These annual migrant flow estimates provide a comprehensive view of how people move between countries worldwide. Using advanced deep-learning techniques, the dataset combines official statistics, international migration databases, and innovative data sources to generate consistent and reliable estimates, including for countries with limited migration data. The result is a valuable resource for analyzing global migration trends and patterns over time.

Global Annual Migration Data Explorer
Details

Migrant flow data estimate the number of people who move from one country (the origin) to another country (the destination) within a specific year. These estimates provide a detailed picture of international migration patterns and help researchers, policymakers, and organizations understand how populations move across borders over time.

Data and Methodology

The estimates are generated using an advanced deep-learning framework developed by Thomas Gaskin and Guy J. Abel (2025). The model integrates and reconciles multiple migration data sources to produce consistent annual migration flow estimates. These sources include migrant stock data from the United Nations Department of Economic and Social Affairs, bias-adjusted migration flow estimates derived from Facebook data, harmonized migration statistics collected through the QuantMig initiative, and official migration records from selected countries.

A key strength of the approach is its ability to estimate migration flows for countries where migration data are scarce or unavailable. To achieve this, the neural network uses a broad set of predictive factors, including geographic characteristics, economic conditions, cultural connections, and political indicators. By learning relationships between these variables and known migration patterns, the model can generate more accurate estimates for data-poor regions and improve the overall global coverage of migration statistics.

To compare these annual results with longer-term five-year migration flow estimates based on demographic accounting models, visit the original:

Global Migration Flows Explorer
Reference

Gaskin, T., & Abel, G. J. (2026). Deep learning four decades of human migration. Published online in Nature. DOI: 10.1038/s41586-026-10611-7.

IIASA