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.

Thursday, 31 December 2009

Mudslides in Rio de Janeiro, Brazil

Various international news agencies are reporting that the year is ending with a number of fatal landslides in and around Rio de Janeiro in Brazil. For example, the BBC report says:

"Heavy rains have caused floods and landslides in Brazil, leaving at least 11 people dead. Officials said the worst incident had occurred in Jacarepagua, in the western part of Rio de Janeiro state, where a family of five died in a landslide. The state has been hit by 24 hours of downpours threatening new year celebrations on Copacabana Beach"

Whilst the report on AP says:
"Heavy rains in the Rio de Janeiro area have triggered mudslides that have killed at least 18 people. Rio de Janeiro state's Office of Civil Defence says that more deaths are likely as more rains are expected."

A quick look at the TRMM cumulative precipitation graphs for this area suggests that it has suffered from a prolonged period of heavy rainfall, mostly falling in heavy rainfall events:

Further rain is forecast for the next 24 hours.

Wednesday, 30 December 2009

Munich Re's list of the largest disasters of 2009

The reinsurance giant Munich Re releases annual statistics on the losses from natural catastrophes each year. They released the data for 2009 yesterday, presumably banking on the balance of probabilities that there would be no more major events in the three days left of the year. The release is available here:

Overall results

The news for 2009 is good, with no disasters causing mass (i.e. >10,000) fatalities. There list of the top ten events in terms of fatalities is as follows (NB using the logical European style date format of
  • 30.09/1.10.2009: Indonesia: Earthquakes, 1,195 deaths
  • 26-30.09.2009: South East Asia, East Asia: Typhoon Ketsana, 694 deaths
  • 07-10.08.2009: China, Philippines, Taiwan: Typhoon Morakot, 614 deaths
  • 03-14.10.2009: South East Asia, East Asia: Typhoon Parma, 469 deaths
  • 25-27.05.2009: Bangladesh, Bhutan, India: Cyclone Aila, 320 deaths
  • 29.09.-15.10.2009: India: Floods, 300 deaths
  • 06.04.2009: Italy: Earthquakes, 295 deaths
  • 21.08-15.09.2009: India: Floods, 223 deaths
  • Aug-Sep 2009: West Africa, Central Africa: Floods, 215 deaths
  • 04-13.11.2009 El Salvador, Nicaragua, Mexico, USA: Hurricane Ida, 204 deaths
Two aspects of these statistics are particularly interesting. First, the lack of a really large event is pleasing, but is probably no more than a serendipitous lack of a large earthquake in a populated area and limited numbers of large land-falling tropical cyclones, especially in the Atlantic basin. Second, the top five events were all associated with large numbers of landslide fatalities, most notably perhaps typhoon Morakot in Taiwan, the landslides triggered by the Indonesian earthquake and the multiple slides in the Philippines caused by typhoon Parma.

Geographical distribution
The geographical spread of these larger events is quite wide, including SE Asia, E. Asia, S. Asia, Africa, Europe and Central America. However, probably the most interesting aspect of this entire release is a map of showing the location of all the natural catastrophes that have occurred through the year:

There are several things to note here. First, you may well have spotted that the highest density of catastrophes appears to have occurred in the United States, Europe and China. This of course reflects the vulnerability of countries with large asset values to geophysical and meteorological processes. Second, the distribution of the event types is quite varied. The climatological events are mostly concentrated in the USA and Australia; Europe and N. America is mostly affected by storms, whilst in Asia the events are primarily floods. Africa probably has far fewer catastrophes than most people would expect.

Economic Losses
Economic losses were also lower than in previous years at $50 billion, compared with $200 billion in 2008. The largest loss-inducing event was a winter storm that affected N. Spain and France in late January, inducing losses of $5.1 billion. The USA was affected by four of the ten events that caused the highest costs in terms of losses.

Climate Change
The effects of climate change on disaster losses is very complex issue. I am increasingly persuaded by the argument that there is now a strong climate change signal in the loss data, primarily due to increased precipitation intensities and increased intensities of the largest tropical cyclones, both of which are supported by strong scientific evidence that has been subjected to peer review. Interestingly, Munich Re are also pretty clear on this point:

'Torsten Jeworrek, Munich Re Board member responsible for global reinsurance business, drew attention to the marked increase in major weather-related natural catastrophes worldwide since 1950, the number now having more or less tripled. Economic losses from weather-related natural catastrophes in the period since 1980 totalled approximately US$ 1,600 bn (in original values). "Climate change probably already accounts for a significant share. In the light of these facts, it is very disappointing that no breakthrough was achieved at the Copenhagen climate summit in December 2009. At Munich Re, we look closely at a multitude of risks and how best to handle them. Risks that change in the course of time are especially hazardous. Climate change is just such a risk of change."

Losses caused by climate change will continue to increase in the future. Jeworrek: "We need as soon as possible an agreement that significantly reduces greenhouse gas emissions because the climate reacts slowly and what we fail to do now will have a bearing for decades to come."'

In the next few days I will review landslide events both for 2009 and for the "noughties", and also the major, game-changing disasters of the last decade.

Wednesday, 23 December 2009

Happy Christmas

A wintry scene to wish readers everywhere a Happy Christmas! I will be back after the weekend.


Monday, 21 December 2009

Landslides in Oregon

The Oregon Geological Survey has a quarterly magazine called Cascadia highlighting some aspect of the geology of that state. Their Fall 2006 edition focuses on landslides, with a series of articles on different aspects of landslides in that state. Some of the articles have a broader interest, such as the use of LIDAR for landslide mapping.

