A new arXiv preprint suggests that some neutral-hydrogen structures seen toward nearby pulsars lie beyond the Local Bubble, while others may trace its wall. The study makes those placements by comparing peaks in three-dimensional dust-derived density profiles with H I absorption components, where H I denotes neutral hydrogen. Its most pronounced case is a variable feature toward B1929+10: the authors associate it with dust about 215 parsecs away, more than 100 parsecs beyond the modeled Bubble structure, but stop short of calling that location definitive.
How the comparison was made
To make the comparison, the researchers used four of five available pulsar sightlines, selected because they fell within the dust map’s 1.25-kiloparsec range. They combined dust-inferred total-neutral-hydrogen information with H I absorption temperatures along those lines of sight.
The dust part of the exercise began with one-dimensional extinction profiles drawn from posterior realisations and interpolated to 1 pc distance resolution. A fixed conversion factor of 3002 pc mag−1 cm−3 turned extinction into a total-neutral-hydrogen density estimate. Twelve realisations supplied statistical uncertainty. The 1 pc³ interpolation also means the density values are lower limits.
Potential structures were isolated by choosing the full width at half maximum—the width of a density peak at half its height—and selecting corresponding density values in Glue, a visualisation tool. The authors describe that step as qualitative rather than a quantitative decomposition. For velocity-based matching, they used a Monte Carlo distance converter, but kinematic-distance ambiguity and non-circular motions produced uncertainties of 50% to 100% in relevant sightlines.
The clearest matches
The clearest matches came from B0823+26 and B1133+16. B0823+26 had one H I component at 4.91 km/s and one clear dust structure at 340 pc, with a peak density of about 13 cm−3. Since the modeled Bubble wall in that direction lies inside 200 pc, the dust feature was placed beyond the Bubble. Toward B1133+16, two components at −2 and −3.5 km/s were associated with a diffuse dusty structure near 140 pc and with the Bubble wall.
B1929+10 produced a particularly important uncertain case. The most pronounced variable component, at 4.8 km/s, was associated with a dusty feature near 215 pc—more than 100 pc beyond the modeled Local Bubble structure. The paper treats an outside-the-Bubble location for this tiny-scale atomic structure, or TSAS, as likely but not settled.
Toward B2016+28, the first density peak probably corresponded to an absorbing structure in the Bubble wall. Four other peaks did not correlate well with the remaining H I Gaussian components, leaving those associations unresolved.
Bounds, not exact readings
The four sightlines also produced markedly different line-of-sight values. Mean total-neutral-hydrogen density ranged from 0.14 ± 0.02 to 1.64 ± 0.05 cm−3, while mean peak density ranged from 1.4 ± 0.5 to 18 ± 2 cm−3. Reported ionisation fractions ranged from 0.008 ± 0.002 to 0.11 ± 0.01. These are model-derived values.
The two pressure approaches gave different scales. Direct pressure lower limits generally fell between 20 and 700 K cm−3, while the geometric method produced values closer to theoretical expectations, including about 3,000 K cm−3 for a prominent B2016+28 structure. For another B2016+28 feature at roughly 760 pc, the reported geometric lower limit was about 1,400 K cm−3.
On ionisation, the authors interpreted the lowest-latitude pulsar sightlines as having upper limits near 1%. They said that scale was consistent with comparison values near 0.003 and with shock-model predictions of 0.01 to 0.03 for a roughly 200 km/s shock. The estimates may be biased high because pulsar dispersion measures can preferentially sample highly ionised regions.
A method with a narrow working range
Only four sightlines were analysed, so the reported values cover a small set of directions. Velocity blending and complex Galactic-plane structure also made it difficult to match every dust peak with an H I component.
The authors therefore recommend focusing future dust/H I matching on simple absorption spectra with fewer than three components and Galactic latitudes above roughly 10 degrees. The recommendation reflects the difficulty of interpreting crowded sightlines such as B2016+28.
Taken together, the results point to a geometry-informed way to compare dust maps with H I absorption on simpler, higher-latitude sightlines. They do not make a probable association definitive: the B1929+10 TSAS placement remains likely rather than certain.
The work is listed as arXiv:2608.26098v1, dated 26 August 2026.
Paper data and sources
Original title: Probing the Properties of Neutral Clouds in the Vicinity of the Local Bubble: Linking three-dimensional Dust with Neutral Hydrogen Absorption
Authors: Aishani Das-Ghosh, Nickolas Pingel, Snežana Stanimirović
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-26
DOI: Not available
Original paper · Full text