Forthcoming meeting at the Royal Geographical Society: Natural disasters: how can we improve?

The RGS will host a meeting on 25th May entitled:

"Natural disasters: how can we improve?"  

Chair:
Martin Bell, UNICEF Ambassador for Humanitarian Emergencies

Speakers: 
Dame Barbara Stocking, CEO Oxfam GB
Cameron Sinclair, Founder of Architecture for Humanity

Date: 25 May 2010, 7pm
Location: Royal Geographical Society, London
Tickets: £10, £7 (RGS-IBG members)
Details: www.21stCenturyChallenges.org

EGU Day 3 - Dave Keefers Sergey Soloviev lecture

On Wednesday morning I attended the presentation by Dave Keefer, the winner of the Sergei Soloviev medal.  Dave has single-handedly defined the field of seismically-induced landslides, an area of great importance.  In the lat five years alone such slides have killed over 50,000 people, and the ongoing landslide activity in the aftermath of big earthquake events means that the legacy of the shaking continues for decades after the main shock.

Dave started by showing that in general losses from natural hazards are increasing both in social and economic terms.  Seismically-induced landslides are a key component of this.  Dave looked at the historic development of research into seismically-induced landslides, dividing the last century into three key phases. The first era was defined as from the beginning of time to 1947, during which time it was increasingly recognised that there was a problem with earthquake induced landslides.  However, documentation of events was patchy in almost every case,

The second era was defined as 1948 to 1988, during which “complete” mapping of landslides was initiated, including the development of quantification of landslide impacts.  He highlighted three key events in this period:

1964 Prince William Sound earthquake in Alaska, which saw landslides over an area of 270,000 square kilometres, causing 56% of the economic losses and at least 45 0f 130 deaths in the earthquake.
1970 Yungay earthquake in Peru that triggered the Huascaran landslide, which killed 6000 people and travelled over 180 km.
1976 Guatemala earthquake ,which triggered over  50,000 landslides over an area 16,000 square kilometres.

Dave used these data to write 1984 paper based upon 40 historic earthquakes with good data across a wide range of magnitudes with a broad geographic distribution.  This paper is of course a classic and, remarkably, subsequent analyses using modern techniques to generate far better datasets have hardly changed the main findings.  It is rarely reported that the study also looked at landslide type, finding that 86% of triggered mass movements were disrupted slides and falls on very steep slopes. 

Dave also highlighted the importance during this period of the development of the Newmark analysis, which then allowed the development of hazard maps over large areas.

Finally, Dave highlighted the modern era, which has seen detailed mapping of landslide events plus site specific data collection, plus some palaeo analyses.  Dave highlighted three key earthquakes:
1989 Loma Prieta, which saw the first GIS based landslide inventory, in this case for the Santa Cruz mountains.  This showed an exponential decay of landslide concentration with distance from epicentre and from fault.
1994 Northridge: similar analyses but more detailed than Loma Prieta, but this time developed to produce a regional scale Newmark analysis based hazard map, calibrated using the actual landslides.
1999 Chi—Chi earthquake in Taiwan, which remains the best studied seismically-induced landslide event of all time.  Most up to date analyses suggest that 13,000 landslides were triggered.  Dave described this events as unleashing an explosion of research on this topic.

Finaly, Dave highlighted the need to overcome our current limitations of knowledge, in particular highlighting a number of research needs:

1.    Urgent need to locate accelographs onto slopes to measure ground motion modification by slope properties;
2.    The need to develop automated mapping of landslides from remote sensing to allow the development of proper inventories;
3.    The need to further develop slope-stability and hazard mapping techniques;
4.    More detailed modelling and analysis of landslide mechanisms
5.    The need for complete mapping and characterisation of a great (M>8) earthquake, when such an event occurs.
In questions Dave highlighted the urgent need for better data from landslide inventories and on site specific effects.

