BASTION — Flood Financial Impact Platform

Problem Statement

The Reality of Climate Volatility. Climate change has transformed extreme weather from a rare occurrence into a frequent, unpredictable reality. In regions like Catalonia and Valencia, intensifying storm patterns are overwhelming traditional infrastructure. When floodwaters rise in these densely built environments, the resulting damage to homes, businesses, and supply chains costs billions of euros and takes years to recover.

The Failure of Current Models

Traditional risk assessment suffers from three critical gaps:

• Visibility Gap: Insurers only realize their true financial exposure after the event, leading to "scrambling" for capital.
• Resolution Gap: Models rely on broad zip-code averages rather than building-specific physical data (slope, drainage, and permeability).

The Financial Consequence. This lack of transparency threatens the financial stability of insurers and leaves communities vulnerable. Without a shift from historical probability to real-time engineering precision, the insurance industry will remain unprepared for the scale of modern climate disasters.

Solution

A physics-based flood risk platform that replaces statistical guesswork with engineering-level precision — building by building, in real time.

Regulatory Baseline + Hyper-Local Intelligence

• Starts with SNCZI (Spain's official flood hazard maps) as the regulatory foundation
• Layers in ICGC terrain data: slope gradients, surface permeability and drainage capacity at street and building level

Satellite Ground Truth

• Sentinel-1 SAR imagery from the ESA Copernicus Programme validates the model continuously

• Penetrates cloud cover and captures flood extent as events unfold, not after the fact

Hyper-Customised Asset-Level Pricing

• Premiums calculated building by building based on measurable physical variables: drainage gradient, upstream catchment, rainfall intensity and structural exposure

• Eliminates cross-subsidisation from zone-averaged premiums

• Policy holders pay for their risk, not their neighbourhood's average

Real-Time Operational Engine

• Arduino-based IoT sensors measure water depth, pressure and temperature across river basins and urban drainage networks

• Meteorological API streams feed live precipitation and storm-front data into the model

• Delivers hours of advance warning before an event materialises

• Insurers can activate reserves, deploy loss adjusters and alert policyholders before damage occurs

• Every hour of anticipation is a direct improvement to the combined ratio

The Core Differentiator

• Every premium, reserve recommendation and risk classification comes with the technical data that justifies i

• No black-box estimates — only traceable, auditable calculation

• First platform to deliver transparency and optimisation simultaneously

Link to the Hackathon Challenges

The CASSINI Hackathon proposes three challenges: securing equitable and efficient access to water, tracking and preventing water pollution, and disaster risk monitoring. Our platform is primarily centred on the third.

Disaster Risk Monitoring

• Traditional models tell you what happened. Our platform tells you what is about to happen — and what it will cost

• Sentinel-1 SAR, SNCZI and ICGC data monitor flood risk continuously at building level

• When Arduino IoT sensors detect abnormal water levels, the system generates an immediate financial impact estimate — before the event materialises

• Insurers receive hours of advance warning to activate reserves, deploy loss adjusters and alert policyholders

• Disaster risk monitoring becomes a real-time financial decision engine

Team

Our Team coordinated by Joan San Martín as the primary referent, consists of five engineering students dedicated to transforming technical rigor into entrepreneurial innovation through precise and deterministic vulnerability modeling. We integrate specialized roles in geospatial analysis and technical architecture to deliver hyper-customized solutions that replace traditional statistical averages with high-resolution topographic evidence.