# Forest Data Partnership Commodity Probability Maps: Cocoa, Coffee, Oil Palm, Rubber at 10 m, 2018 and 2024 (model 2026a)

## Facts

- Provider: Google for the Forest Data Partnership; anonymous mirror by Portolan Mirrors
- License: CC-BY-4.0 (https://developers.google.com/earth-engine/guides/forest_data_partnership_gcs_readme)
- Format: JSON, JPEG, TIFF (in sampled files)
- Stored at: Amazon S3, s3://us-west-2.opendata.source.coop/portolan-mirrors/forest-data-partnership-catalog (us-west-2)
- Access: public, readable without an account
- Category: Agriculture & Environment
- Source: https://developers.google.com/earth-engine/guides/forest_data_partnership_gcs_readme
- Page: https://scigantic.com/archive/bf5f8aac-e25d-5d9a-b349-1cc4289cbb8e
- Keywords: Agriculture & Environment, JSON, JPEG, TIFF, AWS Open Data

## About

Per-pixel probability, from 0 to 1, that land is planted with cocoa, coffee, oil palm or rubber, at 10 m resolution, from the Forest Data Partnership community model release 2026a produced by Google, used for deforestation-free supply chain monitoring (for example under the EU Deforestation Regulation, EUDR). This mirror covers the West African cocoa belt for 2018 and 2024: 224 one-degree tiles per commodity and year from Sierra Leone to Gabon (15 W to 17 E, 1 S to 11 N), eight collections (cocoa-2018, cocoa-2024, coffee-, palm- and rubber-) of about 110 GiB each. Each item is a Float32 Cloud Optimized GeoTIFF of about 11,133 x 11,133 pixels in EPSG:4326 with a thumbnail, ids lng_ _lat_ , and each collection adds a pan-tropical overview COG of all 5,516 tiles at about 320 m and a STAC GeoParquet items.parquet. Tiles were copied unchanged from Google's requester-pays GCS bucket and checked against Google's MD5. Read with rasterio, GDAL /vsicurl/ or DuckDB. Licensed CC BY 4.0; attribute "Produced by Google for the Forest Data Partnership".

## Load the data

```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='portolan-mirrors/forest-data-partnership-catalog/', Delimiter='/')
```

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

## Data structure

File types in sampled files: JSON, JPEG, TIFF

Example file: `portolan-mirrors/forest-data-partnership-catalog/cocoa-2018/lng_-10_lat_10/lng_-10_lat_10.json`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/bf5f8aac-e25d-5d9a-b349-1cc4289cbb8e
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/bf5f8aac-e25d-5d9a-b349-1cc4289cbb8e
- Over MCP: `get_schema_card` with id `bf5f8aac-e25d-5d9a-b349-1cc4289cbb8e` at https://api.scigantic.com/mcp

## Related topics

- [JSON datasets](https://scigantic.com/topics/json)
- [TIFF datasets](https://scigantic.com/topics/tiff)
- [JPEG datasets](https://scigantic.com/topics/jpeg)
- [Agriculture & Environment datasets](https://scigantic.com/topics/agriculture-and-environment)
