Peeking into the next decade in Large-Scale Structure Cosmology with its Effective Field Theory

TL;DR

The paper investigates how the effective field theory of large-scale structure (EFTofLSS) can sharpen cosmological parameter constraints for next-generation surveys by performing Fisher forecasts based on one-loop power spectrum and bispectrum predictions. A central innovation is the perturbativity prior, which constrains loop terms to their theoretically expected size when many EFT parameters are fit, improving robustness against overfitting. Applying this framework to BOSS, DESI, and MegaMapper, the authors forecast substantial gains in measuring neutrino masses, spatial curvature, and primordial non-Gaussianities, with MegaMapper offering particularly strong prospects (e.g., >12σ for nonzero neutrino masses and Ω_k ≈ 0.0012). They carefully validate the Fisher approach against full MCMC analyses, explore the impact of shot noise and priors on EFT parameters, and demonstrate that higher-order statistics coupled with informative priors can unlock tests of inflationary physics beyond current limits. The results highlight the potential of the next decade of LSS surveys to probe fundamental physics, while also identifying shot noise and uncertain EFT parameters as key limiting factors that future work must address.

Abstract

After the successful full-shape analyses of BOSS data using the Effective Field Theory of Large-Scale Structure, we investigate what upcoming galaxy surveys might achieve. We introduce a ``perturbativity prior" that ensures that loop terms are as large as theoretically expected, which is effective in the case of a large number of EFT parameters. After validating our technique by comparison with already-performed analyses of BOSS data, we provide Fisher forecasts using the one-loop prediction for power spectrum and bispectrum for two benchmark surveys: DESI and MegaMapper. We find overall great improvements on the cosmological parameters. In particular, we find that MegaMapper (DESI) should obtain at least a 12$σ$ ($2σ$) evidence for non-vanishing neutrino masses, bound the curvature $Ω_k$ to 0.0012 (0.012), and primordial inflationary non-Gaussianities as follows: $f_{\text{NL}}^{\text{loc.}}$ to $\pm 0.26$ (3.3), $f_{\text{NL}}^{\text{eq.}}$ to $\pm16$ (92), $f_{\text{NL}}^{\text{orth.}}$ to $\pm 4.2$ (27). Such measurements would provide much insight on the theory of Inflation. We investigate the limiting factor of shot noise and ignorance of the EFT parameters.

Figures (18)

• Figure 1: Triangle plots and errors comparing a Fisher forecast (blue) against the data analysis from DAmico:2022osl (red) for base $\Lambda$CDM parameters. For the Fisher forecast the full measured covariance is used including all cross-correlations. We here analyze $\ell=0,2$ power spectrum multipoles and the bispectrum monopole, both at one loop order. We implement the approximate AP effect as discussed in Sec. \ref{['dataeffect']}.
• Figure 2: Triangle plots comparing different Fisher forecasts for base $\Lambda$CDM parameters using the $\ell = 0,2$ power spectrum multipoles and the bispectrum monopole, all at one loop order. The plots differ only in their covariances, where we compare the measured covariance (grey), including all cross-covariances, the diagonal of the measured covariance (red), and the analytical prediction for the diagonal covariance (blue). We implement the approximate AP effect as discussed in Sec. \ref{['dataeffect']}.
• Figure 3: Triangle plots and errors from Fisher forecasts for BOSS including the spectral tilt and spatial curvature (left), massive neutrinos (right), and primordial non-Gaussianity (bottom). The power spectrum monopole and quadrupole, and the bispectrum monopole were used both at one loop order. In the table we also report the upper and lower bounds of the $68\%$ confidence interval for the sum of massive neutrinos, i.e $\mathbb{P}\left[\left(\sum_i m_{\nu_i}- \sum_i m_{\nu_i}^{\text{ref}}\right) \in (\sigma^-,\sigma^+)\right] =0.68$. The covariance used here is the full, measured covariance with all cross-correlations. We implemented the approximate AP effect as discussed in Sec. \ref{['dataeffect']}.
• Figure 4: Triangle plots and errors from several different Fisher forecasts for BOSS using the analytical covariance. We compare base results to results obtained without shot noise (left) and with biases fixed or with a "galaxy-formation prior" (g.p.) (right). In the table, we also show the impact of including higher multipoles on the power spectrum and bispectrum and also see the impact on $f_{\text{NL}}$. For the constraints on $f_{\text{NL}}$, we fix the other cosmological parameters.
• Figure 5: Triangle plots and errors from Fisher forecasts for DESI including the spectral tilt and spatial curvature (left) and massive neutrinos (right) and Non-Gaussianity (bottom). In the table we also report the upper and lower bounds of the $68\%$ confidence interval for the sum of massive neutrinos, i.e $\mathbb{P}\left[\left(\sum_i m_{\nu_i}- \sum_i m_{\nu_i}^{\text{ref}}\right) \in (\sigma^-,\sigma^+)\right] =0.68$. We use all power spectrum and bispectrum multipoles at one loop order for the above results and use the analytical covariance without cross-correlations.