This blog provides a commentary on landslide events occurring worldwide, including the landslides themselves, latest research, and conferences and meetings. The blog is written on a personal basis by Dave Petley, who is the Wilson Professor of Hazard and Risk in the Department of Geography at Durham University in the United Kingdom.

This blog is a personal project that does not seek to represent Durham University.

Wednesday, 1 February 2012

Investigating the Tumbi Quarry landslide in Papua New Guinea

Despite widespread calls for an investigation of the Tumbi Quarry landslide in Papua New Guinea, and a pledge from the PNG Government to undertake such an inquiry, there is little evidence of action.  Given the suggestions, at the moment with no hard evidence either way in the public domain, that quarrying operations for the Exxon-Mobil LNG pipeline project may have played a part in the initiation of the landslide, this is somewhat unfortunate.  RNZI has today posted an article from which this quote is taken:
"Christine Yango denies that the project is to blame for the landslide, saying it was a natural phenomenon which has nothing to do with nearby LNG activity"
Christine Yango is the Liaison Officer with Oil Search, one of the partners in the LNG Project.  If this quote is true then it appears that the company has already come to a conclusion about the landslide.  I wonder as to the basis for this.
So lets just summarise what we know about the landslide.  First, the landslide is located at, and indeed has destroyed, the Tumbi Quarry site.  This is clearly evident from the first image that I posted:

The remains of what I infer to be the quarry roads and cuts are located in the upper right part of the image, although most of the works appear to have been destroyed.  It is interesting to note that there is no real sign of the spoil (e.g. topsoil) from the quarry works - I wonder where the dump sites for this were located (this is peculiar given the indicative locations in tha map below)?  If downslope or upslope of the main quarry area then this may have been a factor in the development of instability.
Second, the LNG project documentation clearly states that a quarry at this site was to be used in the LNG Project for aggregate extraction.  On page 6 it states that "Tumbi Quarry, known as QA1, is required for aggregate to be used by the Project in the Komo area".  The document even provides a map of the proposed quarry (see page 10 of the pdf):

To gain an understanding of the orientation, the photograph above has been taken from the northeast looking southwest onto the site.
This indicates that the quarry site was located right in the middle of the section of slope that has failed.  It is bold to suggest that the two are not linked without undertaking a proper investigation.
Finally, if this is a natural slope failure, I wonder what was the cause?  The images do not suggest that heavy rainfall was occurring at the time, and I have seen no indications of earthquake activity.  It is true that progressive failure sometimes means that a slope fails without a trigger, but if so then the site would have shown  extensive evidence of deformation - e.g. cracks, grabens, lateral scarps, etc.  If the ground showed such features then these would normally be picked up in the ground investigation prior to the extension of the quarry, and during routine monitoring as the quarry was developed.  If such features were present, the quarry is unlikely to have been developed - the risk of failure would have been too high. And if they were picked up in routine monitoring, action would have been taken immediately as slope failures are a major threat to quarry operations.
So there is a real contradiction here that needs addressing:
  • If this was a natural (i.e. progressive) failure, as the liaison officer suggests, then why weren't the surface features of the landslide picked up prior to the construction of the quarry, and indeed during quarrying operations?
  • But, if these indications of failure were not present, then the slope failure was most likely triggered by a short term process - which would normally be one of intense rainfall, seismicity or human activities.   Which of those was occurring at the time of the landslide to trigger it?
Maybe there is a simple explanation for this contradiction, but at the moment I am unsure as to what this might be.  Of course an independent investigation would find this.  Personally, I would have thought that it would be in the interests of Exxon-Mobil to undertake such an investigation in order to clear up the uncertainty regarding the role of the quarry in the initiation of this landslide.
Meanwhile, judging by the media reports a major concern of the pipeline developers appears to be the logistical challenges associated with the loss of the road.  Again, quoting the RNZI article, Christine Yango is quoted as syaing:
“Yeah I think people are waiting. But we can’t deny the fact that the road system is cut off because of the landslide. It affects major infrastructure areas: the Hides gas conditioning plant and the Komo airfield; the road that brings all the supplies and things. I think at the moment that for food rations and things like that, they’re using air transport, helicopters, but they haven’t cleared up the area to allow vehicles to pass through yet.”

