Three new landslide / debris flow videos

Thanks to Lynn Highland at the USGS for drawing my attention to this video of a debris flow in Costa Rica:



This is a slightly strange event. It occurred on the Sarapiqui River in Costa Rica after the 16th January 2009 Cinchona earthquake. It is not all clear why the earthquake should have triggered such a debris flow given that there was no rainfall, but a suggestion is that there may have been mobilisation of shallow groundwater. Another possibility is the creation and then collapse of a barrier lake.

The same event also appears to have been captured in this video:



If anyone has any more information then I would be very interested in hearing about this event.

In Hong Kong last month I saw a video taken of a collapsing hillside during the Wenchuan earthquake. This appears to have been shot in the Wolong Panda Reserve area. The quality is poor, but the footage of the collapsing slope is dramatic:

The National Centre for Landslide Research, Studies and Management in India

Yesterday, the Indian National Disaster Management Authority held a press conference to announce the publication of a new set of "National Disaster Management Guidelines – Management of Landslides and Snow Avalanches". At this press conference it was also announced that the Indian Government will set up a National Centre for Landslide Research, Studies and Management, to be located in one of the most landslide-prone states (I would guess that this will either be in the north of the country or in the south-east). Unfortunately at the moment there is little information available about this centre, other than what was contained in the speech given by the Union Minister of Mines and DONER, Shri Bijoy Krishna Handique:

"The proposal for establishing a National Centre for landslides research, studies and management, as recommended by the guidelines is a welcome step in the direction of capacity building and research and development and I feel that GSI [Geological Survey of India] will be able to host this Centre as part of its core activities. Such a centre of excellence will ensure adequate national coverage, information flow, community participation, networking, and feedback with regard to landslides and snow avalanches, besides coordinating the effort of the states and other concerned central organizations. It will also foster, promote, and sustain a scientific culture in the management of slopes and landslides and encourage the transition to a culture of prevention, mitigation, preparedness and response."

Given that India has a pretty serious landslide problem that appears to be getting worse, this is a very sensible move.

I would like to get hold of a copy of the guidelines, but at the moment they are not available. However, the press conference suggested that they cover nine distinct areas:

• Landslide hazard, vulnerability, and risk assessment
• Multi-hazard conceptualisation
• Landslide remediation practice
• Research and development
• Monitoring and early warning of landslides
• Knowledge network and management
  
• Capacity building and training
• Public awareness and education
• Emergency preparedness and response
  
• Regulation and enforcement.

All of which sounds very sensible!

Chongqing landslide rescue update - 23rd June

Xinhua has a video report about the ongoing rescue attempts at the Chongqing landslide site. The report suggests that remotely sensed imagery allowed a number of locations in which the crags in the head scar area were found to be unstable and threatening the rescue teams. To this end, two artillery pieces were brought in and shells were fired at the crags to try to dislodge the unstable sections. This failed unfortunately, so now dynamite will be used instead.

Unfortunately there is no news of the rescue attempt itself.

Fatal landslides in 2009 so far - a review on the solstice

The summer solstice today seems like a good time to take stock of the position with fatal landslides to date, as collected in the Durham Fatal Landslide Database. Globally the fatal landslide season really starts in early June, the point at which the Asian monsoon starts to generate substantial amounts of rainfall in South and East Asia. Of course the greatest proportion of landslides occur in July and August, so we are before the main event.

So, how does the year look so far. Well, interesting actually. The graph below provides the cumulative number of recorded fatal landslides for 2007, 2008 and 2009, excluding the ones triggered by earthquakes (which obviously don't have a seasonal pattern). Obviously for 2007 and 2008 the full year graph is available.


You should be able to see that for 2007 and 2008 the normal S-shaped curve is clear, with a fairly low rate of increase in the first and last c.120 days. The time before this period, the northern hemisphere summer, shows a much greater rate of fatal landslides (i.e. a steeper line). It is also clear that in 2009 to date the number of fatal landslides is above the previous two years, although it is converging a little of late (of which more below). 2008 was anomalously low - 2007 is far more of an "average" year - although note that this is mostly because of the low number of recorded fatal landslides in the early part of the year.

