The risk we know about, but fail to address

By Heidi Hefre, Head of Section Mountain Hazards at the Norwegian Geotechnical Institute (NGI), and Unni Eidsvig, Lead Engineer and Researcher at NGI. This opinion piece was originally published by nordnorskdebatt.no on 25 August 2026.

In May 2023, a boulder weighing several tonnes crashed into the garden of a family in Spansdalen in Lavangen, coming to rest just eight metres from the house. The terrain above the property was surveyed in early 2025, and the assessment concluded that the landslide hazard was high. Eighteen months later, the family is still waiting for the municipality to decide whether the property will be protected. This summer, residents of Nerlandsøya in Herøy woke to what they thought was thunder. In fact, rocks had broken loose from the mountainside and landed between two homes in the middle of the night, while one of the families lay asleep. Fortunately, no one was injured. National landslide hazard maps had identified the homes as exposed as far back as 2016.

Both stories received extensive media coverage this summer. The cases are different, but they share one feature that matters more than the local details: the hazard was known in advance. What Norway lacks is not knowledge, but a system that determines what that knowledge should lead to.

A responsibility with no clear limits

In previous projects and studies for organisations including Finance Norway and the Ministry of Energy, NGI has reviewed Norway’s legislation, technical requirements and sector guidance on protection against natural hazards. One finding stands out: the Norwegian building regulations set clear safety requirements for floods and landslides, but only for new development. These requirements work well in land-use planning and prevent new construction in hazard zones. They do not, however, apply retrospectively to buildings that are already there.

For homes already located in a hazard zone, there are no defined criteria for what level of risk is acceptable and no specific requirement to reduce a known risk. Homeowners are responsible for their own property, including any consequences for neighbouring properties, but they have no statutory duty to protect it against natural hazards. The authorities, for their part, can order protective measures on private land, but this power is rarely used. The result is that a hazard map may simply remain on file, with no one clearly responsible for acting on what it shows.

Who is actually responsible?

The Lavangen case shows how this works in practice. Funding for protective measures is available from the Norwegian Water Resources and Energy Directorate (NVE), but only the municipality can apply. Residents cannot apply directly. Whether a known landslide hazard is followed up therefore depends on the capacity and priorities of the individual municipality. NVE does not intervene on its own initiative. The incentives are poorly aligned: the cost of prevention often falls on the individual homeowner or the municipality, while the financial consequences of a landslide are largely borne by society as a whole.

This is not a new observation. In an op-ed published by Kommunal Rapport in February 2025, Eli Sandberg and Lene Sandberg described how today’s incentive structure encourages a “wait-and-see” approach, with both landowners and municipalities holding back in the expectation that the central government will assume the practical and financial responsibility. The buck is passed even when the need for protection is specific and well documented.

The scale of Norway’s exposure to natural hazards is well documented. Around 210,000 buildings are at risk, and NVE has estimated the cost of protecting existing development at approximately NOK 85 billion. Weather-related and natural-hazard events already cost society around NOK 5 billion a year. If prevention remains at today’s level, analyses indicate that this figure could almost quadruple by 2100. The 2026 national budget allocates NOK 782 million to flood and landslide work, of which NOK 593 million goes to NVE’s preventive work. The commission appointed after the Gjerdrum landslide recommended spending NOK 1.5 billion annually to protect existing development.How can we close the gap between documented risk and the failure to act? Other countries may offer useful lessons on what a more systematic model for managing such risks could look like.

Iceland has chosen a different path

Iceland resembles Norway in the respects that matter here: a small population, substantial exposure to natural hazards and a state that plays a major role. The difference lies in what happens before disaster strikes. Iceland has established criteria for acceptable risk in existing development. If hazard mapping identifies a high level of risk, the municipality is required to prepare an action plan. In the most exposed areas, the risk must be reduced through physical protection or by purchasing homes.

The measures are financed through a collective scheme. All buildings with fire insurance pay an annual levy equivalent to 0.03 per cent of their insured value. Through a dedicated fund, the state covers up to 90 per cent of the cost of protective measures or of buying and relocating homes. Iceland has therefore linked criteria for when risk requires action to a financing mechanism that makes action possible.

Norway also has a relocation scheme, introduced in 2018 following the floods in Kvam in 2011 and 2013. If a home is damaged in a natural disaster, the authorities may refuse permission to repair or rebuild it because of the risk of further damage. In such cases, insurance can cover both the house and the plot, allowing the owner to establish a new home elsewhere. The scheme may also apply to an undamaged house if the ground beneath it has become unstable as a result of a natural disaster. But mapped risk alone does not trigger this entitlement. This is a crucial difference from Iceland, where a high level of mapped risk can lead to preventive protection or the purchase of a home before a natural disaster strikes.

What should Norway do?

Known risk cannot simply be left unaddressed. Norway should develop specific criteria for what level of landslide risk is acceptable for existing development. It may not be realistic to impose the same safety standards on older homes as on new buildings. We may have to accept a somewhat higher level of risk, provided that we prioritise the measures that deliver the greatest possible safety for the money spent.

The most important question, however, is what happens once the mapping is complete. A hazard map should not be the end product, but the start of a defined process: Is the risk acceptable, or must the home be protected or vacated? The financing must also be in place before action is needed.

As long as the same type of hazard map leads to different action depending on which municipality a person lives in, the mapping is only half the job. Under today’s system, families who already know that they live in a hazard zone are left alone with both the knowledge and the risk.