In New Zealand, landslides are becoming more frequent and destructive as climate change intensifies extreme rainfall events, raising serious concerns for communities and infrastructure across the country. Recent incidents, including widespread slips in Tairāwhiti and fatal landslides in the Bay of Plenty, have highlighted the growing scale of the threat.
Experts note that while New Zealand has always been prone to landslides due to its steep terrain and fragile geological conditions, climate change is now worsening the situation. Increased rainfall intensity is saturating soils more frequently, weakening slopes and triggering sudden and often catastrophic failures.
Landslides are already among the country’s most damaging natural hazards, causing an estimated annual loss of NZ$250–300 million. Over the past two centuries, they have resulted in more fatalities than earthquakes and volcanic eruptions combined, underscoring their long-term impact.
Scientists say the primary challenge now is predicting where future landslides are most likely to occur. Understanding these risk zones is critical for planning safer infrastructure and protecting vulnerable communities. However, the complex nature of landslides has historically made such predictions difficult.
Recent technological advancements are beginning to change that. Researchers are increasingly using machine learning, a form of artificial intelligence, to analyse large datasets that combine past landslide occurrences with environmental factors such as topography, vegetation cover, and rainfall patterns.
The rapid expansion of satellite imaging has played a crucial role in this progress. High-resolution data now allows scientists to map landslide damage immediately after storms and monitor environmental conditions across large areas. These datasets are then used to train algorithms that can identify patterns and predict high-risk zones.
Once developed, these models can be combined with weather forecasts to generate real-time landslide hazard maps. This enables authorities to anticipate risks more effectively, issue timely warnings, and make informed decisions about land use and infrastructure development.
Preliminary findings from ongoing research suggest that the relationship between rainfall and landslide risk is highly sensitive. Even small increases in rainfall intensity could significantly expand the areas vulnerable to landslides. Many slopes are already near their failure thresholds, meaning that minor climatic changes can trigger major impacts.
Natural solutions such as maintaining forest cover can help reduce risk by stabilising soil and absorbing rainfall. However, experts warn that under more extreme future climate conditions, even forested areas may become increasingly vulnerable.
The findings highlight the urgent need for both technological innovation and climate action. While improved data and predictive tools can help reduce immediate risks, limiting global warming remains essential to prevent the situation from worsening.
As New Zealand faces a future of more intense storms and shifting environmental conditions, scientists stress that better planning based on accurate risk assessment will be key. Identifying vulnerable areas today could help minimise damage, protect lives, and build resilience against the disasters of tomorrow.