Data powering the Global Food Twin application
The Global Food Twin is a model that estimates the most likely transportation routes for food flows between producing and consuming regions worldwide. By combining food balance sheets, production maps, trade data, transportation networks, and optimization algorithms, we've created the best guess of how food moves across the globe—from producing to consuming regions—accounting for the complex multi-modal transportation systems. Visit a proof of concept application visualizing this model and global food flows here.
A few facts about the Global Food Twin:
This project was led by Cameron Kruse at Earth Genome working with Zia Mehrabi, Ginni Braich, and Kushank Bajaj Better Planet Laboratory on the model; Development Seed on the front end application; the model built on prior work by Jasper Verschuur Assistant Professor at Delft University.
This repository contains the data produced by the model and used by the Global Food Twin Application. More details about how this data was developed including the code used can be found in this repository GitHub repository.
This repository is organized into inputs and outputs. Below is a diagram of the folder structure. the all_files.md document contains a full file structure.
This data is licensed under the Creative Commons Attribution 4.0 International License.
If you use this data in accordance with the CC 4.0 license, please cite:
Better Planet Laboratory and Earth Genome. (2025). Global Food Twin. Harvard Dataverse. https://doi.org/10.7910/DVN/BLVPFU
Reach out to Cameron Kruse with questions.