astro-legacy-archive/wham-halpha-map

WHAM velocity-integrated H-alpha map

This dataset contains LAMBDA's lambda_WHAM_1_256.fits bintable. Its one configuration is lambda_WHAM_1_256, the source filename stem verbatim. The source column is named UNKNOWN1; that opaque identifier is deliberately preserved. Embedded metadata makes the field, geometry, units, and pixel mapping self-describing without renaming or reshaping the source.

The Wisconsin H-alpha Mapper measured spectra with a one-degree beam and 12 km/s velocity resolution. This released map contains H-alpha intensity integrated from -80 to +80 km/s in the local standard of rest at each observed position. CADE transformed the survey to HEALPix using the interpolation method described by Paradis et al. (2012). LAMBDA made minor changes to the FITS headers to make them compliant with FITS standards.

Structure and interpretation

The source is Galactic, NESTED HEALPix at NSIDE=256, giving 786,432 pixels. It does not store one table row per pixel. Instead, its bintable has 768 rows and each UNKNOWN1 cell is a fixed vector of 1,024 float32 values. The Parquet column therefore remains a fixed-size list with the identical source shape. Element j of row i is HEALPix pixel i * 1024 + j in NESTED ordering. No pixel index, flattened duplicate, or normalised column is added.

The source unit is Rayleighs, mirrored exactly from TUNIT1. The map has a one-degree circular beam recorded by BMAJ=1.0. The source header also records TELESCOP=KittPeak;CTIO and INSTRUME=WHAM. All 786,432 values are finite and nonzero. The FITS header declares BAD_DATA=-1.63750E+30, but no element equals that sentinel, so the metadata records no coverage code. The 204 negative intensities remain source measurements and are not given an inferred meaning.

No extinction correction has been applied. The one-degree beam is much broader than an NSIDE=256 pixel, so neighbouring pixels substantially oversample the interpolated beam and are not independent angular measurements. The embedded physics also records the source RESTFREQ=457108. GHz keyword.

The embedded healpix_map JSON records the source URL and hash, bintable header, configuration naming rule, column TFORM and unit, HEALPix geometry, coverage census, physical interpretation with evidence bases, credit, and the fixed regeneration command. Consumers should read this metadata instead of assuming that every HEALPix Parquet file has scalar rows.

The measured census provides additional integrity checks. The 786,432 elements contain no NaN, infinity, or zero. Their minimum is -1.1300189 Rayleighs, maximum 4008.2158 Rayleighs, median 1.7352594 Rayleighs, and float64 mean 7.394277478340879 Rayleighs. Exactly 204 values are negative. Those values are mirrored without clipping or a derived flag because the archive does not document them as missing.

The primary header's DATAMIN and DATAMAX match the measured extrema. The bintable also records RESTFREQ=457108., BMAJ=1.0 degrees, the two observatory sites, and the WHAM instrument name. These source cards remain in the embedded header text; they do not justify renaming UNKNOWN1 or changing the source unit spelling Rayleighs.

This is a constituent survey product. The separately held Finkbeiner H-alpha dataset is a composite of WHAM, VTSS, and SHASSA with different NSIDE, beam, table structure, processing mask, and coverage. They are distinct scientific products, not resolution variants of one array.

How to use

python -m venv .venv && .venv/bin/pip install datasets huggingface_hub pyarrow

from datasets import load_dataset
dataset = load_dataset("astro-legacy-archive/wham-halpha-map", "lambda_WHAM_1_256", split="train[:1]", columns=["UNKNOWN1"])
row = dataset[0]
print(row["UNKNOWN1"][:3])

[3.2924516201019287, 3.5031063556671143, 3.503103733062744]

import json
from huggingface_hub import hf_hub_download
import pyarrow.parquet as pq
path = hf_hub_download("astro-legacy-archive/wham-halpha-map", "lambda_WHAM_1_256.parquet", repo_type="dataset")
print(json.loads(pq.read_metadata(path).metadata[b"healpix_map"])["schema"])

astro-legacy-archive/healpix-map

The embedded document records Galactic coordinates, NESTED ordering, source unit Rayleighs, coverage, and the source packing rule. Consumers resolve vector cells by applying pixel_rule, which maps element j of row i to the recorded HEALPix pixel.

Provenance, rights, and citation

HEASARC states verbatim:

HEASARC materials are all available freely for your use.

This is a free-use statement rather than a named formal licence.

The source file is https://lambda.gsfc.nasa.gov/data/foregrounds/wham/lambda_WHAM_1_256.fits (3,156,480 bytes; SHA-256 252e43ce47e9b3e8b1179ec665cf9c680ad412eeaf3366941c73c8aa6a7c7e03). The authoritative product page is https://lambda.gsfc.nasa.gov/product/foreground/fg_wham_h_alpha_map_info.html; the download page is https://lambda.gsfc.nasa.gov/product/foreground/fg_wham_h_alpha_map_get.html. The source and output hashes are pinned in facts.yaml.

No named formal product licence is published. The HEASARC free-use statement quoted above is represented as other. Cite Haffner et al. (2003), Haffner et al. (2010), and Paradis et al. (2012).

LAMBDA requests the following acknowledgement:

We acknowledge the use of the Legacy Archive for Microwave Background Data Analysis (LAMBDA), part of the High Energy Astrophysics Science Archive Center (HEASARC). HEASARC/LAMBDA is a service of the Astrophysics Science Division at the NASA Goddard Space Flight Center.

The WHAM project requests:

The Wisconsin Hα Mapper and its Hα Sky Survey have been funded primarily by the National Science Foundation. The facility was designed and built with the help of the University of Wisconsin Graduate School, Physical Sciences Lab, and Space Astronomy Lab. NOAO staff at Kitt Peak and Cerro Tololo provided on-site support for its remote operation.

Errata and requests

To report an erratum, propose a correction, or ask a question about this dataset, open a Discussion or Pull Request in its Community tab; posting requires a free Hugging Face account. Dataset requests and archive-wide questions may be posted in the Community tab of any dataset. Confirmed corrections are made in the source conversion tools and published by rebuilding the dataset.