Utilizing localized fast radio bursts to constrain their progenitors and the expansion history of the Universe

TL;DR

This work uses a sample of 110 localized FRBs to jointly infer the Hubble constant $H_0$ and the host DM distribution by modeling the IGM DM with two cosmological simulations (CoDa II and IllustrisTNG) and treating the host DM as a lognormal variable. By constructing a likelihood that integrates over both host and IGM contributions and performing MCMC over $h$ and $\mu_0$, the authors find $h=0.70^{+0.11}_{-0.07}$ (IllustrisTNG) and $h=0.64^{+0.11}_{-0.09}$ (CoDa), consistent with both Planck and local $H_0$ measurements within uncertainties. They project that ~200 localized FRBs could resolve the Hubble tension at $\gtrsim 4\sigma$, depending on the redshift distribution and IGM modeling. The analysis also constraints FRB progenitor ages, favoring very young ($<1$ Myr) populations and suggesting that scatter broadening is dominated by the immediate FRB environment rather than the host galaxy ISM, with implications for magnetar-like progenitors and FRB host environments.

Abstract

Fast radio bursts (FRBs) are increasingly being used for cosmological applications such as measuring the Hubble constant and baryon abundance. The increasing number of localized FRBs and precise measurement of dispersion measure (DM) make them a suitable probe for such an approach. We use a sample of 110 localized FRBs as well as a small sub-sample of 24 FRBs with scattering timescale measurements or limits. We infer the Hubble constant ($H_0$) and the DM distribution of the host galaxies simultaneously by fitting our model to the FRB DM measurements. With current data, our results are in agreement with both high and low redshift measurements of $H_0$, obtained using Cosmic Microwave Background (CMB) and Type Ia supernovae data respectively. We project that with about 200 localized FRBs, we would be in a position to distinguish between the two scenarios at 4$σ$ confidence. In addition, the host DM is expected to be related to star formation in the host galaxy and the stellar age of the progenitors. We show that young progenitors with an age of less than 1 Myr are consistent with our inferred distribution of host DM at 95 percent confidence. These young sources may be associated with long scatter broadening times and large DM from their source environments. Indeed, we find that scatter broadening times of FRBs are inconsistent with the Milky Way ISM, but at the same time, do not appear to be strongly correlated with the FRBs' redshift or with the SFR or stellar mass of their host galaxies. This suggests that scattering is dominated by the immediate environment of the sources.

Figures (7)

• Figure 1: Marginalized probability distribution function for $h$ and $\mu_0$ by fitting these parameters to our sample of 110 FRBs. We show the 68 percent confidence interval in shaded regions for respective cases with the same color.
• Figure 2: Normalized redshift distribution of the FRB sample. In red, we show the number of FRBs in Table \ref{['tab:FRB_sample']} within $\Delta z=0.1$ intervals along with the Poisson error in gray. The corresponding approximate distribution is shown in dotted blue. We also consider a uniform distribution (dashed green) between $z=0-1$ and a more physically motivated distribution for a hypothetical future large FoV telescope with good localization capabilities (in black; see § \ref{['sec:projections']}).
• Figure 3: Futuristic constraints from an FRB sample of a given size (denoted in the plot). The color coding follows from Fig. \ref{['fig:redshift_distribution']}. In dashed blue, we consider a scenario where half of total population of FRBs is contributed by persistent emission-like sources with $\langle {\rm DM}_{\rm host}\rangle \approx 350\hbox{pc cm}^{-3}$ (see Sec. \ref{['sec:FRB_age']} for details). In our simulation, we create a FRB sample with a background Hubble constant $h=0.74$. We use 200 FRBs for our simulation. With about 200 localized FRBs, we can resolve the Hubble tension at 4$\sigma$ while more number of localized FRBs at higher redshifts will increase the detection significance.
• Figure 4: We plot the distribution of spindown ages as a function of the absolute value of latitude of the galactic pulsars, color coded in their measured DM (left panel). We have not included millisecond pulsars in this sample. We also show the Galactic radio magnetars in large triangles. These magnetars are not included in any of the analysis. (Right panel) We show the 2D distribution of DM and latitude for pulsars with spindown age $<10$ Myr (green) and 10-100 Myr (blue). We also show the linear best fit for the two cases in the log10(DM) and ${\rm \log_{10}(|latitude|)}$ space.
• Figure 5: Comparison of cumulative distribution function (CDF) of our inferred host DM contribution with the DM distribution of Galactic pulsars sampled as a function of viewing angle and as a function of age (See text for details). In dashed and solid black, we consider our inferred best fit value of $\mu_0=4$ and the rescaled version respectively. We consider a sample of 5000 FRBs to obtain a smooth curve. We also plot realizations where we sample 100 FRBs from a lognormal distribution with $\mu_0=4$ and $\sigma_{\rm host}=1$, in grey.