Sample Earth Tree Crop Reference Dataset: Coffee and Cocoa Samples, Colombia and Ghana

Provider
Alliance of Bioversity International and CIAT (Sample Earth)
License
CC-BY-4.0 (source)
Format
JPEG (in sampled files)
Stored at
Amazon S3, s3://us-west-2.opendata.source.coop/alliance-bioversity-international-ciat/sample-earth-2026-co-gh (us-west-2)
Access
Free and public. No account with the data provider is needed.
Category
Earth Observation & Geospatial
Source
https://source.coop/alliance-bioversity-international-ciat/sample-earth-2026-co-gh

About this dataset

Land-cover reference samples for training and validating coffee and cocoa maps, from the Sample Earth initiative of the Alliance of Bioversity International and CIAT, produced with Google support. Experts interpreted very high resolution satellite imagery inside 600 m x 600 m sampling cells and labelled polygons and points with four classes: coffee, cocoa, other tree crops (including bamboo) and seasonal agriculture. ghana.gpkg covers 258 sampling sites with 6,277 polygons and 13,366 points on imagery from 2021-12 to 2025-11; colombia.gpkg covers 147 sites with 6,672 polygons and 12,272 points on imagery from 2020-10 to 2025-09. Each GeoPackage (WGS 84, EPSG:4326) has four tables: land-cover-legend, sampling-area cell centres with imagery dates, and the labelled polygon and point layers. Quality-controlled labels support commodity crop mapping for the EU Deforestation Regulation (EUDR), deforestation-free supply chain monitoring, agroforestry and land-use studies. Read with geopandas or GDAL. Licensed CC BY 4.0; cite Alliance Bioversity-CIAT Sample Earth.

How to access this dataset in Python

import boto3
from botocore import UNSIGNED
from botocore.client import Config
s3 = boto3.client('s3', region_name='us-west-2', config=Config(signature_version=UNSIGNED))
s3.list_objects_v2(Bucket='us-west-2.opendata.source.coop', Prefix='alliance-bioversity-international-ciat/sample-earth-2026-co-gh/', Delimiter='/')

Lists the files with anonymous access. Runs on any machine, with no cloud account.

Data structure

File types in sampled files: JPEG

Files

Example file: alliance-bioversity-international-ciat/sample-earth-2026-co-gh/README.md

Analyze this dataset without downloading it

On Scigantic, Sample Earth Tree Crop Reference Dataset: Coffee and Cocoa Samples, Colombia and Ghana opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Earth Observation & Geospatial datasets