Probing Stellar Kinematics with the Time-Asymmetric Hanbury Brown and Twiss Effect

TL;DR

The paper introduces a time-asymmetric Hanbury Brown and Twiss (HBT) effect in intensity interferometry as a new probe of internal stellar kinematics, formalizing a spatio-temporal visibility framework that incorporates line-of-sight velocities into the II signal. Using a discrete-particle model, the authors simulate dynamical sources—specifically a Keplerian circumstellar disk and a two-star absorption-line binary—and show that ordered velocity gradients imprint a measurable shift in the temporal HBT peak away from zero delay, with the sign and magnitude tied to inclination and rotation. They discuss observational feasibility, highlighting atmospheric turbulence, photon statistics, and filter design, and argue that space-based, picosecond-resolution II experiments could reveal quantitative kinematic information, including low-velocity regimes challenging for spectroscopy. The work provides a theoretical basis and practical considerations for using temporal II as a timing-based probe of stellar dynamics, with potential laboratory analogues and future instrument concepts to enable such measurements.

Abstract

Intensity interferometry (II) offers a powerful means to observe stellar objects with a high resolution. In this work, we demonstrate that II can also probe internal stellar kinematics by revealing a time-asymmetric Hanbury Brown and Twiss (HBT) effect, causing a measurable shift in the temporal correlation peak away from zero delay. We develop numerical models to simulate this effect for two distinct astrophysical scenarios: an emission-line circumstellar disk and an absorption-line binary system. Our simulations reveal a clear sensitivity of this temporal asymmetry to the system's inclination angle, velocity symmetry, and internal dynamics. This suggests that, with sufficiently high time resolution, II can be used to extract quantitative information about internal kinematics, offering a new observational window on stellar dynamics.

Figures (6)

• Figure 1: The Keplerian Disk modelled after the $\gamma$ Cassiopeia decretion disk is displayed here from three different viewing angles. A histogram of the line-of-sight (LoS) velocity distribution is presented for each of the views below. The distribution of velocities would correspond to a Doppler shift of the emission lines of up to $\pm$0.8nm
• Figure 2: The squared visibility as a function of delay time $\tau$ was plotted according to equation \ref{['eq:visib']}. In each plot one can see three distinct curves, color-coded according to the different inclination angles shown in Fig. \ref{['fig:disks']}. Each subplot corresponds to a certain point in the $(u,v)$-plane, of which only one corner is shown, since for negative $v$ we would see an exact copy of the positive parts and for negative $u$ we would observe a mirrored version along the $u=0$ axis.
• Figure 3: Comparison of the same disk viewed with an inclination of 90$^\circ$, but with two different rotations. We see that the peak shifts to a positive time delay for an increasing $u$ in the $u,v$-plane for a disk with right rotation, while it decreases at the same point if the disk is left rotating.
• Figure 4: A histogram of the binary system's velocity distribution. The velocity distribution corresponds to observing the spectrum within a filter centered around a a wavelength with artificially added decreases representing the absorption lines. The differences between 45$^\circ$ and 90$^\circ$ for this model compared to the ones of the disk seen in Fig. \ref{['fig:disks']} seems negligible.
• Figure 5: Similarly as in \ref{['fig:visibs']} the squared visibility of the binary functions as a function of delay time $\tau$ was plotted according to equation \ref{['eq:visib']}. Subplots correspond again to different points in the $u,v$-plane. The peak shape does not change as drastically with increasing $u$ and $v$ as it did for the previous model, but a clear asymmetry remains. The curves for 45$^\circ$ and 90$^\circ$ overlap almost completely. Points with non-zero $v$ are not shown, as they primarily exhibit a decrease in amplitude without significantly altering the temporal shape of the function.