The EU Deforestation Regulation (EUDR) is about to kick into high gear, but a critical flaw threatens to cut hundreds of thousands of smallholder coffee farmers out of the European market. A new open-access dataset called Sample Earth aims to rectify this issue by helping mapmakers accurately classify smallholder farms instead of mislabeling them as forests.
Launched last week by the Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT), Sample Earth is a reference dataset designed to improve map accuracy in deforestation due diligence. The groups behind this initiative say that inaccurate mapping could lead to the exclusion of many smallholder coffee farmers from the European market as EUDR takes effect.
The reference dataset, hosted at Harvard Dataverse, utilizes more than two decades of Alliance research using satellite imagery to monitor land-cover changes across the Global South. According to the groups, current third-party accuracy assessments of maps used in deforestation due diligence lack a universal standard. Sample Earth is designed to set a new transparency and quality benchmark for map-based compliance while including built-in improvement mechanisms that allow mapmakers to access confidential land-use data without exposing individual farmers’ locations.
Thibaud Vantalon, a scientist at the Alliance of Bioversity International, explains that most maps are not accurate at local scales because the data is biased toward regions with a lot of training data. Remote regions are poorly mapped, and Sample Earth aims to fill this gap in training data for smallholders.
Louis Reymondin, a data scientist at the Alliance, notes that maps are needed for due diligence, and buyers will likely steer clear of areas misclassified as ‘high risk’ for deforestation.
As an example, the groups point to the EU’s main reference map for the regulation, published in 2025, which misclassifies more than half of the coffee production zones in Colombia, China, Guatemala, and Mexico as forest, according to Alliance data. Louis Reymondin, a data scientist at the Alliance, notes that maps are needed for due diligence, and buyers will likely steer clear of areas misclassified as ‘high risk’ for deforestation. With Sample Earth, the groups invite governments, companies, NGOs, and research institutions to invest in expanding this inclusive, high-quality land-cover reference to preserve livelihoods and incentivize environmental protection.
The EUDR has become one of the most contentious pieces of legislation affecting the coffee industry. After multiple enforcement delays since the law passed in 2023, EU member states said last week they intend to push enforcement back another year, to December 2026 for large organizations and June 2027 for smaller ones (the delay still requires European Parliament approval). Noting the rapid loss of the world’s tropical forests, certain environmental groups and civil society organizations have criticized these delays. Attitudes toward the legislation in the coffee industry are mixed, with some companies voicing support for the current enforcement timeline based on their investments in compliance, while others maintain that the law is flawed, potentially resulting in unintended consequences for smallholders.
– Sample Earth is a new open-access dataset designed to help mapmakers accurately classify smallholder coffee farms instead of mislabeling them as forests. – The reference dataset utilizes satellite imagery to monitor land-cover changes across the Global South and aims to set a new transparency and quality benchmark for map-based compliance. – By investing in expanding Sample Earth, governments, companies, NGOs, and research institutions can preserve livelihoods and incentivize environmental protection.
CLOSE: What are your thoughts on the EU Deforestation Regulation (EUDR) and its potential impact on smallholder coffee farmers? Share your insights with us in the comments section below.
Questions & Answers
What is the allegation made by international researchers about smallholder coffee mapping?
International researchers claim widespread errors in smallholder coffee mapping, citing inaccuracies and inconsistencies.
What does the offered dataset aim to rectify in the smallholder coffee mapping issue?
The offered dataset aims to address issues with existing smallholder coffee mapping by providing a more accurate representation of coffee farms worldwide.
Information sourced from industry reports and news outlets.
