Forward modelling of quasar light curves and the cosmological matter power spectrum on milliparsec scales

TL;DR

This paper introduces a novel method to analyze strongly lensed quasar light curves to measure the small-scale matter power spectrum, providing the strongest constraints to date on primordial fluctuations at milliparsec scales.

Contribution

The authors develop an optimal MCMC-based technique to extract the temporal power spectrum from quasar light curves, enabling new constraints on cosmological matter fluctuations at unprecedented small scales.

Findings

Constrained the non-linear matter power spectrum at milliparsec scales to Δ²<4×10⁷.

Set an upper limit on the primordial scalar power spectrum at k≈3 pc⁻¹ to P_R<3×10⁻⁹.

Forecasted that future surveys like LSST will vastly improve these measurements.

Abstract

We devise an optimal method to measure the temporal power spectrum of the lensing and intrinsic fluctuations of multiply-imaged strongly lensed quasar light curves, along with the associated time delays. The method is based on a Monte-Carlo Markov Chain (MCMC) sampling of a putative gaussian likelihood, and accurately recovers the input properties of simulated light curves, as well as the "Time Delay Challenge". We apply this method to constrain the dimensionless cosmological (non-linear) matter power spectrum on milliparsec scales (comparable to the size of the solar system), to $\Delta_{\rm NL}^2< 4 \times 10^7$ at $k_{\rm NL} \sim 10^3 {\rm pc}^{-1}$. Using a semi-analytic nonlinear clustering model which is calibrated to simulations, the corresponding constraint on the primordial (linear) scalar power spectrum is ${\cal P}_{\cal R} < 3 \times 10^{-9}$ at $k_{\rm L}\sim$ 3 pc$^{-1}$. This is the strongest constraint on primordial power spectrum at these scales, and is within an order of magnitude from the standard $\Lambda$CDM prediction. We also report measurements of temporal spectra for intrinsic variabilities of quasar light curves, which can be used to constrain the size of the emitting region in accretion disks. Future cadenced optical imaging surveys, such as LSST, should increase the number of observed strongly lensed quasars by 3 orders of magnitude and significantly improve these measurements, even though improvements in modelling quasar accretion and stellar microlensing are necessary.