The magazine is here: Cascadia

Sunday, 20 December 2009

On the perils of Lake Sarez (Usoi) in Tajikistan

Science this week has an article (Stone 2009) on the perils associated with Lake Sarez in the Pamirs. Sarez is a huge lake (56 km long and with a volume of 17 billion cubic metres of water) that was formed by a landslide triggered by the 1911 earthquake in Tajikistan (see image below).

Google Earth image of Lake Sarez. The landslide dam is to west (left).

Google Earth image of the landslide dam at Usoi. The source of the landslide was to the north of the current deposit.

The landslide dam (see image above) stands 567 metres tall. To put that in perspective, the image below shows Taipei 101, until recently the world's tallest building. It is 501 metres tall:

Taipei 101 (source Wikipedia)

Since its creation Lake Sarez has been steadily filling, which has long been a concern. There are an estimated 5.5 million people living downstream of the dam in the Amu Darya river valley, which flows through Tajikistan, Afghanistan, Turkmenistan, and Uzbekistan. There are really three key concerns with this dam:
  1. The dam could fail through seepage - a few years ago water started to seep through the landslide deposit, the concern is that this will erode out the core of the landslide;
  2. The dam could fail in an earthquake - this is a seismically-active zone, but the threat is considered to be quite low as the dam is considered to be quite stable;
  3. The dam could fail as a result of another landslide going into the lake, creating a displacement wave (similar to the Vaiont landslide) that causes the dam to overtop. Of course this is most likely to be triggered by an earthquake landslide.
The article points out that the third of these is the most likely, such that the site has a sizable warning system just in case.

The article points out that given the number of people downstream the risks are now considered to be too high. The dam itself cannot be stabilised, so there is a need to draw down the level of the lake by at least 50 m. However, there can be little doubt that this falls in the "easier said than done" category.

The key component of the article is highlighting that there are a range of views as to the level of danger at this site, both in terms of the possibility of another landslide and of the stability of the dam itself. The article quotes a number of notable landslide experts:
  • Jorg Hanisch is quoted as saying that "the probability is 1 in a million,"of the dam being overtopped by a wave created by a landslide. He also rules out any possibility of the dame being eroded by seepage.
  • Jean Schneider from BOKU in Vienna is quoted as saying that "The risk of even a partial outbreak is exaggerated...the dam will only possibly be overtopped in the far future."
  • On the other hand, Kadam Maskaev (deputy director of the emergency situations committee in Tajikistan) views the seepage in a different way: "The filtration regime of the dam is changing, and that makes me nervous."
  • Kyoji Sassa, the chair of the International Consortium on Landslides, has a different view again. The article claims that he argues that the threat from a further landslide is significant.
The suggested optimum mitigation approach is a diversion tunnel that would be used to generate hydroelectric power, with the water also being made available to downstream communities. However, the costs are high ($500 million) and such a project is not without risks. In the sort term it appears that there will be a research campaign that will culminate in a conference in 2011, the 100th anniversary of the dam. That would be an interesting meeting to attend!

Stone, R. (2009). Peril in the Pamirs Science, 326 (5960), 1614-1617 DOI: 10.1126/science.326.5960.1614

Friday, 18 December 2009

AGU Presentation on the Societal Impacts of Landslides

This is the Powerpoint file from my presentation at AGU Fall Meeting in 2009 on quantifying the impacts of landslides on society, which is scheduled for Friday morning.

Please feel free to download it and to make use of it in presentations etc, but please acknowledge me. If you do use it please reference the presentation / abstract as follows:

Petley, D.N., Rosser, N.J and Parker, R. 2009. Quantifying the impacts of landslides on society. Eos Trans. AGU, 90(52), Fall Meet. Suppl., Abstract NH52A-04.

AGU Day 4: landslide sessions

There were two landslide sessions at AGU today, both in the afternoon. This was the part of the meeting about which I was most excited, but sadly to be honest the reality did not meet my expectations. I won't write about all of the talks here, just a selected few.

First up was Bruce Malamud, who talked about the statistics of landslide clusters. This is great work, showing the remarkable similarity between landslide distributions for different multi-landslide events from different areas with different triggers. Helpfully he sumarised the ket implications of this work:

• It provides thae basis for a multi-landslide magnitude scale (like the Richter scale for earthquakes;
• The average landslide area in a multi-landslide event appears to be about 3000 square metres;
• It is at least theoretically possible to determine a volume of sediment production based on a volume area relationship;
• If large landslides arfe preserved in the landscape it shoukd be possible to determine the number of landslides that have been “lost” from erosion (turned out to be >99% in both Japan and Italy)

Interestingly, rockfalls have a very different statistical distribution - I wonder why.

Next up was Colin Stark, who tried to explain aspects of this distribution, and in particular the so-called "roll-over" component (basically there are fewer small slides than one would expect) but building a model of rupture propagation. This was interesting for two reasons. First, the model makes some assumptions about failure that cause concern (but this may have been misunderstanding; second, he came over as being exceptionally unsure of his own model, which was a little odd. The model appears to me to make some predictions that are very testable indeed - so this should evolve quickly.

Larsen, Montgomery and Korup gave a really interesting paper on the role of materials in controlling volume –area scaling for landslides. They challenged recent papers that have suggested that this scaling is independent of material, showing that in fact the scaling is very different from soil than for rock.

The final paper that I want to mention here was by Goren on the absolutely amazing Heart Mountain landslide. This will be the subject of a future post - but basically she was talked about the mechanics of this landslide - surely the largest terrestrial failure of all time! Wow!

AGU Day 4 - session on Scientists' Communication on Critical Global Environmental Issues

I am going to split my AGU Day 4 report into two posts. This one will cover the session this morning on the communication of the science of environmental change, whilst the second post will cover the landslide sessions.

The first session that I attended this morning was an excellent, experience-led examination of how to communicate environmental science to the public and to policy-makers. The central theme was, perhaps inevitably, aspects of climate change and its associated impacts, with a great array of speakers giving different perspectives on their experiences.

