
I will shortly post the latest update on the monitoring site, which shows that the lake level is continuing to rise at about 5 cm per day as inflow exceeds outflow. The spillway continues to show a remarkable level of stability. These two images compare an image from Sunday (left side) with one taken 24 hours later on Monday (right side). Whilst it is tempting to play spot the difference, the reality is almost no change:
So what to do now? Well, it is clear that there is a huge social cost to the current situation, which probably cannot continue for much longer. It is also surely clear that the current monitoring team cannot maintain the current level of activity for long given their living conditions. So here is my suggestion for a way forward. Please note this is not advice, and should not be treated as such, but rather it is a starting point for discussion.
1. NDMA sets up a detailed, integrated data collection system based upon, but enhancing, the current Focus activities. This should include the following measurements at 6 hour interval:, with the first measurement each day being taken at dawn
- The flow rate in the spillway
- The seepage rate
- The lake level
- Landslide activity
- The inflow rate
- Seismic activity
2. An expert group meets just after dawn each day and integrates that data with the weather forecast to issue guidance on the stability of the dam. If the group considers that conditions are safe, the boat service is allowed to operate, and people are allowed onto the red zone – i.e. the land below the dam – during the hours of daylight, so long as they are within earshot of a siren.
3. This is backed up with real time monitoring, preferably automatic, of inflow, landslide / rockfall activity at strategic points along the lake, and outflow, feeding a warning system. A command centre is established that would operate the alarm system should landslide flow rate increase. Boat crews and people in the red zone are trained as to what they should do if the alarm sounds.
4. The expert group meets at dusk each day to plan for the next day.
5. Alongside this a study is implemented immediately of management options for the spillway. This study should be undertaken properly, with international input, and will be made openly available when complete. A timetable is placed on the production of this report, and on the decisions that will be made on the basis of it.
6. All of this is backed up with a transparent communication strategy that explains the risks and benefits of the current situation, and of the various options.
This would seem to me to be the best balance between maintaining safety and allowing people to start to rebuild their lives. Of course this strategy will increase the risk that people face from the dam – but there are real risks associated with their displaced state too that we need to start to reduce. However, I must emphasise that to do this safely NDMA must use expert advice – this cannot be done any other way.
This is a very unfortunate situation.The dam is quite stable, but it can't be trusted to hold up when the inflow rises strongly because of ice melt or rain.The only reliable option would be to evacuate everyone downstream and blow up the dam IMO.
This is a very unfortunate situation. The dam is holding up well, but it can't be trusted.The only reliable option would be to evacuate everyone downstream and destroy the dam IMO.
I recognize that you are in a difficult position, as one who apparently was officially consulted to assess the situation. It strikes me that your plan is devoid of any specific actions they might take, as (it would appear) you want to be sure that your mere mention of some action wouldn't be misconstrued as a recommendation.But I'm not sure if what is left is really substantive guidance, and as a non-expert, it didn't give me any insights into what the possibilities are.I wonder if you could at least give us a few scenarios of what sort of things might be decided on, and what sort of information would lead the decision-makers to take specific actions.Just some musings:The situation is not good: a dam of unknown composition holding back a huge amount of water that, if released in one major event, would be catastrophic. People mention using explosives, but how exactly would that be done? (Artillery like projectiles? Radio controlled explosives attached to netting that one hopes will get caught where one would like?) And what are the risks? More landslides from above? Smaller landslides from the existing (and now very steep) banks block the spillway? Explosives trigger the collapse of the existing dam?It would appear that in a few weeks the flow through the spillway will increase substantially (3-4 fold?), regardless of what is done. The only question is, what will the dam/spillway be like at that time.It does seem to me that *IF* relatively low risk explosions can be used at the top to help remove specific obstructions (without risking major unintended side effects) that that would be reasonable to do. (After all, had they had the time, it would be little different than had they been able to do that before the water topped the spillway.)Anyway, just thinking out loud
