Flood vulnerability factors for smarter risk modelling

08-10-2025

In flood risk modelling, precision matters. That’s why we’re introducing Flood Vulnerability Factors, an innovation that uses property attributes like typology and geometry to refine damage estimates and support smarter underwriting and lending decisions.

Flood water and houses in Salisbury after heavy rain and River Avon bursts it's banks

SamaneSharifi Monfared

As a backend developer and climate change data analytics specialist at Haskoning, Samane brings 7 years of Python expertise to the Twinn FloodScore team. She specialises in API development, PostgreSQL Big Data processing, flood analytics, statistics generation, and QA, building Windows and web apps that advance climate data understanding. With skills in Python, machine learning, NLP, and image processing, she integrates AI into climate risk solutions to help organisations respond to environmental threats.

What is it?

Flood Vulnerability Factors are percentage-based indicators that quantify how much more – or less – damage a specific property type is likely to suffer in a flood compared to the average. These factors reflect both the typology (e.g., detached house vs. terraced house) and, where available, the actual footprint area of the building.

This means you can now move beyond the “average residential property” assumption used in most flood risk models and gain a more nuanced view of flood risk at the address level.

At its core, the product offers two tiers of insight:

Standard: Uses OS AddressBase property class data (e.g., detached, semi-detached, warehouse) and sector average building footprints.
Premium: Also utilises actual building geometry from OS NGD for even greater local precision, enabling unique total damage estimates for each property.

Why it matters

Many flood hazard models assume a standard property type for simplicity. But flood damage can vary significantly depending on the occupation, structure and size of the building. For example:

A large, detached house may suffer ~150% of the damage that the average residential property has, even if its per-square-metre damage is slightly lower.
Likewise, a warehouse or commercial unit may have a completely different vulnerability curve compared to a residential home.
By incorporating both property type and building size, Flood Vulnerability Factors allow insurers and lenders to fine-tune their risk assessments and pricing strategies by augmenting existing flood model implementations.

How it works

The product is built using:

Ordnance Survey AddressBase Premium for property classification.
Multi-Coloured Manual learnings to define damage curves.
FloodScore depth statistics to guide weighting for vulnerability comparisons.
Ordnance Survey NGD Buildings for geometric precision [Premium].

We calculate vulnerability using two methods:

Unweighted Mean: A simple relative average of vulnerability functions across all flood depth datapoints.
Weighted Mean: A relative average of vulnerability functions, adjusted for the likelihood of different flood depths occurring based on flood type (fluvial, pluvial, coastal).

The result? A single percentage that reflects the relative flood vulnerability of each property, available per square metre or as a total damage estimate:

Aerisl view of properties with flood vulnerability risk scores overlaid

It's incredibly rewarding to see complex research translated into practical, accessible solutions, ones that don't require specialist tools or costly implementation. This kind of innovation is key to advancing how the financial industry understands and manages risk."

David MartinProduct Director

Who benefits most

Insurers: Improve pricing accuracy and reduce blind spots in underwriting.
Banks & Lenders: Assess property-level flood risk for mortgage portfolios.
Catastrophe Modellers: Enhance exposure modelling with more granular data.
Risk Managers: Make smarter decisions using actuals instead of averages.

Flexible access, scalable insight

Flood Vulnerability Factors are available in both offline and integrated formats, direct from Haskoning or via your preferred reseller supplier:

Offline delivery: Send us your UPRNs or property references, and we’ll return the vulnerability data.
API integration: For real-time workflows, connect directly to our service and pay only for what you use.
Whether you’re working with AddressBase Premium or NGD, we’ve designed the product to be accessible and scalable, so you can choose the level of detail that fits your needs.

Part of a bigger picture

This innovation is part of our broader commitment in 2025 to enhancing risk intelligence across the UK. It reflects our drive to move beyond hazard and focus on vulnerability and exposure modelling, giving you the tools to make better decisions, faster.

Ready to upgrade your flood risk assessments?

Get in touch today to learn more, request a demo, or explore how Flood Vulnerability Factors can support your portfolio with smarter, property-level insights.