First up was the man the denialists love to hate (well, perhaps second to Al Gore and on a par with James Hanson), Michael Mann. His theme was on communicating temperature change, and he started with a quip that he has spent a fair amount of time on that theme of late, which raised something of a belly laugh from the audience. His central theme was that the science case is now clear (he noted that AGW is not controversial scientifically, only societally), and the case is not hard to make to an educated and rational audience. So why is there a problem? Well, the issue is that it is clear that there is a group that wish to do no more than sew doubt in the minds of the public, and so repeatedly stir up ill-founded controversies about climate change. He referred to the ongoing hockey stick discussion - noting that as a poster-child of the debate it was inevitable that it would be attacked - but also noting that the case for anthropogenic warming does not rely on this dataset in any way, even though it continues to withstand attacks by its detractors.

So what can we do? Mann argued that all scientists need to get out there to make the case. Basically with a few exceptions the mainstream media has proven incapable of understanding the science, or at least of presenting the argument in a rational way. It is essentially up to the science community to sort this out - something of a rallying call to us all. I hope that the science community will step up to the plate.

Second up was Richard Alley, talking about ice sheets and sea level. As with his Bjerknes lecture there was a sense of extraordinary enthusiasm for his science. It was sad to see that once again he put a disclaimer on the front of his talk, noting that he was not presenting the views of Penn State (it is deeply troubling that senior academics are driven to have to do that). He started by showing a map of the effects of a sea level rise of 6 m on the eastern seaboard of the USA to show why sea level rise matters.

Alley argued that scientists have the best job description going (essentially to find out about things), but that arguing is part of the job – we are required to continually challenge, chase and discuss. Although absolutely correct, this does not play well with policy makers or the public, who don't like to feel that there is uncertainty about science.

He then went on to look at the IPCC predictions from 2001 on CO2, warming and sea level rise. He noted that they were pretty good for the first two, but sea level was too uncertain to be able to predict. In 2007 a better prediction was made, with caveats for the unknown components, but the reality is that sea level is rising faster than expected. He then went on to talk about how melt on the large ice caps can lead to increased rates of collapse either from the loss of buttressing around the margins or from the movement of melt water from the surface to the base. The upshot was that he suggested that a sea level rise of a metre or so by 2100 is not unreasonable from a science perspective, but much higher rates look unlikely unless there is an unknown gorilla in the room.

Whilst he didn't spend a huge amount of time on issues of communication directly, the talk was an excellent summary – essentially a masterclass in how to communicate the science.

Third up was a talk by Serreze on Communicating Arctic Change - essentially focusing on sea ice loss. Again, the key observation was that the ice loss trend was more rapid than the scientific models had forecast. He then spent some time thinking through the key aspects of science communication, which included:
  • The use of analogy – for example he showed some great maps from Donald Perovich showing areas of seasonal ice loss as a proportion of the land mass of Europe or N. America
  • The need to find themes that resonate for the community in question
  • The need to be open and transparent, and to use opportunities to explain properly the scientific process
  • The need to respond to misinformation and challenges quickly, but to do so in a thoughtful manner.
One of the questions asked about how the Arctic Sea Ice looked for next year. His response was that it doesn't look good right now as the surface area is well below normal, and the ice is very thin and young. However, it will all depend upon the weather, which is unpredictable.

Next up was a great talk on changes in hurricane intensity by Elsner. He noted that his recent paper on the increasing strength of the most intense tropical cyclones had caused a storm (I bet he has never used that joke before...). He showed strong evidence to demonstrate that although the overall number of hurricanes had not really changed, the strength of the most intense ones had increased, especially in the N. Atlantic. He noted that arguments that this didn't matter as the most intense hurricanes occur out to sea do not stand up to scrutiny. However, he also noted that in the Caribbean it may well be that the number of hurricanes declines with warming, but that the strength of those that do occur increases.

From a science communication perspective he noted that they had put their data and their code on a website, providing open access to allow people to test their ideas. This has been very effective.

The penultimate talk was on the impacts of growing levels of hypoxia (oxygen deficits) in the ocean, given by Whitney. He noted that higher levels of nutrients from the land mass (mostly from fertilisers and human waste) is driving oxygen deficits in the oceans that are impacting their ecology. He noted in particular that there is a tendency to replace fish with jelly fish and that the squeeze that reduced oxygen levels are placing on the usable habitats is making species such as tuna and sailfish more susceptible to predation. Perhaps the most interesting part of the talk was the demonstration that warming in the seas off eastern N. Asia is affecting oxygen levels off the west coast of the USA, which is now seeing invasions of organisms that favour low oxygen conditions. He then went on to look at issues of communication with stakeholders, noting the need to:
  • Identify issues to which the local community can relate;
  • Make story understandable;
  • Use case studies that the people can relate to;
  • Separate scientific analysis from advice on policy;
  • Prepare for resistance to the scientific message.
Finally, Gleick from Oakland talked about scientific communication. This talk was a neat summary of the key themes that had emerged from the other talks, with one difference - he was unconvinced that it is really easy to separate science from policy, although there is a need to differentiate between science and opinion.

I guess his key point was that policy makers need good scientists. There is only one thing worse than policies based on no science at all, and that is policy that is based on bad science. He noted that the fact is that we don’t communicate science well enough, or enough, or to the right audiences, or to the right audiences enough. He reminded the audience that the level of science education and awareness is low, which allows unnecessary controversies rage out of control. His final point was again that we all need to do our bit, and that we need better journalists!

In questions he made reference to the frustration of denialst blogs, and the need to respond to them, through reference to this cartoon - "someone is wrong on the internet":

Source of the cartoon:

This raised quite a laugh.

