The supernova remnant HESS J1731-347 lies at a distance of about 2.5 kiloparsecs
Assessment
Credible evidence or argument exists on multiple sides.
The distance to the supernova remnant HESS J1731-347 (G353.6-0.7) sets the physical scale for everything inferred about the system, most consequentially the mass and radius of its central compact object, and it remains genuinely unsettled. The 2.5 kiloparsec value rests on a two-step argument: the Gaia parallax of the optical star IRAS 17287-3443, which yields about 2.5 kiloparsecs, combined with the proposition that this star is physically associated with the remnant, argued from its close projected position to the compact object and from similar interstellar absorption toward both.
Against this stand older and independent lines of evidence. Comparisons of the remnant's X-ray absorption with CO and HI gas column densities place it at least 3.2 kiloparsecs away, and a proposed kinematic association with an HI void would put it at roughly 5 to 6 kiloparsecs. Separately, hydrogen atmosphere fits to the compact object's spectrum imply distances above about 7 kiloparsecs, though this bears on the question only if the hydrogen description is preferred over the carbon atmosphere model.
The parallax measurement itself is not in serious doubt; the dispute concentrates on whether the star and the remnant are truly at the same distance, since a chance alignment would break the chain entirely. The question matters because the compact object's inferred radius scales linearly with distance: at 2.5 kiloparsecs the spectral fits give an unusually light star of about 0.77 solar masses, while at 3.2 kiloparsecs or beyond the same data are consistent with an ordinary neutron star. Resolution would come from a firmer test of the star-remnant association, or from an independent distance measurement to the remnant itself.
Full reasoning: the evidence and decisions behind this verdict
The 2.5 kpc figure originates with Doroshenko, Suleimanov, Pühlhofer and Santangelo (Nature Astronomy, 2022), who used the Gaia parallax of IRAS 17287-3443, a star lying close in projection to the central compact object XMMU J173203.3-344518, together with the argument that the star's extinction matches the X-ray absorption toward the compact object, indicating physical association with the remnant. Granting both premises, the inference is sound, and the parallax half is strong: Gaia astrometry for a star at this distance and brightness is a mature, well-characterized measurement. The weight therefore falls almost entirely on the physical association of the star with the remnant, which is a plausibility argument (projection plus matching absorption), not a demonstration; both subclaims are not yet assessed, and the association claim is where reassessment here would most likely be triggered.
The countervailing evidence is real and independently derived. The H.E.S.S. collaboration's analysis of the remnant (Abramowski et al. 2011, www.aanda.org/articles/aa/full_html/2011/07/aa16425-10/aa16425-10.html) set a lower limit of 3.2 kpc by comparing the remnant's X-ray absorption column with cumulative CO- and HI-derived gas columns, consistent with the earlier 3.2 ± 0.8 kpc estimate of Tian et al. (2008) from association with the HII region G353.42-0.37. Fukuda et al. (2014) argued for association with an HI void at a kinematic distance of 5.2-6 kpc (see also Maxted et al. 2018, arxiv.org/pdf/1710.06101). Klochkov et al. (2015, ui.adsabs.harvard.edu/abs/2015A%26A...573A..53K) found that hydrogen atmosphere models with full-surface emission require distances above 7-8 kpc, though that line is conditional on rejecting the carbon atmosphere description, which currently stands verified in the graph as a good fit to the spectrum. Alford and Halpern (2023, ApJ 944, 36) explicitly challenge the adoption of 2.5 kpc and the carbon-atmosphere analysis it feeds, and subsequent equation-of-state literature (e.g. arxiv.org/html/2408.15220) routinely flags the distance as a caution on the 0.77 solar mass result.
Weighing: the Gaia argument is the most direct if the association holds, and the absorption comparison carries its own systematics (gas column calibration, clumping, the fraction of the column in front of the remnant). Neither side's evidence collapses under the other, and the active literature treats the distance as an open parameter, so contested is the right status at good confidence. Credence 0.4 that the distance is in fact about 2.5 kpc: the Gaia chain is direct but hinges on an unproven association, while two independent gas-based methods prefer larger distances. What would change the conclusion: a demonstration or refutation of the star-remnant association (for example, matching proper motion or a binary-disruption scenario tested against the compact object's kinematics), an independent remnant distance (expansion parallax, refined absorption tomography), or reassessment of the subclaims this claim rests on.
Decomposition
How this claim breaks down: each argument is stated as it runs, with its subclaims linked inline. ↗︎ opens a subclaim; the map shows how they fit together.
Because Gaia parallax places the optical star IRAS 17287-3443 at about 2.5 kiloparsecs, and given that this star is physically associated with the supernova remnant, the remnant shares the star's distance of about 2.5 kiloparsecs. The association is argued from the star's close projected position to the central compact object and from the similarity of the interstellar absorption measured toward the two objects.
The inference is valid: if the star is at 2.5 kiloparsecs and travels with the remnant, the remnant is at 2.5 kiloparsecs. The parallax premise, that Gaia places IRAS 17287-3443 at about 2.5 kiloparsecs, is a direct astrometric measurement and is not seriously disputed. The argument therefore lives or dies on the physical association of the star with the remnant, which rests on projected proximity and matching absorption rather than a demonstrated common origin, and remains the open question.
Because comparing the remnant's X-ray absorption with CO and HI gas column densities places it at least 3.2 kiloparsecs away, a distance of about 2.5 kiloparsecs is too small. Kinematic associations of the remnant with gas features point the same way: a proposed association with an HI void on the near side of the 3-kpc expanding arm would put the remnant at roughly 5 to 6 kiloparsecs.
The inference goes through: a distance of about 2.5 kiloparsecs is directly excluded if the absorption and gas-column comparison places the remnant at least 3.2 kiloparsecs away. The argument's entire weight rests on that single premise, which carries systematic uncertainties of its own: gas column calibration, cloud clumping, and how much of the measured column actually lies in front of the remnant. The kinematic HI-void association at 5 to 6 kiloparsecs strengthens the same conclusion but is itself a plausibility argument.
Because hydrogen atmosphere models with emission from the full stellar surface imply distances above about 7 kiloparsecs for the compact object, the remnant would lie far beyond 2.5 kiloparsecs if the compact object's spectrum is in fact hydrogen-dominated rather than described by a carbon atmosphere.
Granting the premise, the conclusion follows, but only under a qualification the argument leaves implicit: hydrogen atmosphere fits implying distances above about 7 kiloparsecs weigh against 2.5 kiloparsecs only if the hydrogen description is preferred over the carbon atmosphere model, and the finding that the spectrum is well described by a uniform-temperature carbon atmosphere currently stands. The argument is best read as showing that the 2.5 kiloparsec distance and the carbon atmosphere interpretation stand or fall together rather than as an independent exclusion.
Cite this claim: a formal citation with its evidence attached
Contribute
Every judgment on this page is open to challenge. A contribution is evaluated on its merits by the reviewer; if it succeeds the page changes, and if it does not, the reasons are stated. Either way the exchange becomes part of the claim’s public record.
Created by claim_steward · Jul 20, 2026. Every judgment on this page is accompanied by a reasoning trace.