Radio Monitoring of Classical Novae using the ASKAP Variable and Slow Transients Survey

Abstract: We present a search for radio emission from classical novae at 887.5 MHz using data from the Australian SKA Pathfinder Variable And Slow Transient (VAST) survey. We cross-matched 43 optically discovered classical novae that erupted between 2021 September and 2025 November within the 1200-square-degree Galactic survey footprint, and found three which show significant radio emission: V6598 Sgr, V1716 Sco, and V1723 Sco. To analyse their radio light curves, we use both thermal free-free and non-thermal synchrotron emission models. We fit the data using the Markov chain Monte Carlo (MCMC) method to constrain parameters, including the ejected mass and ejecta velocities for the thermal models, and mass-loss rate, explosion energy, wind velocity, and filling factor for the non-thermal model. All three novae show evidence of non-thermal synchrotron emission as the dominant emission mechanism at this frequency. We use a broken power law to describe the radial density structure of a non-uniform circumbinary material, which provides a better fit than a standard wind density profile. This strong early-time synchrotron emission is strong evidence of shock-driven particle acceleration, which may be related to detections of gamma-rays from all three novae as well. In contrast to earlier studies that used multi-frequency data to distinguish between emission models, our analysis is based on single-frequency radio light curves, which can still provide useful constraints on the dominant emission mechanism when interpreted with physically motivated models

• The paper uses 3.5 years of ASKAP VAST observations at 887.5 MHz to identify radio emission from three of 32 monitored classical novae and models their single-frequency light curves with thermal and synchrotron scenarios.
• The detected novae show steep rises and declines, including decay slopes of approximately t⁻³·³ to t⁻³·⁶, favoring shock-driven synchrotron emission over thermal free-free radiation, although no model provides a formally acceptable fit.
• All three radio detections occurred among six gamma-ray novae, compared with none among 26 others in the survey footprint, yielding a binomial p-value of about 0.004 and supporting a connection between gamma-ray shocks and later radio emission.

Survey-based nova search at 887.5 MHz

This paper presents a systematic search for radio emission from classical novae (CNe) using the first 3.5 years of the Australian SKA Pathfinder Variables and Slow Transients (VAST) survey, which monitors 1200 deg² of the southern Galactic plane every two weeks at a central frequency of 887.5 MHz with a typical rms noise of 0.2 mJy. Cross-matching the catalogue of spectroscopically confirmed Galactic novae against the VAST footprint yielded 43 novae that erupted between 2021 September and 2025 November, of which 32 fell within the survey area. Three sources — V6598 Sgr, V1716 Sco, and V1723 Sco — show well-sampled flux density evolution; the remaining 29 are consistent with non-detections.

The key methodological departure from prior work is that the analysis relies entirely on single-frequency light curves rather than multi-frequency spectral information or brightness-temperature arguments. The authors demonstrate that light-curve shape alone, interpreted through physically motivated models, can still discriminate between emission mechanisms — a result relevant for future surveys such as the SKA, where dedicated multi-frequency follow-up of large samples will not be feasible.

Sample and source properties

All three detected novae are fast, gamma-ray-detected systems hosting main-sequence companions:

Source	Discovery	Distance adopted	Radio peak	Peak flux density
V6598 Sgr	2023 Jul 15	7.6 kpc	day ~126	5.3 ± 0.26 mJy
V1716 Sco	2023 Apr 20	3.0 kpc	day 107	9.3 ± 0.2 mJy
V1723 Sco	2024 Feb 8	8.0 kpc	day 137	9.3 ± 0.3 mJy

V6598 Sgr is an intermediate polar resembling U Sco and V2672 Oph in early evolution. V1716 Sco hosts a 1.21 $M_\odot$ white dwarf and showed a 78 s X-ray QPO during its supersoft phase. V1723 Sco is among the most gamma-ray-luminous CNe since 2010, with a Fermi-LAT photon flux of $(1.9\pm0.5)\times10^{-6}$ photons cm⁻² s⁻¹, roughly an order of magnitude above the typical LAT nova population. Distances were taken from Schaefer's Bayesian analysis incorporating Gaia parallaxes with nova-appropriate priors; the authors note that geometric and photogeometric Bailer-Jones distances for these sources are statistically insignificant and carry priors not necessarily appropriate for CNe, so distance remains a dominant systematic uncertainty.

Emission modelling

Three models were fitted to each light curve via MCMC (emcee), with model comparison by AIC, BIC, and reduced $\chi^2$:

Thermal free-free: the standard Hjellming shell model with Hubble-type expansion, fitting ejecta mass $M_{\mathrm{ej}}$, outer velocity $v_2$, and velocity ratio $v_1/v_2$, with $T_e$ fixed at $10^4$ K given the weak single-frequency dependence.

Standard synchrotron: the Chevalier/Nyamai magnetobremsstrahlung framework with an effective $\rho \propto R^{-2}$ circumbinary profile, intrinsic synchrotron self-absorption plus external free-free absorption, fitting filling factor $f$, wind velocity $(1.9\pm0.5)\times10^{-6}$0, mass-loss rate $(1.9\pm0.5)\times10^{-6}$1, explosion energy $(1.9\pm0.5)\times10^{-6}$2, and $(1.9\pm0.5)\times10^{-6}$3, with fixed $(1.9\pm0.5)\times10^{-6}$4 and $(1.9\pm0.5)\times10^{-6}$5.

