AuxTel: from spectra with a little telescope to cosmology with LSST
Abstract
Cosmology has experienced a great development during the last decades, being today a precision science with the arrival of large cosmological surveys, such as DES, DESI and Euclid. In the coming years, the Large Survey of Space and Time (LSST) will start to carry out a massive survey which will provide data with statistical errors below the systematic uncertainty due to photometric calibration. Given this increase of statistical precision, systematic effects are becoming the main source of uncertainty. In this regard, measuring colours and calibrating them is a crucial task for current and future photometric cosmological surveys in their way to obtain reliable cosmological measurements. This is particularly important for obtaining cosmological information from supernovae and to compute photometric redshifts. One of the main sources of photometric uncertainty is associated with atmospheric transmission. To deal with atmospheric effects, LSST counts with a companion telescope, the Auxiliary Telescope (AuxTel), whose purpose is to measure the atmospheric transparency and to derive colour corrections based on spectroscopic observations. In this contribution we will go through the latest improvements carried out on AuxTel during September 2023, the methodologies that we are developing to properly extract spectra and the method that we are proposing to derive colour compensations based on spectroscopic observations with AuxTel to be applied to LSST’s photometry.
Full text
AuxTel: from spectra with a little telescope to cosmology with LSST 18th Iberian Cosmology meeting Salamanca, 27-03-2024 Martín Rodríguez Monroy - IFT, Madrid, Spain
● LSST ● Systematic effects in photometry ○ Instrumental systematics ○ Atmospheric systematics ● LSST’s auxiliary telescope: AuxTel ○ Purpose ○ Ongoing work with AuxTel ● Summary Outline 2
The Legacy Survey for Space and Time…. 3
Legacy Survey of Space and Time (LSST) ● Optical telescope: 8.4 m diameter ● Wide-field camera: 3.5 deg, 3.2 Gpix ● 6 broad band filters: ugrizY (320 - 1050nm) ● 10 year coadded limit magnitude: u = 25.6; g = 26.9; r = 26.9; i = 26.4; z = 25.6; Y = 24.8 ● Photometric calibration: ○﹤5 mmag repeatability and colours ○﹤10 mmag absolute ● Final catalogue: ○1010 galaxies ○1010 stars ○ Prospect of O(10⁵) supernovae Jacqueline Ramseyer Orrell/SLAC National Accelerator Laboratory 4
Bianco et al., 2021 Legacy Survey of Space and Time (LSST) Using the current baseline cadence, LSST ten-year survey will take more than five million exposures, collecting over 50 petabytes of raw image data to produce a deep, time-dependent, movie of about 20,000 square degrees of sky → Pushing the limits of statistical error! 5
Systematic effects in photometry…. 6
Statistical vs systematic uncertainties 7 LSST observations: ● Large number densities ● Large number of SNe ●Systematic errors main source of uncertainty ●Can lead to important biases → small statistical error Credit: Thierry Souverin, LPNHE ●For SNe cosmology, photometric calibration will be the main source of uncertainty: ○Instrumental calibration (telescope’s throughput) ○Atmospheric transmission
We wish to recover the information (magnitudes, colours, etc) of the objects (stars, galaxies, etc) as seen from top of atmosphere / through a standard atmosphere ● LSST measures broadband fluxes over a range of frequencies filters: ugrizY ● From magnitudes we define colours: u-g, g-r, r-i, etc…. Spectral energy distribution (SED) Flux in # of photons: Systematics in photometry 8
The amount of signal at each pixel (ADUs) depends on the different “obstacles” on optical path: SED x atm. x mirrors x filter x disperser x entrance window x εCCD x analysis = ADUpassband (Disperser present only in AuxTel) Systematics in photometry Telescope’s throughput 9
LSST’s Auxiliary Telescope: AuxTel ● Auxiliary telescope: 1.2m, f/18 ● Equipped with a spectrograph ● Provides spectra in real time to compare them with space observations (calibration objects) ●Hologram installed in February 2021 to provide improved spectral resolution ●Currently providing the only real data from LSST! More than 3000 reconstructed spectra since september 2022 (and reprocessed according to analysis progress) We use AuxTel to measure the atmosphere and compensate its impact: ●Measure atmospheric components, e.g. PWV, O3, aerosols, that impact the colors that LSST will measure ●Real time atmosphere transmission x LSST throughput →Derive color compensations for each object observed by LSST LATISS camera 16
Spectroscopy with AuxTel = slitless spectroscopy: ● Pros: ○ Several objects at a time ○Fast pointing (no need of great positioning precision) ○Can do spectro-photometry ● Cons: ○ Spectral resolution limited by seeing ○ Sky and stellar background ● Initially, it was planned to use a Ronchi grating ○Problems: should be used with parallel beam + focus depends on λ Square wave From the instruments to the measurements 17
Idea: use a holographic optical element, i.e. hologram, which keeps spatial information Advantages of a hologram: ● Works with a convergent beam ●Each λ is focused near focal plane Moniez et al. (https://doi.org/10.1093/mnras/stab2109) From the instruments to the measurements 18
From the instruments to the measurements Latest technical updates: recent work on Cerro Pachón to install, validate and characterize several new devices: ● Installation and characterization of mask to reduce sky / stellar contaminations 0th order of parasitic field stars Cleaner spectra 19
From the measurements to the atmospheric parameters We characterize the atmosphere by taking spectra: ● CALSPEC standard stars ● Different airmasses and atm. conditions The spectra are extracted with the Spectractor software. It provides: ● Data = spectrum ● Covariance The modeling is given by: ●CALSPEC spectrum ●Throughput ●LibRadtran model of atm Fit atmospheric parameters Spectrum extraction Water absorption Atm. emulator Neveu et al. (arXiv:2307.04898) 20
From atmospheric parameters to color corrections Derive photometric corrections for LSST for varying atmospheric conditions. Different options: ●Baseline: measure atmospheric parameters (clouds, PWV, 𝜏VAOD, O3) and derive model of the atmosphere (if AuxTel points far away from LSST during the night) ● Directly measure atmospheric transparency if AuxTel points to the same field as LSST ● Also, possible prior to other methods (FGCM) 21 Impact of PWV on the colors of different objects characterized by ugrizY
LSST AuxTel spec. obs flats Spectractor spec-flats pixel / spectro flats Atm. Emulator extracted spectrum + sim. atm. transparencies Color correction method Role of AuxTel on LSST Also possible to do photometry with AuxTel (with available filters) Sylvie Dagoret-Campagne: https://github.com/LSSTDESC/getObsAtmo 22
Color correction method: basics Current atm. (Zt,PWVt, τt) Standard atm. (Zstd,PWVstd, τstd) LSST AuxTel AuxTel ⇒ real time atm. transparency Allows to go back to top of atmosphere (TOA) ● Known SED ● Partially known or unknown ⇒ modeling + 23
Summary…. 24
Summary ●Systematic uncertainties are becoming dominant ●This will require very fine calibrations and corrections ●AuxTel is an essential asset for the quality of LSST’s data ○ Working on the measurement of atmospheric parameters ○ Working on new methods to flat-field spectroscopic exposures ○ Developing color compensation method for LSST with AuxTel observations ● Sometimes, the keys to make great progress can be rather simple things, such as a little telescope or even a piece of plastic 25 S. Perlmutter et al 1999 ApJ 517 565