Urban Heat Index

Overview

The NatureScore Urban Heat Index combines CAPA Strategies’ high-resolution, ground-level temperature measurements with NatureQuant’s natural and built-environment intelligence to create a comprehensive indicator of urban heat. CAPA’s mobile sensing approach, capturing temperatures at 2-meter heights throughout the day, offers far greater detail than stationary sensors or satellite methods. These measurements are integrated with environmental attributes such as biomass, impervious surfaces, building footprints, water bodies, and roadway density, then processed through an AI-driven model to identify where and why urban heat islands form. In short, the index measures how well an area can regulate the elevated temperatures typical of dense urban environments.

Codebook

The codebook provides a standardized reference for all variables in this dataset, including each column’s name, format, definition, notes, and (when applicable) its valid values. Use this section to understand how fields are structured, how to interpret individual variables, and to ensure consistent analysis across different tools and workflows. The codebook is designed to support quick discovery, reduce ambiguity, and make it easier to work confidently with the dataset

Coverage

The Urban Heat Index has been collected in the U.S. and Canada on an annual basis since 2024. Historical versions are not available, only the most recent snapshot is available on Dewey.

Key Concepts

What is the NatureScore Urban Heat Index

• The NatureScore Urban Heat Index is a composite, data-driven metric that combines measured air temperatures with environmental and socio-economic context to evaluate urban heat risk at a hyperlocal level.
• It leverages ground-level air temperature data collected by CAPA Strategies across U.S. cities, together with NatureQuant’s detailed mapping of natural and built elements, such as vegetation, impervious surfaces, building geometry, water bodies, and infrastructure, to model where and why urban heat islands form.

Modeling Approach

• Machine learning integration — The UHI model uses machine learning to weight natural and built environment features by how strongly they correlate with observed temperature data, producing a predictive “urban heat surface.”
• Inclusion of socio-economic vulnerability — The UHI not only measures physical heat risk, but also integrates a socio-economic dimension using the Area Deprivation Index (ADI). This ensures that neighborhoods with both high heat risk and social vulnerability (e.g. lower income, poor housing, lack of resources) are flagged for priority.
• Scalability & applicability — Although trained on U.S. cities, the framework is designed to be scalable to urban environments globally, enabling assessments and interventions via green infrastructure irrespective of location.

Interpreting the Heat Index Score

• Scores range from 1 to 10. A score of 1 represents the lowest priority (low heat risk and relatively favorable natural + socio-economic context), while 10 indicates the highest priority: areas that combine high heat risk, low natural environment coverage, and socio-economic vulnerability.
• Scores are city-relative. That means each neighborhood is ranked relative to other neighborhoods within the same city, not compared across different cities or regions.
• A higher score signals a priority area for urban greening or heat-mitigation interventions (e.g. planting trees, creating green spaces, increasing shade, reducing impervious surfaces, etc.) to reduce heat risk and support environmental justice.

Documentation & Further Reading

Full White Paper: NatureScore Urban Heat Index — Technical Documentation and Methodology Download the full PDF documentation