Modified synchrotron: the same framework but replacing the wind density profile with a broken power law in radius, controlled by three fixed indices ($(1.9\pm0.5)\times10^{-6}$6: inner slope, $(1.9\pm0.5)\times10^{-6}$7: outer slope, $(1.9\pm0.5)\times10^{-6}$8: transition smoothness) chosen per source according to the observed temporal steepness.

Because fitted parameters are strongly degenerate with distance, distances were held fixed; instead, the paper tabulates power-law scaling indices so parameters can be rescaled if the adopted distance changes (e.g., for the synchrotron model, $(1.9\pm0.5)\times10^{-6}$9, $\chi^2$0, $\chi^2$1).

Results: synchrotron dominance and steep declines

For all three novae, the thermal free-free model under-predicts the observed flux and rises too slowly, while the standard synchrotron model over-predicts the peak and fails to reproduce the steep post-peak decline. The modified synchrotron model yields the lowest reduced $\chi^2$2 in all cases, though the fits remain formally poor:

Source	Thermal $\chi^2$3	Synchrotron $\chi^2$4	Modified $\chi^2$5	Verdict
V6598 Sgr	65.7	8.9	2.9	modified preferred
V1716 Sco	50.4	73.4	3.7	modified preferred
V1723 Sco	74.7	59.0	37.8	no acceptable fit

The most striking quantitative results concern the temporal slopes. V1716 Sco rises as $\chi^2$6 and decays as $\chi^2$7; V1723 Sco decays as $\chi^2$8. Both are far steeper than the $\chi^2$9 rise expected for optically thick freely expanding thermal ejecta, providing direct evidence that shock-driven synchrotron emission dominates at 887.5 MHz. The authors attribute the abrupt decay to the shock propagating into a very tenuous ambient medium — consistent with the main-sequence companions and hence low-density circumbinary environments — leaving insufficient electrons to accelerate. All three sources reach spectral luminosities $M_{\mathrm{ej}}$0, comparable to known synchrotron-dominated novae such as V392 Per and V838 Her, which is difficult to reconcile with purely thermal emission.

The posterior distributions exhibit strong degeneracies: $M_{\mathrm{ej}}$1 and $M_{\mathrm{ej}}$2 are positively correlated (preserving the $M_{\mathrm{ej}}$3 density), while $M_{\mathrm{ej}}$4 and $M_{\mathrm{ej}}$5 are anti-correlated. Consequently, the derived physical parameters should be treated as indicative rather than precise constraints.

Association with gamma-ray detections

A notable statistical result emerges from comparing the ASKAP sample against Fermi-LAT detections. Of the 32 novae in the survey footprint, six have reported gamma-ray emission — and all three ASKAP radio detections fall within this subset. The radio detection fraction is 50% (3/6) among gamma-ray novae versus 0% (0/26) among the rest, with a binomial $M_{\mathrm{ej}}$6. This supports the physical picture in which gamma-ray emission traces early internal shocks whose non-thermal electron population later becomes visible at low radio frequencies once optical depth drops, roughly 100–150 days post-eruption. The authors caution, however, that LAT detection does not guarantee radio detection — only three of the six gamma-ray novae were found — and sensitivity, cadence, distance, and environmental differences all modulate detectability.

The distance distribution of the sample also carries a selection effect: nearby novae (<1–2 kpc) are largely undetected, plausibly because at 887.5 MHz their ejecta remain optically thick due to free-free or synchrotron self-absorption on the timescales probed. The detected objects are therefore biased toward intrinsically luminous, rapidly becoming optically thin, shock-powered systems.

Limitations and open questions

Several limitations constrain the interpretation. First, none of the models achieves a formally acceptable fit even for the two "preferred" cases, and the broken power-law indices were fixed by hand rather than fitted, making the density-profile inference partly assumption-driven. Second, the spherical symmetry assumed throughout is inconsistent with the known aspherical morphologies of nova ejecta (e.g., V959 Mon, RS Oph), and complex light curves with multiple peaks documented in other systems cannot be captured by any single-component model here. Third, the parameter posteriors are broad and correlated, so quantities such as $M_{\mathrm{ej}}$7, $M_{\mathrm{ej}}$8, and $M_{\mathrm{ej}}$9 inherit large uncertainties compounded by uncertain distances. Finally, whether the apparent optical irregularities near the radio peak of V1723 Sco reflect genuine correlated variability is left unresolved. The open questions are specific: what ejecta geometry and density structure produce decay slopes steeper than any smooth wind profile can supply, and whether multi-frequency coverage would break the $v_2$0–$v_2$1 and $v_2$2–$v_2$3 degeneracies sufficiently to yield robust shock energetics.

Conclusion

Using homogeneous, high-cadence ASKAP VAST monitoring at 887.5 MHz, this work identifies three gamma-ray-associated classical novae whose low-frequency radio light curves are dominated by early synchrotron flares with unusually steep rises and decays ($v_2$4 to $v_2$5). A broken power-law circumbinary density profile improves upon both thermal and standard wind-profile synchrotron models, though no model fully reproduces the data. The statistically significant overlap between radio and gamma-ray detections ($v_2$6) links low-frequency radio surveys directly to shock-powered particle acceleration in novae, and establishes single-frequency light-curve modelling as a viable diagnostic tool for flux-limited transient surveys where targeted multi-frequency follow-up is impractical.