Microlensing of lensed supernovae Zwicky & iPTF16geu: constraints on the lens galaxy mass slope and dark compact object fraction

TL;DR

This study uses two galaxy-scale lensed Type Ia supernovae to jointly constrain the lens galaxy mass slope $η$ and the fraction of dark compact objects $f_{ m dc}$ by combining microlensing magnification maps with standard-candle magnifications. Employing a PEMD macro lens model, Sérsic light profiles, microlensing simulations, and a Bayesian framework, the authors extract tight constraints on $η$ and place upper limits on $f_{ m dc}$, with a stronger bound when combining both SNe assuming a common $f_{ m dc}$. They find no evidence for a dark compact-object population and measure $η$ values around 1.70–1.81, consistent with previous lens samples but on the lower end for these specific systems. The results demonstrate the potential of strongly lensed SNIa to probe the inner mass distribution of lens galaxies and the abundance of compact dark matter, with future surveys likely to yield much tighter constraints. The approach carefully accounts for microlensing, stellar mass, and light, and highlights the value of standard candles in breaking degeneracies between macro-lens structure and substructure.

Abstract

To date, only two strongly lensed type Ia supernovae (SNIa) have been discovered with an isolated galaxy acting as the lens: iPTF16geu and SN Zwicky. The observed image fluxes for both lens systems were inconsistent with predictions from a smooth macro lens model. A potential explanation for the anomalous flux ratios is microlensing: additional (de)magnification caused by stars and other compact objects in the lens galaxy. In this work, we combine observations of iPTF16geu and SN Zwicky with simulated microlensing magnification maps, leveraging their standardizable candle properties to constrain the lens galaxy mass slope, $η$, and the fraction of dark compact objects, $f_{\rm dc}$. The resulting mass slopes are $η= 1.70 \pm 0.07$ for iPTF16geu and $η= 1.81 \pm 0.10$ for SN Zwicky. Our results indicate no evidence for a population of dark compact objects, placing upper limits at the $95\%$ confidence level of $f_{\rm dc} < 0.25$ for iPTF16geu and $f_{\rm dc} < 0.47$ for SN Zwicky (for compact objects with masses above $ 0.02 M_{\odot}$). Assuming a constant fraction of dark compact objects for both lensed SNe, we obtain $f_{\rm dc} < 0.19$. These results highlight the potential of strongly lensed SNIa to probe the innermost parts of lens galaxies and learn about compact matter.

Figures (12)

• Figure 1: JWST image of the field of SN Zwicky post-explosion, showing its lens and host galaxies. The upper right inset shows SN Zwicky when it was active in earlier HST observations. The lower right inset provides a zoomed-in view of the JWST field, showing the lens galaxy and the host galaxy, which is not strongly lensed. The SN image positions are derived from the HST observations and projected onto the observations as white circles. The dashed circle with a $1"$ radius indicates the aperture within which the stellar mass measurement of the lens galaxy has been conducted.
• Figure 2: Flowchart of the observations and inferred quantities of this work. The left shaded panels show the specific observations and filters that go into modelling the derived quantities in the middle column, which in turn serve as input for the microlensing simulations and our Bayesian model. The final outputs, shown on the right, are the lens galaxy mass slope, $\eta$, fraction of dark compact objects, $f_{\rm dc}$, and the stellar mass fraction at the SN image positions, $f_{*,i}$.
• Figure 3: The macrolens model fit and residuals to Keck $J$-band observations of SN Zwicky, using a PEMD model with a mass slope of $\eta = 1.8$. From left to right, the panels show: 1.) Keck $J$-band observations of SN Zwicky. 2.) The best-fit macro lens model with the SN image fluxes scaled to match the observed fluxes. Only the SN image positions are used to fit the macro lens model, because the fluxes are affected by microlensing. 3.) Normalised residuals between panels 2 and 1, demonstrating an excellent fit when SN fluxes are scaled. 4.) The predicted fluxes from the best-fit macro model, which deviate from the observed fluxes. This discrepancy highlights the effects of microlensing, which is not captured by the macro model.
• Figure 4: Microlensing magnification maps and distributions for SN Zwicky, assuming a lens mass slope $\eta = 1.8$ and compact object fraction $f_c = 0.2$. The microlensing magnification is given in units of magnitudes, $\Delta m = -2.5\log_{10}(\mu_{\rm obs}/\mu_{\rm th})$. Left: Microlensing magnification maps for images A, B, C, and D. The microlensing caustics show the areas of magnification and demagnification in the source plane due to compact objects in the lens galaxy. $\theta_*$ refers to the Einstein radius of an individual microlens. Right: The magnification distributions for images C and D, obtained by convolving the SN size with the microlensing maps. The microlensing magnification distributions for images A and C are similar and show a wide spread in magnifications, since they are both saddle points in the time-delay surface. Images B and D show similarly narrow distributions, because they represent minima in the time-delay surface.
• Figure 5: Stellar mass estimates for iPTF16geu (top) and SN Zwicky (bottom) obtained with prospector. The left panels show the stellar mass posteriors, and the right panels the best-fit SEDs and photometric measurements used for the analysis. For iPTF16geu, the observations consist of six filters from HST, and for SN Zwicky, three filters from HST and four from JWST.