Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Strongly lensed repeating fast radio bursts as precision probes of the universe

Abstract

Fast radio bursts (FRBs), bright transients with millisecond durations at GHz and typical redshifts probably >0.8, are likely to be gravitationally lensed by intervening galaxies. Since the time delay between images of strongly lensed FRB can be measured to extremely high precision because of the large ratio 109 between the typical galaxy-lensing delay time $$\sim{\cal O}$$ (10 days) and the width of bursts $$\sim{\cal O}$$ (ms), we propose strongly lensed FRBs as precision probes of the universe. We show that, within the flat ΛCDM model, the Hubble constant H0 can be constrained with a ~0.91% uncertainty from 10 such systems probably observed with the square kilometer array (SKA) in <30 years. More importantly, the cosmic curvature can be model independently constrained to a precision of 0.076. This constraint can directly test the validity of the cosmological principle and break the intractable degeneracy between the cosmic curvature and dark energy.

Introduction

Fast radio bursts (FRBs) are bright transients with millisecond durations at ~GHz frequencies, whose physical origin is subject to intense debate1,2. Most FRBs are located at high galactic latitudes and have anomalously large dispersion measures (DMs). Attributing DM to an intergalactic medium origin, the corresponding redshifts z are typically >0.8. Up to now, more than 30 FRBs have been published3. One of them, FRB 121102, shows a repeating feature4. The repetition of FRB 121102 enables high-time-resolution radio interferometric observations to directly image the bursts, leading to the localization of the source in a star-forming galaxy at z = 0.19273 with sub-arcsecond accuracy5. The cosmological origin of the repeating FRBs is thus confirmed. For other FRBs, although no well-established evidence has been published, they are also strongly suggested to be of a cosmological origin, due to their all-sky distribution and their anomalously large values of DMs2. It is possible that all FRBs might be repeating, and only the brightest ones are observable. On the other hand, it is also possible that repeating and non-repeating FRBs may originate from different progenitors6. Interestingly, these transient radio sources are likely lensed from small to large scales, for example, through plasma lensing in their host galaxies7, gravitational microlensing by an isolated and extragalactic stellar-mass compact object8,9, and strong gravitational lensing by an intervening galaxy10,11. Here we only focus on the possibility of strongly gravitationally lensed FRBs and their applications to conduct cosmography. Therefore, in our following analysis, “lensed FRBs” only refers to the case that an FRB is strongly gravitationally lensed by an intervening galaxy. For a lens galaxy with the mass of dark matter halo $$\sim10^{12}M_ \odot h^{ - 1}$$ (h is the Hubble constant in units of 100 km s−1 Mpc−1), the typical time delay and angular separation between different images of lensed FRBs are $$\sim{\cal O}$$ (10 days) and of the order arcseconds, respectively. These multiple images of lensed FRBs cannot be resolved by radio survey telescopes since their typical angular resolution is of the order of 10 arcmin. Therefore, a lensed non-repeating FRB source may be observed as a repeating FRB source, showing two to four bursts with respective time delays of several days. The DM value and the scatter broadening of each burst could be slightly or even significantly different from each other depending on the plasma properties along different lines of sight (LOS). It is therefore difficult to identify the lensed non-repeating FRBs. However, if a repeating FRB is strongly lensed by an intervening galaxy, a series of image multiplets from the same source will exhibit a fixed pattern in their mutual time delays, appearing over and over again as we detect the repeating bursts11. Observations of FRB 121102 in radio and its counterpart in optical indicate that this repeater is not lensed (no intervening lens galaxy or multiple images of the host are observed) and the intrinsic repetition happens randomly. Therefore, a fixed temporal pattern associated with a future repeating FRB source would be a smoking-gun signature that it is strongly lensed. Each burst emitted from the source would travel through different paths to reach to the observer with time delays. If these lensed bursts can be imaged, they should appear as different images in the sky. Their spectra and lightcurves might be slightly different from each other because of different paths they traveled through, so that the morphology of bursts may not be the main feature to identify lensed FRBs. For a series of randomly generated repeating bursts, the intrinsic time difference between two adjacent bursts should be the same for all lensed (two or four) images. Therefore, a fixed time pattern of all the repeating bursts is the most robust evidence for identifying a lensed FRB system. Once a survey telescope registered a fixed time pattern repeating two or four times with a delay $$\sim{\cal O}$$ (10 days), one could then employ more powerful radio telescopes such as very large array (VLA) or the future SKA to observe more repetitions and resolve multiple images of the bursts. Meanwhile, one could observe the source using optical and near-infrared (IR) telescopes to identify an intervening lens galaxy near the LOS as well as the multiple images of the host galaxy (Einstein ring or arcs) with angular separations of the order of arcseconds. If the image locations in both radio and optical (or near-IR) bands match each other, in combination with the fixed time-delay repetition pattern mentioned above, a lensed FRB system can be confirmed and the host and the lens galaxies identified.

