II Jornadas Red URBAN MOME. Barcelona 4 Noviembre 2025

Vulnerability map:  Climate-Ready BCN

Source: ICLEI Action Fund 2.0
Budget: €1M
Objective: Support citizens and public authorities in adapting to extreme climate events and reducing energy poverty
Implemented: From May 2023 to December 2025

Climate Vulnerability in Barcelona at building level

The Climate Vulnerability Map of Barcelona is a geospatial analysis tool designed to identify the buildings most at risk during extreme heat events.

It evaluates key performance indicators (KPIs) for buildings and supports climate change planning

It provides an assessment of all residential buildings in Barcelona(61,000)

What is it?

CVI

How we estimate the Climate Vulnerability Index (CVI)

Vulnerability index to asses heat wave resilience

Climate vulnerability is typically framed within three key dimensions defined in the IPCC's Third Assessment Report (Intergovernmental Panel on Climate Change) in 2001:

Exposure
Sensitivity
Adaptive capacity

Although most studies classify indicators using these three categories, our index introduces additional levels to provide a more nuanced analysis while remaining aligned with the traditional framework.

Climate variability and extreme events

• Track changes over time through indicators such as temperature, precipitation, and vegetation cover. Local data helps assess risks related to heatwaves, droughts, and floods.

Energy indicators

• Explore energy use patterns (gas and electricity consumption). Indoor temperature data is crucial for identifying households at risk during extreme heat events.

Building characteristics

• Assess features such as age, use, and size to understand their impact on energy use, indoor comfort, and structural vulnerability.

Indicators groups (KPIs)

Vulnerability map

Infrastructure indicators
Assess the availability of essential public services (schools, climate shelters, social housing) and the strength of social networks to understand community support capacity.

Health indicators
Analyze the relationship between climate and health outcomes to identify vulnerable populations and prepare healthcare systems.

Demographic indicators
Not all groups are equally vulnerable. Analyzing factors such as gender, age, income, migration status, and unemployment allows for the development of more detailed CVIs.

Socioeconomic indicators
Understand urban resilience through factors such as housing costs, energy poverty, household debt, and gaps in social protection, which worsen during economic crises.

Vulnerability map

Indicators groups (KPIs)

 Climate vulnerability Index (CVI)

1. Data preprocessing – Select input data to calculate the indicators.

2. Framework selection – Determine which indicators positively or negatively affect the CVI.

3. Granularity definition – Select the spatial scale.

4. Normalization & weighting – Harmonize indicators and assign weights.

5. Aggregation – Combine indicators into groups to obtain a final value.

⚠️ Challenge: Each study adapts its CVI to its specific context and priorities.

CVI Construction

Vulnerability map

Our approach:

Building the CVI

Vulnerability map

Use of empirical cumulative distribution function to assign a value to each building for a given KPI:

🎯 Outcome: A practical tool for both policymakers and citizens to explore climate vulnerability and support decision-making.

 Interactive map and CVI: 

• View individual KPI layers
• Customized weights
• Adaptable CVI

Use Calculation of aggregated value for a typology T

Use Calculation of aggregated value for a typology T

Data sources & harmonization

BEEMind: Data ontology

The ontology at the core of our solutions

Applied semantic web technologies:

Understanding and organizing data is as important as the algorithms themselves

Massive data integration

3. Data at building level

2. Data at census code level

1. Data at postal code level

How to integrate and link data?

Main data sources

Heterogenous data ingestion

Massive data integration

Data sources

big numbers

Massive data integration

Interoperable web semantic frameworks

Knowledge graph

Modelling

1. Overall predictive model: heterogenous temporal graph

MindCity:  Modelling

Vulnerability map

Graph Convolutional Neural Network (GCNN) models

To enable learning patterns, fill knowledge gaps and predict scenarious from geo-structured data

2. Weather downscaling microclimate model

Complementary models

Weather 

Cat boost decision tree

2.1. Heat waves prediction and communication

Complementary models

Weather 

3. Thermal demand models of buildings

Simulation of the thermal energy demand of 37,000 residential buildings.

The simulation engine is based on Reduced Order Grey Box Models (RC models) trained with TRNSYS simulations:

• Building-level predictions for space heating and cooling
• Total energy needs under 3 typologies and 17 archetypes and 5 indoor set point temperatures
• Predicting 5 retrofitting scenarios and 2 HVAC systems

Complementary models

big data IT architecture: ENMA

Software architecture

Interactive visualization

MindCity:  Climate Vulnerability Map Visualization

Export and data visualization

Vulnerability map

Vulnerability map

https://maps.climatereadybcn.eu/

Index evolution by streets and buildings

Queries and forecasts

Vulnerability map

MindCity:  Climate Vulnerability Map Visualization

Thank you

Jordi Cipriano
Director of  the Innovation Unit BEEGroup

cipriano@cimne.upc.edu