Recent weeks saw Japan's south-western prefecture of Kumamoto being stuck by two earthquakes – a foreshock of magnitude 6.2 and the main shock with a magnitude of 7.0. According to Catastrophe modeling firm AIR Worldwide, the main shock is estimated to cause an insured property loss of 1.7 to 2.9 billion USD. One of the notable thing about this event was triggering of numerous landslides. One of the major landslide occurred at Minamiaso, had washed away parts of a highway, railway track and a bridge. Many of the landslides have caused the debris flow to reach downstream regions, which are far away from the source of landslides. Apart from the damage caused by the landslides, it has isolated some of the regions and delayed the relief and rescue operation.
Some of the images of landslides can be viewed at : http://blogs.agu.org/landslideblog/2016/04/18/kumamoto-earthquake-1/.
Most of the catastrophe models adopt a lean approach to model landslides susceptibility. There are many reasons for this – (i) modelling landslides are bit complex and (ii) the insurance exposure in landslide susceptible regions (hilly areas) will be relatively low. While developing landslide susceptibility models, many parameters needs to be considered- mainly related to soil properties, level of soil moisture, nature of terrain, land use pattern etc. In many cases, adequate data on these parameters may not be available and hence they will be ignored. However, once triggered, landslides can create havoc. One of the peculiar thing about landslide is that the debris of a landslide can reach regions, which are far away, and can cause damage. Moreover, the highways, bridges or railway lines, which are along a hilly terrain, can get affected very easily and this will severely affect relief and rescue operations.
Swiss Re's latest earthquake model for Japan includes landslide susceptibility layer as well. This was developed considering the terrain slope and the model takes into account the effects of debris flow as well. This landslide susceptibility layer clearly identifies the region, which has suffered landslide, as a landslide prone area.
Some of the structures which will be passing through landslide susceptible areas (and highly vulnerable to landslides as well) are roads, railways lines, tunnels etc. Since the impact of landslide will be confined to a relatively smaller area, when compared to earthquake shake, a lot depends on how the exposure data of these structures are captured, so that their landslide susceptibility can be modelled accurately.
Category: Climate/natural disasters: Disaster risk, Earthquakes
Location: Kumamoto, Japan