The biggest landslide of them all - Saidmareh, Iran

As the summer begins and my mind starts to move over from administration to research, I was pondering really large landslides. As a result I thought that it was about time that I posted about the biggest known sub-aerial landslide - surprisingly it has received comparatively little attention to date.

The landslide itself was identified and written up by Harrison and Falcon in 1938 is a paper in the Journal of Geology that is freely available online through JSTOR. The landslide is located on the Kabir Kuh anticline in Southwest Iran at 33N, 47.65E :

(Click on the image for a better view in a new window)

This slide, which is called the Saidmareh landslide (sometimes also the Saidmarreh, the Seymareh or the Kubir Kuh landslide) is big...really, really big. The statistics defy imagination to be honest - it has a volume of about 20 cubic kilometres, a depth of 300 m, a travel distance of 14 km and a width of 5 km. This means that about 50 billion tonnes of rock moved in this single event!

Fortunately the slide is well covered by Google Earth - this is a perspective overview:


A slide this large is quite hard to understand, so I have annotated the image below. Note the scale!

So lets take a closer look at the source area of the landslide:

The image above shows that this is essentially a dip slope failure on a tectonic ridge - in other words, the landslide came off along an inclined bedding plane. You will see that, as is often the case in fact the slip plane stepped from one bedding plane to the next to exploit the weakest parts of the rock mass, which is mostly limestone with some marls. The maximum fall height was about 1600 m according to Harrison and Falcon (1938).

The deposit is huge, covering an area of about 166 square kilometres:

It is formed from very angular blocks of limestone, some of which are large enough to be seen on the Google Earth imagery:


The highly fragmented nature of the deposit and the long travel distance both suggest that this was a very energetic, high velocity landslide - a rock avalanche (sometimes called a sturzstrom).

The landslide blocked two rivers, allowing a pair of lakes to form, both of which have now drained away. However, the remains of one of them is clearly evident as the deposited sediment provides fertile farm land as shown below on the south side of the landslide. The lake appears to have breached across the landslide debris, creating a channel that has now been weathered. Subsequently the modern river has found a new course off the landslide mass:

The other lake is much larger, being located in the main valley that was blocked by the landslide:

This lake deposit is 39 km long and about 150 m thick close to the landslide.

The age of the landslide is not clear, but there is an ancient ruined Sassanid town bridge on the bed of the larger lake. The Sassanid era extended from 224 to 651 AD, so the landslide must be considerably older than this. One date was reported by Shoaei and Ghayoumian (1998) of 10,370+/-120 years BP.

Finally, what caused such a huge landslide? This is an area that is subject to intense seismic activity so it is highly likely that this was earthquake triggered. There is no evidence that the slide occurred as anything other than a one single, massive failure.

References:
Shoaei, Z. and Ghayoumian, J. 1998. Seimareh landslide, the largest complex slide in the world In: Moore D and Hungr O (eds) EIGHTH INTERNATIONAL CONGRESS OF THE INTERNATIONAL ASSOCIATION FOR ENGINEERING GEOLOGY AND THE ENVIRONMENT, PROCEEDINGS, VOLS 1-5 , 1337-1342.

J. V. Harrison and N. L. Falcon 1938. An Ancient Landslip at Saidmarreh in Southwestern Iran
The Journal of Geology, 46 [3], 296-309.

The Yigong Rock Avalanche, Tibet

Occasionally I like to take a look back at a significant landslide. Today I thought I'd write about the amazing Yigong rock avalanche in Tibet. The location of the landslide is shown on Figure 1: it's location is 30°14' N, 94°59' E, located high on the Tibetan Plateau.

Figure 1: Google Earth image showing the location of the Yigong landslide. Click on the image for a better view.

The landslide itself occurred at midday (UT) on 9th April 2000, when a wedge failure at about 5,500 m elevation (3,300 m above the valley floor) detached a volume of rock of over 100 million cubic metres.

the landslide occurred in Zhamu Creek. It took only 10 min to travel down a horizontal distance of 8 km through a vertical elevation difference of 3330 m from its source at 5520 m a.s.l. to its sediment fan at 2190 m a.s.l.(Fig. 3). The trigger for the failure is not clear, but no obvious rainfall or seismic event was recorded. The debris travelled down the valley (Fig. 2) at velocities of up to about 14 m/sec (Shang et al. 2003). Over the next 10 minutes the landslide travelled about 10 km, entraining debris, snow and ice as it flowed to generate a deposit with a volume of about 300 million cubic metres. The landslide came to rest on the valley floor at an elevation of 2190 m, blocking the Yigong River to a height of 60 m over a distance of about 1000 m (Fig. 3).

