Bright supernovae reveal how surrounding gas shapes their light

An exploding star can shine more brightly when its expanding debris crashes into gas expelled before the explosion. The collision converts motion into radiation, while the surrounding material leaves signatures in the escaping light. But do the brightest hydrogen-rich supernovae form a distinct physical class, or share the same processes as fainter interacting explosions?

In a preprint, Yang Hu and colleagues investigate that question using 304 optical spectra of 106 hydrogen-rich superluminous supernovae observed by the Zwicky Transient Facility. Their main finding is that extraordinary brightness does not come with a unique hydrogen-line shape in the available comparisons.

Reading the surrounding gas

A spectrum separates light by wavelength. Hydrogen produces a series of features called Balmer lines, including Hα and Hβ. Their widths and shapes help researchers investigate the material emitting and scattering the light.

Of the 106 objects, 105 show narrow hydrogen emission associated with the supernova, consistent with the classification SLSN-IIn. These narrow features arise in relatively slowly moving, hydrogen-rich circumstellar material—the gas surrounding the exploding system. Scattering by free electrons in that gas can broaden the features into extended wings.

The remaining object, SN 2022lvm, shows only securely identified broad supernova hydrogen emission in its available spectra. However, a long gap in its observations means an earlier narrow-lined stage cannot be excluded.

The overwhelming narrow-lined majority is not a measurement of the fraction among all hydrogen-rich superluminous supernovae. The parent sample deliberately excluded 13 previously studied broad-lined events, strongly favouring narrow-lined objects.

Similar ingredients, different profiles

After maximum brightness, the hydrogen features become strikingly diverse. Among 45 objects whose spectra were good enough to assess between peak and 200 rest-frame days later, 29 showed no clear excess on the blue side of Hα. Ten showed a mild blue excess, five a more extended excess, and one a prominent, relatively flat-topped blue shoulder.

Blue-side emission can reveal fast material moving toward the observer, while emission from the receding side is obscured or otherwise suppressed. But an asymmetric distribution of surrounding gas can produce similar effects. The profiles therefore offer clues rather than a unique reconstruction of the explosion.

The researchers modelled Hα using Monte Carlo calculations that follow photons through electron scattering. Within those models, faster photon-emitting shells strengthen the blue shoulder, whereas greater scattering optical depth broadens the exponential wings.

Objects with clear blue excesses generally favoured higher fitted shell velocities and tended to be fainter than objects retaining approximately symmetric profiles. One possible explanation is that denser or more massive surroundings slow the ejecta more efficiently and convert more kinetic energy into light.

That interpretation remains tentative. The models describe individual snapshots, not the full evolving interaction. For symmetric profiles dominated by scattering, the fitted shell velocity need not reliably measure the actual shock speed. Geometry, obscuration and uneven observational coverage also complicate the comparison.

How much material was lost?

For 13 objects with suitable observations, the team estimated the mass loss needed to supply the interacting gas. Assuming a surrounding-gas speed of 100 kilometres per second and conversion of half the processed kinetic power into radiation, the reference estimates span approximately 0.028–0.30 solar masses per year.

These are order-of-magnitude estimates, not precise histories. Most use an assumed shock speed of 5,000 kilometres per second, and the calculation assumes a steady, spherical wind. Changing the shock speed or radiation efficiency changes the inferred rates substantially.

Under those assumptions, ordinary steady stellar winds are insufficient. Eruptive mass loss or binary interaction are possible ways to supply dense surroundings, but the study does not identify a single responsible mechanism.

Hidden explosions and an uncertain boundary

Six narrow-lined objects also show characteristics resembling Type Ia supernovae interacting with hydrogen-rich gas. Their iron-dominated spectral blends and other features raise the possibility of thermonuclear explosions hidden within the sample. They do not establish that origin or rule out core collapse.

Excluding those six candidates, the median Hα-to-Hβ flux ratio rises from 2.85 before 100 days after peak to 4.40 afterward. That ratio probes changing conditions, but cannot by itself disentangle collisional excitation, optical-depth effects and extinction.

Across the available lower-luminosity comparisons, the team finds counterparts to the main hydrogen-line shapes seen in the superluminous sample. Some exceptionally bright events also show prolonged high-ionisation emission, though too few are available to establish a systematic difference.

The comparison sample is small and heterogeneous, so the findings do not prove that all these explosions belong to one physical population. They instead weaken the case for treating a brightness threshold as a fundamental dividing line—and strengthen the need for spectra both shortly after discovery and late enough to reveal ejecta hidden by the interaction.