Current FRB observations suggest a sufficiently high all-sky FRB rate of ~103–104 per day2,12. Upcoming surveys such as the Swinburne University of Technology’s digital backend for the Molonglo Observatory Synthesis Telescope array (UTMOST)13, the Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX)14, the Canadian Hydrogen Intensity Mapping Experiment (CHIME)15, and especially the SKA project16 will map a considerable fraction of the sky with a detection rate of FRBs of >100 per day17. For an FRB happening at z > 1, the probability for it to be strongly lensed is ~a few ×10−4 18. As a result, future radio surveys, such as SKA, will have the ability to discover >10 strongly lensed FRBs per year9,10,11. According to the current data, at least 3% (1/30) observed FRBs are repeating FRBs. With a conservative estimate, ~10 strongly lensed repeating FRBs are expected to be accumulated within <30 years with the operation of SKA.

Owing to the small ratio (~10−9) between the short duration of each burst $$\sim{\cal O}$$ (ms) and the typical galaxy-lensing delay time $$\sim{\cal O}$$ (10 days), time delays between images of these systems can be measured to great precision. Moreover, due to overwhelmingly accurate localizations of lensed FRB images from deep VLA observations (or future SKA observations) and clean high-resolution images of the host galaxy without a dazzling active galactic nucleus (AGN), the mass profile of the lens can be also modeled with high precision. Therefore, we propose that lensed FRB systems can be a powerful probe for studying cosmology. Lensing theory predicts that the difference in arrival time between image A and image B, that is, the “time delay” ΔτAB, is expressed as

$${\mathrm{\Delta }}\tau _{{\mathrm{AB}}} = \frac{{{\mathrm{\Delta \Phi }}_{{\mathrm{AB}}}}}{c}D_{{\mathrm{\Delta }}t} = \frac{{{\mathrm{\Delta \Phi }}_{{\mathrm{AB}}}}}{c} \cdot (1 + z_{\mathrm{l}})\frac{{D_{\mathrm{l}}^{\mathrm{A}}D_{\mathrm{s}}^{\mathrm{A}}}}{{D_{\mathrm{ls}}^{\mathrm{A}}}},$$
(1)

where ΔΦAB is the Fermat potential difference between the two image positions, c is the speed of light, DΔt the so-called “time-delay distance” is just a multiplicative combination of the three angular diameter distances ($$D_{\mathrm{l}}^{\mathrm{A}}$$: from the observer to the lens, $$D_{\mathrm{s}}^{\mathrm{A}}$$: from the observer to the source, $$D_{\mathrm{ls}}^{\mathrm{A}}$$: from the lens to the source), and zl is the redshift of lens. This quantity has the dimension of distance and is inversely proportional to the Hubble constant, H0, which sets the age and length scale for the present universe and is one of the most important parameters for cosmology. Therefore, the time-delay distance DΔt is primarily sensitive to H0 and that measured from lensed quasar systems has been used to measure the Hubble constant. Moreover, the relations among these three angular diameter distances are highly dependent on the geometric properties of the space. We introduce dimensionless comoving angular diameter distances, $$d_{\mathrm{l}} \equiv d(0,z_{\mathrm{l}}) \equiv (1 + z_{\mathrm{l}})H_0D_{\mathrm{l}}^{\mathrm{A}}/c$$, $$d_{\mathrm{s}} \equiv d(0,z_{\mathrm{s}}) \equiv (1 + z_{\mathrm{s}})H_0D_{\mathrm{s}}^{\mathrm{A}}/c$$, and $$d_{\mathrm{ls}} \equiv d(z_{\mathrm{l}},z_{\mathrm{s}}) \equiv (1 + z_{\mathrm{s}})H_0D_{\mathrm{ls}}^{\mathrm{A}}/c$$ (zs is the redshift of source), to illustrate this. For example, qualitatively and intuitively, ds is greater, equal to, or smaller than dl + dls if the space is open, flat, or closed, respectively (see Methods section for quantitative details). Therefore, in combination with distances from type Ia supernova (SNe Ia) observations, the time-delay distance can be used to directly measure the spatial curvature Ωk in a cosmological model-independent manner. Decades of observations have ushered in the era of precision cosmology. The flat Λ CDM model is found to be consistent with essentially all the conservational constraints. Yet, recent direct local distance ladder measurements of H0 have reached a 2.4% precise measurement: H0 = 73.24 ± 1.74 km s−1 Mpc−1 19, which greatly increased the tension with respect to the latest Planck-inferred value (H0 = 67.27 ± 0.66 km s−1 Mpc−1)20 to 3.4σ. Lensed FRBs, as a powerful probe and completely independent dataset based on a different physical phenomenon, would provide complementary information and therefore are of vital importance to clarify this issue.

