KL Dra as a Benchmark Laboratory for Accretion-Disk Physics: Constraints from TESS and Ground-Based Surveys

TL;DR

This work delivers the longest-term, high-cadence optical study of KL Dra, an AM CVn system, by unifying nearly a decade of TESS and ground-based surveys. The authors implement a comprehensive space- and ground-based analysis, including detailed parametric fits to superoutbursts, rebrightenings, and normal outbursts, and quantify the long-term supercycle evolution with a mean duration of $60.4 \pm 0.1$ d. They identify three regimes in the supercycle O-C evolution, reveal a consistent SO structure with a persistent precursor, and document 3–4 NOs per SC whose amplitudes and durations grow over the cycle, with NO asymmetry evolving from $f_r \sim 0.2$ to $0.4$–$0.6$. The results challenge a pure tidal-thermal instability picture, highlighting the likely influence of time-variable mass transfer and helium-disk opacity effects, and establishing KL Dra as a critical benchmark for testing disk-instability models in accretion disks.

Abstract

We present the longest-term optical analysis of the AM CVn system KL Dra using $\sim11$ years of monitoring from TESS and wide-field ground-based surveys. The continuous TESS coverage allows us to characterise its frequent outbursts with unprecedented detail, providing the first comprehensive study of an AM CVn during outbursts and enabling detailed modelling of these systems. The superoutbursts in KL Dra generally include a precursor, and are followed by a series of rebrightenings after which a sequence of 3-4 large amplitude normal outbursts is observed. We fit parametric profiles to each superoutburst component (precursor, rise to plateau, plateau, decay), to rebrightenings, and to normal outbursts, which let us quantify every high state feature and investigate correlations with the system's long term supercyle evolution. Our continuous coverage reveals an average value for the supercycles, superoutbursts and normal outbursts of $60.4 \pm 0.1$ d, $5.67\pm0.03$ d and $1.17 \pm0.01$ d, respectively. The supercycle duration may be correlated with the rebrightenings duration and superoutburst amplitude, and anticorrelated with the plateau length. Within a supercycle, normal outbursts grow in amplitude and duration, and the first normal outburst is usually highly asymmetric, while subsequent normal outbursts are more symmetric. We detected superhumps in TESS superoutbursts but not in the rebrightenings or normal outbursts. We interpret the results within the disk instability model, considering additional effects, such as changes in the donor mass transfer rate.

Figures (17)

• Figure 1: Comparison of TESS and ground-based light curves of KL Dra. The left column displays the full time span of available data, while the right column zooms into a 400-day window starting from JD 2459573, highlighted in grey in the left panels. The top row presents the cleaned and corrected TESS light curves, where black lines indicate 1-hour binned data. The second, third, and fourth rows show cleaned and nightly-binned light curves from ZTF, ASAS-SN, and ATLAS, respectively, in different filters. The high-cadence, continuous monitoring of TESS is evident in the right column, while the long-term but lower-cadence coverage from ground-based surveys is prominent in the left column.
• Figure 1: Light curves and models for all SCs included in this study, plotted as a function of time from the SO onset. The colour/symbol scheme follows Figure \ref{['fig:alldata']} and the panel layout follows Figure \ref{['fig:SC_selected']}. TESS (2 hour bins) = white circles with black edge; ZTF $g/i$ = blue/red; ASAS-SN $g/V$ = grey/brown; ATLAS $o/c$ = orange/cyan. The red solid curve is the best-fit model; the red dashed line marks the end of the SC. Vertical solid lines indicate NOs and vertical dotted lines indicate mini-NOs. The label of each SC appears in the upper-right corner.
• Figure 2: Example of a NO (left) and a SO (right) from TESS data. The blue dots represent the 20 sec full-cadence data, where the colour intensity indicates point density based on a Kernel Density Estimator, with darker shades for lower densities and lighter shades for higher densities. The white circles with black edges show 1-hour error-weighted binned data. The red solid lines correspond to the best-fit models, while the vertical red dotted lines mark the different evolutionary stages of the NO/SO.
• Figure 2: Continuation of Figure \ref{['fig:allSCs']}.
• Figure 3: Sample of KL Dra SCs with complete TESS coverage, plotted as a function of time from the onset of each SC. Each row corresponds to a different SC. The observed TESS data has been downsampled into 2-hour bins, with the best-fit model overplotted as a solid red line. Red vertical dashed lines mark the start and end times of each SC, while small solid and dotted red lines indicate the positions of NOs and mini-NOs, respectively. SCs are sorted in ascending order of duration from top to bottom but are not sequential in time. All are displayed on the same time scale for consistency.