Satellite Data Handbook Aims to Improve Farm Stats
A new Handbook on Remote Sensing for Agricultural Statistics has been released by the United Nations, offering guidance on how satellite data can be used to produce reliable agricultural statistics. NASA Harvest contributed to the publication and highlighted its importance for improving how governments convert satellite imagery into trustworthy agricultural data.
The handbook addresses a growing challenge in agricultural monitoring. Climate disruptions, geopolitical conflicts, and the rising cost of field surveys have made traditional data collection methods more difficult. At the same time, accurate crop information has become increasingly important for policy decisions, food security planning, and global agricultural markets.
Although satellite imagery is now widely accessible, turning that data into credible statistics remains complex. Maps generated from satellite analysis often appear highly precise, yet the uncertainty behind them is not always clearly explained. In many cases, these maps are used to guide policy and trade decisions even when the underlying data may contain significant limitations.
To address this issue, the handbook establishes standards for how satellite-based agricultural evidence should be produced, documented, and evaluated.
Designed primarily for national statistical agencies responsible for official agricultural reporting, the guide includes practical methods, case studies, and recommendations for communicating uncertainty in data. Its goal is to help countries transition from experimental uses of satellite imagery to dependable statistical systems.
One notable contribution from NASA Harvest appears in Chapter 23, which examines sunflower production in Ukraine during the 2023 growing season. Ongoing conflict had made on-the-ground surveys nearly impossible, yet understanding sunflower acreage remained vital because of the crop’s role in both Ukraine’s economy and global vegetable oil markets.
Researchers used imagery from the Copernicus Programme satellites Sentinel-1 and Sentinel-2 to identify potential sunflower fields. However, rather than relying solely on the generated maps, analysts applied a statistical sampling process.
Independent photo-interpreters reviewed sample locations multiple times, resolving disagreements to produce more accurate estimates of planted area while also calculating uncertainty.
“The challenge isn’t producing a map," said Becker-Reshef, NASA Harvest’s Co-Director. "It’s being able to stand behind the number that comes from it, especially when conditions are far from ideal.”
The handbook emphasizes that clearly acknowledging uncertainty can make satellite data more valuable for official statistics. By encouraging transparent methods, it helps governments determine when satellite-based estimates are reliable enough to support important agricultural decisions.
Photo Credit: un-handbook-on-remote-sensing-for-agricultural-statistics