Accelerating rate of filling at the Attabad landslide in Hunza, Pakistan

The rate of inflow at the Attabad landslide has now started to increase appreciably as warming and snowmelt really get underway.  The graph below shows the daily rate of water increase, using data from the Focus measurements (the daily data are available on my monitoring website):

The rate of increase for the last two days with data has been 89 cm per day.  At that rate the water will overtop on about 24th May, although clearly there remains considerable uncertainty around this date.  It is unlikely that we have yet achieved maximum inflow rate, so the date could easily move forward again.

As of yesterday the remaining freeboard was just 17.64 metres, as this Pamir Times image from a couple of days ago shows:

Just above the boats you will see that there is a small terrace on the landslide deposit.  This is the water level point at which I recommended in my report that should trigger the move to an increased alert state.  We should now start to see an increase in seepage rate as the thickness of the landslide barrier decreases markedly.

EGU Day 2 - afternoon sessions on vulnerability, plus Italian landslide statistics

In the afternoon of Day 2 I decided to focus more on the human / vulnerability side of natural hazards, and went to two sessions on Social Sciences in Natural Hazards Research.  First up was Katie Oven from Durham, with a number of co-authors , who talked about landslide hazards in Nepal .  Now I was the principal supervisor of this research, but I am going to write about it anyway, so forgive me but understand my bias!  Katie set out to look at some fundamental questions around the interactions between landslides and people in upland Nepal. 

Katie started with the work that we published a few years ago on the occurrence of fatal landslides, which we wrote up for Natural Hazards (paper available here), and in particular on the reasons why Nepal is suffering an increase in landslides.  The question is whether this is due to physical problems – e.g. climate change, deforestation – or human factors – e.g. population growth, urbanisation.  The key issue that underpins this is that landslide vulnerability is far less well understood than is the physical hazard.  So Katie wanted to know who is vulnerable to landslides and why are they vulnerable.   To do this she looked at the upper Bhote Kosi landslide in Nepal.  In this paper she presented three case studies, all of landslide prone communities.  She found that in recent years (the last 20-30 years or so) there has been a major migration of people from  upper hillslope communities to live by the  road.  This changes their exposure to hazard, in particular putting them into the path of high energy flows that are very capable of killing people.  It is clear that people have made a conscious decision to increase their vulnerability to landslide risk because the location decreases vulnerability to other risks – i.e. they have better access to health, education, income-generating opportunities.

So, the key thing is that there is a tendency in the science community to over-estimate the role of natural hazards compared with other risks in the lives of the vulnerable people.  Decisions might appear to be illogical viewed through only the prism of hazards, but once the range other risks are taken into consideration moving into an area of higher hazard looks entirely sensible.  This of course means that we probably need to think differently about how to manage hazards in a way that really works for people.  Unfortunately, all bets will be off when Nepal suffers the earthquake that is well overdue; that is a truly terrifying possibility, but again the risks to individuals need to be considered alongside the everyday risks that the people in this are face. 

The next talk that caught my eye was that of  Marjory Angignard and her co-author looked at now affected people perceive natural hazards.  This study came out of the Mountain Risks training network project of 17 PhD students and post-doctoral researchers across Europe, working together to try to understand differences between public and expert cultures of risk.  The study presented here compared Italian and French perceptions of risk from natural hazards, and found quite large differences not only in the hazard perception (not surprising given that the hazards are different) but also that people’s source of information differed substantively between the two, presumably representing deeper cultural differences. This means that there is no silver bullet that solves the communication problems; for each area the approach needs to be locally tailored.

Eric Lindquist and Katrina Mosher-Howe looked at natural hazard problem and solution definition in the news media, based upon a study of reporting of Tropical Storm Allison in the USA.  Alison brought  five days of rainfall from 5th to 9th June 2001, inducing 22 fatalities, 70,000 houses flooded, the death of 30,000 lab animals in a flooded basement, and $5 billlion of damage.  In particular the storm caused very serious flooding in Houston.  The study looked at 500 news articles on the event from the Houston Chronicle, examining them for variables such as cause, impact, damage, blame, solutions, manifestations of harm, etc. 

A number of things came out.  First, as expected there was a very clear exponential decay in coverage of the event over several years.  Most media attention focussed on home-owners, even though this group was not the most affected.  News articles tended to blame local government, there was little blame to developers because this is a very powerful group in Houston.