Excellent thoughts on what to do next.Pamir Times has reported that the local people might resort to self-help to widen the spillway. I take that as a measure of their perfectly understandable level of frustration.However sending large numbers of people to widen the spillway by digging will expose them to the hazards of unstable slopes above that are continuously giving way. People would very likely be lost or injured.What else could be done?1. A cofferdam would allow the underlying slabs to be exposed, drilled and blasted. But in the interim the lake level will raise considerably and the question of deploying an adequate number of drilling rigs to do the job quickly before more villages are submerged and the cofferdam is challenged by rapidly increasing water level would have to be addressed. With the flow at the current rate, I seriously doubt this is a viable option. That said, it's always possible another landslide could block the spillway and produce the same results and hazards as a cofferdam.2. Bridging the channel with gantries (provided the difficult problem of support on unstable slopes could be solved) and drilling from above again raises the question of getting sufficient equipment there. The amount of equipment supplied to dig the spillway testifies eloquently to the low capabilities available. 3. Given the paucity of viable alternatives, bombing the streambed slabs with precision guided munitions seems about all that's left. I would start at the top, observe results for a few days, then work the lake level down through the slabs a bit at a time. Once the last slab goes, retrogressive erosion could drain the remaining lake in very little time.George
Measurements every six hours; experts meet at dawn; experts meet at dusk? How do we know this level of effort is necessary? Would once a week be good enough? How was the balance of safety to cost measured?And of course we are only talking short-term. What's the long-term plan?
Found this clip of the spillway which was probably recorded around the start of the water flow through the spillway. I could not tell how steep the spillway actually is and this clip clearly illustrates that. Does that mean there is more down-cutting rather than expansion of the spillway? http://www.youtube.com/watch?v=evWn-gR3YU0
Some Idea from the Alps …If the dam stay stable till next winter, here are some idea for a long term plan for a global intergrated risk managment.I think that a by pass tunnel should be envisaged. The main goal is to keep the level of the lake well below the level of the dam and to avoid a second jamming that we could not exclude from the landslide.As I know the Hunza Valley, it is sure that the cost of such a tunnel is very expensive but keeping the lake (with a stable dam) may also be a chance :it could be use for irrigation reservoir upstream (with pumps) as well as downstream (by gravity). The level of the tunnel should be also dimensionned in this sense.Second "gain" the tunnel may be used retablishing the link with the Karakoram highway. We could even produce some electricity with the by-pass tunnel.Account of these gains must be integrated to the overall cost.Such solution were already done in the Alps, even in the past when the difference of income was not as high as today, with implementing the possibilty to irrigate crop fieldsPeople of Hunza, keep hope many mountain people are in minds with you.
In response to 1 and 2 of Anonymous I'd like to say that the small number of earth moving machinery and the short time they were digging everyday was not due to lack of equipment in general but to suspected political corruption, meaning the choice of the contractor was made not for his ability to do the job efficiently but to his willingness to pay kickbacks. This is what the people say in the Hunza valley (not my idea).As to nr2 I proposed a while ago in a Youtube comment that missiles be launched from US drones and aimed at the dam since they are in the vicinity (Swat Valley is close by to the South and the Af-Pak border too) and they could do some good with them for a change.Khamomil
