17 March 2020
The catastrophic lahars from Mount Kelud in 1919
On 19 and 20 May 1919 a catastrophic eruption occurred on Mount Kelud in East Java, Indonesia. This major eruption, one of the most deadly of the 20th Century, is estimated to have killed 5,160 people. The vast majority of those died in lahars (volcanic debris flows) triggered by the eruption. This was also one of the worst landslide disasters of the 20th Century.
There is a really interesting paper about the social impact of this event (Mawiyato and Sasmita 2019) available online, open access. They point out that this is a particularly interesting lahar event in part because of the detailed accounts of the impacts. The underlying problem at Mount Kelud is the presence of a crater lake, which at the time had an estimated volume of 40 million cubic metres. The eruption displaced this lake, which entrained large volumes of ash, generating the enormous lahars. On Researchgate there is a map of the deposits left by these landslides:
Mawiyato and Sasmita (20189) provide an account of the lahars, with a focus on those that travelled southwards to Blitar:
Bladak Dam of Kali Lahar, which was built to reduce the lahar floods of Mount Kelud, was destroyed. With a speed of 60 km/h, the flows of lahar reached Blitar in less than an hour and destroyed everything standing in its way. The scale of damage reached dozens of kilometers of the volcano. Many villages were reported to have been flattened to the ground and even some of them were wipe out from the map. Among the destroyed vilages were Sumbersari, Salam, Ngoran villages in Udanawu district. Villages along the road stretching from Panataran Temple to Blitar were badly devastated and also Omboh, Sidareja and Sumberejo villages of Wlingi.
To illustrate the magnitude of the impact of these lahars, Mawiyato and Sasmita (2019) describe the events in a prison located in the path of the lahars. About 900 prisoners were trapped in their cells. Of these only about 100 were able to escape the lahar, but they were quickly caught by the flows and were killed. There is perhaps no worse prospect than that of being trapped in a locked cell as it fills with boiling mud.
The devastation of the lahars are also well described. For example, this is the account of the damage in the town of Blitar:
Instantly Blitar lost its form as a town, turned into a sea of lahar. It was dark as being covered by cloud resulting from the swift ash rain. The height of the lahar was approximately 1.6 m, houses around the town square were all damaged, many walled houses were collapsed. The Chinese, Dutch, Javanese settlements whose house-buildings between were somewhat distant could be said to be non-existent. Only the houses of brick walls that coincided and hand-in-hand were not so damaged, but the walls were broken down, such as the house of the resident master, the district house, the post office, the bank office, the clinic, the hotel, the detention center and the other.
Apart from the loss of life, the lahars and ashfall caused massive damage to the extensive coffee farms on the flanks of the volcano, and to other agricultural areas.
Mount Kelud erupted again in 1966, killing more than 200 people. To manage the hazard, a set of tunnels, known as the Ampera Tunnels, were constructed to manage the water levels in the crater.
Nawiyanto and Sasmita, N. 2019. The Eruption of Mount Kelud in 1919: Its Impact and Mitigation Efforts. In: 1st International Conference on Social Sciences and Interdisciplinary Studies (ICSSIS 2018). Atlantis Press. https://doi.org/10.2991/icssis-18.2019.25.
16 March 2020
An analysis of the Piz Cengalo landslide
In August 2017 a large landslide initiated on the Piz Cengalo mountain in Switzerland. It travelled 6.5 km before striking the village of Bondo. In total it was responsible for eight fatalities.
In an article just published in the journal NHESS, which is open access, Mergili et al. (2020) provide a detailed description and simulation of the landslide. The first part of the paper provides a really good description of the event. The landslide was initiated at 09:31 local time on 23 August 2017 when about 3.6 million cubic metres broke off from the eastern face of Piz Cengalo. The sequence of events is shown in the images below, from Mergili et al. (2020):-
After detaching, the landslide impacted the glacier below and entrained about 600,000 cubic metres of ice. In a manner that was similar to the Mount Haast landslide, the mass behaved initially as a rock avalanche. However, downstream it evolved into a debris flow that swept down the valley to strike the village of Bondo. Over the next day or so a series of nine debris flow surges struck the village.