Overall, it was a great session with good speakers and an audience was engaged and interested. The sense across the hall was of frustration that the strength of the science of environmental change, and the threat that this poses, just is not getting through. The lack of scientific controversy on all the key planks that underpin our understanding of anthropogenic climate change is a clear theme of the conference. We really do need to try to find ways to ensure that policy makers and the public at large understand this.

Thursday, 17 December 2009

AGU Day 3

There were two highlights of day 3 for me, which to be honest did not have much of interest on landslides. The first was the Geobloggers lunch, which was attended by about 30 bloggers. It was interesting to put names to faces and also to learn about some new blogs. I was struck by two things: 1. the range of ages of bloggers - far less oriented towards the young than I had expected; and 2. the diversity of reasons for blogging. All were pretty sure that they got a lot out of it and were enthusiastic about the role that they play. I sense that organisations like AGU have a sense that blogs and bloggers need to be a part of their future, but don't know what that role would be. I am not sure that this session will have helped though!

The second was a session on coastal processes, mostly focused on erosion. In most cases erosion is a mass movement (landslides or rockfalls) process of course, so this was of direct relevance. I presented the work of my PhD student Emma Norman, and a post-doc in my team, Mike Lim, presented his work too. There was one other hard rock talk, a presentation from Limber and colleagues on the development of a model for coastal erosion, with an emphasis on how bays and headlands form. The headline message was interesting - that headlands can only be preserved if erosion is enhanced in bays by small pocket beaches, which provide sediment that increases the rate of abrasion on the cliffs. This is all very well, but the model appeared to over-simplify to the point of irrelevance. I sensed that the audience was pretty frustrated too, with a quite pointed question about the assumption that the geology is homogeneous.

After that attention turned to erosion in the Arctic. This session was started with a quite surreal presentation. Jesse Walker is one of the gods of Arctic geomorphology, so the convenors asked him to attend to give a presentation on this theme as background. Unfortunately he was unwell - I hope he recovers quickly - so one of the session chairs gave it on his behalf. A presentation of over 50 slides in 15 minutes by someone who did not put together the talk was not a great experience, although the presenter did the very best he could in the circumstances.

Afterwards there were four rather mixed presentations on the processes and rate of coastal erosion in Alaska. The underlying message is startling - erosion rates of 10-15 m per year are common at these latitudes. Given that most of the erosion occurs in a very short summer season, this gives retreat rates of 10 to 15 cm per day!!! One of the speakers, Cameron Wobus, presented some fabulous time-lapse photography over a full summer season to show this erosion. Just to give you an idea of how fast this erosion occurs, the image below is from his website, showing just a few years of retreat:

The real concern must be the response of this system to the rapid and massive changes in temperature and summer ice cover at these high latitudes. The eroding bluffs primarily consist of ice (dirty icebergs was one description of them). The warming is causing increases in erosion because:
1. The temperatures are warmer so the bluffs are thawing;
2. Under normal conditions at the start and end of the warm season the shore is protected by ice berms - basically marine ice that has blown onshore. These berms are smaller and are melting faster;
3. The loss of sea ice in the summer means that the waves have a larger fetch, so that they are larger and more powerful;
4. The melting season has become longer;
5. Sea level is rising;
6. The time period of ice free marine conditions each year is longer, so the sea erodes for more time;
7. The sea is warmer so can erode more quickly (much of the erosion is caused by the warm sea melting a notch in the bluff).

Overall the picture painted was alarming. As Tom Ravens said, most coastal erosion in the US is undertaken in the south where rates are generally small. The reality is that the serious problems are in the high latitudes, so there is a desperate need to reorient coastal research.

Tomorrow is the main landslide day. This appears to have a mass of stuff of interest. I can't wait. Tonight I have to work on a research grant proposal - what joy!

Wednesday, 16 December 2009

daveslandslideblog is two years old today!

Very kindly AGU decided to hold a birthday lunch to celebrate the second anniversary of daveslandslideblog (actually the geobloggers lunch, but who's counting?). I started the blog two years ago as a bit of an experiment to see what would happen, and have really enjoyed putting it together (and I have learnt a huge amount too). During this time I have made 437 posts. I put a web counter on the site in early April 2008, since when the site has had >236,000 page requests from >125,000 visitors. In 2009 to date there have been visitors from 177 distinct countries. Unfortunately, like almost all blog sites Blogspot/Blogger is currently blocked in China, which is the world's most landslide-prone country. This is a real regret, and may mean that I need to set up a satellite site on a dedicated website at some point.

The pattern of visitors through time is quite interesting too. I track this data on a daily basis (I am such a nerd!), and the long term graph looks like this (this is smoothed with a seven day filter):

The huge peak early on is the Wenchuan earthquake and then Tangjiashan. The subsequent peaks can all be traced to specific large landslide events that have attracted lots of page searches. The overall trend is strongly upwards through time, but is also very seasonal. The key factor appears to be university term/semester dates (especially in the US and Europe), with visitor numbers picking up when term/semester starts (see this autumn for example), and then declining towards the end of the semester/term. Finally, the recent drop off may look alarming, but is just part of the normal cycle - there were lots of landslides in the autumn this year, hence the big peak, which was inevitably followed by a decline, plus numbers have dropped off as the university semesters have finished.

So what is the future of the blog? For now I anticipate keeping it going, although at times it can be a struggle. I notice that the other landslide blog, the strangely-named Landslides under Microscope, appears to have ceased (last post in October), which is a shame. I don't anticipate any major changes (and I am continuing to resist putting adverts onto the site), not least due to time constraints, but we'll see. The most bizarre aspect of this has been the recognition over the last year that I am far better known for my blog than for my research - and I have really appreciated all of the positive comments that I have received from people I have met around the world.