Tuesday, 31 January 2012

An update on the Tumbi Quarry (Papua New Guinea) landslide

The landslide a week ago in Papua New Guinea continues to generate a great deal of noise in that part of the world.  The confirmed death toll is now 25 people, but that number is expected to rise.  However, given the size of the landslide, I wonder whether it is really possible to recover all of the victims.  Only four bodies have been recovered to date.

There are now some better images of the landslide available online.  This one, from the Australian, appears to show the head-scarp area:

http://www.theaustralian.com.au/news/world/fears-for-life-as-landslide-hits-png/story-e6frg6so-1226253126412
The article suggests that the landslide occurred at the site of a limestone quarry (of which more below).  Certainly the rock would appear to be limestone with quite a thick soil layer.  The very planar rear scarp would suggest a large, unfavourable joint as the release plane.  The block at the top on the left is still sitting on this joint and may be unstable.  The joint either has a very weathered surface and/or is draped by soil from above.

There is also a good overview photograph from the Namorong Report website:



It is interesting to see the way in which the lower part of the  landslide has bifurcated down two drainage paths - unusual but not at all unprecedented.  Of more interest is the apparently large lateral margin scarp on the far side of the landslide, below the trees, which suggests that the landslide has removed a considerable volume of material on that side.  It is hard to tell, but this does not seem to be replicated on the nearside.  To me this suggests quite a complex landslide process (although this does not mean that the triggering process itself was complex). This does suggest that if a meaningful investigation of this landslide is to be undertaken then it will need someone who really knows how to interpret landslide processes.  It is not likely to be straightforward in any way.

At the time of the landslide there were suggestions that it was linked to the LNG pipeline being built through the area by ExxonMobil.  It now appears that the hillslope was a limestone quarry associated with the project, which fits with my observations of the area at the crown of the landslide.  There is a pdf online here,with an Esso Highlands logo, entited "TUMBI QUARRY (QA1) Resettlement Action Plan", which states that "Tumbi Quarry, known as QA1, is required for aggregate to be used by the Project in the Komo area. The Project will be leasing the quarry from Hides Gas Development Corporation (HGDC) for an estimated two year period, depending on aggregate requirements of the Project, after which the quarry will be returned to HGDC."

The LNG watch blog has a post by Dr Kristian Laslett of Ulster University, which notes that the site had been blasted in the days leading up to the landslide.  This seems to be supported by various news reports, although of course there is no proof either way.  It seems entirely plausible, though not certain, that the landslide could have been triggered by blasting.  There are many previous examples of landslides being triggered by quarries.
It seems to me that an investigation of this event is urgently required.  However, the investigation does need to be undertaken by an open-minded, independent and very experienced team as the landslide is complex.  IF the quarry was the cause or trigger for the landslide, and if it was providing rock for the Exxon-Mobil project, then this is surely somewhat significant.

Given the focus on the Deepwater Horizon accident a couple of years ago, I am amazed that the mainstream media are not pursuing this story with some alacrity.  One would also hope that Exxon-Mobil will fund a proper investigation of this, and that they will look again at any other quarries that are providing rock for the pipeline project.

Tuesday, 1 November 2011

Intriguing landslide of the week – catastrophic bluff collapse at WE Energies plant at Oak Creek in Wisconsin

Summary: An initial report of a landslide that occurred yesterday at the WE Energies electricity generating station in Wisconsin, USA.  Some thoughts on possible causes are included.