Interestingly, two different scenarios have been proposed to explain the behaviour of the landslide, and in particular the formation of the first debris flow. These are described in the schematic diagram below, from Mergili et al. (2020):-
In the first scenario (S1), the front of the rock avalanche became highly saturated, and transitioned into the first debris flow. In the second (S2), the rock avalanche stalled mid-valley, but was overtopped by an avalanching flow of entrained ice and rock fragments, to form the debris flow.
Mergili et al. (2020) have attempted to simulate these two scenarios. Interestingly, the models indicate that both are plausible, and the simulations do not allow us to determine which is likely to be correct.
The behaviour of these complex landslides remains challenging to understand and model, but is really important if future hazards are to be understood. The impacts of global heating, which are increasing the frequency of these large events, means that this remains a really important research question.
Mergili, M., Jaboyedoff, M., Pullarello, J. and Pudasaini, S. P. 2020. Back calculation of the 2017 Piz Cengalo–Bondo landslide cascade with r.avaflow: what we can do and what we can learn. Natural Hazards and Earth System Sciences, 20 (2), 505-520. DOI: 10.5194/nhess-20-505-2020.
12 March 2020
Salkantay: Planet Labs imagery of the upper portions of the flow
Capturing imagery of the source area of the ice/rock avalanche and subsequent debris flow at Salkantay in Peru is extremely difficult. This is an area that is heavily affected by cloud cover, especially in the upper portions of the catchment. Earlier this week I was able to post some images of the track of the debris flow that had been captured by Planet Labs using their PlanetScope constellation, but the upper part of the landslide was not captured. On 10 March 2020 a further pass by one of the satellites has captured the upper portion of the landslide, even though the actual source remains frustratingly obscured.
This is an image of the area around the source, captured before the landslide on 24 October 2019 by Planet Labs:-
This is the Planet Labs image of the same area, captured on 10 March 2020:-
Unfortunately the imagery does not really provide much additional information about the landslide source, although it can be seen that the lateral extent of the collapse was quite limited in extent on the mountainside. The landslide mass has travelled into the lake basin and has then emerged on the west side. Interestingly, there is still water in the lake, suggesting that it was not displaced or that the lake has refilled after the landslide. This is quite surprising – I would welcome views on how this has happened.
The track of the landslide is clear. Directly below the lake the landslide has eroded the vegetation but does not appear to have generated a great deal of scour. Further down the track there appears to be considerable scour, and on the west side of the image the landslide track has become wider. This is probably controlled by the pre-failure topography.
Reference and acknowledgement
Planet Team (2020). Planet Application Program Interface: In Space for Life on Earth. San Francisco, CA. https://www.planet.com/
9 March 2020
Downstream impacts of the Salkantay rock avalanche and debris flow
On 8 March 2020 Planet Labs captured an image showing the downstream impacts of the Salkantay rock avalanche and debris flow. Cloud-free days are rare in this part of the world are rare, so this is a good catch. Unfortunately the area of the actual failure is just off the edge of the image, but the downstream effects of the debris flow are clear.
This is an image of the area captured before the landslide on 7 January 2020 with the Planet Labs Planetscope system:
This is an image collected with the same system on 8 March 2020 showing the aftermath of the debris flow:-
The impact of the landslide is clear. On the east side of the image a small proportion of what is presumably the landslide deposit can be see, before the channelisation occurred. The debris flow has eroded the channel flowing initially towards the west, and then towards the north. In some places erosion of the banks appears to have triggered failures on the hillsides. In this example, a new landslide has been triggered, and an existing one has been reactivated. In the latter case, a road across the landslide appears to have been destroyed:-
Hopefully images will be captured in the next few days showing the failure itself, although this is of course dependent on cloud cover.
Planet Team (2020). Planet Application Program Interface: In Space for Life on Earth. San Francisco, CA. https://www.planet.com/
6 March 2020
The landslide-induced TGV (high speed train) accident in France yesterday
Yesterday, 5 March 2020, a TGV (high speed train) struck a landslide between Strasburg and Vendenheim in the Bas-Rhin area of France. The train remained upright, not least because it appears that it was a glancing blow rather than a direct collision, but 22 people were injured, one seriously. The best news report, with thanks to Scott Johnson, is in L’Usine Nouvelle. The article is in French, but Google Translate does a fine job.
