Understanding the evolution of the cosmic star formation rate density (SFRD) is a central goal of observational galaxy evolution. A major uncertainty in its reconstruction arises from dust-obscured star formation, which can be strongly underestimated by ultraviolet-selected surveys, and from the possible contribution of active galactic nuclei (AGNs) to the infrared emission used as a tracer of star formation. This thesis investigates these effects by revisiting the infrared properties of ALMA-detected galaxies in the COSMOS field using updated multiwavelength information, including recent JWST-based COSMOS2025 data. The analysis starts from the A3COSMOS catalogue and adopts a blind-like selection designed to minimize the biases associated with targeted ALMA observations. Updated photometry and redshift information are incorporated, giving priority to spectroscopic measurements, followed by COSMOS2025 photometric redshifts and, when necessary, photometric redshifts derived through CIGALE. The final analysis is performed on a sample of 189 galaxies. Their physical properties are derived through panchromatic spectral energy distribution fitting with CIGALE, including stellar emission, dust attenuation and re-emission, and an AGN component. Two different AGN prescriptions, skirtor2016 and fritz2006, are independently employed in order to quantify the dependence of the inferred quantities on the adopted torus model. The resulting total infrared luminosities are used to reconstruct the infrared luminosity functions through the 1/V_max method and, after integration, to estimate the dust-obscured SFRD. The analysis is carried out in two broad redshift intervals, 0.5=< z<3.0 and 3.0 =< z < 7.0, while the complete sample is also analysed using the finer redshift subdivision adopted in the previous A3COSMOS study. The two AGN models provide closely consistent infrared luminosities, luminosity functions and full-sample SFRD estimates, indicating that the global dust-obscured star formation history is only weakly affected by the adopted AGN prescription. The recovered evolution shows a high SFRD around Cosmic Noon followed by a decline towards higher redshift, broadly consistent with previous infrared and millimetre determinations. The role of AGN-hosting galaxies is further investigated by selecting sources with f_AGN>=0.2 and f_AGN>=0.5, where f_AGN is defined over the rest-frame 5-40um interval. Galaxies with moderate AGN contributions account for a substantial fraction of the infrared-derived SFRD, whereas the contribution associated with the most AGN-dominated systems is smaller and significantly more model dependent. This dependence becomes particularly strong at high redshift, where the two AGN prescriptions select substantially different high-AGN populations. These results show that the total dust-obscured SFRD is relatively robust against the adopted AGN model, while its decomposition according to AGN strength remains a major source of systematic uncertainty.

Understanding the evolution of the cosmic star formation rate density (SFRD) is a central goal of observational galaxy evolution. A major uncertainty in its reconstruction arises from dust-obscured star formation, which can be strongly underestimated by ultraviolet-selected surveys, and from the possible contribution of active galactic nuclei (AGNs) to the infrared emission used as a tracer of star formation. This thesis investigates these effects by revisiting the infrared properties of ALMA-detected galaxies in the COSMOS field using updated multiwavelength information, including recent JWST-based COSMOS2025 data. The analysis starts from the A3COSMOS catalogue and adopts a blind-like selection designed to minimize the biases associated with targeted ALMA observations. Updated photometry and redshift information are incorporated, giving priority to spectroscopic measurements, followed by COSMOS2025 photometric redshifts and, when necessary, photometric redshifts derived through CIGALE. The final analysis is performed on a sample of 189 galaxies. Their physical properties are derived through panchromatic spectral energy distribution fitting with CIGALE, including stellar emission, dust attenuation and re-emission, and an AGN component. Two different AGN prescriptions, skirtor2016 and fritz2006, are independently employed in order to quantify the dependence of the inferred quantities on the adopted torus model. The resulting total infrared luminosities are used to reconstruct the infrared luminosity functions through the 1/V_max method and, after integration, to estimate the dust-obscured SFRD. The analysis is carried out in two broad redshift intervals, 0.5=< z<3.0 and 3.0 =< z < 7.0, while the complete sample is also analysed using the finer redshift subdivision adopted in the previous A3COSMOS study. The two AGN models provide closely consistent infrared luminosities, luminosity functions and full-sample SFRD estimates, indicating that the global dust-obscured star formation history is only weakly affected by the adopted AGN prescription. The recovered evolution shows a high SFRD around Cosmic Noon followed by a decline towards higher redshift, broadly consistent with previous infrared and millimetre determinations. The role of AGN-hosting galaxies is further investigated by selecting sources with f_AGN>=0.2 and f_AGN>=0.5, where f_AGN is defined over the rest-frame 5-40um interval. Galaxies with moderate AGN contributions account for a substantial fraction of the infrared-derived SFRD, whereas the contribution associated with the most AGN-dominated systems is smaller and significantly more model dependent. This dependence becomes particularly strong at high redshift, where the two AGN prescriptions select substantially different high-AGN populations. These results show that the total dust-obscured SFRD is relatively robust against the adopted AGN model, while its decomposition according to AGN strength remains a major source of systematic uncertainty.

