Understanding the New Nature Study on Landslide Risks

A groundbreaking study published today in Scientific Reports, a Nature journal, has quantified how climate change could dramatically amplify landslide hazards in Aotearoa New Zealand. Titled 'Increasing landslide susceptibility and intensity under climate change for Aotearoa New Zealand,' the research led by scientists from the University of Canterbury reveals stark projections based on the devastating Cyclone Gabrielle event of 2023. This work not only analyzes the mechanics of rainfall-induced landslides (RILs) but also forecasts future risks under warming scenarios, offering critical insights for researchers, policymakers, and communities across New Zealand's North Island, where the impacts are most pronounced.

The study's timing is particularly poignant, coinciding with ongoing recovery efforts from recent extreme weather and heightened public awareness of natural hazards. Landslides remain New Zealand's deadliest natural disaster, claiming more lives historically than earthquakes, tsunamis, or volcanic activity combined. By leveraging advanced modeling, the researchers bridge the gap between past events and future threats, emphasizing the role of academic institutions like the University of Canterbury in advancing geohazard science.

Cyclone Gabrielle: A Benchmark for Extreme Landslide Events

Cyclone Gabrielle struck the North Island in February 2023, unleashing torrential rains that triggered an estimated 800,000 landslides, covering roughly 100 square kilometers of terrain. This ex-tropical cyclone, one of the most intense landslide-triggering storms globally, caused 11 deaths—four directly from landslides—and economic losses exceeding NZ$14.5 billion. Regions like Hawke's Bay and Gisborne/Tairāwhiti bore the brunt, with over 116,000 human-mapped landslides documented in detailed inventories by GNS Science and partners.

The event highlighted vulnerabilities in steep, deforested hill country, where intense rainfall overwhelmed soil stability. Antecedent soil moisture, land cover, and rainfall duration played key roles, but the sheer volume of precipitation—up to 500mm in 24 hours—proved overwhelming. Universities such as Canterbury and Auckland contributed to rapid post-event mapping using LiDAR and satellite data, enabling precise inventories that form the backbone of this new research.

Methodology: Cutting-Edge Modeling from University of Canterbury

The research employs a sophisticated data-driven framework, integrating generalized additive models (GAMs) with rainfall thresholds and national susceptibility maps. Researchers delineated geomorphological slope units across New Zealand using open-source tools like r.slopeunits, then calibrated models against Cyclone Gabrielle's landslide inventory (Version 1.0 by Leith et al., 2023).

Static factors (topography, geology, land cover) combine with dynamic triggers (rainfall intensity from MetService quantitative precipitation estimates, or QPE) and spatiotemporal variables. Future scenarios draw from downscaled CMIP6 climate models under a +2°C warming storyline, simulating a Gabrielle-like storm. Validation via receiver operating characteristic area under the curve (ROC-AUC) exceeded 0.94, underscoring reliability. This approach, pioneered at the University of Canterbury's geohazards team, advances beyond traditional physics-based models by capturing complex interactions at national scale.

Key Projections: Up to 90,000 More Landslides Under Warming

Under +2°C warming—a scenario aligned with current global trajectories—a replicated Cyclone Gabrielle could spawn up to 90,000 additional landslides, a 11% increase over 2023 totals. Areas of extreme density (over 10% slope coverage) might expand by 34%, particularly in northeastern North Island. This escalation stems from intensified rainfall and altered soil responses, with susceptibility rising disproportionately to precipitation gains.

Complementary work by Oliver Wigmore at Canterbury uses machine learning (gradient boosted decision trees) on Gabrielle data from Hawke's Bay and Gisborne, confirming forest cover's protective role and projecting national RIL surges under shared socioeconomic pathways (SSPs). These findings align with NIWA's high-intensity rainfall design system (HIRDS), forecasting more frequent extremes.

Photo by Alexandre Lecocq on Unsplash

University of Canterbury's Pivotal Role in Geohazard Research

Lead author Livio Dreyer and colleagues Thomas R. Robinson, Marwan Katurji, and James H. Williams hail from the University of Canterbury, a hub for earthquake and landslide engineering. Collaborating with Kerry Leith from Earth Sciences New Zealand (GNS Science), their interdisciplinary team exemplifies higher education's impact on national resilience. UC's post-Gabrielle LiDAR surveys quantified erosion volumes, informing global models.

This study builds on UC's legacy, including susceptibility mapping post-2016 Kaikōura earthquake. Academic output supports tools like GNS's landslide forecasts, integrated into emergency management. For aspiring researchers, UC offers robust programs in geological sciences, fostering expertise vital for climate adaptation.Read the full open-access study here.

Regional Vulnerabilities and Historical Context

Northeastern regions face heightened risks due to soft sedimentary geology and pastoral land use. Past events like 2017 Edgecumbe (10,000+ landslides) and 2022 Nelson underscore patterns: cyclones deliver prolonged, intense rain on preconditioned slopes. Climate models predict 20-50% rainfall intensification by 2100, exacerbating antecedent moisture effects.

Māori communities in Tairāwhiti, stewards of ancestral whenua, highlight colonial land practices' role in vulnerability. Universities like Auckland engage in indigenous-led research for holistic risk assessment.

Mitigation Strategies: From Nature-Based Solutions to Policy

Forest cover emerges as a key mitigator, preventing thousands of landslides per Manaaki Whenua models during Gabrielle. Recommendations include targeted reforestation, erosion controls, and zoning reforms. Integrating susceptibility maps into district plans could avert billions in losses.

• Retire high-risk pastoral land to native bush.
• Enhance early warning via NIWA-GNS systems.
• Invest in resilient infrastructure, informed by UC models.

Government's Cyclone Gabrielle recovery incorporates these insights, with universities leading capacity-building.

Broader Implications for New Zealand's Research Landscape

This publication elevates NZ's profile in global geohazard science, amid rising international interest post-IPCC reports. UC's contributions position it as a leader, attracting funding and talent. For higher education, it underscores demand for earth sciences graduates skilled in AI-driven modeling and climate projection.

Photo by Johan Mouchet on Unsplash

Stakeholder Perspectives and Future Research Directions

Experts like GNS's Chris Massey note rainfall intensity's primacy, while iwi leaders advocate cultural integration in planning. Future work may refine dynamic soil moisture models and sea-level interactions. Collaborative hubs at UC and Victoria University of Wellington are pivotal.

Actionable Insights for Communities and Academics

Communities should consult GNS landslide guidance; academics can access datasets for replication. Proactive adaptation safeguards lives and livelihoods as climate pressures mount.