Agricultural field boundaries produced by Trazo models for key geographies in South America for the 2023-2024 planting season. Post-processed.
# Trazo1 Field Boundaries — Early Adopters Program (2025)
> **Superseded — not the recommended data.** For current work use
> [`trazo3-fields`](../../trazo3-fields/). Part of the
> [EAP release](../llms.txt); see the [catalog root](../../catalog.json).
Field boundaries from the Trazo1 model, shared through the Early Adopters Program in late 2025. Trazo1 uses heavy boundary weighting (class weights 0.88 boundary / 0.04 background / 0.20 interior), making it the most sensitive model to soft boundaries — those defined by two abutting field interiors rather than a hard feature such as a road — and the most likely to fragment fields finely.
`brazil_rio_grande_do_sul_2024.parquet` is the "grassland left in" variant: grassland was not filtered out of this region as it was elsewhere, which is why it is the largest file here.
This is the only collection covering Paraná.
## Files (10)
- `argentina_chaco_2024`
- `argentina_cordoba_2024`
- `brazil_goias_2024`
- `brazil_mato_grosso_do_sul_2024`
- `brazil_para_2024`
- `brazil_parana_2024`
- `brazil_piaui_2024`
- `brazil_rio_grande_do_sul_2024`
- `brazil_tocantins_2024`
- `paraguay_chaco_2024`
## Schema
| Column | Meaning |
|---|---|
| `det_method` | detection method; constant `auto-imagery` |
| `id` | feature id within the file; not stable across models or releases |
| `area` | field area in **hectares** |
| `perimeter` | field perimeter in **metres** |
| `geometry` | MultiPolygon boundary, EPSG:4326 |
Narrower than the released data: no MapBiomas class (`mbmode24`) and no Hansen forest-loss
columns. All files in this collection share this schema, so they glob:
```sql
SELECT count(*) FROM read_parquet('eap-trazo1/*.parquet');
```
## Caveats
Superseded data from late 2025 — prefer [`trazo3-fields`](../../trazo3-fields/) for current
work. Boundaries are model-derived and have not been field-verified. Field counts are not
comparable across models: Trazo1 and Trazo2 fragment fields more finely than Trazo3, so they
report more boundaries over identical ground. Not intended to be used alone to support
deforestation- or conversion-free claims.
## References
- [Technical note (Grupp et al. 2026)](https://www.wri.org/research/field-boundaries-south-america)
- [EAP release overview](../llms.txt)
- [Model card](https://huggingface.co/worldresourcesinstitute/Trazo1)