The second graph, below, shows the recorded number of fatalities (again excluding those caused by earthquake-induced landslides). The coincidence between the patterns in 2007 and 2009 is very striking; again 2007 was very close to an average year on this measure. The comparatively low total at this point in 2008 is notable - but it is also clear that the rainy season resulted in a large number of fatalities (the large step is mostly due to hurricane-induced landslides in Haiti, plus a very active monsoon at this time).

One other aspects of this years data is really interesting. You may have noticed that in both the number of fatal landslides and the number of fatalities the trend in the last few weeks is a little anomalous. In particular, I would expect to see the steepening trend becoming clear as the Asian rainy season starts. However, this is clearly not the case as yet, as the graph below shows more clearly:

Here the blue line is the cumulative total of fatalities and the green line is the trend for fatal landslides. The recent flattening off is clear. This is mostly a clear indication of the rather strange pattern in the development of the South Asian monsoon this year. The northern passage of the monsoon appears to have stalled, as this map from the Indian Meteorological Agency shows:


Here the red lines are the average position of the monsoon front, whilst the green are the conditions observed this year. It appears that the northward advance is two weeks or so late at present - and it appears that this is being reflected in the fatal landslide data. However, the northward advance of the monsoon has apparently restarted this weekend, so the more usual trend may start to be displayed shortly.

Interestingly, there is some evidence that the SW Monsoon is weaker in El Nino years, and conditions are changing from La Nina to El Nino present. We are also recording more fatal landslides in South America than in recent years. However, it is far to early to tell whether the patterns that we are seeing this year are really associated with the possible El Nino conditions. It will be an interesting summer!

Chongqing landslide - NASA satellite image

NASA have provided the first decent overview of the Chongqing landslide via a high resolution satellite image. This can be found here.


The morphology of this is decidedly odd - I will spend some time trying understand it better. According to the NASA page the source is to the north and the landslide has moved southwards and then spread out to the southwest.

Meanwhile, the search for victims continues according to Xinhua.

Cayton Bay / Knipe Point landslide - report on options

A year or so ago I posted on an ongoing landslide at Cayton Bay in North Yorkshire, just an hour or so down the road from my base Durham. This landslide, which is no commonly called the Knipe Point landslide, was threatening 50 or so houses (see image below) - in the end three were demolished, although the rest are still under threat from the slide. Since my post the BGS have created a nice summary website here, from which this picture is taken:

The local council, Scarborough Borough, managed to find from a range of sources about £300,000 to pay for an investigation of the site, which was undertaken by Halcrow. This investigation was completed this week and will be discussed by the Council in a few days time. The Council has put out a press release here, although the report is not available online. The key findings are as follows:

a. The landslide is a deep seated, ancient landslide system. consisting of:
1. A main deep seated failure for which ground movement is controlled by the residual strength of the clay and a deep confined natural groundwater table;
2. Shallow mudslides in the overlying glacial tills for which the ground movement is controlled by an upper natural groundwater table. This slide is highly sensitive to small changes in the groundwater conditions.

I find the above quite surprising as the failures affecting the houses do not move that often - so this apparent high level of sensitivity is a little odd. I will need to read the report (I will try to get a copy).

The press release then discusses stabilisation issues - which is I am sure what the householders are worried about. It briefly mentions the constraints (minimal impact on the environment as the landslide contains a Site of Special Scientific Interest (SSSI), but a 50 year design life).

Three key actions are recommended:

1. "Maintain and continue monitoring of the surface and subsurface ground movement, groundwater and weather station network."
2. "Liase with stakeholders to review the findings of the report and discuss the way forward for managing the cliff instability risk at the site in the short and long term."
3. "Review funding options for promotion of the preferred engineering stabilisation options, and prepare an application for funding under the relevant and most appropriate legislation."

