A portolan/stac catalog of every known available non-AI generated field boundary.
The Crop Sequence Boundaries (CSB) developed with USDA's Economic Research Service, produces estimates of field boundaries, crop acreage, and crop rotations across the contiguous United States. It uses satellite imagery with other public data and is open source allowing users to conduct area and statistical analysis of planted U.S. commodities and provides insight on farmer cropping decisions.
NASS needed a representative field to predict crop planting based on common crop rotations such as corn-soy and ERS is using this product to study changes in farm management practices like tillage or cover cropping over time.
CSB represents non-confidential single crop field boundaries over a set time frame. It does not contain personal identifying information. The boundaries captured are of crops grown only, not ownership boundaries or tax parcels (unit of property). The data are from satellite imagery and publicly available data, it does not come from producers or agencies like the Farm Service Agency.
License and Liability <https://gee-community-catalog.org/projects/csb/#license-and-liability>)Browse this collection in the data browser, or start from the AGENTS.md for tested queries.
The latest edition is also available at a stable path: us_usda_cropland/latest/us_usda_cropland.parquet. All editions together through the S3 glob s3://ftw/harmonized-field-data/us_usda_cropland/year=*/*.parquet (see the AGENTS.md for the DuckDB setup; plain https cannot expand *).
Properties that are the same for every field are stored once, in the GeoParquet file's collection metadata rather than as columns (latest edition shown; a client reading only the table will not see them):
admin:country_code: USdetermination:datetime: 2024-01-01T00:00:00Zcrop:code_list: https://fiboa.org/code/us/usda/cropland.csvQuery the published files in place with DuckDB; nothing needs downloading first.
This catalog is a mirror: the data is produced and licensed by United States Department of Agriculture and republished here as cloud-native GeoParquet and PMTiles by Fields of the World. Each edition was downloaded from the source and converted with fiboa-cli 0.21.0, vecorel-cli 0.2.15:
The conversion is deterministic and lives in fiboa-cli; changes to how a column is mapped are made there, not in this catalog.
other — License and Liability (converter: License and Liability <https://gee-community-catalog.org/projects/csb/#license-and-liability>). Attribute the data to United States Department of Agriculture.
| — |
| us_usda_cropland-2020.json |
| 2021 | 7,590,146 | 3.9 GB | — | us_usda_cropland-2021.json |
| 2022 | 7,659,738 | 3.9 GB | — | us_usda_cropland-2022.json |
| 2023 | 7,637,878 | 3.9 GB | — | us_usda_cropland-2023.json |
| 2024 | 7,522,713 | 3.9 GB | 658.3 MB | us_usda_cropland-2024.json |
collection | large_string | The identifier of the collection. (spec) |
hcat:code | uint32 | The 10-digit HCAT code indicating the hierarchy of the crop. The first 4, 6, 8 digits select increasingly specific crop groups. (spec) |
crop:name | large_string | Crop name in the original language. (spec) |
bbox | struct<xmin: double, ymin: double, xmax: double, ymax: double> | The bounding box of the field. Per-feature covering column (GeoParquet 1.1), in the source CRS. (spec) |