Comments and feedback welcome - in particular, how can I make it better?

Ten years ago today - the Vargas landslide disaster

Today is the tenth anniversary of probably the greatest landslide disaster in living memory - the 1999 Vargas landslide catastrophe in Venezuela. On 14th-16th December 1999 a coastal storm in Vargas triggered multiple debris flows that swept onto the densely populated river deltas below. It is estimated that 30,000 people died.

The USGS summary of the event, which remains the best starting place to understand the disaster, is here, which is the source of both photos.

Note the size of the boulder wedged into the side of this building.

Having a bad day with a drilling rig!

Most of us who work with landslides have on occasions had to work on drilling - dangerous, dirty, difficult to deal with - and that's just the drillers! One of the most difficult problems is knowing where the drilling head is actually going. In this context there is an extraordinary story that I found through GeoPractNet, on a The Local, a Swedish news site, from 11th December:

Driver in shock as drilling crushes subway train
The red line in Stockholm’s subway was stopped for one hour on Thursday after a work team’s drilling punctured the subway tunnel and damaged a train.

The work team was drilling at Wollmar Yxkullsgatan, on Södermalm, to prepare for the installation of geothermal heating for a nearby hotel, but their drilling punctured the subway line and crushed the side panels of the driver’s carriage of a train that was waiting on the tracks. “They drilled right down onto a subway train,” Lars-Erik Baarsen, station officer at Södermalms Police, told news agency TT. After the workers had drilled to a depth of 20 to 25 metres, the team noticed that the resistance to the drill disappeared. “They then withdrew the drill and discovered that two-and-a-half metres of the drill was missing,” Baarsen said. Meanwhile, down in the tunnel, the driver of the subway train was shocked when the side panel of his carriage was suddenly crushed by something from above.

Fortunately the driver was unhurt. Of course this is not quite as spectacular as probably the greatest drilling rig error of all time, the extraordinary 20th November 1980 Lake Peigneur accident in Louisiana (read the link - you won't regret it), but it is quite amazing.

AGU Day 2

NB: I have posted a review of the session on Scientists' communication of critical global environmental change issues, in which both Michael Mann and Richard Alley spoke here:
Communication of critical global environmental change issues

There wasn't much in the Natural Hazards area of interest to me on Day 1, so I decided not to blog yesterday. There was a little more today - and lots later in the week - so here goes.

I spent the first part of the morning in the underground geoscience session, for which I gave the first (8 am) invited paper on our Boulby Geoscience project. The rest of the papers were pretty good, but beyond the scope of this blog. After coffee I went to a paper by Karnawati and her colleagues from Gadjah Mada University, the topic being the landslides triggered by the recent earthquake in West Sumatra. This was pretty interesting, although I was a little frustrated by the lack of detail. She noted that the area around Maninjau lake, to the north of the epicentre, was most seriously affected by landslides. The lake is the remains of a volcanic crater, so I guess the materials may well be pretty susceptible to failure, but she noted that the main landslides all lay close to faults or lineaments, which is an ingteresting observation in this setting (I wonder whether essentially everything is close to a lineament in such a dynamic setting?). There is now a serious threat from post-seismic re-activation of the slides - not a surprise - and the author gave some examples showing that this has already started to happen. The rainy season starts about now. Dr Karnawati noted that a key task is therefore identifying safe areas for relocation of affected villages - it appears that this is going well - and community engagement with disaster risk reduction.

I then moved sessions to a presentation on the Slumgullion landslide by Schulz from USGS (see my earlier post on this work here). The first part of this talk reiterated the extraordinary observation that the daily movement of the slide is correlated with atmospheric tides, as per their recent paper. In the second half Schulz noted that the processes and morphology of large landslides and faults are similar, but that it is much easier to study landslides. To gain a better understanding of movement processes (presumably relating to both systems) they had positioned an array of seismometers on the landslide for week in the summer. Although they had power problems early on (so often the case with seismometers in my experience), the later data showed that the seismic events also correlated with the movement events and thus the atmospheric tides. The initial analysis has focused on the harmonic events that they recorded - which are likely to be related to fluid flow. As these occurred during periods of low atmospheric pressure it may be that their hypothesis about atmospheric tides inducing pore water movements is on the right lines.

This was followed by an entertaining talk by Dick Iverson, also from the USGS, who sought to explain stick-slip movements on landslides. These movements - in which the slide appears to move and then stop repeatedly - are something of a conundrum. The problem is that stick-slip in faults is usually explained by the storage of energy in elastic deformation, but in a slope with no cohesion it is hard to do this. Iverson demonstrated mathematically that the behaviour is related to the relationship between pore pressure and movement - basically when the slide starts to move diffusion allows the pore pressure to drop, which slowly reduces movement. When the slide stops pore pressure builds again until movement restarts. I think that the idea that this is the key mechanism has been around for a while, but this was the first time I have seen it modelled properly.

After lunch I went to the Bjerknes Lecture - one of the big set pieces - which was given by Richard Alley from Penn State. He brought a geological perspective to the link between atmospheric carbon dioxide and temperature. He started the lecture by noting the ongoing harassment of climate scientists by the denialist camp - giving an example of a demand made to his university that he be fired for continuing to claim that carbon dioxide causes temperature change. He presented this with great humour and grace, but the underlying message about the way that scientists are being treated was clear, and was a great concern. At a time when the denialist scientist de jour, Iam Plimer, has been embarrassed in a debate with a journalist, and the high profile campaigner Christopher Monckton has been caught on video calling campaigners "The Hitler Youth", and then has brazenly denied it, the sense that the science community is under siege is clear.

