# Global Fields of The World (FTW) 2nd Edition: Field Boundary Predictions 2017-2025

## Facts

- Provider: Fields of The World / Taylor Geospatial
- License: CC-BY-4.0 (https://source.coop/ftw/global-data-2e)
- Format: JSON (in sampled files)
- Stored at: Amazon S3, s3://us-west-2.opendata.source.coop/ftw/global-data-2e (us-west-2)
- Access: public, readable without an account
- Category: Earth Observation & Geospatial
- Source: https://fieldsofthe.world
- Page: https://scigantic.com/archive/08f4db23-99c9-5bbe-8c57-96cd5ca3df97
- Keywords: Earth Observation & Geospatial, image segmentation, satellite imagery, JSON, AWS Open Data

## About

Second edition of the global agricultural field boundary predictions from Fields of The World (FTW), sponsored by Taylor Geospatial. An FTW instance segmentation model is run on the TGE Labs Sentinel-2 Level 3 quarterly cloudless mosaics for 2017 to 2025, producing per-pixel field probability Cloud Optimized GeoTIFFs and vectorized field polygons as GeoParquet, plus PMTiles for display. The 2018, 2022, 2023, 2024 and 2025 collections use the updated 2nd-edition processing, each with about 14,590 raster tiles and 205 to 225 million predicted parcels (2025: 225,519,173); other years keep their earlier published collections. Folders: raster/, vector/, index/, assets/ and intermediate/, with a STAC catalog.json walked by the Portolan browser and an AGENTS.md guide. Read GeoParquet with DuckDB or geopandas over HTTP range requests and mosaic COG tiles with rasterio. The README's Limitations section explains that these are model outputs, not survey data. Supports cropland area estimation, field size statistics and agricultural change analysis. Data licence CC BY 4.0; cite Kerner et al., Fields of The World.

## Load the data

```python
import os
from PIL import Image
im = Image.open(os.path.join(os.environ['SCIGANTIC_MOUNT_PATH'], 'vector/2025/thumbnail.png'))
print(im.size, im.mode)
```

Run in a Scigantic notebook on 2026-11-18 against `vector/2025/thumbnail.png`: PIL image (1024, 571) mode RGBA format PNG.

From any machine:

```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='ftw/global-data-2e/', Delimiter='/')
```

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

## Data structure

File types in sampled files: JSON

Example file: `ftw/global-data-2e/_transfer-preflight/20261106T151256Z-access.json`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/08f4db23-99c9-5bbe-8c57-96cd5ca3df97
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/08f4db23-99c9-5bbe-8c57-96cd5ca3df97
- Over MCP: `get_schema_card` with id `08f4db23-99c9-5bbe-8c57-96cd5ca3df97` at https://api.scigantic.com/mcp

## Related topics

- [Earth Observation & Geospatial datasets](https://scigantic.com/topics/earth-observation-and-geospatial)
- [JSON datasets](https://scigantic.com/topics/json)
- [Satellite imagery datasets](https://scigantic.com/topics/satellite-imagery)
- [Image segmentation datasets](https://scigantic.com/topics/image-segmentation)
