A new methodology leveraging satellite imagery and machine learning is offering a more granular understanding of human development within countries, moving beyond traditional national averages. The research, published in Nature Communications, promises to reveal disparities often obscured by national-level statistics and could reshape how policymakers allocate resources.
For over three decades, the United Nations Development Programme’s (UNDP) Human Development Report Office (HDRO) has been a leading voice in assessing national progress through the Human Development Index (HDI). The HDI, a composite statistic incorporating income, education, and health, has become a widely used alternative to gross domestic product as a measure of societal wellbeing. However, the HDRO acknowledges that significant variations in human development can exist *within* countries, and obtaining consistent local estimates has historically been a challenge due to limited subnational data.
The new study addresses this limitation by utilizing advanced technology to generate localized HDI estimates. Researchers developed a model that analyzes satellite imagery – encompassing factors detectable through remote sensing – and applies machine learning algorithms to infer levels of human development at the municipal and county levels. The study provides results for 61,530 municipalities and counties globally, and further assesses these differences using a 0.1° x 0.1° grid, demonstrating how conclusions can shift with increased geographic detail.
Heriberto Tapia, Research and Strategic Partnership Advisor at the HDRO, co-authored the research alongside academic collaborators from the Stanford Doerr School of Sustainability, the California Institute of Technology (Caltech), and the University of British Columbia (UBC). This collaboration underscores the HDRO’s commitment to innovation in human development metrics and its partnerships with leading research institutions to bolster the evidence base for policy decisions.
The implications of this research extend beyond academic interest. By providing a more detailed picture of human development, the localized HDI estimates can help identify areas within countries that are lagging behind, allowing for more targeted interventions. Here’s particularly crucial in large and diverse nations where national averages may mask significant regional inequalities. The study’s findings could inform resource allocation, infrastructure development, and social programs, ensuring that assistance reaches those who need it most.
Researchers emphasize that these estimates are designed to *complement*, not replace, official national HDI reporting. The model generates estimates based on satellite imagery, rather than direct measurement in every location, and the precision of the estimates can vary depending on the context. However, the study demonstrates the potential of leveraging readily available data sources – such as satellite imagery – to overcome data gaps and provide a more nuanced understanding of human development.
The development of the Nature Relationship Index (NRI), currently under development by the HDRO and expected to be featured in the 2026 Human Development Report, reflects a broader trend towards expanding the definition of national progress. Inspired by the HDI, the NRI seeks to recognize the importance of mutually beneficial relationships between people and the natural world. The conceptual foundations of the NRI were outlined in a paper published in Nature in .
The ability to assess human development at a highly granular level also has implications for understanding and addressing global challenges such as climate change and disaster risk. A recent study published in Nature Communications () identified regions particularly vulnerable to high-impact heatwaves, noting that some of the most at-risk areas are also those with limited healthcare and energy resources, as defined by the UN Human Development Index. More detailed HDI estimates could help pinpoint specific communities within these regions that require urgent assistance and adaptation measures.
the methodology employed in this research – utilizing machine learning and satellite imagery – has broader applications beyond human development. The researchers have published the satellite features used to generate the HDI estimates, allowing others to increase the spatial resolution of administrative data detectable via imagery. This opens up possibilities for improving data collection and analysis in a wide range of fields, from environmental monitoring to urban planning.
The HDRO’s continued investment in innovative metrics and partnerships with research institutions signals a commitment to refining our understanding of human progress. As the world faces increasingly complex challenges, the ability to measure and monitor development at a local level will be crucial for ensuring that no one is left behind. The new localized HDI estimates represent a significant step forward in this effort, offering a more nuanced and actionable picture of human wellbeing across the globe.
