Independent Benchmarks
for
Deforestation Maps
Know which maps you can trust. Science-based benchmarks that help mapmakers prove accuracy and give companies confidence in sourcing and compliance.
Map A
Forest detected
Map B
No forest detected
Accuracy
94%
Verified
Trusted by mapmakers and sustainability teams worldwide
Gaps in your maps put sustainability at risk
Companies, governments, and mapmakers rely on satellite data to demonstrate deforestation-free sourcing. Yet different monitoring tools often report conflicting results for the same land. When accuracy is uncertain, risk increases.
Without independent benchmarks, it is impossible to know which maps to trust or how to improve them.
Data conflicts
Different maps produce different classifications for the same location.
Compliance risk
Misclassifications can lead to fines, shipment delays, or blocked contracts.
Smallholder exclusion
Incorrect labelling can unfairly exclude farmers from global markets.
Trusted, Science-Based Benchmarks
Sample Earth provides the independent source of truth for land-cover and deforestation maps, helping you meet global standards while protecting social equity.
Organisations gain clarity
Make sourcing and compliance decisions with confidence.
Mapmakers gain credibility
Make sourcing and compliance decisions with confidence.
Farmers gain fair representation
Ensure complex agroforestry systems are accurately captured.
1
st
89
Satelligence
Commercial
Precision
72%
Recall
72%
70%
COMPLETE
65%
DATA MGMT
66%
INCLUSIVE
2
nd
82
ETH-Cocoa
ETH Zurich
Precision
72%
Recall
72%
70%
COMPLETE
65%
DATA MGMT
66%
INCLUSIVE
3
rd
78
FDaP
Forest Data Partnership
Precision
72%
Recall
72%
70%
COMPLETE
65%
DATA MGMT
66%
INCLUSIVE
1
Satelligence
90%
2
ETH-Cocoa
87%
3
JRC-EU-V2
82%
4
JRC-EU-V1
79%
5
Dynamic World
72%
6
GFC-10%
53%
7
FDaP
48%
8
ESA-WOC
48%
9
GFC-30%
45%
10
GLC_FCS30D
41%
View Leaderboard
Helping the Right People Make
Confident Decisions
Mapmakers
Demonstrate accuracy, prove
credibility, and benchmark your tools.
Companies
Make responsible sourcing and EUDR
compliance decisions confidently.
Governments
Access transparent, independent data
to inform policy and enforce standards.
NGOs & Advocates
Ensure fair representation and prevent
misclassification of farmers.
Two Ways to Drive Verified Impact
Benchmark your maps
Independent evaluation to prove performance and support due diligence.
Confidential, unbiased assessment
Your maps are evaluated using a neutral protocol with no commercial conflicts of interest.
Consensus-validated reference data
Each data point is reviewed by multiple trained interpreters, with group review where consensus is not reached.
Local representation of smallholder systems
Reference data captures complex agroforestry landscapes often missed by automated models.
Improve your map accuracy
Locally calibrated datasets designed to support next-generation Earth Observation models for deforestation monitoring and ethical sourcing.
High-quality reference dataset
Millions of hand-annotated GPS points for training and validating land-cover models.
Commodity-specific data
Built for applications such as coffee and cocoa supply chain monitoring.
Community-powered collection
Built and continuously updated by a trained global network of image interpreters to reduce geographic and social bias.
Open access for non-commercial use
Free for non-commercial use to support producer governments, researchers, and industry collaboration.
Why organizations choose us
Trusted by global leaders committed to building resilient, climate-smart agriculture.
For EUDR to succeed, we need to lower the burden of monitoring and reporting, and we need to ensure that longstanding smallholder farms can be reliably reported as non-deforested areas. AI combined with satellite imagery is a powerful tool that can help address these challenges, but AI systems are only as good as their training and validation data.
Dan Morris
Research Scientist, Google AI for Nature and Society
Global forest maps have advanced, but without open, standardized reference data, progress in disambiguating forest land use from other land use like cacao and coffee agroforestry remains limited. Today, initiatives like the Forest Data Partnership and DIASCA are putting efforts such as Sample Earth high on the global agenda as we work to define and standardize guidelines for open reference data collection.
Rémi d'Annunzio
Forestry Officer at FAO and product manager of Whisp
High-quality data and data-based action are the foundation for compliance with deforestation-free rules and net-zero carbon emission targets. However, highly accurate public data is rare. This poses a significant risk to all stakeholders involved. A standard to deliver highly accurate and transparent data in partnership with governments and farmers is of critical importance more than ever.
Michael Matarasso
Impact Director and Head of North America at the World Cocoa Foundation (WCF)
The coffee sector faces a real challenge in mapping millions of smallholders — a task that is critical for EUDR compliance but often out of reach for cooperatives working in remote, high-risk regions. We're excited to partner with CIAT to produce the world's first map of coffee-growing areas on a smallholder level, as a public good. By combining local ground truth data with cutting-edge satellite and AI technology, we will ensure smallholders aren’t left behind, but are instead leading the way in proving their coffee is deforestation-free — securing access to global markets while protecting their forests and livelihoods.
Niels Wielaard
CEO Satelligence
Built by CGIAR and its research partners
Sample Earth is developed by the Terra-i / Alliance Bioversity-CIAT team with over 20 years of experience in remote sensing, forest monitoring, and ethical supply chain research. We combine scientific rigour with open, inclusive tools for global impact.
Powered by CGIAR: The world's largest public agricultural research partnership.
Choose the right plan for your climate resilience journey.
Tailored insights to match your needs, from global overview to hyper-local asset protection.
Benchmarking
from
$ 40, 000
per crop & country
Get independent, science-based benchmarking to showcase your map’s performance, build trust with clients, and support responsible sourcing and compliance.
Key features:
Independent benchmarking report for clients, auditors, regulators
Confidence intervals and spatial error map
Consensus-validated reference data
Local smallholder agroforestry representation
Request a Demo
Training Dataset
FREE
Access locally calibrated, hand-annotated datasets to train your models, reduce smallholder misclassification, and enhance deforestation monitoring.
Key features:
Evaluation of organization-specific climate risks (quantifiable)
Secure proprietary data integration
Quantifiable yield impact assessments
Strategic planning support for adaptation programs
Leverage shareable proprietary & open data for collaboration
Dedicated expert support & guidance
Access Training Data
Ready to verify your monitoring data?
Book a demo to benchmark your solution or explore available datasets.
Book a demo