Speaker
Description
Modern large galaxy surveys have reached a level of statistical precision where observational systematics dominate the error budget of large-scale structure measurements. Accurately quantifying and mitigating these effects is therefore essential to avoid biased cosmological constraints and to fully exploit the potential of current and upcoming surveys.
Observational systematics are typically handled using random catalogues, which represent the distribution of unclustered galaxies. These catalogues enter the computation of the $n$-point statistics, which encode precious cosmological information about the large-scale structure of the Universe. Because cosmological information is sensitive to the accuracy of the random catalogues, their construction is crucial for obtaining robust results.
In this talk, I will review the role of random catalogues and present a strategy to create them using a source-injection approach applied to Euclid spectroscopic data. Simulated spectra are injected into real observations and processed through the full spectroscopic pipeline. By construction, this method learns instrumental and survey-related effects from the data and imprints them into the random catalog.
I will conclude by discussing the impact of the characterized systematics on clustering measurements.