Sensing erosion before the storm arrives

Appling rapid erosion hazard modelling to assess post-storm erosion risk.

Key messages

  • Extreme rainfall events like Cyclone Alfred can rapidly increase soil erosion risk.
  • Using weather rainfall data (5-minute, 1 km resolution), the Rapid Erosion Hazard Model predicts storm-driven erosion hazards at near real-time with high accuracy.
  • The resulting erosion risk maps and time-series outputs support councils, landholders and WaterNSW in planning timely recovery and mitigation.
  • This approach provides decision-ready insights that strengthen resilience before, during and after extreme rainfall events.

Context

In March 2025, ex-Tropical Cyclone Alfred brought record rainfall and flooding to south-east Queensland and north-east New South Wales, the region’s first cyclone in 50 years.  

Beyond the visible flooding and infrastructure damage, fertile soil was washed away – a hidden loss with lasting impacts. The NSW Department of Climate Change, Energy, the Environment and Water (DCCEEW) applied its rapid erosion hazard modelling tool, funded through the NSW Climate Change Adaptation Strategy, to assess post-storm erosion risk.  

By combining radar rainfall data with vegetation, soil and terrain information, the team produced decision-ready maps that help councils, Local Land Services and water managers prioritise recovery actions, guide investment, and strengthen long-term climate adaptation planning across the region.

Key findings

Our analysis revealed significant erosion hazards across several regions in south-east QLD and northern NSW. The highest erosion risk areas during the peak rainfall periods were in the steep slopes with loose soils in Lismore and Brisbane. Identified findings showed:

  • High-risk areas: Small catchments in Lismore and steep slopes near Brisbane show the highest erosion risk.
  • Capacity of weather radar data: A strong correlation between radar-derived erosivity and on-ground observations confirmed the value of near real-time radar data.
  • Improved accuracy: The use of 5-minute rainfall intervals allowed the model to capture sudden erosivity surges.
  • Actionable products: erosion risk maps, time-series analysis, post-event assessments to inform strategies.
  • Why it matters: Findings support councils, Local Land Services NSW and WaterNSW in designing emergency and long-term actions.
Figure 1. Rapid erosion hazard assessment using weather radar rainfall data and erosion modelling for ex-Tropical Cyclone Alfred (8-9th March, 2025).

Strengthening climate adaptation

The study demonstrates how integrating weather radar data with erosion hazard modelling strengthens climate adaptation.

  • For landholders: clear information on erosion risks for timely, targeted land management.
  • For soil and land managers: evidence-based planning for erosion control, monitoring change and improving land resilience.
  • For WaterNSW and councils: rapid assessments to protect water quality, infrastructure and community assets.

The use of this modelling capability in routine post-disaster assessments will help NSW adapt to more frequent and intense storm events under climate change and the ability to predict and map erosion hotspots before impacts occur ensures better allocation of resources and reduced long-term damage.

Applications

The Rapid Erosion Hazard model can be applied by local councils, Local Land Services (LLS) and WaterNSW in response to bushfires, flooding and severe weather.

By identifying regions with high erosion risk, LLS and National Parks can target resources and interventions more effectively, ensuring that short-term erosion control measures are applied where they are most needed. 

The rapid erosion assessment tool assists WaterNSW to manage water quality by identifying areas where soil erosion could affect drinking water catchments at extreme rainfall events.  

Our modelling helps councils plan erosion mitigation, from short-term actions like barriers and planting to long-term solutions like land restoration and sustainable farming. 

Further reading 

Linked Datasets