Presaging Doppler beaming discoveries of double white dwarfs during the Rubin LSST era

TL;DR

This work tackles the problem of discovering short-period double white dwarfs (DWDs) via relativistic Doppler beaming in the Rubin LSST era. It develops a forward-modeling framework by coupling SeBa-based binary population synthesis with a three-component Milky Way model and realistic LSST cadences to predict beaming-detected DWDs, including LISA verification binaries. The results indicate LSST can identify at least 287 DWDs through Doppler beaming, with 47 of them detectable by LISA, and reveal biases toward unequal-mass, short-period systems that offer constraints on mass-transfer physics. The approach provides a quantitative forecast for the LSST era, enabling targeted follow-up and enabling robust tests of stellar binary evolution models.

Abstract

Double white dwarfs (DWDs) are by far the most common compact binaries in the Milky Way, are important low-frequency gravitational-wave sources, and in some cases merge to become Type Ia supernovae. So far, no DWD has been identified solely through relativistic Doppler beaming, even though the beaming amplitude directly relates to the radial velocity semi-amplitude. In this work, we initiate a comprehensive binary population synthesis using SeBa and incorporate the resulting binaries into a tripartite Galaxy model. Our proof-of-concept simulations demonstrate that the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) can reliably recover relatively bright ($r \lesssim20~$mag) unequal-mass binaries in compact orbits with P $\approx$ 10-600 minutes with moderate to high inclinations. We find that LSST can detect at least 287 short-period DWDs, of which 47 are LISA-detectable gravitational wave sources. LSST lightcurves allow us to readily determine the period and fully characterize the orbit, in contrast with the challenges of orbit determination for DWDs in spectroscopic searches. The formation of unequal mass, short-period DWDs strongly depends on the assumptions regarding the mass-transfer phases during binary population synthesis, and the total number and characteristics of Doppler-beamed DWD systems observed in LSST will provide new tests of models of stellar binary evolution. Here, we lay the foundation for the comprehensive integration of synthetic Galactic binary population into realistic LSST survey simulations, thereby enabling quantitative forecasts of the number and characteristics of any binary sub-population during the LSST era.

Figures (7)

• Figure 1: Left: Our adopted model for the SFR of the Milky Way (black line) composed of the thin disk (blue dashed line), thick disk (green dashed line), and the bulge (orange dashed line). Right: Cumulative stellar mass of these components as a function of time.
• Figure 2: Doppler beaming amplitude is shown for a binary with $\mathrm{(M_1,T_{eff,1})}$$\mathrm{=(0.3~M_{\odot},10000~K)}$, $\mathrm{(M_2,T_{eff,2})}$$\mathrm{= (0.6~M_{\odot},7500~K)}$, inclination of 60 degrees, and source magnitude of $20~\mathrm{mags}$ in $r$ band. The dotted line corresponds to a single exposure error, and the dashed line is the estimated error after stacking 70 exposures or approximately 10% of the total LSST exposures for a typical target.
• Figure 3: Examples of period recovery for different values of false alarm probability. The systems in the top row would be classified as binary candidates, while those in the bottom row would be discarded based on our chosen cutoff. We have combined lightcurves from all available bands to reduce computational cost of periodogram search, even if the beaming amplitude varies for different bands, which can be seen in the inset plots as tracks with differing amplitudes.
• Figure 4: The distribution of parameters of recovered DWDs is shown. We show the 2D histograms in the background and the actual data points as a scatter plot on top.
• Figure 5: The distribution of periods (left), total mass (middle), and the mass ratio (right) of the recovered DWDs (blue) as compared to the rest of the DWDs with $r<20~$mag and period within 10 hours (grey). The figure illustrates the biases in the DWD sample selected through the Doppler beaming effect. We find that unequal mass binaries with period between 2--5 hours and total mass of 1 M$_{\odot}$ are preferentially recovered.