Results

In order to investigate the constraining power of lensed FRBs on some fundamental cosmological parameters, we perform a series of simulations with the proper inputs in the following three aspects (see Methods section for details): (i) the redshift distribution of incoming FRBs; (ii) for a source at redshift zs, the lens redshift zl to produce the maximal differential lensing probability; (iii) the uncertainty of each factor contributing to the accuracy of time-delay distance measurement. Since the time-delay distance is very sensitive to the Hubble constant, we estimate the constraining power on H0 by simulating 10 lensed FRBs in the flat ΛCDM with the matter density being fixed as Ωm = 0.3. With the assumed fiducial model (flat ΛCDM model with H0 = 70 km s−1 Mpc−1 and Ωm = 0.3) and three factors outlined above, we performed 10,000 simulations, each containing 10 lensed FRB systems and obtained the probability distribution of the estimated H0. Two different redshift distributions, Nconst(z) and NSFH(z), are considered (see Methods section for details). They do not lead to significant differences in the constraint on H0 and consistently give stringent constraints with a ~0.91% uncertainty. Results are shown in Fig. 1. It is suggested that compared to the currently available results, ~10 lensed FRBs will have obvious predominance in precision in constraining H0. For instance, it improves by a factor ~5 with respect to the current state-of-the-art case of lensed quasars21.

In addition to constraining the Hubble constant within the flat ΛCDM model, one can also give a model-independent estimate of the cosmic curvature using lensed FRBs. As one of the most important parameters in cosmology, a precise estimation of the spatial curvature is essential for justifying whether the Friedmann–Lemaître–Robertson–Walker (FLRW) metric could exactly characterize the background of the universe22, and for studying some fundamental issues such as cosmic evolution and dark energy property23. In the FLRW metric, it was found that distances along null geodesics satisfy a specific sum rule24, which has been proposed to test the validity of the homogeneous and isotropic background25. More recently, with an upgraded distance sum rule, time-delay distance measurements from lensed quasars were proposed to test the FLRW metric and estimate the cosmic curvature (see Methods section for details)26. Here in combination with ~4000 SNe Ia observations from the near-future dark energy survey, we examined the ability of the lensed FRBs for constraining the cosmic curvature. We found that the constraints from lensed FRBs with the two considered redshift distributions are very similar and the spatial curvature parameter can be constrained to a precision of ~0.076. Results from lensed FRBs and other currently available model-independent methods are presented in Fig. 2. Again, in this model-independent domain, lensed FRBs are the most promising tools for constraining the cosmic curvature. Moreover, the precision of the results from lensed FRBs potentially approaches that inferred from the Planck satellite observations within the standard ΛCDM model, where Ωk = −0.004 ± 0.015 was obtained20.

Discussion

Here we propose strongly lensed repeating FRBs as a precision cosmological probe. Representatively, we investigate the constraining power of lensed FRB systems observed in the near future on two of the most important cosmological parameters, Hubble constant H0 and cosmic curvature Ωk. For H0, we obtained that it can be constrained with a relative ~0.91% uncertainty from 10 lensed FRB systems. This promising constraint with sub-percent uncertainty level suggests that lensed FRBs, as a powerful probe and completely independent dataset based on a different physical phenomenon, would provide complementary information and therefore are of vital importance to clarify the tension between the latest Planck-inferred H0 and the one from direct local distance ladder observations. For Ωk, it can be constrained to a precision of ~0.076 from 10 lensed FRB systems in a model-independent way on the basis of the distance sum rule. This result is the most precise one in the field of model-independent estimations for the cosmic curvature and, on the one hand, will provide a stringent direct test for the validity of the FLRW metric. On the other hand, this constraint with considerable precision is helpful to break the degeneracy between the cosmic curvature and dark energy, and thus is conducive for investigating the nature of dark sectors of the universe. Moreover, having model-dependent and direct measurements of the same quantity is of utmost importance. In the absence of significant systematic errors, if the standard cosmological model is the correct one, indirect (model-dependent) and direct (model-independent) constraints on this parameter should be consistent. If they were significantly inconsistent, this would provide evidence of physics beyond the standard model or unaccounted systematic errors. Strongly lensed FRBs can help to reach such a goal.

Methods

In order to examine the potential of using lensed FRBs as cosmological probes, three related aspects need to be addressed: (i) the redshift distribution of the incoming FRBs; (ii) for a source at redshift zs, the lens redshift zl to produce the maximal differential lensing probability; (iii) the uncertainties of different factors contributing to the accuracy of time-delay distance measurements. We discuss these three items one by one. In addition, we also introduce the distance sum rule for estimating the cosmic curvature.

FRB redshift distribution

We consider two possible scenarios suggested in ref 9. The first one invokes a constant comoving number density, so that the number of FRBs in a shell of width dz at redshift z is proportional to the comoving volume of the shell dV(z)27. By introducing a Gaussian cutoff at some redshift zcut to represent an instrumental signal-to-noise threshold, the constant-density distribution function Nconst(z) is expressed as

$$N_{{\mathrm{const}}}(z) = {\cal N}_{{\mathrm{const}}}\frac{{\chi ^2(z)}}{{H(z)(1 + z)}}e^{ - D^{{\mathrm{L}}^{\mathrm{2}}}(z)/\left[ {2D^{{\mathrm{L}}^{\mathrm{2}}}(z_{{\mathrm{cut}}})} \right]},$$
(2)