The preferred stabilisation option is unfortunately large and complex:

• Installation of deep drainage to reduce and control groundwater levels in the deeper water table.
• Construction of bored piles at Knipe Point to isolate the lower Cayton Cliff landslide system from the land above the cliff top.

The press suggest that this would cost £12 million (some reports suggest £20 million!). I would think that finding this sum of money is going to be tricky given the limited number of houses involved, although maybe the threat to the main road will help here.

The Chongqing landslide - update 16th June

The latest update on the Chongqing landslide rescue is provided by Xinhua. Key points:

• Rescue operations continue, with 3,000 people working on site;
  
• Ten small-scale landslides have occurred at the landslide site since Friday;
• southwest China's Chongqing Municipality from Friday, and rescuers continued to search for 64 missing for the 10th day on Tuesday.
• The water level in the barrier lake has dropped to 10 m below the top of the temporary embankment;
• The rescuers have not yet found the mine entrance.

Chongqing landslide update - 15th June 2009

An update on the Chongqing landslide rescue is provided by a CCTV report. Latest news:

• Contrary to the reports yesterday, rescue operations are continuing;
• The pumps to lower the barrier lake are now in operation and are succeeding. It will take about a week to empty the lake, whereupon construction of the channel will start.

The Chongqing landslide - the rescue is probably about to be abandoned

Xinhua is reporting that the rescue at the Chongqing landslide is likely to be abandoned shortly. Given the length of time that has elapsed since the failure the chances of rescuing the miners alive are now very small. More worryingly, there is an increasing risk of further failures at the site:

"According to a geologist with the rescue headquarters who declined to be identified, cracks with the span of eight to ten meters wide and up to 30 meters deep have appeared on the landslide-induced hills. "If the 1.75 million cubic meters of soil and rocks fell down from 80 to 100 meters high, another massive landslide shall occur," he said. Four remaining big stones that stood on the edge of the mountains are also likely to slip off when it rains, he said. Two landslides of smaller scale forced suspension of rescue work twice early Friday morning. The landslide site has entered the flood season, and landslide-induced lakes are likely to trigger mud-rock flows, said Zhu Xiansheng, head of the water conservancy bureau of Chongqing."

Such rescues are always a balance between the benefits of successfully rescuing the victims and the risks to the search teams. Given the time that has elapsed, the difficulties involved in the operation and the lack of indications that the miners are alive, the increasing risks to the rescue teams do suggest that stopping the operation is probably prudent, even though it is a very difficult decision to make. Thirty metre deep cracks that are widening are a strong indication that all is not well on the hillside, although failure is certainly not inevitable.

The latest update on the Chongqing landslide rescue

Xinhua and CCTV have published updates on the Chongqing landslide rescue. The picture is looking increasingly grim given that the estimated survival time of the miners was five to seven days, and it is now a week since the landslide. Highlights of the reports are as follows:

• Drilling continues but to date no signs of life have been detected. Drilling (or is this tunnel construction - see below) is only progressing at 5 m per day as they are creating an inclined hole to try to prevent secondary failures;
• Unfortunately the mine plans were buried in the landslide, so the exact location of the mine entrances is unknown. The chances of being able to build a tunnel that will intersect the mine entrance is considered to be less than 20%;
  
• A small landslide that a the drilling platform interrupted rescue operations today. No-one was injured. However, it is now believed that over a million cubic metres of the slope above the site is unstable, and the possibility of more landslides is considered to be "very high".
• The heavy lift helicopter is transporting heavy equipment onto the site (see image above). This is expected to increase the rate at which excavation can occur;
• The barrier lake now contains 40,000 cubic meters of water (this is a big increase in estimated volume compared with yesterday - but is still not a huge amount of water as these things go); The lake level rose by 0.5 m in the last 24 hours.
  
