Improved safety with landslide inventory

– The idea of further developing an event based database for landslides stems from a Klima 2050 thematic meeting in January 2016. In the future, we must expect more frequent landslides, triggered primarily by extreme precipitation and snowmelt, says Jose Cepeda from the Norwegian Geotechnical Institute (NGI) and leader of this sub-project of Klima 2050. He considers Klima 2050 to be an important arena for finding practical solutions to the consequences of climate change. A better understanding of what causes landslides and how the risks associated with them can be reduced are important topics within climate adaptation. Landslides can cause significant damage to infrastructure such as railways and roads as well as to buildings. As precipitation increases so also does the danger of landslides and the need for cost effective safety measures.

Debris flows and floods in Kvam
In cooperation with Norges vassdrag og energidirektorat (NVE), NGI has been involved in implementing safety measures in Kvam following landslide incidents caused by heavy rainfall and floods in 2011 and 2013. The decision to start collection of data from these events was a result of both the scale of the damage and the extent of the landslides.

The village of Kvam in Gudbrandsdal is built on sediments produced by previous floods and landslides (alluvial fans). Loose sediments in the sides of the steep valley upstream from Kvam make the area especially hazard prone.

In 2011, large quantities of rainfall triggered landslides in Veikvedalen, upstream from Kvam. The landslide masses flowed along the river at the bottom of the valley and resulted in flooding which caused a large amount of damage. Although, on estimate, such events occur on average once in every hundred years, a similar event occurred only two years later in 2013.

Water triggered landslides cost most
A warmer climate and more extreme weather will lead to more water triggered landslides, particularly soil slides, which can be triggered by intense rainfall, long periods of lower intensity precipitation or by rapid snow melting and thawing of frozen soil.

In addition to NGI and NVE, the team lead by Cepada involved the University of Rome and also made use of scientific input from the Geological Survey of Norway (NGU). Starting with the work that had previously been carried out to register and evaluate the events in Kvam, the team set about developing the database.

– In order to understand the area and what actually happened in Kvam, we have done a lot of fieldwork. We have documented several additional, previously unmapped landslides and based on field observations, interpretations of results from remote sensing and geophysical surveys, we have calculated a model for bedrock depth. We have also separately identified both initiation and runout zones over the whole landslide footprint for each event, which has not been done before.

Responsibility for the landslide warnings
In Norway, NVE issues warnings for both floods and landslides. The warning system is based on conditions such as precipitation and snow melting.

In addition, NVE has a national landslide database comprising 60,000 registered events, very few of which, however, contain details about the landslide footprint. Many events come from Norwegian Public Roads Administration records of landslides which blocked roads.

– We want to improve landslide warning methodology for debris slides and debris flows. A complete database like the one developed here will be useful for that,’ says Odd Are Jensen from NVE.

– The type of data gathered in Kvam is the kind we would like to see more of in the national landslide database, he adds, and mentions too that the structural mitigation measures in Kvam are scheduled for completion in spring 2017.

Local knowledge for landslide mapping
The damage caused by landslides and floods in Kvam shows how important it is to have complete mapping of the consequences of expected climate effects in an area, in order to adapt and secure infrastructure and buildings.

– Of course, no database will ever be complete but we have come a long way, says Cepeda. He believes that the Kvam database will provide an important basis for future predictions and for identifying necessary safety measures.

– When mapping landslide hazards, one of the most fundamental things to investigate is landslides which have previously occurred in the area. It is useful to speak to local farmers who perhaps know about landslides which occurred in the 18th century or just three years ago. Nowadays, such local knowledge is about to vanish, so incorporating information about historical events in the database has become more and more important, says Jensen. He encourages people to register landslides on www.skredregistrering.no.

Predicting landslides at other locations
The Kvam database can be useful in risk analyses for landslide prone areas in other parts of the country which are similar to Kvam in terms of terrain and ground conditions. It should be possible to use analyses based on the data to predict landslides and recommend mitigation measures.

– More complete databases can help us to prepare for future landslide events and to optimise mitigation measures, says Cepeda.