where χ(z) is the comoving distance and DL is the luminosity distance. $${\cal N}_{{\mathrm{const}}}$$ is a normalization factor to ensure that the integration of Nconst(z) is unity and H(z) is the Hubble parameter at redshift z. The second distribution requires that FRBs follow the star-formation history (SFH)13, so that

$$N_{{\mathrm{SFH}}}(z) = {\cal N}_{{\mathrm{SFH}}}\frac{{\dot \rho _ \ast (z)\chi ^2(z)}}{{H(z)(1 + z)}}e^{ - D^{{\mathrm{L}}^{\mathrm{2}}}(z)/\left[ {2D^{{\mathrm{L}}^{\mathrm{2}}}(z_{{\mathrm{cut}}})} \right]},$$
(3)

where $${\cal N}_{{\mathrm{SFH}}}$$ is the normalization factor and is chosen to have NSFH(z) integrated to unity. The density of SFH is parametrized as

$$\dot \rho _ \ast (z) = h\frac{{\alpha + \beta z}}{{1 + (z/\gamma )^\delta }},$$
(4)

with α = 0.017, β = 0.13, γ = 3.3, δ = 5.3, and h = 0.728,29. For redshifts of currently available FRBs, different from previous estimation using a simple relataion between DM and z proposed by Ioka30, we re-estimate them with a more precise DM–z relation given in ref 31. It is found that the inferred z values are systematically greater than previously estimated ones, which are typically >0.8 and with several FRBs having z > 1 even after properly subtracting the DM contribution from the FRB host. In this case, for these two FRB distribution functions, a cutoff zcut = 1 is chosen to match redshifts of currently detected events. In our analysis, Nconst and NSFH are employed to investigate whether cosmological implications from lensed FRBs are dependent on the assumed redshift distributions.

Lensing probability

According to the lensing theory32, the probability for a distant source at redshift zs lensed by an intervening dark matter halo is

$$P = {\int}_0^{z_{\mathrm{s}}} dz_{\mathrm{l}}\frac{{dD^{\mathrm{p}}}}{{dz_{\mathrm{l}}}}{\int}_0^\infty \sigma (M,z_{\mathrm{l}})n(M,z_{\mathrm{l}})dM,$$
(5)

where dDp/dzl is the proper distance interval, σ(M, zl) is the lensing cross-section of a dark matter halo with its mass and redshift being M and zl, respectively, n(M, zl)dM is the proper number density of the deflectors with masses between M and M + dM. For a singular isothermal sphere (SIS) lens, the cross-section producing two images with a flux ratio being smaller than a given threshold r is33

$$\sigma ( < r) = 16\pi ^3\left( {\frac{{\sigma _{\mathrm{v}}}}{c}} \right)^4\left( {\frac{{r - 1}}{{r + 1}}} \right)^2\left( {\frac{{D_{\mathrm{l}}^{\mathrm{A}}D_{\mathrm{ls}}^{\mathrm{A}}}}{{D_{\mathrm{s}}^{\mathrm{A}}}}} \right)^2,$$
(6)

where σv is the velocity dispersion. Moreover, the comoving number density of dark matter halos within the mass range (M, M + dM) at redshift z is

$$n(M,z)dM = \frac{{\rho _0}}{M}f(M,z)dM,$$
(7)

where ρ0 is the present value of the mean mass density in the universe, and f(M, z)dM is the Press–Schechter function34. For any FRB at redshift zs following the distribution Nconst or NSFH, we determine the lens redshift zl by maximizing the differential lensing probability, dP/dz. Assuming an SIS-like lens halo of mass $$M = 10^{12}M_ \odot h^{ - 1}$$ (h = H0/100 km s−1 Mpc−1) and r ≤ 5, the function of the lens redshift zl producing the maximal differential lensing probability with respect to the source reshift zs was shown in Fig. 2 of ref 10. In our analysis, this function is used to determine the lens redshift zl for any given source at redshift zs.

Uncertainty contribution

In order to estimate the time-delay distance from individual lensing systems for an accurate cosmography, as suggested in Eq. (1), it has been recognized that three key analysis steps should be carried out35,36: that is, (1) time-delay measurement, (2) lens galaxy mass modeling, which can be used to predict the Fermat potential differences, and (3) the line of sight (LOS) environment modeling, which is adopted to account for the weak lensing effects due to massive structures in the lens plane and along the LOS.

The differences in the arrival time between images can be precisely measured for a lensed FRB system since the short duration of each burst $$\sim{\cal O}$$ (ms) is much smaller (~10−9) than the typical galaxy-lensing time delay $$\sim{\cal O}$$ (10 days). Therefore, errors of time-delay measurements for lensed FRBs are negligible. Compared to the best 3% uncertainty of time-delay measurements in traditional lensed quasars21,37, the precision for the case of FRBs is greatly improved.

For the lens galaxy mass modeling, it requires a high-resolution, good-quality image of the lensed host galaxy and accurate localizations of the lensed FRB images. The advantage of a lensed FRB system is that it does not have a bright AGN, so that clean host images can be obtained before or after FRB. In practice, once a strongly lensed repeating FRB is identified by a large field-of-view radio survey program (e.g., CHIME), images of the lensed FRBs can be accurately localized from the deep follow-up observations with VLBI or SKA. High-quality optical images of the host galaxy can be obtained from follow-up facilities such as HST or the near-future James Webb Space Telescope (JWST), which can be used to study the mass distribution of the deflector with lens modeling techniques.