• The team continues to build an embankment to keep the water away from the rescue site. The water level is now 1 m below the embankment
• However, the drainage pipe is now in place and will start pumping today. With a capacity of 15,000 cubic metres per day this should keep the level below the maximum under current conditions;
• In due course a proper drainage channel will be needed, but at the moment the focus is on the rescue.

Three observations:
1. The news that the authorities do not know the actual location of the mine entrances is new - and very candid. One should not be critical of them for this - a characteristic of the Guinsaugon (Leyte) landslide was the huge difficulties that the rescuers, including the US Marine Corps, had in determining the former location of buildings and infrastructure;
2. It seems to me that there are two operations occurring simultaneously here - one to drill bore holes to allow detection equipment to be located and, I suspect, to try to find the mine entrances. This explains why 40 rigs are being used.
3. There is a double race occurring here - one associated with the limited survival time of the miners and the other to beat the seasonal rains. Unfortunately the slope could collapse even without further rain (i.e. through a progressive failure), so the danger to the rescuers is real and very immediate.

Latest update on the Chongqing landslide

The desperate rescue attempts continue at the site of the Chongqing landslide, but time is now clearly running out for the miners. The heavy lift helicopter, the same one that assisted at Tangjiashan, is now in full operation, and the drilling continues using 20 machines according to this report. The drilling operations have now reached 100 m below the surface and thus have reached the position of the portal. Unfortunately this appears to have collapsed (unsurprisingly). Additionally, the rescue teams have had to abandon their use of explosives due to the instability of the landslide mass.

There can be little hope for the trapped miners now, although it is still worth trying to get to them.

Meanwhile, according to this report the barrier lake volume has now reached 10,000 cubic metres (this is not a huge volume). Operations to lay a drainage pipe are ongoing.

Chongqing landslide update - 9th June

Image source: http://uk.news.yahoo.com/19/20090609/img/pwl-china-landslide-death-9-6f14af6c0f4b.html

The latest news from the Chongqing landslide site is as follows:

1. Attempts to build a tunnel to the trapped miners continue. The heavy lift helicopter is now in operation. Electric cables and a generator have been installed to provide power to the ongoing operations;
2. Heavy rain has now reduced to drizzle, but the water depth in the barrier lake increased by a metre in 24 hours;
3. Operations to create a channel to drain the lake continue. The image above appears to show the barrier lake in the background;
4. The rescue teams are using sandbags to try to prevent water from the barrier lake entering the mine - this would of course be disastrous for the trapped miners.
5. Satellite imagery was expected to be available from today (I would like to see this!). Meanwhile, monitoring points have been established to provide warning of any further landslides.

Time must be running out for the miners, but it sounds as if the Chinese are undertaking a huge effort to rescue them.

Meanwhile, some additional pieces of background information have been published in China Daily:

• The iron ore mine, which was called Sanlian, produced only 100 tonnes per day. This is very low.
  
• A similar accident occurred 0n 30th April 1994, creating a landslide deposit that blocked the river;
  
• Two days before the accident rockfalls were observed from the slope;
• The local authorities warned the villagers about the potential for a landslide in 2003. They were offered 5,000 yuan (c.£450 or 520 Euros) to move. Most villagers refused to leave.

Background to the Chongqing landslide

According to this report, rescue operations for the 27 miners trapped in a mine beneath the Chongqing landslide have been suspended due to bad weather. As I mentioned in my earlier post, working on a fresh rockslide deposit is exceptionally dangerous, so this is a wise decision. However, the outlook for the trapped miners must be looking increasingly bleak, given that it now appears that they are trapped 200 m underground and heavy machinery is not being used.

The above article has some very interesting and pertinent information about the context of the landslide. It quotes local residents as saying that the mine in question, which opened in 1949, was closed in 2000 "after being labelled dangerous by an official geological survey team". However, work resumed in 2004 under private ownership, whereupon landslides started to occur from the slope. In 2004 "masses fell from the mountain into the valley... after which the Tiekuang government offices, local school and circa 70 residencies were moved. But 40 of those people buried by the landslide decided to stay in the area, after local officials assured them that there were no problems."

