Using satellite imagery to understand and promote sustainable development

TL;DR

This paper reviews how satellite imagery combined with machine learning advances the measurement of sustainable development outcomes, addressing data scarcity, model accuracy, and future research directions.

Contribution

It synthesizes recent literature on satellite imagery and machine learning for sustainable development, highlighting challenges and proposing future research directions.

Findings

Satellite imagery's resolution and availability are rapidly increasing.

Machine learning models often struggle with noisy and scarce training data.

Current models show promising performance but need improvements for policy use.

Abstract

Accurate and comprehensive measurements of a range of sustainable development outcomes are fundamental inputs into both research and policy. We synthesize the growing literature that uses satellite imagery to understand these outcomes, with a focus on approaches that combine imagery with machine learning. We quantify the paucity of ground data on key human-related outcomes and the growing abundance and resolution (spatial, temporal, and spectral) of satellite imagery. We then review recent machine learning approaches to model-building in the context of scarce and noisy training data, highlighting how this noise often leads to incorrect assessment of models' predictive performance. We quantify recent model performance across multiple sustainable development domains, discuss research and policy applications, explore constraints to future progress, and highlight key research directions for the field.