USDA NASS Cropland Data Layer

Use in your agent

Inventory a keyless USDA Hawaii CDL archive in Parquet. Checked against the provider's documentation on Nov 1, 2026.

Access

• No account or key is required for the national-download files.
• The checked Hawaii ZIP contains 2023, 2024 and 2025 10-meter GeoTIFFs plus metadata and documentation.

Request

curl -O 'https://www.nass.usda.gov/Research_and_Science/Cropland/Release/datasets/Hawaii_CDL_2023_2024_2025.zip'

Format

ZIP archive containing three 10-meter Hawaii GeoTIFF rasters, auxiliary files, FGDC and ISO XML metadata, documentation and a readme.

Load into pandas and Parquet

Needs Python 3 with requests, pandas and pyarrow.

"""USDA NASS Hawaii CDL archive inventory -> pandas -> Parquet. No key."""
import io
import zipfile
import requests
import pandas as pd

URL = "https://www.nass.usda.gov/Research_and_Science/Cropland/Release/datasets/Hawaii_CDL_2023_2024_2025.zip"
headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64) SignalLabCheck/1.0"}
r = requests.get(URL, headers=headers, timeout=120)
r.raise_for_status()
with zipfile.ZipFile(io.BytesIO(r.content)) as zf:
    rows = [{"name": x.filename, "bytes": x.file_size, "compressed_bytes": x.compress_size}
            for x in zf.infolist() if not x.is_dir()]
df = pd.DataFrame(rows)
df.to_parquet("usda_nass_hawaii_cdl_archive.parquet", index=False)
print(len(df), "archive members", int(df["bytes"].sum()), "uncompressed bytes",
      int(df["name"].str.endswith(".tif").sum()), "GeoTIFF files")

Run by us on 2026-11-01 (python 3.13.5 | pandas 3.1.0 | pyarrow 26.0.0 | requests 2.34.2). Result: 18 archive members 187453219 uncompressed bytes 3 GeoTIFF files

Limits

• Rate limit: Not stated.
• National archives range into gigabytes; the checked Hawaii ZIP was 28,133,175 bytes compressed.
• The current 10-meter confidence layer is not offered as a direct download because of file size.

Quirks

• The sample code inventories archive contents; reading raster pixels requires a GeoTIFF library such as rasterio or GDAL.
• Do not compare raw pixel totals with official NASS acreage without accounting for classification and counting bias.
• Resolution, classifier, training inputs and class codes change across vintages; retain year-specific metadata.

Terms and attribution

• NASS calls the CDL public domain and free to redistribute.
• The data are provided as is, without a warranty for conclusions or results.

Sources

• https://www.nass.usda.gov/Research_and_Science/Cropland/sarsfaqs2.php: “The USDA NASS Cropland Data Layer and the data offered on the CroplandCROS website is provided to the public as is and is considered public domain and free to redistribute.”
• https://www.nass.usda.gov/Research_and_Science/Cropland/sarsfaqs2.php: “Continental US coverage (2008 to current), partial US coverage (1997-2007)”
• https://www.nass.usda.gov/Research_and_Science/Cropland/sarsfaqs2.php: “The CDL data can be downloaded in a raster-based GeoTIFF file format and used in most common GIS software.”
• https://www.nass.usda.gov/Research_and_Science/Cropland/Release/: “Download the national 2025 10-Meter CDL zipped file”