The mine owner has been arrested, but the local people are critical that the local media is reporting the landslide as having been natural: “it wasn’t a natural disaster; human error is the only thing to blame for the tragedy”.

The race against time to save 27 miners trapped below the Chongqing landslide

Image from: http://topiclinks.boston.com/photo/07Nr7zUf5l1W1

In China there is now a dramatic race against time being played out as rescuers race to reach 27 miners trapped below the Chongqing landslide. Xinhua is reporting that the two entrances to the Jiwei Mountain Iron Ore mine were blocked by the landslide. The trapped miners are likely to be able to survive for five to seven days, so there is very limited time available. Rescuers are now trying to blast a 40 m deep shaft through the landslide debris to reach the miners, as this Xinhua image shows:


To facilitate this the rescuers have build a road to bring in heavy machinery and now have access to a heavy lift helicopter.

However, we should not under-estimate the difficulties of this task. First, the landslide debris looks to be incredibly coarse, which will mean that digging a shaft or a tunnel is very difficult indeed, as this Xinhua image shows (note the rescuers for scale):


Third, the debris will be at its dry angle of repose and so is only just stable. Therefore, digging into will potentially destabilise the mass above. Supporting a tunnel or shaft in this material is not going to be easy. Finally, of course, rainfall would be very dangerous. The rainy season is just starting. Therefore, there are substantial risks to the rescuers as well as the miners.

Meanwhile, the slide, which is now estimated to have a volume of 12 million cubic metres, has also blocked the valley, such that a lake is forming. Once again the Chinese are having to build a drainage channel and evacuate people downstream and within the lake area.

The number killed by the landslide is very unclear at present, but the best estimate seems to be 79 buried by the landslide plus the 27 miners.

Illustration of the scale of the Chongqing rockslope failure

The landslide at Chongqing is very large - it has an estimated volume of 3.5 million cubic metres. The scale of this is well-illustrated by this image from Suomen Kuvalehti, which shows rescuers walking across the landslide debris:

Update on the Chongqing landslide

Xinhua is now reporting that 79 people have been killed in the catastrophic landslide in Chongqing. In addition, 27 miners are trapped in a mine whose adit has been buried.

Xinhua has also released an image of the source zone of the landslide (showing continued rockfall activity):




And also the landslide deposit. It is now appears that this is a massive and catastrophic rock slope failure:

UPDATE 2: Media reports of a very large landslide in China

The Chinese State news agency Xinhua is reporting that a landslide in Chongqing Municipality has buried about 60 people.

UPDATE 2: The Chinese state media are now reporting 80 fatalities:

"At least 80 people are feared buried in a landslide at an iron ore mining area in southwest China's Chongqing Municipality on Friday, according to the local government. Rescuers had pulled out seven injured people, including four seriously hurt, from the debris as of 8:30 p.m., according to the publicity department of Wulong County, the site of the accident. The landslide happened at about 3 p.m. in the Jiwei Mountain area, in Tiekuang Township, about 170 kilometers southeast of the downtown area. Millions of cubic meters of rock filled a valley and buried an iron ore plant and six houses."

Interestingly, TRMM 3 day rainfall data does not suggest that the Chongqing area is subject to heavy rainfall at the moment (see image below - Chongqing is in central west China). I wonder whether this is another flowslide?



I'll post again when more info is available.

Are satellite-based landslide hazard algorithms useful?

In some parts of the world, such as the Seattle area of the USA, wide area landslide warning systems are operated on the basis of rainfall thresholds. These are comparatively simple in essence - basically the combination of short term and long term rainfall that is needed to trigger landslides is determined, often using historical records of landslide events. A critical threshold is determined for the combination of these two rainfall amounts - so for example, it might require 100 mm of rainfall in hours after a dry spell, but 50 mm after a wet period. These threshold rainfall levels have been determined for many areas; indeed, there is even a website dedicated to the thresholds!