In Wisconsin yesterday a landslide occurred at the WE Energies power station complex near to Oak Creek in southern Wisconsin.  This is the site, which is located on the western edge of Lake Michigan:


The landslide has occurred on the southern side of the complex in an area that is being developed for pollution control equipment (presumably sulphur scrubbers or suchlike).  It appears to be a fairly mobile flow slide that covered an area that is reported to be “the size of a football field”:

Read the rest of this post on the new site

Friday, 21 October 2011

Aberfan - 45 years ago today



Introduction
The story of the Aberfan disaster is seared into the memories of a generation of people in South Wales, and it remains a tragedy of huge proportions.  Today, 45 years on from the disaster, there is much to learn from the events leading up to, and that occurred on, the day.  In this article, we seek to explain the events that occurred in Aberfan on 21st October 1966, to review why the disaster occurred, and to examine the aftermath.  Finally, we briefly examine the legacy that this disaster has left in many spheres of life.

Tuesday, 22 February 2011

First news of the Mw=6.3 earthquake in Christchurch, New Zealand

Because of the severity of this event I have as a one-off posted on this blog.  All updates will appear on the new AGU website.

Christchurch in New Zealand was once again hit by an earthquake today, with the event happening at 12:51 pm local time (23:51 UT on 21st Feb).  As usual, the Geonet site is the best place for up to date information about this event.  At the time of writing (just over two hours after the earthquake, the location report is as follows:

A Google Earth image of the location of the epicentre shows why this earthquake is likely to have had a huge impact:

Note the reported location of the earthquake at the foot of the image.  In essence this is a bulls-eye hit of a substantial event on the city.  The Geonet reported magnitude is 6.3 (note that this is likely to change over the next few hours as the data improves), but the depth is very shallow (only 5 km).  The consequence is high levels of shaking in the city centre.  The Geonet shaking intensity map is based upon instrument data and reports by people on the ground.  As I write the map looks like this:

The squares are the instrument data, and the circles the eye-witness reports.  The red squares indicate MM=8 and the orange ones MM=7.  Christcurch itself is in the MM=7 zone, which in terms of buildings is "damaging".
Christchurch is probably the most "English" city in Australasia, which means actually that there are at least some masonry buildings.  An example is the cathedral - the picture pair below shows after (from here) and a file image from before the event:
The likely impact of the earthquake is likely to have been exacerbated by the timing in the middle of the day, which means that many people will have been in the street (unlike the September 2010 (Darfield) event, which occurred very late at night).  This means that many more people will have been in the way of collapsing buildings, breaking glass, etc.  Fatalities are very likely as a consequence.  Liquefaction was also a major issue in the Darfield earthquake, and is likely to be a major problem in this case too.

Readers may be interested in my images from the aftermath of the September 2010 (Darfield) event, which can be found here (the famous railway line images), here (the fault rupture) and here (building damage).

Because of the severity of this event I have as a one-off posted on this blog.  All updates will appear on the new AGU website.

Tuesday, 7 December 2010

Last post on this site

This is the last post on this, the Blogspot / Blogger site.  From now on everything will be on the new AGU blog site - you can access it here:

http://blogs.agu.org/landslideblog/

My first post was on 16th December 2007, almost exactly three years ago.  The first image that I posted showed the La Honda landslide in California, so I will finish here with a Google Earth image of the same site, taken at about the same time.  Spot the displaced road...
 

So long and thanks for all the fish,

Dave

Friday, 3 December 2010

My new position from January: Executive Director of the Institute of Hazard, Risk and Resilience

Apologies for the paucity of posts over the last fortnight – I promise that normal service will shortly be returned! I am currently somewhat rushed off my feet. You may know that my current job is split between two roles – nominally I spend 30% of my time on my own research, including fieldwork; conferences; supervising PhD students and post-docs; writing papers; refereeing; replying to emails; and, of course, writing this blog! The other 70% of my time is spent helping to run the Faculty of Social Sciences and Health (Faculty in this context is the European not US meaning – i.e. it is what is often referred to in the US as a College, although confusingly that word has a different meaning for us as well). In particular, I look after research and enterprise activities across nine academic units, including the Medical School, Archaeology, Anthropology, Education, Law, Geography, Government and International Affairs and Applied Social Sciences, and various associated units and centres.


From the middle of January I will give up this role to move to something new and exciting. At this time I will take over as the Executive Director of the Durham University Institute of Hazard, Risk and Resilience.

Read the remainder of this post at the new AGU hosted home of this blog.