The landslide is a large rotational slip in a slope in a cutting. The displacement of the mid-section is quite large, but little of the debris appears to have reached the tracks. This prevented a more serious accident. The train, which had 348 passengers on board, was travelling 270 kilometres per hour (170 miles per hour) at the time of the collision.
The line is quite new – Wikipedia indicates that it was constructed in the period between 2010 and 2016. A failure on this scale will inevitably cause concern, and is surprising. News reports indicate that the landslide was triggered by heavy rainfall. Interestingly, this is being described as an “accident intolerable” – i.e. an unacceptable accident – by the local trade union.
After the accident, the train came to a stop at about 48.729, 7.514, based on matching images to Google Earth. The accident must have been to the southeast of this point. The most likely location appears to be 48.719, 7.538, but this is very tentative.
Landslide-induced train accidents occur fairly often around the world, sometimes with very serious consequences. It is very unusual for an accident to affect a modern high speed line, especially in well-designed earthworks, which would typically have been constructed with a c.125 year design life. Thus, understanding the causes of this accident will be a priority.
5 March 2020
A geological detour: the Geotimes pillow lavas at Wadi Jizzi in Oman
Earlier this week I made a brief trip to Oman as part of my (non-geological) day job, in this case attending the opening ceremony of Intaj Suhar, the Oman equivalent of the wonderful University of Sheffield Advanced Manufacturing Research Centre, which sits within my portfolio. On Sunday I took a detour inland to visit the famous Geotimes pillow lavas at Wadi Jizzi, about 30 km inland from the city of Sohar.
These rock are famous as they appeared on the front cover of the Geotimes Magazine in 1975 (the magazine is now known as Earth Magazine). I visited courtesy of a somewhat bemused taxi driver, who clearly could not really understand why I’d drive out into the desert to look at rocks. However, these are truly spectacular:-
Cutting across the outcrop are two well exposed altered basalt dikes:-
These altered dikes seem to be much more susceptible to weathering, so form distinctive gullies within the outcrop:-
These pillow lavas were formed within the Semail Ophiolite at the bottom of the Tethys Ocean. The William and Mary Blog has a nice explanation as to how they formed:-
“Pillows commonly form when lava is extruded under water. As lava disgorges from its vent on the sea floor it comes in contact with the surrounding seawater that rapidly quenches the lava to a glassy solid, thereby partially clogging the conduit and forcing to lava to ooze out nearby. This repetitive process of extrusion and rapid quenching produces the tube to pillow-like morphology.”
These pillow lavas are exceptionally well-preserved. The explanation for this might lie in the deposits on the other side of the valley:-
I suspect that these alluvial deposits might have covered the pillow lavas until recently, preserving them.
Pillow lavas are very hard, and the interlocking structure will give the rock mass considerable strength. As a consequence they are likely to be resistant to failure, and thus not a good place to go looking for landslides.
4 March 2020
Baixada Santista: significant landslides in Brazil this week
Heavy rainfall on 2 and 3 March 2020 has triggered many landslides in the urban area of Baixada Santista in SE Brazil. Globo has a good report that provides details of the losses – it was detailing 18 fatalities at the time of publication, with a further 30 people reported missing. Rainfall totals in Guaruja are reported to have been in the region of 300 mm, with many other areas receiving over 100 mm. The same article has two maps showing the distribution of the recorded landslides and the losses to date:-
In one case, in Morro do Macaco Molhado (shown on the map above) a landslide buried a 25 year old woman and her son aged 10 months. Both were killed. Sadly, two members of the fire and rescue service were killed in a second landslide at the same location whilst trying to save the victims.
There are some images online showing the nature of the landslides in Baixada Santista. For example, this is a landslide in Guaraja:-
Meanwhile, Globo has published this image of a landslide at Morro do Guaraja, which has started as a shallow slip on the hillside (note the precarious houses built on platforms on the slope), and has then channelised through the urban area, probably following a roadway:-
Heavy rainfall in the south of Brazil frequently occurs in the early months of the year, triggering significant numbers of landslides. In January 2011 for example heavy rainfall in Rio de Janeiro killed 903 people, including 424 people in Nova Friburgo and 378 people in Teresópolis. The majority of the fatalities were the result of landslides and the resultant channelised debris flows.