The last ingredient of uncertainty contribution for time-delay cosmography is the one from LOS environment modeling. In this issue, a troublesome point is that the external convergence (κext), resulting from the mass distributed outside of the lens galaxy but close in projection to the lens system along the LOS, does not change observable quantities of a lens system (i.e., image positions, flux ratios for point sources, and the image shapes for extended sources) but affects the predicted time delays by a factor of (1 − κext) when we fit the lens and source models to observations. Consequently, the true DΔt is related to the modeled one via $$D_{{\mathrm{\Delta }}t} = D_{{\mathrm{\Delta }}t}^{{\mathrm{model}}}/(1 - \kappa _{{\mathrm{ext}}})$$. For the lens HE 0435-122348, by using various combinations of relevant informative weighing schemes for the galaxy counts49 and ray tracing through the Millennium Simulation50, it was found that the most robust estimate of κext has a median value $$\kappa _{{\mathrm{ext}}}^{{\mathrm{med}}} = 0.004$$ and a standard deviation σκ = 0.025, which corresponds to a 2.5% uncertainty on the time-delay distance51. More recently, using deep r-band images from Subaru-Suprime-Cam and an inpainting technique and Multi-Scale Entropy filtering algorithm, a weak gravitational lensing measurement of the external convergence along the LOS to HE 0435-1223 was achieved $$\kappa _{{\mathrm{ext}}} = - 0.012_{ - 0.013}^{ + 0.020}$$, which corresponds to ~1.6% uncertainty on the time-delay distance52. Furthermore, the distribution of κext is robust to choices of weights, apertures, and flux limits, up to an impact of 0.5% on the inferred time-delay distance53. Here we assume that for lensed FRB systems, the LOS environment contributes a systematic uncertainty at an averaged 2% level to the inferred time-delay distance, expecting that every lensed FRB system is given enough attention by the community so that we can combine auxiliary follow-up data from facilities at different wavelengths and other available simulations with convergence maps. It is still worth noting that larger LOS systematic uncertainty could lead to larger error bars of cosmological parameters. In addition to the distribution of mass external to the lens, the mass distribution in the outskirt of the lens halo, in which there are little optical light traces, might lead to errors in the inferred time-delay distance at the percent level and it was shown that weak gravitational lensing and simulations may help to reduce these uncertainties54.

Distance sum rule

According to the cosmological principle, the space of the universe is homogeneous and isotropic at large scales. In this case, the spacetime can be described by the following FLRW metric:

$$ds^2 = - c^2{d}t^2 + a^2(t)\left( {\frac{{{d}r^2}}{{1 - Kr^2}} + r^2{d}{\mathrm{\Omega }}^2} \right),$$
(8)

where a(t) is the scale factor and K is a constant characterizing the geometry of three-dimensional space. In this metric, the dimensionless comoving angular diameter distance of a source at redshift zs as observed at redshift zl is written as

$$d(z_{\mathrm{l}},z_{\mathrm{s}}) = \frac{1}{{\sqrt {\left| {{\mathrm{\Omega }}_k} \right|} }}S_K\left( {\sqrt {\left| {{\mathrm{\Omega }}_k} \right|} {\int}_{z_{\mathrm{l}}}^{z_{\mathrm{s}}} \frac{{dx}}{{E(x)}}} \right),$$
(9)

where $${\mathrm{\Omega }}_k \equiv - K/H_0^2a_0^2$$ (a0 = a(0) is the present value of the scale factor),

$$S_K(X) = \left\{ {\begin{array}{*{20}{l}} {{\mathrm{sin}}(X)} \hfill & {{\mathrm{\Omega }}_k < 0}, \hfill \\ X \hfill & {{\mathrm{\Omega }}_k = 0}, \hfill \\ {{\mathrm{sinh}}(X)} \hfill & {{\mathrm{\Omega }}_k > 0,} \hfill \end{array}} \right.$$
(10)

and E(z) ≡ H(z)/H0. In addition, we denote d(z) ≡ d(0, z), dl ≡ d(0, zl), ds ≡ d(0, zs), and dls ≡ d(zl, zs), respectively. If the function of redshift z with respect to cosmic time t is single-valued and d′(z) > 0, these three distances in the FLRW scenario satisfy a simple sum rule24

$$d_{\mathrm{ls}} = d_{\mathrm{s}}\sqrt {1 + {\mathrm{\Omega }}_kd_{\mathrm{l}}^2} - d_{\mathrm{l}}\sqrt {1 + {\mathrm{\Omega }}_kd_{\mathrm{s}}^2} .$$
(11)