Major causes articulated the causes as being associated with physical/topographic factors and drainage problems – i.e. the hazard was seen as an external process imposed on a vulnerable population.  Lots of emphasis was placed on poor infrastructure – e.g. poor drainage, etc.  However, the solutions proposed in the articles focussed on “soft” approaches – i.e. economic, development, social, infrastructure, political changes, and suchlike.

 They then looked at what actually happened in the aftermath of the event.  Interestingly, most of the changes were associated with “hard” engineering – for example Harris County spent $750 million on improvements and the City of Houston spent $250 million.  The soft solutions were poorly resourced.  Thus, there was a mismatch between the proposed solutions and actual answers. 

Very interesting stuff, but I wonder whether that newspaper has a particular political perspective that might have influenced its coverage?

Alessandro Trigila and colleagues focussed on the impact of landslides on urban areas and infrastructure in Italy, based upon the nationwide, amusingly titled but hugely impressive IFFI project.   Some of the statistics on landslide hazards in Italy defy belief:

• The project has documented 480,000 landslides covering 20,700 square km
• Landslides represent 6.9% of the land mass of Italy
• 5708 Italian municipalities are affected by landslides, representing >70% of the total number.  
• 992,403 people are at risk, representing 1.7% of population).
• There are 43,621 hazardous locations along the transportation network.

The Attabad landslide: well, that's me told then!

The Pamir Times is running an interesting story today that clearly puts me in my place!  Here it is:

“Hunza River lake would not burst, David Petley’s report is wrong”: 

Asad Zamin, DC Gilgit
Posted on May 5, 2010 by Pamir Times

PT Report

Gilgit, May 4: The report prepared by Dr. David Petley is wrong and the lake formed on Hunza River is not going to burst. Government experts have rejected any chance of lake burst.

These views were expressed by the Deputy Commissioner of Gilgit district while briefing the media about evacuation plan set up for downstream areas, at his office. He said that necessary measures are being taken because 1300 households in Hunza – Nagar and Gilgit district would be affected by the a flood that is likely to triggered by over-topping of the lake.

“There is no chance of lake outburst and we request the local populations to stay calm”, the DC, who also is the director general of district disaster management authority.

“Government of Pakistan, FOCUS Pakistan and other relevant institutions have teamed up to ensure safety of lives and property to the optimum in case of heavy flood that might be caused by over – topping”, Asad reportedly said. He also said that the 10, 000 internally displaced people will be sheltered at nine relief camps after evacuation to safer locations.

Commenting on removal of the Bailey Bridge in Danyore and the resulting isolation of Danyore and other localities he said that the Jalalabad – Danyore Road will be paved and expanded within a week to ensure supplies for Hunza – Nagar district.

The commissioner emphasized that the measures are being taken as precaution.

It is pertinent to note here that Dr. David Petley has calculated June 9 as a probable date for the spillover to take place, while also saying that the actual date might be earlier depending on volume of the water in the Hunza River. He has also said that there is a substantial risk of dam burst with the passage of time.

Three points to note.  First, my report does not say that a burst is inevitable, but that the risk is substantive.  Second, my report recommends precautionary actions.  Third, it is a brave man who says "there is no chance" about any natural system.

There are many readers here who are well-qualified to comment on the situation, so I invite you to do so.  To help I reproduce here an image from the Pamir Times yesterday showing the spillway:

EGU Day 2 - morning landslide sessions

For this year's EGU General Assembly I intend to only blog on talks that really catch my eye.  This morning there was a fabulous session on Landslide Forecasting with series of great talks.  Of these,  Samuele Segoni and colleagues presented a very interesting paper on a hugely ambitious project to develop a regional landslide warning system for Tuscany based on rainfall thresholds.  The project appears to be extraordinarily successful – it appears to work with very few false alarms or missed forecasts.  However, to do this the area had to be split into 25 warning zones, using c.330 rain gauges.  As most such systems do, the approach uses an intensity – duration power law relationship. Thus, to make these systems work requires a huge infrastructure.  Interestingly, now that it is clear that such a system can work from a technical perspective, the emphasis needs to shift to the societal problems of trying to disseminate warnings effectively, and getting people to react appropriately to them.  That is a real challenge.