Anyway, back to the lecture. Alley reviewed climate changes in the geological record, going back 4.5 billion years. His central point was that in almost every case carbon dioxide has emerged as the smoking gun in terms of causation, and indeed that it is essentially impossible to explain the observed changes without carbon dioxide acting as the key forcing. This is true for the "faint young sun paradox" (4.6 billion years ago), the snowball earth period, the late Permian extinction period, the mid-Cretaceous "Saurian sauna" period and the Palaeocene-Eocene thermal maximum for example.

Alley did note that although his own data shows that for the last 500,000 years carbon dioxide and temperature have moved in lockstep together, sometimes that CO2 lags behind temperature. He observed with great amusement that this has led some denialists to claim that CO2 is a thermometer not a cause of temperature change. Of course the glacial - interglacial cycle is driven by the cyclicity in the orbit, but Alley noted that the magnitude of the temperature changes can only be explained using a carbon dioxide feedback mechanism. He ridiculed the idea that because CO2 sometimes lags temperature it cannot be the cause of warming. He compared this to debt associated with credit cards, saying that if he went out and spent a modest amount on his card he would end up owing the credit card company money. If he didn't pay this off then punitive interest rates would mean that his debt rapidly spiralled out of control. In the end his bankruptcy would be the result of the debt increasing due to the interest rates. The application of the interest rates lagged behind the spend, but still caused the insolvency. He likened the orbital cycle to the initial spend but the carbon dioxide to the interest. So, clearly carbon dioxide can drive the temperature, even if it didn't act as the initiator of the process. Of course you need an explanation as to why this would be the case - but that explanation is simple as its is an application of straightforward radiative physics, as demonstrated two centuries ago.

Next he noted the huge progress that has been made in understanding the link between temperature and CO2 in recent years. He noted that when he was a young scientist there were several examples in the geological record in which there was global warmth but no apparent corresponding high CO2 level in the atmosphere. As sampling and measurements have improved it has been shown that almost all of these cases do in fact have high carbon dioxide concentrations. Just a few anomalies remain - most notably the warm period in the Miocene. However, very recent (peer reviewed) research is now showing that CO2 levels were high, although more work is needed.

Finally, he took a few minutes to show that arguments for other causal factors for global temperature change in the geological record just don't hold water, He used the example of cosmic ray flux, a favourite of the denialist blogosphere a year or two ago. He showed that the geological record globally demonstrated a huge spike in cosmic ray flux at 40,000 years BP, but that temperature did not respond at all.

He finished with a very simple message - the geological record shows that carbon dioxide is the key factor that controls temperature. Other factors do operate, but the CO2 signal consistently dominates. He noted that the geological record shows that over a timescale of few centuries timescale a doubling of CO2 results in a warming of about 2.8 C, which is consistent with the IPCC figure for climate sensitivity. He made a very bold statement that if the key factor that explains the temperature record in the geological record is CO2 - without carbon dioxide concentration changes it is impossible to explain the observed behaviour.

Overall it was a great lecture - highly entertaining, informative, challenging and coherent. The audience for the talk was huge, and greatly appreciative of the quality of the presentation.

The lecture should be available as a podcast in due course here:

Do take a look - you won't regret it (not least for the amusing first five minutes).

NB: I have posted a review of the session on Scientists' communication of critical global environmental change issues, in which both Michael Mann and Richard Alley spoke here:
Communication of critical global environmental change issues

Sunday, 13 December 2009

Poor mining management was the cause of the Sigou Village landslide in China

Back in August I posted a couple of times on a fatal landslide in at Lifan (sometimes written Loufan) in Sigou County in Shanxi Province in China (the posts are here and here), which appeared to be a mining related flowslide that killed 45 people. Yesterday the People's Daily website ran a story explaining the results of the enquiry into the landslide:

"Illegal production and slack safety oversight were main causes for the landslide in north China which killed 45 in early August, investigation result showed Friday. Some 37 had been detained for investigation as they were allegedly responsible for the fatal landslide in Shanxi Province which left another one injured, and resulted in an economic losses of 30.8 million yuan (4.53 million U.S. dollars), according to a statement issued by the investigation team led by the State Council, or cabinet.

Thirteen people including Yan Quewa, head of a local iron mine where the tragedy happened, have been prosecuted, while 24 others have received administrative and Party disciplinary punishment. The local mine was imposed a fine of 5 million yuan. On Aug. 1, the landslide toppled a waste dump of the mine and buried Sigou Village of Loufan County in the suburbs of the provincial capital Taiyuan. The waste dump, within 200 meters from the village, belongs to Jianshan Iron Mine. It was operated by Taiyuan Iron and Steel (Group) Company Ltd.

Chinese law stipulates such dumps should be at least 500 meters away from residential areas and should have embankments or walls to contain dust or prevent landslides. The dump near Sigou, however, had none."

The mining and quarrying industry in China has a dreadful safety record.

Statement from the UK Science Community on climate change

In the aftermath of the UEA email hack, the UK science community have issued a statement regarding climate change. This is hosted by the UK Met Office web site:

The statement, upon which I am proud to be listed, says:

"We, members of the UK science community, have the utmost confidence in the observational evidence for global warming and the scientific basis for concluding that it is due primarily to human activities. The evidence and the science are deep and extensive. They come from decades of painstaking and meticulous research, by many thousands of scientists across the world who adhere to the highest levels of professional integrity. That research has been subject to peer review and publication, providing traceability of the evidence and support for the scientific method.

The science of climate change draws on fundamental research from an increasing number of disciplines, many of which are represented here. As professional scientists, from students to senior professors, we uphold the findings of the IPCC Fourth Assessment Report, which concludes that ‘Warming of the climate system is unequivocal’ and that 'Most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations'."

This statement has been signed by >1700 scientists from the UK.

