Periodic Reporting for period 3 - EWC (Enabling Weak lensing Cosmology)
Berichtszeitraum: 2020-04-01 bis 2022-03-31
Enabling Weak lensing Cosmology (EWC) was a team of cosmologists working across Europe to prepare for the next generation of astronomical data from the European Space AgencyEuclid mission. The consortium closed on 31st March 2022, when the EU H2020 funding ended.
Euclid will map three-quarters of the extra-galactic sky back in time over three-quarters the age of the Universe. It’s science objective is to determine the nature of dark energy – that is causing the expansion of the Universe to accelerate – by mapping how dark matter grows over time. One of the primary ways that Euclid will do this is by measuring an effect called weak lensing: the effect where dark matter structure distort the observed images of galaxies. Euclid will measure this signal by imaging 1.5 billion galaxies with a resolution similar to that of the Hubble Space Telescope.
Although Euclid is designed to minimize observational systematic effects the observations are still limited by two factors. Various instrumental effects need to be corrected for, and the tremendous improvement in precision has to be matched with comparable advances in the modelling of astrophysical effects that affect the signal. The objective of our team was to make significant progress on both fronts. To do so, we (i) quantified the morphology of galaxies using archival HST observations; (ii) carried out a unique narrow-band photometric redshift survey to obtain state-of-the-art constraints on the intrinsic alignments of galaxies that arise due to tidal interactions, and would otherwise contaminate the cosmological signal; (iii) integrated these results into the end-to-end simulation pipeline; (iv) performed a spectroscopic redshift survey to calibrate the photometric redshift technique. The Euclid Consortium identified these as critical issues, which needed to be addressed before launch, in order to maximise the science return of this exciting mission, and enable the dark energy science objectives of Europe
Euclid will map three-quarters of the extra-galactic sky back in time over three-quarters the age of the Universe. It’s science objective is to determine the nature of dark energy – that is causing the expansion of the Universe to accelerate – by mapping how dark matter grows over time. One of the primary ways that Euclid will do this is by measuring an effect called weak lensing: the effect where dark matter structure distort the observed images of galaxies. Euclid will measure this signal by imaging 1.5 billion galaxies with a resolution similar to that of the Hubble Space Telescope.
Although Euclid is designed to minimize observational systematic effects the observations are still limited by two factors. Various instrumental effects need to be corrected for, and the tremendous improvement in precision has to be matched with comparable advances in the modelling of astrophysical effects that affect the signal. The objective of our team was to make significant progress on both fronts. To do so, we (i) quantified the morphology of galaxies using archival HST observations; (ii) carried out a unique narrow-band photometric redshift survey to obtain state-of-the-art constraints on the intrinsic alignments of galaxies that arise due to tidal interactions, and would otherwise contaminate the cosmological signal; (iii) integrated these results into the end-to-end simulation pipeline; (iv) performed a spectroscopic redshift survey to calibrate the photometric redshift technique. The Euclid Consortium identified these as critical issues, which needed to be addressed before launch, in order to maximise the science return of this exciting mission, and enable the dark energy science objectives of Europe
WP1: The PAU Survey has been designed to measure accurately the redshift of galaxies using photometric techniques with narrow band filters. Its wide area coverage, the depth reached and the wavelength rage coverage make it a unique survey that can be used to calibrate the intrinsic alignment signal in weak lensing surveys like Euclid. We took observations at the William Herschel Telescope of the Observatorio del Roque de los Muchachos and automatically transfered them to our institutions and archive them. We have developed the algorithms to be able to measure the galaxy fluxes in all the narrow bands observed. We have also implemented the processing into a big data platform hosted at the Port d'Informació Científica (PIC) as the data volume makes it impossible to run in simple machines.
WP2: Photometric redshifts provide an efficient tool to locate galaxies and therefore make them available as tracers of the large-scale structure of the universe. Here, we explored how to best measure galaxy redshifts using the narrow band data collected from PAUCam. With PAUCam, we sample the spectral energy distribution at 40 different wavelengths. Such fine resolution allows us to determine the galaxy redshifts accurately. We have developed template-based photo-z algorithms as well as machine learning techniques. In particular, the inclusion of emission lines in the galaxy spectra improves considerably the performance for emission line galaxies. Our results are consistent with the prediction we obtained in simulations.
WP 3: We measured how galaxies – big clouds of stars - orientate themselves in space. In particular how the 3D shape of a galaxy can be “aligned” (i.e. point towards) big clumps of matter as a result of the effect of gravity. Its really important to measure this effect so we can account for it in the Euclid mission. In the Euclid mission we want to measure the gravitational lensing effect, that changes the observed shapes of galaxies and can be used to determine what the Universe is made of; but without removing the effect of the 3D alignments this measurement would no be possible.
WP 4: During the reporting period the simulation-based tests of the HST Euclidization procedure were expanded, revealing that standard applications for the generation of HST-based Euclid-like weak lensing image simulations can lead to incorrect calibrations of multiplicative shear measurement biases at the few per-cent level. We delivered a refined tile-wise reduction of HST/ACS observations in the CANDELS-Wide fields including complementary PSF models. Revised HST correction software for charge transfer inefficiency has been released at https://github.com/jkeger/arctic(öffnet in neuem Fenster). The routines used to calibrate the model have additionally been released at https://github.com/jkeger/warm_pixels(öffnet in neuem Fenster).