Thanks Dave for your blogs. I am following these since late April. I have a few comments and suggestion here:Using explosive or enhacing the breach in any way is not a good idea. Blowing up the dam is a recepie for bringing a disaster home, which mother nature may not bring if the dam holds on till the peak flows of the season are passed.Huge boulders in the landslide matrix is what, I think, is preventing the dam from bursting. They are probably not only providing stability to the dam but also acting as energy breakers in the outflow channel after erosion of 'fines'. I think this slide matrix presents a grand oppoutunity which can be taken advantage of. I think it is possible to make an estimate if the matrix contains enough of these larger rocks by visual inspection of eroding spillway channel and by doing some kind of geophysical investigation on the intact portions of the slide (I might recommend resistivity imaging or EM). With such data, we may be able to predict with more certainty if the dam matrix could hold till the end of high flow season.When past the high flow season (there is an "IF" in this case though), we may think of constructing a second spillway channel between the more stable rocks on the left bank of the slide and the slide itself. (From the pictures I have seen, I guess there are more bigger boulders on this side of the slide). The crest of this channel should be keptlower than the crest of the first spillway. While building this channel, the excavated material should be dumped on the downstream slope of the slide to make it gentler and erosion resistant.The second spillway should only start operating when lake level is almost at the crest of the first spillway. It will, then, hopefully lower the then existing lake level below the sill of the first spillway. When that happens, the sill of the first spillway could be cut lower than that of the second spill way so the water starts flowing from it again. This way the spillways can be operated in tandem till the lake is emptied to a safe levelHowever, such an operation is very tricky in the presense of a potentail 'water bomb' in the form of the existing lake. It should NOT be attempted without the close supervion of qualified experts. Even if the current spillway holds till the end of high flood season, the lake would still be standing a 100m or so of vertical head – quite a hydrostatic force and a potentail threat. Therefore, a strategy must be deliberated and put in place to empty the lake in a gradual manner.I would stongly oppose any suggestion which might trigger a large breach in the dam at this stage. Maybe mother nature turns kind on us and the spillway holds till the end of the peak season.If the dam breaks in due course of current flow regime, the only measures that could be taken are 'reactive'. Proactive measures at this stage (i.e. fiddling with the current spillway) could be dangerous. If it has to breach, it will breach anyway! But now there seems a possibility that the dam/spillway might hold. The planning for this (optimistic) scenario must start now. Building a second spillway is a 'massive' task by volume, and requires experts and very carefull planning and excecution.The above suggestions are based on quite a loud thinking. We can always think and rethink and deliberate on what may be possible. Right now there is an unfortunate situation in Ataabad and the rest of the valley, but there are also opportunities. A well planned lake drainage work can create lots of jobs (even if for a short time) to benefit the affected families – to say the least. We must look at the opportunities that this situation provides us, and how we can turn our misfortunes into fortunes.NB: The author of above suggestions is a Hydrogeologist and Civil Engineer – with 21 years of experience on mega projects spanning in 5 countries across 3 continents and may be contacted at:[email protected]
This makes is sound like the lake actually adds economic value: "A well planned lake drainage work can create lots of jobs (even if for a short time) to benefit the affected families – to say the least. We must look at the opportunities that this situation provides us, and how we can turn our misfortunes into fortunes."Once the lake is drained there is nothing to sell. I've heard someone suggest auto theft is a good thing, as it creates lots of jobs: First those who steal gain, and then the insurance company pays for a new car. The workers who build the new car win. Everyone wins, right? Oh, did I forget about the insurance company who had to pay?
There is a possible mechanism for accelerating the cut through of the dam yet maintaining a level of control. 1. Cautious use of explosives to remove controling features.2. Cautious use of new "landslides" to reduce the flow when the situation looks to be on the verge of getting out of hand.There are some rather large rocks perched above the current spillway that might be shifted into the channel at strategic moments to help control the flow. Requires considerable preparation because you don't want to be caught one landslide short. Just a concept for operations.Thoughts??
