Beschreibung
The next-generation of wide-field galaxy surveys, known as Stage IV experiments, aims to investigate the "dark sector" by mapping a significant fraction of the observable Universe at unprecedented depth. In doing so, these surveys will also contribute to transform our understanding of galaxy formation and evolution. Their vast statistical power is however limited by systematic effects that dominate over the statistical uncertainties of the measurements, with redshift calibration being one of the dominant observational systematics for Stage IV large scale structure cosmology. In this talk I will describe the two complementary approaches I am leading to tackle the redshift calibration problem for next generation surveys. On one side, the 4MOST Complete Calibration of the Colour Redshift Relation survey, which has recently entered its survey verification phase and is designed to obtain more than 300k spectroscopic redshifts for a sample of galaxies that cover the colour SOM representative of the manifold spanned by Euclid and Rubin-LSST. Complementary to this, the forward-modelling of photometric and spectroscopic galaxy surveys, a method that bridges cosmology with galaxy evolution, by generating physically motivated galaxy populations, mapping them to multi-band observables, and producing realistic mock survey images and spectra that can be passed through the same detection, measurement, and selection steps as the data. I will discuss the past (Tortorelli+18,20,21) and on-going efforts (Tortorelli+24, Tortorelli+25) in forward-modelling galaxy surveys, from the modelling of the galaxy population (GalSBI-SPS) to the simulation of images (UFig) and fiber-based spectra (USpec), targeting 4MOST and DESI spectra, but extendable to other fiber-based spectrographs (e.g. PFS). The model is currently being constrained against HSC Deep observations, whose depth and blending regime make them a valuable precursor of Stage-IV surveys. This approach allows us to use a physically grounded population model to reproduce the observed colour, magnitude, and size distributions precisely in the regime where conventional redshift calibration is most challenging, thus obtaining redshift distribution with the precision required for Stage IV cosmological analyses.