28 February 2020
More information about the Salkantay landslide and mudflow
Over the last 24 hours more information has become available about the Salkantay landslide and mudflow. Oscar Vilca has kindly contacted me to say that the event occurred on 23 February 2020, and not 24 February as had been widely reported. I will correct my original post.
The triggering event is being described as an ice / rock avalanche with an initial volume of 400,000 cubic metres. This has clearly bulked up to form a mudflow with a much higher volume, presumably through entrainment of ice and saturated debris in the channel. This is similar to the Seti River rock avalanche and debris in Nepal in 2014, which also had devastating effects. On this occasion the initial collapse may have been smaller, but the mudflow was on a similar scale.
On Twitter, Julio Montenegro G. has posted an interpretation of the event, based upon an image of the scar, which has then been located on pre-event imagery:-
I am not sure as to the origin of the image that shows the scar of the initial failure, but a better version was posted to Twitter by Turismo Peru:-
If this is indeed the scar then my interpretation is that this is a classic wedge failure in the rock mass, with a near vertical fall onto the ice and moraine at the toe of the slope. The rock slope would have been a mixture of rock and ice, both on the surface and within fractures. On impact the mass has probably fragmented to form an ice / rock avalanche, which has then entrained debris and ice / snow / water, transitioning to become the mudflow seen in the videos. This has behaved in a manner that is akin to a lahar, with a large volume, high velocity and long runout.
There are of course some real human tragedies in this disaster. The estimated human cost appears to be 13 people.
27 February 2020
Breaking News: a catastrophic glacier collapse and mudflow in Salkantay, Peru
On 23 February 2020 (corrected – this was erroneously reported as 24 February 2020) an enormous, catastrophic debris flow tore down the Salkantay River in Santa Teresa, Peru. This event has killed at least four people, with a further 13 reported to be missing. Given the magnitude of the flow, this number is probably uncertain.
The mudflow was captured in an extraordinary video posted to Youtube:-
A mudflow on this scale usually requires an extraordinary cause. Diario Correo in Peru has an explanation – this event was caused by glacial collapse on Salkantay mountain. This hypothesis is proposed by Oscar Vilca Gómez, who the article describes as a specialist in Hydrology and Glaciology. He visited the site site of the detachment as part of a research team from the National Institute for Glacier Research of the Ministry of Environment. They propose that an ice / rock avalanche detached from the mountain, crossed the Salkantay Cocha lake, and generated the huge debris flow.
The article includes the following image of the site:-
In first inspection this appears to be a wedge failure in the rock mass that has fragmented to generate the rock / ice avalanche. The photographer appears to be standing on the landslide deposit.
Salkantay (which also appears to be named Salcantay at times) is located at -13.340, -72.540. Salkantay Cocha lake appears to be at -13.342, -72.569. At the moment it is not clear as to which slope has failed to generate this ice-rock avalanche and debris flow. There is excellent Google Earth imagery of this area, so it should be possible to get a better understanding in due course.
At the moment details of this very significant event are somewhat unclear; I hope that more details will emerge. This event is reminiscent of the 2012 Gayari ice and rock avalanche in Pakistan and the 2017 Villa Santa Lucia landslide in Chile.
26 February 2020
The 1954 Prospect Point Rockfall at Niagara Falls
Regular reader George Haeh has very kindly pointed out a remarkable video on Youtube showing the 1954 Prospect Point Rockfall at Niagara Falls:-
This is an archive British Movietone newsreel, complete with dramatic music and rather elegant commentary. The associated text suggests that the rockfall had a mass of about 168,000 tonnes (185,000 tons). Niagara Frontier has this image of the rockfall:-
There is a good article about the Niagara Falls rockfall on the Niagara Frontier website. The rockfall occurred on 28 July 1954 on the American Falls side of the site. It appears that cracks at the site, shown in the video above, were first noticed by a nine-year old child the day before the failure. The rockfall left a large debris pile, which then became a tourist attraction in its own right. In the aftermath, the Niagara Frontier State Parks Commission undertook blasting to remove rock left in a precarious position, with large blasts on 6, 12 and 16 August 1954. Thereafter the site was deemed to be safe.