Validation of galaxy clustering and weak lensing analysis pipelines for spectroscopic and photometric surveys

Understanding the late-time acceleration of the Universe’s expansion is a central challenge in modern cosmology. To investigate the mechanisms behind this acceleration, missions using space telescopes like Euclid and ground-based telescopes like the Vera C. Rubin Observatory are collecting large galaxy survey datasets. Developing efficient data analysis pipelines is crucial for extracting valuable cosmological informations. This Thesis validates data analysis pipelines for galaxy clustering in spectroscopic and photometric surveys, as well as weak gravitational lensing in photometric surveys. Various likelihood models were implemented for this purpose. We tested the implementation by comparing cosmological functions computed with CosmoBolognaLib (CBL), Cosmological Likelihood Observable of Euclid (CLOE), and Core Cosmology Library (CCL). The comparison showed excellent agreement, with differences consistently below 0.4% for fundamental cosmological functions. Similar results were found for spectroscopic galaxy clustering model, while the photometric model differences ranged from 0.01% to 5%. A Bayesian MCMC analysis was then used to validate constraints on seven cosmological parameters: Hubble constant H0, the logarithm of the primordial scalar amplitude log As, spectral index ns, baryonic matter density Ωbh2, cold dark matter density Ωch2, and dark energy equation of state parameters w0 and wa. The results from CBL, CLOE, and CCL were compared, modelling spectroscopic galaxy clustering through the Legendre multipoles of galaxy power spectrum, and angular clustering and weak lensing through the angular and shear power spectra in harmonic space and the Limber approximation. The results are presented as the estimates of the seven parameters, given by the median and 1σ of the marginalised 1-D posterior distributions obtained through the MCMC. In all our analysis, the results show good agreement between the parameter estimates obtained from CBL, CLOE, and CCL.