21 August 2017

Regent landslide in Sierra Leone: the causes of the disaster

Posted by Dave Petley

Regent landslide in Sierra Leone: the causes of the disaster

The latest tally of losses from the Regent landslide in Sierra Leone has been released by the coroner.  To date 499 bodies have been recovered, with about 600 people still reported missing.  As expected, this is now the worst landslide of the calendar year to date.

It is increasingly clear that this disaster was caused by a large, but not huge, failure on the bedrock – regolith interface on a slope above the town.  This event would undoubtedly have caused mass fatalities, but the true magnitude of the disaster was the result of this failure transitioning to become a high mobility channelised debris flow.  This image, from Ultimate FM online, is the most revealing:

Regent landslide

The track of the Regent landslide in Sierra Leone, via Ultimate FM online.

.

Note the broad landslide track heading downstream from the camera.  Given that victims from this disaster have been recovered from the sea, it is likely that this landslide travelled about 6.5 km as the crow flies, probably transitioning into a mudflow en route.  The slope that failed is about 300 m long and about 130 m from toe to crest.  This is a March 2017 Google Earth image of the slope that failed in the Regent landslide disaster:-

Regent landslide

March 2017 Google Earth imagery of the slope that failed in the Regent landslide disaster.

.

Whilst this slope does show signs of some distress, it is highly unlikely that it could have been identified as being dangerous to this degree from visual imagery alone.  Identifying potentially hazardous slopes in such settings remains a major challenge.  Of greater interest perhaps is to compare the above image with the first high resolution image available in the Google Earth archive:-

Regent landslide

Google Earth imagery from 2005 showing the location of the Regent landslide.

.

The slope has lost some forestry, but the scale of deforestation here is not huge.  What is more notable is the change in land use at the toe of the slope.  In 2005 this area was mostly unpopulated; immediately prior to the Regent landslide it had dozens of houses.  Needless to say it is this growth of urban areas in unsuitable sites that is driving an increase in losses from these events.  The combination of this effect with increased rainfall intensities as the climate warms is lethal.