The second paper that caught my attention was by Peter Lehmann and Dani Or, looking at precursor events in the initiation of landslides using concepts of self-organised criticality.  Their starting point was that in the home country, Switzerland, 6% of landscape is prone to instability. In 2005 a rainfall event triggered over 1000 landslides, causing damage estimated at over $3 billion.  Essentially they seek to explain landslide initiation by considering processes that provide a cascade effect, as in the sand pile models of criticality.  Here they model the landslide as being controlled by fibres and fibre bundles, which are analogues of the loss of strength of the landslide material.  They showed that precursor events can be observed as weakening and breakage of the bundles occurs, replicating observed behaviour.  It was very neat and very interesting, and provided potentially important insights into failure initiation.

Next up was Nejan Huvaj-Sarihan from Turkey, presenting her doctoral work undertaken at the University of Illinois.  This was an experimental investigation of failure time prediction in landslides, using creep-rupture as the basic concept.  She built a simple direct shear machine to investigate the creep-rupture process, and showed two key things:
1.    She observed creep at even very low shear stresses (<20% peak strength).  She used this to infer that all slopes creep, which is correct;
2.    She could initiate creep rupture failure when factor of safety was greater than one, but that the time to failure depended upon how close to FoS = 1 the system is. 
Using this data she then explored whether failure prediction can be undertaken using the range of techniques available, concluding that it can in all three cases.  This is a very neat study, although I am surprised to see creep-rupture in pre-sheared materials.

During questions someone made the point that there is a “geotechnical disease” to ignore the time element.  This was a very provocative statement, but is quite correct.

Oded Katz and his colleagues followed this up with a model based study that sought to look  material disintegration in controlling the geometry and size of landslides.  Again, this was a neat bit of work that came to some key conclusions.  For me the most important one was that the power law roll-over in landslides is indicative of the change in material properties, and that the collapsing of the power law relationships onto each other occurs because there is such a narrow range of residual strengths available in natural systems.  However, the talk also demonstrated beautifully that failure is associated with breakage of inter-particle bonds.  Initiation of movement in the models occurred only when shear surface was fully developed through bond breakage, which links with the previous three presentations.

Those three talks (Lehmann, Huvaj-Sarihan and Katz) together present an extraordinary level of insight into landslide processes that on their own justifies my attendance at the meeting.

In the after coffee session I would like to highlight just one talk, that of Monique Fort and her colleagues on debris flow initiation in the Ghatte Khola watershed of Nepal.  This watershed suffers extraordinary pre-monsoon rainfall events – she quoted a storm in 1974 that had over 300 mm of rainfall in an hour – can this really be right?  Anyways this small ( 7.8 square kilometre) catchment generates debris flows that result from shallow slides that block the tributary valley, then collapse, creating flows.  These in turn enter the main valley, which is then blocked in turn, and another flow occurs down the main channel.  There are two interesting things here – first, how a small failure can initiate a bigger flow that in turn blocks the main valley, generating an even larger one – who would try to forecast hazards when this sort of situation occurs?  Second, she highlighted the ignorance of road builders in Nepal to these processes, resulting in inappropriate designs that then fail in the next storm.  This is a hobby-horse of mine; I could not agree with Monique more.

New photograph of the spillway at Attabad

The Pamir Times today has some images of the spillway and dam at the landslide dam at Attabad, in Hunza, Pakistan.  This picture shows the spillway itself:

Constructing such a spillway has been a Herculean effort by all involved, who deserve considerable credit.  However, the structure is narrower and smaller than is ideal, and the materials at the base in this section at least appear to be the clays. 

Once again I advise that a very cautious approach needs to be adopted when the water flows through this spillway, especially during a flood.

Meanwhile, an article in The News in Pakistan (careful - the page tries to automatically print, which is very irritating) draws heavily upon my recent report about the site.