In a new article, published in Proceedings of the National Academy of Sciences, it is proposed to open archives of commercial satellite images to improve research on the Sustainable Development Goals (CSR) United Nations. Access to very high resolution satellite images (VHR) is crucial for monitoring smallholder farms, especially in low- and middle-income countries, where such agriculture is vital for food security.
"Small farmers, who often cultivate less than two hectares of land, produce more than 30% world food. However, their agricultural practices and productivity are poorly documented, partly due to the lack of available VHR satellite imagery, says Felicia O. Akinyemi, Associate Professor of the Department of Geomatics, Karlstad University. These pictures, the cost of which can reach 2 euros per square kilometer, are often too expensive for research institutions, working in these regions".
Researchers note, what are these initiatives, like US Landsat, EU Copernicus and NICFI, showed, how open satellite data can drive innovation in sustainability research. They suggest applying similar models to make VHR data available for non-commercial research, which can significantly improve the monitoring of smallholder farms and contribute to the achievement of several sustainable development goals, including Target 2: Zero hunger.
“My research focuses on Earth observation using satellite imagery and machine learning, to find out, how changes in land use are related to degradation processes in agroecosystems", says Akinyemi. "My interest was sparked by the expansion of agricultural land and the simultaneous loss of high-quality agricultural land due to urbanization in many parts of the world".
Focusing on West Africa, where agricultural extension fronts are of global importance, Felicia u 2022 received an individual scholarship from the EU named after Maria Sklodowska-Curie to conduct research within the framework of the LucFRes project.
Monitoring smallholder farming systems using satellite data poses several challenges. One of the main difficulties is that, that machine learning algorithms need field-tested data, which are often lacking in the regions, where small-scale agriculture predominates. Without reliable training data, model predictions become weak. "Besides, field sizes on small farms are often extremely small — from 0,25 hectare to 5 hectares, — and that means, that many publicly available satellite images are too low resolution to accurately represent crop types, especially in systems, where several cultures intertwine", says Akinyemi.
Akinyemi's research contributes to development goals, related to sustainable agriculture, in particular, the Zero Hunger and Land Degradation Neutrality goals.
"Due to frequent cloudiness during the growing season, we combined Sentinel-2 optical and Sentinel-1 radar data to analyze the spectral-temporal patterns on a monthly and bimonthly basis during the growing season. The use of satellite images with sub-meter resolution could significantly improve mapping", - she explains.
In a pilot study in southwestern Nigeria, the project is investigating, how changes in land use affect the sustainability of agricultural systems in the face of climate change. It combines satellite-based analysis of land-use change with local stakeholders' perceptions of future land use. This approach provides a deeper understanding of that, how to increase the adaptability of agriculture.
Source: https://phys.org/news/2025-04-satellite-image-archives-boost-sustainable.html