Apparently, one has ds > dl + dls, ds = dl + dls and ds < dl + dls for Ωk > 0, Ωk = 0 and Ωk < 0, respectively. Furthermore, in order to compare this sum rule with observations, we can rewrite Eq. (11) as

$$\frac{{d_{\mathrm{ls}}}}{{d_{\mathrm{s}}}} = \sqrt {1 + {\mathrm{\Omega }}_kd_{\mathrm{l}}^2} - \frac{{d_{\mathrm{l}}}}{{d_{\mathrm{s}}}}\sqrt {1 + {\mathrm{\Omega }}_kd_{\mathrm{s}}^2} .$$
(12)

Recently, the sum rule in this form was proposed to model-independent test, the FLRW metric, by comparing the distance ratios dls/ds measured from strongly lensed quasar systems with distances estimated from SNe Ia observations25. More recently, we upgraded the distance sum rule and rewrote it as26

$$\frac{{d_{\mathrm{ls}}}}{{d_{\mathrm{l}}d_{\mathrm{s}}}} = T(z_{\mathrm{l}}) - T(z_{\mathrm{s}}),$$
(13)

where

$$T(z) = \frac{1}{{d(z)}}\sqrt {1 + {\mathrm{\Omega }}_kd^2(z)} ,$$
(14)

to test the FLRW metric and estimate the cosmic curvature with time-delay distance measurements.

Statistical analysis

In order to estimate constraining power from 10 lensed FRB systems, we propagate the relative uncertainties of time delay (δΔτ = 0), Fermat potential difference (δΔΦ = 0.8%), and LOS contamination (δκext = 2%) to the relative uncertainty of DΔt, and then to the (relative) uncertainties of cosmological parameters: $$(\delta {\mathrm{\Delta }}t,\delta {\mathrm{\Delta \Phi }},\delta \kappa _{{\mathrm{ext}}})\sim \delta D_{{\mathrm{\Delta t}}}\sim (\delta H_{\mathrm{0}},\sigma _{{\mathrm{\Omega }}_k})$$. Then, we performed Markov Chain Monte Carlo (MCMC) minimization of the following χ2 objective function:

$$\chi ^2 = \mathop {\sum}\limits_{i = 1}^{10} \left( {D_{{\mathrm{\Delta}t},i}^{{\mathrm{th}}}(z_{{\mathrm{d}},i},z_{{\mathrm{s}},i};{\mathbf{p}}) - D_{{\mathrm{\Delta}t},i}^{{\mathrm{sim}}}} \right)^2/\sigma _{D_{{\mathrm{\Delta}t},i}}^2,$$
(15)

where $$D_{{\mathrm{\Delta}t}}^{{\mathrm{th}}}$$ is the theoretical time-delay distance in the assumed cosmological model or from the combination of distance sum rule and SNe Ia observations, while $$D_{{\mathrm{\Delta}t}}^{{\mathrm{sim}}}$$ is the corresponding simulated distance with its uncertainty being $$\sigma _{D_{{\mathrm{\Delta}t},i}} = \delta D_{{\mathrm{\Delta}t},i}D_{{\mathrm{\Delta}t},i}$$. p represents cosmological parameters (H0, Ωk) to be constrained and they are sampled in ranges H0 [0, 150], Ωk [−1, 1].

We perform 10,000 simulations each containing 10 lensed FRB systems. For each dataset, we carry out the above-mentioned MCMC minimization to obtain the best-fit value of corresponding parameters. Then, we plot the probability distributions of the best-fit H0 and Ωk in Figs. 1, 2, respectively.

Data availability

The data that support the findings of this study are available from the corresponding author upon request.

References

1. 1.

Lorimer, D. R. et al. A bright millisecond radio burst of extragalactic origin. Science 318, 777 (2007).

2. 2.

Thornton, D. et al. A population of fast radio bursts at cosmological distances. Science 341, 53 (2013).

3. 3.

Petroff, E. et al. FRBCAT: the fast radio burst catalogue. Publ. Astron. Soc. Aust. 33, e045 (2016).

4. 4.

Spitler, L. G. et al. A repeating fast radio burst. Nature 531, 202 (2016).

5. 5.

Chatterjee, S. et al. A direct localization of a fast radio burst and its host. Nature 541, 58 (2017).

6. 6.

Palaniswamy, D., Li, Y. & Zhang, B. Are there multiple populations of fast radio bursts? Astrophys. J. Lett. 854, 1 (2017).

7. 7.

Cordes, J. M. et al. Lensing of fast radio bursts by plasma structures in host galaxies. Astrophys. J. 842, 35 (2017).

8. 8.

Zheng, Z. et al. Probing the intergalactic medium with fast radio bursts. Astrophys. J. 797, 71 (2014).

9. 9.

Muñoz, J. B. et al. Lensing of fast radio bursts as a probe of compact dark matter. Phys. Rev. Lett. 117, 091301 (2016).

10. 10.

Li, C. & Li, L. Constraining fast radio burst progenitors with gravitational lensing. Sci. China Phys. Mech. Astron. 57, 1390 (2014).

11. 11.

Dai, L. & Lu, W. Probing motion of fast radio burst sources by timing strongly lensed repeaters. Astrophys. J. 847, 19 (2017).

12. 12.