Saturday, 12 December 2009

The annual cycle of fatal rainfall-induced landslides

On the flight to San Francisco yesterday I spent some time preparing for one of my AGU presentations - the one on Friday on the impact of landslides on society. I have been looking at my landslide fatality database, which now stretches back for over seven years. There is still some way to go with this to really understand long term trends in fatal landslides, but the dataset is now big enough to do some quite interesting things.

I have divided each year into five-day blocks (we often call one of these blocks a bin), and then taken the average number of recorded fatal landslides within that bin over the seven year period. So, bin number one is 1st to 5th January, the second is 6th to 10th, etc. I have then looked at the cycle through time by plotting a graph in which I have smoothed the data using a 25 day filter - this is a noisy dataset, so this is needed given the comparatively short window:

The resulting graph is pretty interesting I think. First, the peak in aggregate recorded fatal landslides (the black line) clearly occurs in the northern hemisphere summer, the minimum is around about now. The peak is actually on about 25th July. This is coincident with the peak of the SW Monsoon over the Indian subcontinent. Notice though that the graph is asymmetric - i.e. it rises to a peak more quickly (about 100 days) than it then declines (about 150 days). I assume that this is because in this post-SW monsoon peak the influence of the monsoon in East Asia and of tropical cyclone landfalls becomes significant.

The minimum period coincides I think with the onset of winter in the northern hemisphere (which is a dry period for many of the most landslide prone areas) but is before the rainy season really gets going in SE Asia. By early January the rains in for example Indonesia are really under way, and the occurrence of landslides increases.

The standard deviation is a measure of variability between years. So, if for a specific bin the number of fatal landslides was always three then the standard deviation would be low. If however one year there were none, the next six, the next ten and the next two then the standard deviation would be much higher. It is interesting that as the average number of fatal landslides increases in the N. Hemisphere summer so does the standard deviation - this is to be expected. However, in the post-peak period the standard deviation remains high for a while before declining. I think that this probably reflects the influence of tropical cyclones in this period, which tend to landfall rather sporadically but then to cause many landslides over a small area. Over the seven year period many of the bins in this period have been affected by a tropical cyclone.

I hope to see you at the session!

Thursday, 10 December 2009

Landslides in Art Part 3

This is the latest in my occasional series on the depiction of landslides in art. The previous parts of this series are here:
Part 1
Part 2

Today I turn to a classic painting from the 19th Century, painted by Lord Frederic Leighton, entitled "On the Coast, Isle of Wight":

This oil sketch, which hangs in the National Gallery in London, shows a stretch of the Isle of Wight coastline in southern England. The view is from Ventnor looking towards the west towards Saint Lawrence. The coast here is a famous area of landslides known as The Undercliff - the landslide morphology is clearly visible in the painting.

Wednesday, 9 December 2009

Tuesday, 8 December 2009

Landslide in the Lake District (Cumbria) in the recent floods

Thanks to Rhiannon George of Meridian Geoscience for her help with this - the images and most of the information have been provided by her.

The press coverage of the recent (English) all time record rainfall event in Cumbria (in the area most commonly known as the Lake District in NW England) has focused upon the disastrous floods at Cockermouth. However, less well reported is the extensive devastation up in the headwaters of the catchment, and in adjacent areas, where landslides and floods occurred extensively. Rhiannon kindly provided information about two shallow landslides that occurred at Howtown, near to Ullswater. The pictures that Rhiannon has provided are pretty impressive:

This is one of the two slides, which appears to have split into two runout tracks.

This, and the image below, is the other.

These are comparatively small, shallow failures. They appear to have initiated quite low on the slope as shallow slips that have then eroded and entrained the top metre or so of the ground downslope. The cessation of movement appears to be gradient controlled - i.e. the point at which the slope angle decreased to the point that erosion ceased and deposition started.

So why did these slides occur at this point at this time? Well, the rainfall that this area received broke the English record for one day and five day precipitation, so it is not surprising that many floods and slides occurred. The Google Earth image of the site of the second slip, taken prior to the landslides, also helps:

The landslide initiated in the highly disturbed and gullied area on the lower slope. This image suggests to me that the slope was under some distress prior to the failure. In other cases we have found that such gullying and disturbance, which is not seen on the adjacent slopes,
has indicated that for whatever reason, water was flowing into this area underground, maybe through a natural soil pipe network. In the heavy rainfall the flow of water can exceed the capacity of such pipes, causing the local ground to saturate and the the landslide to initiate.

This is an image of the head scarp of the second slide. If the above mechanisms is correct then it might be possible to see pipes in this area, but in the aftermath of the slide they may have become clogged, so some digging around would be needed.

Monday, 7 December 2009

Landslides in Art Part 2: "Landslide" by Chiura Obata

This is the second of my occasional series on landslides in art. Part One can be found here.

In 1941 the Japanese-American landscape painter Chiura Obata produced the following painting, simply named "Landslide":

For me this is a powerful image for two reasons. First, it manages to communicate the horror of landslides, and in particular the chaotic engulfing of everything in its path. Second, of course 1941 was a very difficult time to be a Japanese-American; indeed between 1942 and 1943 Obata was detained by the authorities, during which time he was attacked and hospitalised for a fortnight.

Friday, 4 December 2009

Landslides along the Panama Canal

The Panama Canal is truly one of the engineering wonders of the 20th Century. This 80 km long waterway was completed in 1914, cutting 13,000 km on the sea route between New York and San Francisco. However, the fearsome cost of the canal in human lives is jaw-dropping - it is estimated that 27,500 people lost their lives during construction, primarily as a result of malaria.

However, it is less well-known that probably the largest technical problem arose from landslides that occurred in a section of the canal route known as the Gaillard Cut (sometimes called the Colebra Cut), which is the continental divide, about 13 km from the Pacific Ocean. Unfortunately, at the time of construction the strength of the rock was over-estimated, primarily because the engineers (understandably) did not anticipate that the materials would weaken due to oxidation after excavation.

