What can we learn about Reionization astrophysical parameters using Gaussian Process Regression?

TL;DR

The paper develops a Gaussian Process Regression framework to non-parametically reconstruct the UV luminosity density during the epoch of reionization and to infer key astrophysical parameters (the ionizing photon budget via ⟨f_esc ξ_ion⟩ and the IGM clumping factor C_HII) from UVLFs, Lyα, and Planck optical depth data. By solving the ionization equation with GP-derived UV histories, it extracts a model-independent reionization history Q_HII(z) and propagates it into global 21-cm observables (ΔT_b) and the 21-cm power spectrum, using MCMC to explore parameter spaces. The study demonstrates that GPR can provide robust, informative inferences that improve when including SARAS and SKA data, offering tighter constraints on reionization parameters and consistent cosmological bounds within ΛCDM. It also highlights the complementarities of global-signal and power-spectrum analyses and outlines future extensions to more complex models and additional observational probes.

Abstract

Reionization is one of the least understood processes in the evolution history of the Universe, mostly because of the numerous astrophysical processes occurring simultaneously about which we do not have a very clear idea so far. In this article, we use the Gaussian Process Regression (GPR) method to learn the reionization history and infer the astrophysical parameters. We reconstruct the UV luminosity density function using the HFF and early JWST data. From the reconstructed history of reionization, the global differential brightness temperature fluctuation during this epoch has been computed. We perform MCMC analysis of the global 21-cm signal using the instrumental specifications of SARAS, in combination with Lyman-$α$ ionization fraction data, Planck optical depth measurements and UV luminosity data. Our analysis reveals that GPR can help infer the astrophysical parameters in a model-agnostic way than conventional methods. Additionally, we analyze the 21-cm power spectrum using the reconstructed history of reionization and demonstrate how the future 21-cm mission SKA, in combination with Planck and Lyman-$α$ forest data, improves the bounds on the reionization astrophysical parameters by doing a joint MCMC analysis for the astrophysical parameters plus 6 cosmological parameters for $Λ$CDM model. The results make the GPR-based reconstruction technique a robust learning process and the inferences on the astrophysical parameters obtained therefrom are quite reliable that can be used for future analysis.

Figures (5)

• Figure 1: The reconstructed UV luminosity density function in the redshift range $z \sim 4-12$ obtained from Gaussian Process regression using the UV17(A) and UV17(B) data.
• Figure 2: Comparison between the constraints obtained on the astrophysical parameters employing the (a) UV17(A)+QHII vs UV17(B)+QHII, (b) with the clumping factor kept constant at $C_{\rm HII}$=5, (c) UV17(A)+QHII+Planck vs UV17(B)+QHII+Planck, and (d) with the clumping factor kept constant at $C_{\rm HII}$=5, during the MCMC. The black dashed line and shaded region represent the Planck best-fit with 1$\sigma$ CL.
• Figure 3: Evolution of the ionization fraction ($Q_{\rm HII}$) and global 21-cm signal ($\Delta T_b$) from reconstruction. Plots (a) and (b) show ionization fraction obtained using UV17+QHII+Planck (blue curve) vs UV17+QHII+Planck for a fixed value of the clumping factor $C_{\rm HII}=5$ (red dashed line) from the GP reconstruction of the UV luminosity density with the UV17(A) (left panels) and UV17(B) (right panels) compilation. Plots (c) and (d) show the global 21-cm signal at the reionization epoch from respective data sets. The QHII data is plotted with references for comparison. Mock A and B represent the simulated $\Delta T_b$ vs $z$ catalogues.
• Figure 4: Comparison between the bounds obtained on the astrophysical parameters using UV17(A)+QHII+Planck+SARAS vs UV17(B)+QHII+Planck+SARAS and (b) with a constant clumping factor of $C_{\rm HII}=5$ (panel (b)). Evolution of the ionization fraction in (c) and the global 21-cm signal at reionization epoch in (d) obtained using UV17(A)+QHII+Planck+SARAS (blue curve) and UV17(B)+QHII+Planck+SARAS (red dashed line) from the GP reconstruction of the UV luminosity density with the UV17(A) and UV17(B) compilations respectively
• Figure 5: Forecasting the reionization power spectrum of 6 $\Lambda$CDM parameters with 2 astrophysical parameters: ionization efficiency and clumping factor, using the Planck+SKA+Ly$\alpha$+UV17(A) data sets and the Planck+SKA+Ly$\alpha$+UV17(B) data sets. Here for this forecast and MCMC analysis, we have used fake Planck realistic data. The analysis is undertaken using the mean reconstructed $Q_{\rm HII}$ function, without incorporating the errors.