Champion, D. J. et al. Five new fast radio bursts from the HTRU high-latitude survey at Parkes: first evidence for two-component bursts. Mon. Not. R. Astron. Soc. 460, L30 (2016).

13. 13.

Caleb, M. et al. Fast radio transient searches with UTMOST at 843 MHz. Mon. Not. R. Astron. Soc. 458, 718 (2016).

14. 14.

Newburgh, L. B. et al. HIRAX: a probe of dark energy and radio transients. Ground-Based Airborne Telesc. VI 9906, 99065X (2016).

15. 15.

Bandura, K. et al. Canadian Hydrogen Intensity Mapping Experiment (CHIME) pathfinder. Ground-Based Airborne Telesc. V 9145, 914522 (2014).

16. 16.

Macquart, J. P. et al. Fast transients at cosmological distances with the SKA. In Adv ancing Astrophysics with the Square Kilometre Array, Proceedings of Science (AASKA14), 55, arXiv:1501.07535 (Giardini Naxos, Italy, 2015).

17. 17.

Fialkov, A., Loeb, A. & Fast Radio, A. Burst occurs every second throughout the observable universe. Astrophys. J. Lett. 846, L27 (2017).

18. 18.

Hilbert, S. et al. Strong-lensing optical depths in a ΛCDM universe—II. The influence of the stellar mass in galaxies. Mon. Not. R. Astron. Soc. 386, 1845 (2008).

19. 19.

Riess, A. G. et al. Determination of the local value of the Hubble constant. Astrophys. J. 826, 56 (2016).

20. 20.

Ade, P. A. R. et al. Planck 2015 results XIII. Cosmological parameters. Astron. Astrophys. 594, A13 (2016).

21. 21.

Suyu, S. H. et al. H0LiCOW—I. H0 Lenses in COSMOGRAIL’s Wellspring: program overview. Mon. Not. R. Astron. Soc. 468, 2590 (2017).

22. 22.

Clarkson, C. et al. A general test of the Copernican principle. Phys. Rev. Lett. 101, 011301 (2008).

23. 23.

Clarkson, C. et al. Dynamical dark energy or simply cosmic curvature? J. Cosmol. Astropart. Phys. 8, 011 (2007).

24. 24.

Peebles, P. J. E. Principles of Physical Cosmology (Princeton University Press, Princeton, NJ, 1993).

25. 25.

Räsänen, S. et al. New test of the Friedmann–Lemaître–Robertson–Walker Metric using the distance sum rule. Phys. Rev. Lett. 115, 101301 (2015).

26. 26.

Liao, K. et al. Test of the FLRW metric and curvature with strong lens time delays. Astrophys. J. 839, 70 (2017).

27. 27.

Oppermann, N., Connor, L. & Pen, U. L. The Euclidean distribution of fast radio bursts. Mon. Not. R. Astron. Soc. 461, 984 (2016).

28. 28.

Cole, S. et al. The 2dF galaxy redshift survey: near-infrared galaxy luminosity functions. Mon. Not. R. Astron. Soc. 326, 255 (2001).

29. 29.

Hopkins, A. M. & Beacom, J. F. On the normalization of the cosmic star formation history. Astrophys. J. 651, 142 (2006).

30. 30.

Ioka, K. The cosmic dispersion measure from gamma-ray burst afterglows: probing the reionization history and the burst environment. Astrophys. J. Lett. 598, L79 (2003).

31. 31.

Deng, W. & Zhang, B. Cosmological implications of fast radio burst/gamma-ray burst associations. Astrophys. J. Lett. 783, L35 (2014).

32. 32.

Schneider, P., Ehlers, J. & Falco, E. E. Gravitational Lenses (Springer, Berlin, 1992).

33. 33.

Li, L. X. & Ostriker, J. P. Semianalytical models for lensing by dark halos. I. Splitting angles. Astrophys. J. 566, 652 (2002).

34. 34.

Press, W. H. & Schechter, P. Formation of galaxies and clusters of galaxies by self-similar gravitational condensation. Astrophys. J. 187, 425 (1974).

35. 35.

Treu, T. Strong lensing by galaxies. Annu. Rev. Astron. Astrophys. 48, 87 (2010).

36. 36.

Treu, T. & Marshall, P. J. Time delay cosmography. Astron. Astrophys. Rev. 24, 11 (2016).

37. 37.

Liao, K. et al. Strong lens time delay challenge: II. Results of TDC1. Astrophys. J. 800, 11 (2015).

38. 38.

Ding, X. et al. H0LiCOW VI. Testing the fidelity of lensed quasar host galaxy reconstruction. Mon. Not. R. Astron. Soc. 465, 4634 (2017).

39. 39.

Sérsic, J. L. Influence of the atmospheric and instrumental dispersion on the brightness distribution in a galaxy. Bol. Asoc. Argent. Astron. 6, 41 (1963).

40. 40.

Oguri, M. The mass distribution of SDSS J1004 + 4112 revisited. Publ. Astron. Soc. Jpn. 62, 1017 (2010).

41. 41.

