Free, open daily exports of non-specialty VDL2/ACARS application messages received by the vdliq feeder network. One Release per UTC month holds one Parquet file per day.
The public derivative contains recognized position, weather, OOOI, and ATS/operational messages. Private collection is separate and is not represented by this reduced public dataset.
Each monthly release (tag YYYY-MM) holds one file per UTC day:
| Asset | Description |
|---|---|
vdl2_messages_<date>.parquet |
That day's public-safe non-specialty messages (zstd Parquet) |
Per-day row counts and SHA-256 checksums are listed in the release notes.
Download a specific day at
.../releases/download/<YYYY-MM>/vdl2_messages_<date>.parquet.
One row per public-safe decoded application message. Empty/link-only frames, engine/APU/ACMS/maintenance content, full raw decoder objects, station/feeder identity, RF measurements, and source UUIDs are intentionally excluded.
timestamp (UTC), icaoHex, fromHex, toHex, tail, flightNumber,
label (ACARS label), mode, sourceType, text, public_category, and
policy_version.
public_category is one of position, weather, oooi, or
ats_operational. The publication policy fails closed: unclassified content is
not included.
import duckdb
# read a specific day straight from GitHub (no download needed)
duckdb.sql("SELECT timestamp, tail, label, text FROM read_parquet('https://github.com/Sky-Power-Services/vdliq-data/releases/download/2026-06/vdl2_messages_2026-06-29.parquet') LIMIT 10")import pandas as pd
df = pd.read_parquet("vdl2_messages_2026-06-29.parquet")Open Database License (ODbL-1.0). Attribution: "vdliq feeder network".