In 1997 NASA and JAXA launched a satellite known as TRMM (Tropical Rainfall Monitoring Mission), which uses a suite of sensors to measure rainfall in the tropical regions. Given that it orbits the Earth 16 times per day most tropical areas get pretty good coverage. A few years ago Bob Adler, Yang Hong and their colleagues started to work on the use of TRMM for landslide warnings using a modified version of rainfall thresholds. Most recently, this work has been developed by Dalia Bach Kirschbaum - and we have all watched the development of this project with great interest. The results have now been published in a paper (Kirschbaum et al. 2009) in the EGU journal Natural Hazards and Earth Systems Science - which is great because NHESS is an open access journal, meaning that you can download it for free from here.

Of course a rainfall threshold on its own doesn't tell you enough about the likelihood of a landslide. For example, it doesn't matter how hard it rains, if the area affected is in a flat, lowland plain then a landslide is not going to occur. To overcome this, the team generated a simple susceptibility index based upon weighted, normalised values of slope, soil type, soil texture, elevation, land cover and drainage density. The resulting susceptibility map is shown below, with landslides that occurred in 2003 and 2007 indicated on the map:


A simple rainfall threshold was then applied as shown below:

Thus, if an area is considered to have high landslide susceptibility and to lie above the threshold line shown above based upon an analysis using 3-hour data from TRMM, then a warning can be issued.

Kirschbaum et al. (2009) have analysed the results of their study using the landslide inventory datasets shown in the map above. Great care is needed in the interpretation of these datasets as they are derived primarily from media reports, which of course are heavily biased in many ways. Examination of the map above does show this - look for example at the number of landslide reports for the UK compared with New Zealand. The apparent number is much higher than in NZ, even though the latter is far more landslide prone. However, in New Zealand the population is small, the news media is lower profile, and landslides are an accepted part of life. However, so long as one is aware of these limitations then this is a reasonable starting point for analysing the effectiveness of the technique.

So, how did the technique do? Well, at a first look not so well:



In many cases the technique failed to forecast many of the landslides that actually occurred, whilst it also over-forecasted (i.e. forecasted landslides in areas in which there were none recorded) dramatically. However, one must bear in mind the limitations of the dataset. It is very possible that landslides occurred but were not recorded, so at least to a degree the real results are probably better than the paper indicates. Otherwise, the authors admit that the susceptibility tool is probably far too crude and the rainfall data to imprecise to get the level of precision that is required. However, against this one should note that the algorithm does very well (as indicated by the green pixels on the map above) in some of the key landslide-prone areas - e.g. along the Himalayan Arc, in Java, in SW India, the Philippines, the Rio de Janeiro area, parts of the Caribbean, and the mountains around the Chengdu basin. In places there is marked under-estimation - e.g. in Pakistan, Parts of Europe and N. America. In other places there was dramatic over-estimation, especially in the Amazon Basin, most of India, Central Africa and China.

All of this suggests that the algorithm is not ready for use as an operational landslide warning system. Against that though the approach does show some real promise. I suspect that an improved algorithm for susceptibility would help a great deal (maybe using the World bank Hotspots approach), perhaps together with a threshold that varies according to area (i.e. it is clear that the threshold rainfall for Taiwan is very different to that of the UK). Kirschbaum et al. (2009) have have produced a really interesting piece of work that represents a substantial step along the way. One can only hope that this is developed further and that, in due course, an improved version of TRMM is launched (preferably using a constellation of satellites to give better temporal and spatial coverage). That would of course be a far better use of resource than spending $4,500 million on the James Webb Space Telescope.

Reference
Kirschbaum, D. B., Adler, R., Hong, Y., and Lerner-Lam, A. 2009. Evaluation of a preliminary satellite-based landslide hazard algorithm using global landslide inventories. Natural Hazards and Earth System Science, 9, 673-686.