Figure 2: The track of the Yigong landslide. Click on the image for a better view. Image from Yueping, Y. 2008. Landslides in China: selected case studies. China Land Press, Beijing.

Figure 3: The deposit the Yigong landslide. Click on the image for a better view. Image from Yueping, Y. 2008. Landslides in China: selected case studies. China Land Press, Beijing.

The landslide itself is rather well shown in Google Earth, although unfortunately the source area has a slightly lower resolution than does the track (Figure 4). It is clear that the debris travelled down a comparatively linear tributary valley until the main valley was reached. In the latter stages of the movement the landslide was technically a distal debris flow, essentially meaning that it entrained a large volume of water (from snow and ice). This would have been immensely destructive in an inhabited area, but fortunately the area is sparsely populated.

Figure 4: The track the Yigong landslide as shown on Google Earth. Click on the image for a better view.

Of course the landslide did create a major problem in that it blocked the valley. A lake rapidly started to fill, and over the next two months a volume of about 3 billion cubic metres of water accumulated. The authorities sought to mitigate the problem by relocating the 4,000 living in the path of the flood whilst simultaneously constructing a drainage channel across the landslide deposit. This lowered the peak height of the dam to 44 m above the river level. The water broke through the channel on 10th June and the lake drained over a period of two days. 17 km downstream a peak discharge of 120,000 cumecs (cubic metres per second) was recorded, with the river level rising to 32 m above the deck of the bridge across the river. The trimline created by the flood is clearly visible in Fig. 5. The area that had been inundated by the lake is clear on Figure 6.

Figure 5: The trimline created by the flood resulting from the overtopping of the Yigong landslide dam. Click on the image for a better view. Image from Yueping, Y. 2008. Landslides in China: selected case studies. China Land Press, Beijing.

Figure 6: The area inundated by the landslide dammed lake at Yigong shown on Google Earth. Click on the image for a better view.

The emergency measures were effective in preventing loss of life in Tibet, although there was considerable damage to infrastructure downstream. However, the flood did cause an international incident as the Chinese authorities failed to warn their counterparts in India that the flood was coming. As a result, the people of Arunachal Pradesh were not prepared for the event, leading to about 130 fatalities and 50,000 being rendered homeless (Yin and Wang 2005).

Finally, as an aside, the team that successfully built the drainage channel in this case was kept intact after the event and were thus ready to respond to the landslide dams created by the Wenchuan (Sichuan) earthquake this year. Their experience at Yingong was undoubtedly critical in their success in mitigating the major landslide dammed lakes that were created. It appears that their skills continue to be needed.

References:
Shang, Y. et al. 2003. A super-large landslide in Tibet in 2000: background, occurrence, disaster, and origin. Geomorphology. 54 (3-4), 225-243.
Yin, Y. and Wang, S. 2005. Landslide hazard and reduction strategy in China. In: Hungr et al. Landslide Risk Management. Tayor and Francis, 423-6.
Yueping, Y. 2008. Landslides in China: selected case studies. China Land Press, Beijing.

Mount Adams rock and snow avalanche

The online site Oregon Environmental News has today carried a rather nice article on a new, large, rock and snow avalanche on Mount Adams (Fig. 1) in the Cascade Range in Washington State. Mount Adams is a stratovolcano that has a long history of such events. Many readers will be aware for example of the 31st August 1997 event, which had a volume of about 5 million cubic metres, and is well documented by the USGS.

Figure 1: Google Earth view of the southwest face of Mount Adams, upon which the most recent event occurred.

This event was first picked up at about 14:50 UT on 1st August at the seismometers at the Cascades Volcanic Observatory, who have kindly put the seismic record online (Fig. 2).

Figure 2: Seismic record from the Cascades Volcanic Observatory showing the 1st August 2008 rock and snow avalanche on Mt Adams (click on the image for a better view).

The seismologists quickly recognised that this event was not volcanic in origin, but was in fact associated with a mass movement. As a result, they contacted a local photographer, who went up there and confirmed that a rock and snow avalanches about 2 miles (3 km) has occurred. H.C Tupper has been able to collect a photograph of the event (Fig. 3).

Figure 3: H.C Tupper's's photograph, posted on Oregon Environmental News, of the 1st August 2008 Mt Adams rock and snow avalanche. Note the annotation is from the source and is thus not by me (click on the image for a better view).