Suncorp and BoM unveil advanced hail forecasting model

Initiative enhances prediction accuracy

Suncorp and BoM unveil advanced hail forecasting model

Catastrophe & Flood

By Roxanne Libatique

Hailstorms remain challenging to predict in terms of timing, location, and potential damage. In response, a partnership between Suncorp Group and the Bureau of Meteorology (BoM) has developed a new forecasting model designed to improve accuracy.

This initiative, combining advanced radar technology with machine learning, aims to bring more precision to forecasting one of Australia’s most unpredictable weather perils.

Hail prediction model

Radar systems are essential in observing precipitation types, including rain, snow, and hail. While they play a critical role in tracking storms, distinguishing hail from rain remains difficult.

To tackle this challenge, Suncorp’s Natural Perils team, led by Dr Rhys Whitley, teamed up with BOM’s Science and Innovation Research Group to enhance radar data’s capabilities.

Published in the scientific journal “Atmospheric Measurement Techniques,” the team’s research outlines a hail prediction model that could pinpoint risk areas within storm systems more accurately.

Whitley explained that radar improvements can only go so far without additional data to capture hail’s complex dynamics.

“If there are very large hailstones, they’ll typically fall straight down. But other sizes of stones might deflect elsewhere within the storm, or high winds might accelerate their falling pattern. Melting can also occur before it hits the ground,” he said.

More precise hail forecasts on suburb level

Historically, weather forecasts could only issue broad hail warnings across wide areas. However, the study created a Hail Damage Estimate (HDE) score by combining BoM’s radar data with historical insurance claims, aiming for more precise hail forecasts on a suburb level.

The project uses machine learning to examine large datasets, covering past instances of damaging hail across the country. By identifying patterns within these data sources, the new model seeks to track hail with a higher level of detail, ultimately enabling real-time alerts for potentially affected areas.

Reducing damage and improving claims management

Australia frequently experiences severe hail events that can damage property and vehicles, often with high economic impact.

When a major hailstorm struck areas south-west of Brisbane in October 2020, residents in suburbs like Springfield Lakes and Greenbank saw substantial property damage and power outages. Such events underscore the need for more specific hail forecasts that can help residents prepare for potential impacts.

Suncorp said these advancements align with the company’s broader resilience strategy, designed to support communities and improve disaster preparedness.

With this new model, insurers can potentially refine their response and readiness, deploying resources where they are most needed.

Related Stories

Keep up with the latest news and events

Join our mailing list, it’s free!