Constraining the AGN contribution to the cosmic star formation rate density up to z~7: the ALMA and JWST synergy

SALION, FRANCESCO MARIA
2025/2026

Abstract

Understanding the evolution of the cosmic star formation rate density (SFRD) is a central goal of observational galaxy evolution. A major uncertainty in its reconstruction arises from dust-obscured star formation, which can be strongly underestimated by ultraviolet-selected surveys, and from the possible contribution of active galactic nuclei (AGNs) to the infrared emission used as a tracer of star formation. This thesis investigates these effects by revisiting the infrared properties of ALMA-detected galaxies in the COSMOS field using updated multiwavelength information, including recent JWST-based COSMOS2025 data. The analysis starts from the A3COSMOS catalogue and adopts a blind-like selection designed to minimize the biases associated with targeted ALMA observations. Updated photometry and redshift information are incorporated, giving priority to spectroscopic measurements, followed by COSMOS2025 photometric redshifts and, when necessary, photometric redshifts derived through CIGALE. The final analysis is performed on a sample of 189 galaxies. Their physical properties are derived through panchromatic spectral energy distribution fitting with CIGALE, including stellar emission, dust attenuation and re-emission, and an AGN component. Two different AGN prescriptions, skirtor2016 and fritz2006, are independently employed in order to quantify the dependence of the inferred quantities on the adopted torus model. The resulting total infrared luminosities are used to reconstruct the infrared luminosity functions through the 1/V_max method and, after integration, to estimate the dust-obscured SFRD. The analysis is carried out in two broad redshift intervals, 0.5=< z<3.0 and 3.0 =< z < 7.0, while the complete sample is also analysed using the finer redshift subdivision adopted in the previous A3COSMOS study. The two AGN models provide closely consistent infrared luminosities, luminosity functions and full-sample SFRD estimates, indicating that the global dust-obscured star formation history is only weakly affected by the adopted AGN prescription. The recovered evolution shows a high SFRD around Cosmic Noon followed by a decline towards higher redshift, broadly consistent with previous infrared and millimetre determinations. The role of AGN-hosting galaxies is further investigated by selecting sources with f_AGN>=0.2 and f_AGN>=0.5, where f_AGN is defined over the rest-frame 5-40um interval. Galaxies with moderate AGN contributions account for a substantial fraction of the infrared-derived SFRD, whereas the contribution associated with the most AGN-dominated systems is smaller and significantly more model dependent. This dependence becomes particularly strong at high redshift, where the two AGN prescriptions select substantially different high-AGN populations. These results show that the total dust-obscured SFRD is relatively robust against the adopted AGN model, while its decomposition according to AGN strength remains a major source of systematic uncertainty.
2025
Constraining the AGN contribution to the cosmic star formation rate density up to z~7: the ALMA and JWST synergy
Understanding the evolution of the cosmic star formation rate density (SFRD) is a central goal of observational galaxy evolution. A major uncertainty in its reconstruction arises from dust-obscured star formation, which can be strongly underestimated by ultraviolet-selected surveys, and from the possible contribution of active galactic nuclei (AGNs) to the infrared emission used as a tracer of star formation. This thesis investigates these effects by revisiting the infrared properties of ALMA-detected galaxies in the COSMOS field using updated multiwavelength information, including recent JWST-based COSMOS2025 data. The analysis starts from the A3COSMOS catalogue and adopts a blind-like selection designed to minimize the biases associated with targeted ALMA observations. Updated photometry and redshift information are incorporated, giving priority to spectroscopic measurements, followed by COSMOS2025 photometric redshifts and, when necessary, photometric redshifts derived through CIGALE. The final analysis is performed on a sample of 189 galaxies. Their physical properties are derived through panchromatic spectral energy distribution fitting with CIGALE, including stellar emission, dust attenuation and re-emission, and an AGN component. Two different AGN prescriptions, skirtor2016 and fritz2006, are independently employed in order to quantify the dependence of the inferred quantities on the adopted torus model. The resulting total infrared luminosities are used to reconstruct the infrared luminosity functions through the 1/V_max method and, after integration, to estimate the dust-obscured SFRD. The analysis is carried out in two broad redshift intervals, 0.5=< z<3.0 and 3.0 =< z < 7.0, while the complete sample is also analysed using the finer redshift subdivision adopted in the previous A3COSMOS study. The two AGN models provide closely consistent infrared luminosities, luminosity functions and full-sample SFRD estimates, indicating that the global dust-obscured star formation history is only weakly affected by the adopted AGN prescription. The recovered evolution shows a high SFRD around Cosmic Noon followed by a decline towards higher redshift, broadly consistent with previous infrared and millimetre determinations. The role of AGN-hosting galaxies is further investigated by selecting sources with f_AGN>=0.2 and f_AGN>=0.5, where f_AGN is defined over the rest-frame 5-40um interval. Galaxies with moderate AGN contributions account for a substantial fraction of the infrared-derived SFRD, whereas the contribution associated with the most AGN-dominated systems is smaller and significantly more model dependent. This dependence becomes particularly strong at high redshift, where the two AGN prescriptions select substantially different high-AGN populations. These results show that the total dust-obscured SFRD is relatively robust against the adopted AGN model, while its decomposition according to AGN strength remains a major source of systematic uncertainty.
SFR density
AGN
JWST
ALMA
Infrared emission
File in questo prodotto:
File Dimensione Formato  
Salion_FrancescoMaria.pdf

accesso aperto

Dimensione 17.18 MB
Formato Adobe PDF
17.18 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/114562