Alessandro, S. On the choice of lens density profile in time delay cosmography. Mon. Not. R. Astron. Soc. 474, 4648 (2018).

42. 42.

Schneider, P. & Sluse, D. Mass-sheet degeneracy, power-law models and external convergence: impact on the determination of the Hubble constant from gravitational lensing. Astron. Astrophys. 559, A37 (2013).

43. 43.

Xu, D. et al. Lens galaxies in the Illustris simulation: power-law models and the bias of the Hubble constant from time-delays. Mon. Not. R. Astron. Soc. 456, 739 (2015).

44. 44.

Tagore, A. S., et al. Reducing biases on H 0 measurements using strong lensing and galaxy dynamics: results from the EAGLE simulation. Monthly Notices of the Royal Astronomical Society, 474, 3403 (2018)

45. 45.

Jaffe, W. A simple model for the distribution of light in spherical galaxies. Mon. Not. R. Astron. Soc. 202, 995 (1983).

46. 46.

Keeton, C. R. A catalog of mass models for gravitational lensing. Preprint at http://arxiv.org/abs/astro-ph/0102341 (2001).

47. 47.

Wong, K. C. et al. H0LiCOW IV. Lens mass model of HE 0435-1223 and blind measurement of its time-delay distance for cosmology. Mon. Not. R. Astron. Soc. 465, 4895 (2017).

48. 48.

Wisotzki, L. et al. HE 0435-1223: a wide separation quadruple QSO and gravitational lens. Astron. Astrophys. 395, 17 (2002).

49. 49.

Fassnacht, C. D., Koopmans, L. V. E. & Wong, K. C. Galaxy number counts and implications for strong lensing. Mon. Not. R. Astron. Soc. 410, 2167 (2011).

50. 50.

Springel, V. et al. Simulations of the formation, evolution and clustering of galaxies and quasars. Nature 435, 629 (2005).

51. 51.

Cristian, E. R. et al. H0LiCOW III. Quantifying the effect of mass along the line of sight to the gravitational lens HE 0435-1223 through weighted galaxy counts. Mon. Not. R. Astron. Soc. 467, 4220 (2017).

52. 52.

Tihhonova, O. et al. H0LiCOW VIII. A weak lensing measurement of the external convergence in the field of the lensed quasar HE 0435-1223. Mon. Not. R. Astron. Soc. 477, 5657 (2018).

53. 53.

Bonvin, V. et al. H0LiCOW V. New COSMOGRAIL time delays of HE 0435-1223: H0 to 3.8% precision from strong lensing in a LCDM model. Mon. Not. R. Astron. Soc. 465, 4914 (2017).

54. 54.

Muñoz, J. B. & Kamionkowski, M. Large-distance lens uncertainties and time-delay measurements of H 0. Phys. Rev. D. 96, 103537 (2017).

55. 55.

Freedman, W. L. et al. Carnegie Hubble Program: a mid-infrared calibration of the Hubble constant. Astrophys. J. 758, 24 (2012).

56. 56.

Reid, M. J. et al. The Megamaser Cosmology Project: IV. A direct measurement of the Hubble constant from UGC 3789. Astrophys. J. 676, 154 (2013).

57. 57.

Li, Z. et al. Model-independent estimations for the cosmic curvature from standard candles and clocks. Astrophys. J. 833, 240 (2016).

58. 58.

Yu, H. & Wang, F.-Y. New model-independent method to test the curvature of the universe. Astrophys. J. 828, 85 (2016).

59. 59.

Mörtsell, E. & Jönsson, J. A model independent measure of the large scale curvature of the universe. Preprint at https://arxiv.org/abs/1102.4485 (2011).

Acknowledgements

We would like to thank Liang Dai, Sherry H. Suyu, and Zong-Hong Zhu for reading the manuscript and their constructive suggestions. We are indebted to Kai Liao for his great help in lens modeling. This work was supported by the Ministry of Science and Technology National Basic Science Program (Project 973) under Grant No. 2014CB845806, the National Natural Science Foundation of China under Grants Nos. 11505008, 11722324, 11603003, 11373014, and 11633001, and the Strategic Priority Research Program of the Chinese Academy of Sciences, Grant No. XDB23040100. X. Ding acknowledges support by China Postdoc Grant No. 2017M622501.

Author information

Authors

Contributions

Z.-X.L. and H.G. initiated the idea using accurate measurements of time delays between images of lensed FRBs as a powerful probe for precision cosmology and drafted the paper. B.Z. contributed in discussion about this idea and in writing the manuscript. X.-H. D. and Z.-X.L. contributed in discussions and simulations for lens modeling. G.-J.W. and Z.-X.L. contributed in calculations for cosmological parameters inferences.

Corresponding author

Correspondence to He Gao.

Ethics declarations

Competing interests

The authors declare no competing interests.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

Li, ZX., Gao, H., Ding, XH. et al. Strongly lensed repeating fast radio bursts as precision probes of the universe. Nat Commun 9, 3833 (2018). https://doi.org/10.1038/s41467-018-06303-0

• Accepted:

• Published: