Letters of Intent received in 2019

LoI 2021-2083
GA Focus Meeting: Photometric Redshifts in the era of all sky panchromatic astronomy

Topics

Photometric redshifts via SED fitting and machine learning
Calibrating and validating redshift distributions in cosmology
Utilising multi-wavelength data in redshift estimation
Photometric redshifts for AGN, independent of selection method.
Computing physical galaxy quantities (e.g. masses, star formation rates, AGN contribution)

Rationale

Deep imaging surveys, spanning X-ray (e.g. eROSITA), optical (e.g. HSC, DES, KiDS), NIR/IR (VISTA, WISE), sub-mm (SPT) and radio wavelengths (e.g. Apertif, ASKAP, JVLA, LOFAR, MeerKAT), have expanded to cover much of the sky and will shortly be joined by the next generation of dark energy experiments: Euclid and LSST. With activity across all continents, these surveys seek to answer a wide breadth of questions in cosmology and extragalactic astrophysics but share a common challenge in deriving galaxy and AGN redshifts and redshift distributions from photometry. These galaxies and AGNs will be used to understand the expansion history of our Universe, reionisation, large-scale structure, the cosmic star formation history and AGN feedback but are far too numerous and faint for spectroscopy, leaving photometric redshifts as a critical aspect of experimental design across multiple subfields.

By the time of the IAU in Busan, the analysis of the third-generation of optical / near-IR weak lensing experiments (HSC, DES, KiDS) will be nearing completion, while the fourth generation (Euclid, LSST) will be on the cusp of main survey data collection. Meanwhile, eROSITA will have already observed the entire sky four times while LOFAR and ASKAP will have surveyed much of the radio sky, providing a map of AGN and clusters at a depth never available before. This focus meeting is perfectly timed to distill the lessons learned from the third generation of dark energy experiments and to prepare for the imminent analysis of the enormous new datasets that will be produced in the following years, for which a factor of five improvement in redshift accuracy over current analyses will be required.

A particular challenge for cosmology with imaging surveys is that weak gravitational lensing by cosmic structure can only be used to test models of dark energy and modified gravity if the redshift distributions of galaxies is precisely known. For multi-wavelength imaging surveys the challenge is to effectively measure photometric redshifts for complex sources with spectral energy distributions including contributions from starbursts, dust and AGNs. As photometric redshifts are uncertain (and often non-Gaussian), there are also challenges when determining subsequent galaxy masses, star formation rates, dust masses and AGN contributions to the spectral energy distribution.

Furthermore, the forthcoming experiments constitute a genuine “big data” challenge. New methods and approaches are being developed for photo-z and physical parameter estimation, including the use of advanced machine learning techniques, with the aim of meeting the requirement that results are not only sufficiently accurate, but also able to run efficiently at scale. The community will benefit greatly from bringing together the different teams working on this fast-moving subject area, the fruits of which will be used for a plethora of legacy science for many years to come.

We seek to engage with the broad and international subsection of the astronomical community that will need photometric redshifts for their science during the 2020s. The meeting will also build upon the success of smaller photometric redshift workshops during the past 5 years, that have accelerated the development of new photometric redshift techniques and new collaborations.