The dual spillway suggestion sounds prudent.There is a massive amount of debris above the left bank to remove that raises the recurrent question of capability to deploy sufficient resources. That debris piled up high against the left side of a narrowing of the gorge. Will there be enough room for a second spillway if somehow sufficient equipment is put to work? Pre-slide surveys may provide the answer.George
Waiting out the Summer high flows would require Winter quality accommodation for all the displaced people. Lowering the lake soon would allow the people to rebuild in time for the Winter. But the decision properly belongs to those who live there. Many downstream are resistant to lowering the dam, but their problem is that it might suddenly fail at any time, even though that could be years from now.George
As an architect, i think some people comments in this blog don't take in account the enormous scale of the landslide and the forces they are dealing with. Next i comment some previous proposals:COFFERDAM:Making a cofferdam is impossible in a site with continous landslides, and technically overwhelming in this site.BYPASS TUNNEL:It is not possible to make a bypass tunnel when the lake is already there. It's something you do before filling the reservoir. The length of the tunnel, the nature of the mountain rock and the needed equipment makes it a non-sense.SECOND SPILLWAY:It's not possible to make a second spillway, while the existing one is at the low point. You will need to evacuate millions of m3 of debris, given the low slope angle you can achieve in the landslide material. And workers will be exposed to continous landslides, and the erosion of the first spillway under them.DRONES:Drones are advised in enemy territory to avoid losses of pilots. In this case, a conventional bomber with precision bombs can do the task.WATER AND ENERGY SUPPLY:Using the dam in long term for water supply and energy production is a non-sense. It is very expensive (and dangerous) to build over the dam the needed facilities for that purpose, in a dam that can burst at any time. For people downstream and upstream this implies living permanently in danger.THE KKH:The time and cost to rebuilt the KKH in higher altitude to avoid the dam is a very big concern for the hole contry.OPINION:In my humble opinion, the dam needs to be removed, by natural means or with little help from humans (just removing some key boulders). And let's hope the burst is progressive.
Yes I thought about conventional bombing by helicopter after posting. Apache type helicopters are equipped to deliver different types of ammunition. Problem is a lot of red tape would be involved. Even the Pakistni army has this type of helicopters, and is already in charge of the work at the landslide. There would be no need to ask the US.Another option I mentioned earlier and would like to have feedback on, since I am not a professional, is siphoning the water out of the lake. It would require a pump to get the siphoning action started but afterwards it would just be the law of physics operating.Such pumps are available in the oil industry, and also the Japanese offered to help with pumping. I wonder why nobody has been studying this option which, it seems to me, would empty the lake in a gradual manner that would not stress the dam. Once the water level is low enough the spillway could be cut down more.The urgency is in the threat to the life of all the orchards which are currently submerged. They represent a vital economic resource.Khamomil
tropical… "Once the lake is drained there is nothing to sell. "Please consider the post flooding landscape if the water leaves the dam in a burst, a 60m wave (?), washing every thing in its path. The poor displaced people will return to their abandoned places with virutally nothing. Not just their houses and properties, the alluvial fans they use to grow crops would be gone too. However, if somehow we can manage a gradual drainage of lake and avoid a devastating flood peak, the displaced people would return to thier houses still intact and their lands still useable.
Gabster… "It's not possible to make a second spillway, while the existing one is at the low point…"It is possible, if found feasible after expert investigations and professional analyses. Yes, I fully agree that the scale of work is 'massive' to say the least. However, organizing a massive earthwork (after the feasibility is established) is not impossible. I have the first hand experience of earthworks of even greater proportions (several hundred million cubic meters in a single opertation). So just by looking at the scale is not a valid reason to conclude that its 'impossible'. Just to quote from Nelson Mandela "Every thing looks impossible until it is done".I am rather surprized that many are wanting the dam to burst (or to make it burst…). That would create the big flood wave of destruction. The strategy at this point should be to somehow "break" the possible flood peak down to a non-destructible level.
tropical… "did I forget about … who had to pay?If you think government of Pakistan is too poor to finance it, please think again. The whole opeartion will cost less than a couple of Presidential Visits to Europe and N. America. A country where even the lesser executives are using Gulfstream and Citaion type business jets should not worry too much about the bill… it will be a small one when measured against the wealth we have 🙂
I agree with this: "However, if somehow we can manage a gradual drainage of lake and avoid a devastating flood peak, the displaced people would return to thier houses still intact and their lands still useable."The point I was trying to make was not about the potential damage of the lake, but rather that the "job creation" theory was not based on sound economic theory.Ultimately this situation is tragic. First minimize the loss of human life and personal injury. Beyond that, let's consider the best option, rather than try and sell a solution based on "jobs creation." That's a play toward desperation.
