What Do Deep Narrowband Images Reveal About Knots in the Pa 30 Supernova Remnant?

The filaments are chains of knots, not smooth streaks

The central result is morphological: in deep [S II] imaging, the radial filaments of Pa 30 are not continuous, smooth streaks. They consist of chains of tiny knots or clumps near the resolution limit of the [S II] image, which is 0.5 arcsec. The number of detected filaments is at least an order of magnitude greater than in earlier images, such as those of Fesen et al. (2023).

The knot separation and frequency are remarkably similar along each filament and across different filaments. The paper reports a characteristic knot spacing of roughly 1 to 6 arcsec along the radial extent of the filaments. This quasi-regular spacing is one of the features that any formation model must reproduce, and the authors note it contrasts with the morphology expected from evaporating clumps.

[O III] and [S II] trace the same filamentary structure

Earlier work had detected diffuse [O III] emission in Pa 30, but it was not established whether that emission came from the same structures seen in [S II]. The new [O III] images, astrometrically aligned with the [S II] images, show the same filamentary structure, and a cross-correlation analysis finds that [O III] filaments are cospatial with [S II] filaments. This is the first confirmation that the previously detected diffuse [O III] emission originates from the same [S II] filaments.

The [O III] data have a lower signal-to-noise ratio, so the paper cannot test whether the [S II]/[O III] ratio is constant or varies along or across the filaments. The authors state that such a test would require deeper [O III] imaging or a high angular resolution integral field unit spectroscopy campaign. This is a limit on what the current dataset can say about the physical conditions inside the filaments, not on the cospatiality result itself.

Knot sizes are measured against the image resolution

The paper measures cross-sectional profiles of the filaments in both [S II] and [O III]. The profiles show a central peak on top of a nebular background. A Gaussian fit to the median profile gives an FWHM of 0.81 ± 0.03 arcsec for [S II] and 1.07 ± 0.22 arcsec for [O III]. The [S II] value is slightly broader than the 0.50 arcsec seeing, which the authors interpret as some evidence that the filaments are partially resolved. Deconvolving the point-spread function yields a typical angular size of about 0.6 arcsec, or roughly 10^16 cm at the adopted distance of 2.3 ± 0.1 kpc.

The [O III] width of about 1 arcsec is consistent with the poorer seeing of the oxygen images, which had a median seeing of 0.85 ± 0.02 arcsec. The paper also notes a slight shift in the [O III] filament profile relative to the centered Gaussian fit, but states that the shift is likely not significant given the low signal-to-noise ratio in the oxygen data. Readers should treat the [S II] knot size as the better-constrained measurement and the [O III] width as largely set by image quality.

A nebula background sits beneath the filament peaks

The cross-sectional profiles do not decline directly to the sky level outside the filament cores. Instead, they flatten to a diffuse local background component that is brighter than the sky. This nebula background decreases toward the sky level only at a distance greater than about 90 arcsec from the center of the nebula.

The authors consider two explanations for this background: diffuse emission filling the regions between the bright filaments, or a large population of faint filaments that collectively appear diffuse. They argue against nearby resolved filaments as the cause, because the characteristic profile is derived via the median statistic, which should be robust to contaminants, and because in most instances the typical profiles do not intersect additional resolved filaments. The paper does not decide between the two remaining possibilities.

The nebula center limits any transverse stellar kick

Using a rolling Hough transform to extract filament orientations, the paper fits for the point about which the filament orientations are most nearly radial. The measured nebula center is consistent with the position of the central star, which the authors take to indicate that the remnant has not received a large transverse kick. The 3 sigma upper limit on the velocity of the stellar kick is about 100 km/s.

This is a quantitative constraint where previously only symmetry arguments and a low proper motion had been used. The fit is robust against the choice of rolling Hough transform parameters: the best-fitting center coordinates remain consistent across a broad grid of window lengths and smoothing radii, though suboptimal parameter combinations broaden the posterior. The most stable solutions come from window lengths of about 10 to 20 arcsec and smoothing radii of about 2 to 3 arcsec.

Which filament formation models survive the knot structure

The paper evaluates three proposed pictures for the filaments against the new knot morphology. In the central-star-wind picture, the fast 16,000 km/s wind ablates clumps in a slower ejecta shell. The authors set this aside because it does not naturally explain the ballistic motion of the filaments, in which expansion velocity is proportional to distance from the central star, and because the quasi-regular knot frequency contrasts with the expected morphology of evaporating clumps.

In the Rayleigh-Taylor picture with strong radiative cooling, dense ejecta encounter the ambient interstellar medium and strong cooling suppresses the Kelvin-Helmholtz instability at the tips of the fingers. With a density contrast of order 5000 and a velocity difference of 1000 km/s, the paper estimates a Kelvin-Helmholtz growth time of about 160 yr at scales around 10^16 cm, which the authors describe as somewhat marginal but able to produce knots along the filaments. The cooling time of the filament is much shorter, around 1 month, so thermal instability might accelerate condensation of knots initiated by Kelvin-Helmholtz instability.

In the wind-CSM interaction picture, a higher-density, lower-speed wind from the merged white dwarf interacts with circumstellar material at a distance of order 10^15 cm. Here the velocity contrast between filaments and the gas between them is modest, so the Kelvin-Helmholtz timescale is on the order of the age of the remnant, making it difficult for this model to account for the knots. The authors note that knots could still have formed by thermal instability at an earlier stage, but that this model does not include an energy equation and so makes no prediction about thermal instabilities.

The knots are the new constraint, not a settled origin

The discovery of clumpy substructure is presented as a new observational constraint on models for the origin and evolution of Pa 30's filamentary ejecta, not as a determination of which model is correct. The most decision-relevant comparison in the paper is between the quasi-regular knot spacing of about 1 to 6 arcsec and the morphologies predicted by the three formation pictures. A model that cannot produce regularly spaced knots along nearly ballistic radial filaments is disfavored by this dataset.

The paper's own ranking of the options is that the central-star-wind picture is set aside, the Rayleigh-Taylor model with strong radiative cooling can produce knots through Kelvin-Helmholtz instability with a somewhat marginal growth time, and the wind-CSM interaction model struggles to account for knots through Kelvin-Helmholtz instability alone. The clearest next step implied by the evidence is deeper [O III] imaging or high angular resolution integral field unit spectroscopy, which would test whether the [S II]/[O III] ratio varies along or across the filaments and would help distinguish diffuse emission from unresolved faint filaments in the nebula background.

Sources used in this answer