Photometric redshift prediction: using measurement errors

• Grant number: 24/00868-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2024
• End date: April 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Applied Probability and Statistics
• Principal Investigator: Rafael Izbicki
• Grantee: Lucas Sala Battisti
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

Redshift is an astronomical measure that describes the shift of an electromagnetic wave toward the red end of the spectrum. This measure is crucial for quantifying the distance between the observer and astronomical objects (such as galaxies and quasars) and also for measuring how the universe is expanding. The most accurate method for redshift estimation is through spectroscopy. However, due to cost and time constraints, an increasingly investigated alternative is estimation through photometry. In this context, the amount of light emitted by an astronomical object is measured for certain wavelength intervals. Additionally, it is possible to measure the degree of uncertainty associated with these quantities of light. Traditionally, these photometric measurements are used as covariates for predicting the redshift of an object; however, error measurements are often disregarded in these studies. Therefore, this project focuses on investigating machine learning methods that aim to incorporate measures of uncertainty for redshift estimation, using data from quasars in the S-PLUS (Southern Photometric Local Universe Survey).