13 May 2010
I have today revamped the monitoring website of the Attabad landslide in Hunza, Pakistan to make the display of information more straightforward as we enter the end game of this situation. In the last day or so there have been three critical developments:
1. Seepage is increasing rapidly
Today’s seepage figure is 2.53 cubic metres per second – that is a substantial increase on yesterday, as the graph below shows:
2. A new seepage point has developed
Part of the reason for the increase in seepage is the development of a new seepage point below the spillway. This is not unexpected, though the rate of seepage is increasing quite quickly at this point. The location is shown in this image, provided by Focus:
Note two things here. First, the seepage is probably allowing water to flow beneath the spillway, which will be reducing the resistance to erosion. Second, and more importantly, even though the water flow here is just 0.23 cubic metres per second, a substantial amount of scour and erosion has already been initiated. Readers will not need me to describe the implications of this in terms of the likely performance of the spillway when water flow commences.
3. The authorities are now admitting that the dam will probably breach
For the first time, government agencies are admitting that the dam is likely to erode away. For example, Onlinenews is reporting that Lt General Shahid Niaz stated today that the water will be flushed out within 2 to 3 days. This is a good development. Indeed, he is also reported to have said ““Such types of lakes have to ultimately breach and there is no chance for its preservation as a dam or a permanent water reservoir,”
I also note that he stated that “international experts’ help is also being taken on the issue”. I wonder who is providing this help, as they haven’t spoken to me. I am delighted that they are getting some expert input though.
I’ll post again on this tomorrow. The current probable window of overtopping is 22nd to 26th May, though this could easily change to an earlier date.