27 May 2011

The Attabad landslide – tensions mount as the water level rises again

Posted by Dave Petley

This time last year the water level rapidly rose at Attabad as snowmelt in the high mountains drove substantial increases in the discharge of the Hunza River.  Overtopping of the landslide dam occurred at the end of May (see the graphs in this presentation).  It is therefore perhaps unsurprising that a year later the level of the lake is again rising, and as a result tensions are increasing amongst the local population.  The Pamir Times has an article today about the lake level, together with some images of the state of play.  It reports that the lake level has risen by 10 feet ( 3 metres) over the last week, driving concern that houses will again be inundated:




Unfortunately, it is really hard to know how the current lake level compares with the maximum of a year ago, but the above image demonstrates the continued vulnerability of some houses are to any further increases in the water level.  The upshot is an increase in political tension, in particular focusing on the unsuccessful attempts of the government, working through the Frontier Works Organisation (FWO), to deepen the spillway.  The NDMA newsletter for March 2011 (NB this is a pdf file) boldly forecasts that the works over the winter would lower the lake level by 30 metres by the end of May, which appears not to have happened (reports suggest 3 metres was achieved?).  Another article in the Pamir Times reports on demands that the Chinese are given the task of excavating the spillway.

It is hard to judge the risk associated with the condition of Attabad now without direct field observation.  There is little doubt that the risks are considerably lower than they were a year ago.  The failure to lower the spillway may well indicate that the core of the dam is very coarse, which means that it will resist downcutting and thus remain stable.  However, we should not be fooled into thinking that there is no risk.  There remains three obvious modes of failure:

  1. A further large landslide into the lake, which could cause the dam to overtop through the creation of a wave;
  2. A large glacial lake outburst flood, which would overtop the dam through increased water volume;
  3. An earthquake, which could trigger collapse of the dam, a further landslide in the lake, or the formation of a barrier lake upstream that could trigger a domino-style failure once breached.

All of these remain comparatively low probability events, but of course the consequences are rather high.  As a result, it is important to remain very vigilant and to ensure that contingency plans are in place.  I am sure that NDMA are doing exactly this, despite the lack of information on their website.