Step 2 · Data Library

The project's datasets, level by level

Every observation, product, figure and calibration record in the research archive, catalogued with its processing level, calibration scale, source availability and known limitations. Datasets that are known to exist but are not yet in the archive are listed too.

A · Explore data

Which observations exist, which telescope produced them, what question they support and what students can explore. See the processing levels, the recommended dataset and all datasets.

B · Scientific Archives

Which files and products exist, at what processing level, in which formats, and what can eventually be downloaded. See the archives by telescope.

A · Explore data · Processing levels

L0 → L8: from recorded spectrum to reproduced result

Each level is made from the ones below it. The chips show where the datasets in the archive sit. Click one to open its record.

Terminology. L0 here means the averaged power spectra written by the acquisition software. DISH222 does not store raw I/Q radio samples, so no dataset in the archive is "raw SDR data".

Recommended starting dataset

26 Aug 2026 · outer-Galaxy raster with DISH222

An automated raster from 02:48 to 05:48 UTC across Galactic longitudes 146°–226° and latitudes −10° to +20°. It is currently the strongest reproducible chain in the archive: the native averaged spectra (776 files) and the pointing manifest (313 positions) are both available, alongside every product made from them.

All products of this session are on the relative C0 (dB above baseline) scale. The external processing code that produced the tables and FITS files is not in the archive, so re-deriving them from the native spectra is itself a research task (Track A).

Acquisition
Processing
Science product
Visualization

Sky chart in right ascension and declination with the Galactic plane as a red curve and the ecliptic in violet; a contour map of H I line peak in dB covers the outer-Galaxy region near RA 4–8 h. — L7 · visualizationC0 · relative dB **What it shows:** the 26 Aug 2026 raster projected on the sky: 313 beam-stacked positions with the H I line peak in dB above the baseline. **When:** generated by the legacy stitcher on the day of the observation. **What to learn:** how individual pointings become a map, and why a relative scale is enough to see the Galactic plane but not to compare brightness with other telescopes.

Showcase product

The all-sky "TRUSTED" composite

This composite combines six observing sessions between 26 Aug and 3 Sep 2026 into one map of peak brightness and velocity, overlaid on the LAB reference survey. It keeps only pixels that pass a selection rule (40–160 K, velocities within ±35 km/s, a dawn Sun band and speckles removed).

Showcase, not a starting point. It is a derived, curated (L6) product whose native lineage is incomplete in the research archive: only one of its six source sessions has native spectra available. Its selection cuts remove faint and high-velocity gas by construction. Its calibration is not recorded in the product metadata; the historical C2 scale is inferred from project chronology (C2 was in use when the composite was made).

Open its dataset record

Two sky maps in right ascension and declination: DISH222 velocity and brightness values that passed the TRUSTED selection, drawn over the faded LAB survey, with boxes marking the sessions of 26 Aug, 30 Aug, 1 Sep, 2 Sep and 3 Sep. — L7 · visualizationShowcaseCalibration not recorded · C2 inferred **What it shows:** DISH222 values that passed the selection rule, over the LAB survey. **When:** about 4 Sep 2026. **Calibration:** not recorded in the product metadata; C2 inferred from project chronology. **What to learn:** how multi-session maps are assembled and how selection rules shape what a map can say.

Browse

All catalogued datasets

Open a record for its purpose, processing, calibration, lineage and known limitations. Research files are held in the project's research archive and are not published on this site. Ask the project team about access.

B · Scientific Archives

What actual files and products exist

Archives are organised by telescope, then by processing stage. Placeholders mark stages whose files are not yet ingested. Nothing is downloadable unless it says so.

Live archiveLive data explorer

PSU fixed radio telescope archive

Historical observations from the 5 m zenith-pointing telescope at Prince Sultan University.

Browse saved spectra, complete observing sessions, time–frequency views, available historical Right Ascension products, and original archived data files. This is the primary interface for exploring the fixed-telescope data.

Real saved spectra
Full observing sessions
Time–frequency views
Historical Right Ascension products
Original archived files, downloadable

Explore the live archive →

The archive is hosted separately from this portal; no observation files are duplicated here.

Live explorer
Available
Awaiting ingestion
Planned

PSU Fixed TelescopeLive explorerFixed · historical
Historical observations are browsable in the fixed-telescope Explorer. Archive lineage from original spectra to a planned modern archive:
- TXT spectra
- MAT bundles
- RA/LST-corrected MAT bundles
- master MATLAB structure
- future FITS / CSV / JSON
PNU Fixed TelescopeAwaiting ingestionFixed · historical

The same archive structure as a set of placeholders: original, aggregated and processed data, FITS, metadata and derived products.
PSU Moving Telescope / DISH222Available on requestMoving · active
Products catalogued from the registry:
- NPZ spectra
- scan manifests
- derived products (cubes, maps, figures)
- future FITS / CSV / JSON

Archive processing levels

What was measured versus what was derived later

Historical lineage (A), modern reprocessing (P) and the DISH222 levels (L) are separate vocabularies. In every one, only the lowest stage is a measurement.

Level	Meaning	Measured or derived
A0	Original recorded spectrum (fixed telescopes)	Measured
A1	Aggregated dataset	Repackaged from A0
A2	Historical coordinate-corrected dataset	Derived
A3	Station master dataset	Derived
P0	Faithfully decoded historical observation	Measured, re-encoded
P1 – P3	Modern celestial coordinates · Doppler / spectral frame · absolute calibration	Derived
P4	Community / FITS archive product	Derived
L0 – L8	DISH222 levels, from native averaged spectra (L0) to independently reproduced products (L8). See Data Library.	L0 measured; L1 – L8 derived