WP 5: The methodology developed and the results of the calibration procedure will be presented in two papers which are in progress at the time of writing. This will represent a fast and efficient way to get rid of the colour gradient bias respecting the requirements set on this kind of bias in the Euclid systematics budget.
WP 6: Synthetic galaxy catalogues play a key role in galaxy surveys. In this project we built a replica of the survey made with PAUCam. We did this by combining a computer model of galaxy formation, with a simulation which models the structure in the dark matter component of the Universe, and its growth over time due to the force of gravity. The output of these simulations is a series of snapshots of the population of model galaxies at fixed times in the history of the universe. We then processed these snapshots with another code to mimic the process of observing the model galaxies with PAUCam.
WP 7: In this WP, we have conducted a survey to obtain the spectroscopic data required for Euclid. As result, we have obtained the redshits for the galaxies included in our programme, improved the coverage of the colour-redshift mapping and simulated the improvement in the bias correction of the tomographic bins.
WP2: Photometric redshifts provide an efficient tool to locate galaxies and therefore make them available as tracers of the large-scale structure of the universe. Here, we explored how to best measure galaxy redshifts using the narrow band data collected from PAUCam. With PAUCam, we sample the spectral energy distribution at 40 different wavelengths. Such fine resolution allows us to determine the galaxy redshifts accurately. We have developed template-based photo-z algorithms as well as machine learning techniques. In particular, the inclusion of emission lines in the galaxy spectra improves considerably the performance for emission line galaxies. Our results are consistent with the prediction we obtained in simulations.
WP 3: We measured how galaxies – big clouds of stars - orientate themselves in space. In particular how the 3D shape of a galaxy can be “aligned” (i.e. point towards) big clumps of matter as a result of the effect of gravity. Its really important to measure this effect so we can account for it in the Euclid mission. In the Euclid mission we want to measure the gravitational lensing effect, that changes the observed shapes of galaxies and can be used to determine what the Universe is made of; but without removing the effect of the 3D alignments this measurement would no be possible.
WP 4: During the reporting period the simulation-based tests of the HST Euclidization procedure were expanded, revealing that standard applications for the generation of HST-based Euclid-like weak lensing image simulations can lead to incorrect calibrations of multiplicative shear measurement biases at the few per-cent level. We delivered a refined tile-wise reduction of HST/ACS observations in the CANDELS-Wide fields including complementary PSF models. Revised HST correction software for charge transfer inefficiency has been released at https://github.com/jkeger/arctic(öffnet in neuem Fenster). The routines used to calibrate the model have additionally been released at https://github.com/jkeger/warm_pixels(öffnet in neuem Fenster).
WP 5: The methodology developed and the results of the calibration procedure will be presented in two papers which are in progress at the time of writing. This will represent a fast and efficient way to get rid of the colour gradient bias respecting the requirements set on this kind of bias in the Euclid systematics budget.
WP 6: Synthetic galaxy catalogues play a key role in galaxy surveys. In this project we built a replica of the survey made with PAUCam. We did this by combining a computer model of galaxy formation, with a simulation which models the structure in the dark matter component of the Universe, and its growth over time due to the force of gravity. The output of these simulations is a series of snapshots of the population of model galaxies at fixed times in the history of the universe. We then processed these snapshots with another code to mimic the process of observing the model galaxies with PAUCam.
WP 7: In this WP, we have conducted a survey to obtain the spectroscopic data required for Euclid. As result, we have obtained the redshits for the galaxies included in our programme, improved the coverage of the colour-redshift mapping and simulated the improvement in the bias correction of the tomographic bins.
We have published 69 papers in refereed journals. 10 more papers are submitted to journals and at the refereeing stage. These papers have already accrued 606 citations in 4 years.
The following provide links to our public data and code repositories:
• PAUS Photo-z catalogue https://www.pausurvey.org/pauscosmos-photo-z-catalog/(öffnet in neuem Fenster)
• PAUS early data release https://www.pausurvey.org/edr/(öffnet in neuem Fenster)
• PAUS simulation data https://www.pausurvey.org/pau-millgas-mock/(öffnet in neuem Fenster)
• Galaxy-Galaxy lensing code https://github.com/KiDS-WL/KiDS-GGL(öffnet in neuem Fenster)
• Photo-z codes: https://github.com/PAU-survey/bcnz , https://github.com/PAU-survey/deepz ,
• Photometry code https://github.com/PAU-survey/lumos(öffnet in neuem Fenster)
• Background estimation code https://github.com/PAU-survey/bkgnet(öffnet in neuem Fenster)
The following provide links to our public data and code repositories:
• PAUS Photo-z catalogue https://www.pausurvey.org/pauscosmos-photo-z-catalog/(öffnet in neuem Fenster)
• PAUS early data release https://www.pausurvey.org/edr/(öffnet in neuem Fenster)
• PAUS simulation data https://www.pausurvey.org/pau-millgas-mock/(öffnet in neuem Fenster)
• Galaxy-Galaxy lensing code https://github.com/KiDS-WL/KiDS-GGL(öffnet in neuem Fenster)
• Photo-z codes: https://github.com/PAU-survey/bcnz , https://github.com/PAU-survey/deepz ,
• Photometry code https://github.com/PAU-survey/lumos(öffnet in neuem Fenster)
• Background estimation code https://github.com/PAU-survey/bkgnet(öffnet in neuem Fenster)