The upshot was a series of large and very damaging landslides that caused the required amount of excavation to increase dramatically. There are two wonderful presentations available online of the problems that landslides caused along the Gaillard Cut, including contemporary photographs of the various failures. The first, prepared by J. David Rogers of the University of Missouri - Rolla is available here:

The second, prepared by Maung Myat, is hosted at the same site as the above:

To give you a taste of the contents of these presentations, here are two images from the former, showing slides along the Gaillard cut:

More recently, the Panama Canal underwent a widening programme consisting of widening the watercourse to 192 m along the straight stretches and up to 222 m on the bends. This work was completed in 2001 to budget and on time. A 1988 review of the geotechnical challenges posed by this work is available here:

Today the satelite imagery gives no indication of the magnitude of problems posed by the Gaillard Cut:

Thursday, 3 December 2009

Landslides at the AGU Fall meeting

Like 16,000 other Earth and environmental scientists, I am heading off next week to the Fall Meeting of the American Geophysical Union, which is held in San Francisco. I have attended several times before - it is an amazing meeting - but this year there is added spice provided by the new Natural Hazards grouping, which means that there is far more of interest for me. The landslide specific session is entitled "Landslide Prediction and Assessment: Mechanisms and Controls on Size Distribution", wich consists of two oral sessions and a large poster session, all on Thursday 17th December. The oral sessions have a great line-up, including:
  • A General Landslide Distribution for Triggered Event Landslide Inventories from 100-10,000 Landslides (Invited). B. D. Malamud; F. Guzzetti; D. L. Turcotte
  • Landslide scaling and magnitude-frequency distribution (Invited). C. P. Stark; F. Guzzetti
    Material Matters in Landslide Volume-Area Scaling . I. J. Larsen; D. R. Montgomery; O. Korup
  • Erosion effects on the size and mobility of granular avalanches and landslides. A. Mangeney; O. Roche; O. Hungr; N. Mangold; A. Lucas
  • Acoustic Emissions During Progressive Failure of Earth Materials – Can we “Hear” The Onset of a Landslide?. G. K. Michlmayr; D. O. Cohen;
  • A framework for spatial and temporal prediction of shallow landslides induced by rainfall (Invited). J. W. Godt; N. Lu; B. Sener-Kaya; R. L. Baum
  • The early warning system of landslides and sediment runoffs using meteorological condition including rainfall-soil moisture index (Invited). T. Kubota; I. C. Silva; H. Hasnawir
  • New techniques for landslide hazard assessments: opportunities, methodology, and uncertainty. D. B. Kirschbaum; C. D. Peters-Lidard; R. F. Adler; Y. Hong
  • Prediction of Rainfall-Induced Landslides . F. Nadim; F. Sandersen
That is quite a line-up!

I am giving three papers at the meeting. These are:

Session Title: EP33C. Coastal Geomorphology and Morphodynamics III
Location: 2004 (Moscone West)
Start time: Wed, Dec 16 - 1:40 PM
EP33C-01. Energy Delivery to Cliffs from Waves, Tides and Storms. E. C. Norman; N. J. Rosser; M. Lim; D. N. Petley

Another paper in the same session will be delivered by a researcher from my team, Mike Lim:

EP33C-03. A quantitative analysis of rock cliff erosion environments. M. Lim; N. Rosser; D. N. Petley; E. C. Norman; J. Barlow

Session Title: H21J. Rocks, Fractures, Fluids, and Life: Insights From Subsurface Observatories I
Location: 3005 (Moscone West)
Start time: Tue, Dec 15 - 8:00 AM
H21J-01. On the development of an underground geoscience laboratory at Boulby in NE England (Invited). D. N. Petley; N. Rosser; J. Barlow; M. J. Brain; M. Lim; M. Sapsford; D. Pybus View Pres.

Session Title: NH52A. Extreme Natural Hazards: Risk Assessment, Forecasting, and Decision Support III
Location: 2006 (Moscone West)
Start time: Fri, Dec 18 - 10:20 AM
NH52A-04. Quantifying the impacts of landslides on society. D. N. Petley; N. Rosser; R. Parker View Pres.

And I am involved in a series of posters too:

Session Title: EP43A. Coastal Geomorphology and Morphodynamics IV Posters
Location: Poster Hall (Moscone South)
Start time: Thu, Dec 17 - 1:40 PM

Session Title: H23E. Rocks, Fractures, Fluids, and Life: Insights From Subsurface Observatories II Posters
Location: Poster Hall (Moscone South)
Start time: Tue, Dec 15 - 1:40 PM
H23E-0997. The Boulby Geoscience Project Underground Research Laboratory: Initial Results of a Rock Mechanics Laboratory Testing Programme. M. J. Brain; D. N. Petley; N. Rosser; M. Lim; M. Sapsford; J. Barlow; E. Norman; A. Williams; D. Pybus View Pres.

Session Title: NH43C. Remote Sensing of Natural Hazards III Posters
Location: Poster Hall (Moscone South)
Start time: Thu, Dec 17 - 1:40 PM
NH43C-1350. Controls on the spatial distribution of landslide hazards triggered by the 2008 Wenchuan Earthquake, Sichuan Province, China . R. N. Parker; N. J. Rosser; D. N. Petley; A. L. Densmore View Pres.

See you there I hope!


Forthcoming conference in Rapid Mass Movements

CCES in Switzerland will host a conference entitled:

"Triggering of Rapid Mass Movements in Steep Terrain (Mechanisms and Risks)"

The meeting is to be held from 11th to 16th April 2010 at Monte Verità in Switzerland.

Details are available here: