MarineCadastre AIS Vessel Transit Data

Use in your agent

Index MarineCadastre AIS transit rasters in Parquet. Checked against the provider's documentation on Nov 1, 2026.

Access

• No account or key. The annual GeoTIFF directory and files are public downloads.

Request

curl 'https://ocmgeodatastor1.blob.core.windows.net/marinecadastre/ais/aistransit/index-aistransit.html'

Format

HTML file index listing annual AIS transit-count GeoTIFF names, modification dates and sizes.

Load into pandas and Parquet

Needs Python 3 with requests, pandas and pyarrow.

"""MarineCadastre annual AIS transit-count file inventory -> pandas -> Parquet."""
import re
import requests
import pandas as pd

URL = "https://ocmgeodatastor1.blob.core.windows.net/marinecadastre/ais/aistransit/index-aistransit.html"
r = requests.get(URL, headers={"User-Agent": "SignalLabCheck/1.0"}, timeout=60)
r.raise_for_status()
rows = re.findall(
    r'href="([^"]+\.tif)"[^<]*</a></td>\s*<td[^>]*>([^<]+)</td>\s*<td[^>]*>([^<]+)</td>',
    r.text,
)
df = pd.DataFrame(rows, columns=["name", "last_modified", "size"])
df["year"] = df["name"].str.extract(r"(\d{4})").astype("int64")
df["url"] = URL.rsplit("/", 1)[0] + "/" + df["name"]
df.to_parquet("marinecadastre_ais_transit_inventory.parquet", index=False)
print(len(df), "annual rasters", int(df["year"].min()), "to", int(df["year"].max()),
      "latest", df.loc[df["year"].idxmax(), "size"])

Run by us on 2026-11-01 (python 3.13.5 | pandas 3.1.0 | pyarrow 26.0.0 | requests 2.34.2). Result: 11 annual rasters 2015 to 2025 latest 534.46 MB

Limits

• Rate limit: Not stated.
• The checked annual files range from 284.48 MB to 534.46 MB each.
• The inventory skill writes file metadata to Parquet; it does not download the rasters.

Quirks

• Transit products are derived annual rasters, not raw AIS messages or vessel tracks.
• Use a raster reader for pixel values and coordinate reference metadata after selecting a year from the inventory.
• The current point-data year is partial, while the transit directory ends at the latest completed annual product.

Terms and attribution

• Government-server information is public domain unless specifically annotated otherwise.
• The provider says that public-domain information may be used freely.

Sources

• https://noaa.maps.arcgis.com/sharing/rest/content/items/22fb5273ccb54f6aa423f6667a856761/data?f=json: “Track and Transit Count data were derived from the annual broadcast points and are available for download in the table to the right.”
• https://noaa.maps.arcgis.com/sharing/rest/content/items/22fb5273ccb54f6aa423f6667a856761/data?f=json: “The data available in the table contains broadcast point data for U.S. coastal waters for calendar years 2009 through 2026.”
• https://noaa.maps.arcgis.com/sharing/rest/content/items/a25a72b4f3034f04b3130321b7577037/data?f=json: “The information on government servers is in the public domain, unless specifically annotated otherwise, and may be used freely by the user”