An interesting development reported in the Parmir Times: local "volunteers" trying to take matters into their own hands and widen the spillway themselves.http://pamirtimes.net/2010/06/17/volunteers-start-widening-of-the-hunza-lake-spillway/I'm not sure what they would be able to accomplish with just the tools and impliments they can carry by hand to the spillway, other than to just put themselves in harms way. But it does indicate that the frustration of some people is getting pretty high.
Has any one, NDMA or Focus, made an estimate of the total volume of water stored in the lake for the current lake level? Is there a lake-level to lake-volume relationship derived for this site?
Divalent… 'local "volunteers" trying to take matters into their own hands 'This underpins the importance of involving locals in the management of the this crisis. If we let them feel involved (rather than isolated and dictated) in various tasks and activities within the overall framework of a carefully thoughtout strategy, these motivated individuals could be an asset.
There is no point in allowing the lake to grow if it offers no solution instead growing hazards. How about this?1. Get the boulders out by blowing them with anti tank guided missiles. These weapons can be airlifted and then fired from a safe distance at the boulders. Being hollow charge, these warheads will cause little collateral damage and cause no land slides. However, due to their size, we may need scores of warheads. Then wait for a few days to determine if the flow was really being stopped by boulders or not. If it was, the spillway will deepen and widen also eroding the saddle.2. If not then go for the saddle.Use the same missiles at regular intervals to deepen the saddle, a few meters below the lake level. The rush of water would cause more erosion and the lake may start emptying. This process will have to be repeated when the inflow outflow become the same, till such time the whole lake is emptied.If this is not done, we live in a permanent hazard.
Welle Brigadier I think this is an excellent idea. Now that the people are putting their lives on the line to save their land, perhaps the Army will be moved to help a little more and do some controlled blasting in the way you described. This could be done very quickly which is what is needed now.Khamomil
I am doubtful about anti-tank missiles as they are designed to melt through metal armor plating after which a modest explosion is enough to do in the crew, but not destroy the tank completely as many armed forces like the opportunity to rehabilitate the tank and reuse it. There is a large above water slab where an ATM could be tested to see if it's up to the job but I remain doubtful about ATMs as something heavier seems indicated. Perhaps the locals might be able to bypass the controlling underwater slab as the surrounding ground at the upper part of the spillway is less steep and not as deeply piled up as in the saddle. If the upper slab is bypassed, further downcutting may happen, providing other underlying slabs do not interfere.Ideally it would be determined just what lies below the spillway all the way down to the original streambed. I suspect it would be reasonably safe to blast down one slab level at a time so long as there's another massive slab immediately underneath. We have seen that so far nature has not been able to move these slabs or erode around them — at least until the next earthquake when they might all let go all at once tomorrow or several years from now.George
A few days ago I read a very interesting comment from Colonel(Retd) Kaiser Hameed Khan, who had proposed using water jets ( Giant Water Cannons) like the ones used in Hydraulic mining to widen the spillway and wash away the Debris which is choking the spillway.Increased water flow would then expedite the draining of the lake. If this is a feasible option, it should not be as dangerous or risky as the explosive option, and also a cheaper option for our country.Though I have no experience of anti Tank Missiles I do not thnk missiles do any damage to rocks otherwise bunkers would not be defended with earth and boulders.I have heard of Dam Burster Bombs but never of Anti Tank Missiles bursting dams.Any way I wish someone conveys these comments to NDMA who seem to be completely stunned with the behavior of the DAM.Saleha Raza
Col Qaisar's idea is based on the Egyptian operations of 1973 in which they swept away the Israeli Bar Lev Line with pressure water jets. But that was pure sand and loam. It may partially work.ATMs are aso used for bunker bursting which is pure concrete. There is another anti tank munition called squash head and kinetic energy SABOTS. They smash concrete with pure kinetic energy and no explosives. Army needs to put some tanks in place and blow off these boulders and granite beds.