TOPz: Photometric redshifts using template fitting applied to the GAMA survey

TL;DR

The paper introduces TOPz, a Bayesian template-fitting code for photometric redshift estimation that leverages physically motivated templates generated with CIGALE, flux and flux-uncertainty corrections, and a luminosity-function–based prior. It demonstrates that TOPz achieves accurate photo-zs on the GAMA nine-band data (σ_NMAD ≈ 0.012 for bright galaxies, rising to ≈0.021 for fainter ones) with low outlier rates and results consistent with EAZY and SFM. The authors also show that TOPz provides stellar masses from the templates and validate the method with DESI DR1 data, underscoring robustness beyond the GAMA spectroscopic limits. The code and data products are publicly released, and the approach is positioned for application to upcoming surveys like J-PAS and WAVES, highlighting the value of template-fitting with priors and template optimisation for future cosmological analyses.

Abstract

Context. Accurate photometric redshift estimation is crucial for cosmological and galaxy evolution studies, especially with the advent of large-scale photometric surveys. Aims. We developed a photo-z estimation code called TOPz (Tartu Observatory Photo-z) and applied it to the GAMA photometric catalogue. Using nine-band photometric data from the GAMA project, we assessed the accuracy of TOPz by comparing its photo-z estimates to available spectroscopic redshifts from GAMA and DESI. The latter extends to z < 2 and m_Z < 24, allowing the photo-z accuracy to be validated beyond the GAMA limits. Methods. TOPz employs a Bayesian template-fitting approach to estimate photo-z from marginalised redshift posteriors. We generated synthetic galaxy spectra using the CIGALE software and ran template set optimisation. We improved the photometry by applying flux and flux uncertainty corrections. An analytical prior was then imposed on the resulting posteriors to refine the redshift estimates. Results. The photo-z estimates produced by TOPz show good agreement with the spectroscopic redshifts in the low-redshift regime (z < 0.5). We demonstrate the redshift accuracy across various magnitude bins and tested how the flux corrections and posteriors reflect the actual uncertainty of the estimates. For the GAMA sample, the sigma_NMAD = 0.012 for m_Z <18 and increases to sigma_NMAD = 0.021 for m_Z >19. The outlier fraction (|dz|/(1 + z)>0.1) in the same magnitude bins increases from 1% to 5%. We show that the TOPz results are consistent with those obtained from other photo-z codes (EAZY and SFM) applied to the same data set. Conclusions. TOPz is an advanced photo-z estimation code that integrates flux corrections, physical priors, and template set optimisation to provide state-of-the-art photo-z among competing template-based redshift estimators.

Figures (30)

• Figure 1: Illustration of the template-fitting approach. The model spectrum is shown with a grey line, and red circles denote the model fluxes in different passbands. The passband transmission curves used for the GAMA data are shown as blue lines. The blue points with error bars are observed fluxes with 2-$\sigma$ errors for a given galaxy. The template spectrum is normalised to best match the observed fluxes.
• Figure 2: Schematic overview of the TOPz code, illustrating key inputs and outputs. The two-way arrows indicate that an iterative approach can be applied for further optimisation. Refer to the sections indicated in the corresponding boxes for additional information. See Fig. \ref{['fig:topz_posterior']} for the detailed description of the TOPz posteriors and output.
• Figure 3: TOPz redshift posterior for a source with many maxima and a spectroscopic redshift of $z \approx 0.21$ ($\zeta \approx 0.19$). The upper panel displays a 2D posterior distribution, with all templates arranged in order of increasing weighted $\zeta$ values. The colour of each point represents the photo-z likelihood of the corresponding template. The lower panel illustrates the marginalised posterior distribution for the same source. Dashed vertical lines indicate the $\zeta$ values for all peaks along with $\zeta_{w}$. For details on the definitions of the three peaks and the probability-weighted $\zeta_{w}$, refer to Sect. \ref{['sect:topz_output']}. The shaded area in the lower panel represents the range of the best peak, bounded by the minimum ($\zeta^{\mathrm{best}}_{\min}$) and maximum ($\zeta^{\mathrm{best}}_{\max}$) values.
• Figure 4: Spectroscopic redshift distribution in the GAMA catalogue (blue line) and a subset of the spectroscopic sample (red line) used for template set optimisation (see Sect. \ref{['sec:templates']}).
• Figure 5: Galaxy redshift $\zeta$ as a function of observed magnitude $m_r$ for the full sample (blue dots) and for the galaxy sample (red dots) used in template set optimisation.