The DES and DESI projects

Ramon Miquel

Since 2005, a group at IFAE, together with a group at ICE (Institut de Ciències de l’Espai), and another at CIEMAT (Centro de Investigaciones Energéticas, Medio Ambientales y Tecnológicas) and Universidad Autónoma de Madrid (UAM), collaborates in the DES (Dark Energy Survey) international project, led by Fermilab (USA) and, since 2015, in the development of DESI (Dark Energy Spectroscopic Instrument), led by LBNL (USA). In late 2016 the group joined the Large Survey Synoptic Telescope (LSST), led by SLAC (USA).


The Dark Energy Survey started in 2013 and is the foremost photometric galaxy survey in this decade, having, in 2019, finished imaging 5000 sq. deg. (an octant of the sky) in five optical and near-infrared bands (grizY) to unprecedented depths (iAB ~ 24), and measuring the position on the sky, distance and shape of almost 300 million galaxies up to a redshift z ~ 1.4. Starting in summer 2020, the Dark Energy Spectroscopic Instrument (DESI) will collect spectra for over 30 million galaxies and quasars covering the whole redshift range 0.2 < z < 3.5, and will become the foremost spectroscopic galaxy survey in the new decade. IFAE is a founding member of both collaborations. In DES, IFAE designed and produced a large fraction of the readout electronics of the DES camera, while in DESI, IFAE designed and produced the 10 (+2 spares) GFA cameras necessary for the guiding, focusing and alignment of the 5000 fibers in the DESI focal plane. Both surveys have as their main goal to unveil the nature of the mysterious dark-energy component of the universe that powers its current accelerated expansion.

The DESI Instrument sees first light

In 2019, the DESI instrument saw its first light, including the GFA cameras designed and produced at IFAE.

During early 2019, the IFAE group finished the delivery of all the GFA cameras, complete with mechanical enclosures, filters, CCDs, readout electronics, thermal control, etc. Ten of them were then mounted in the focal plane of the DESI instrument. The commissioning of the instrument started in late summer 2019, with first light observed on October 22, 2019. Figure 1 shows a spectrum observed on that date, while Fig. 2 shows typical images of a guiding star on- and off-focus as observed with the ten GFA cameras.
Figure 1: A spectrum of the Triangulum Galaxy (M33) as taken by the DESI instrument on October 22, 2019 (first light).

Figure 2: The 10 GFAs observing a guiding star. In the middle, the on-focus images provided by the six guiders. Above and below, the ”donut” images provided by the four focusing cameras, each of which provides two images: one ahead of focus and one behind.

DES cosmological analyses in 2019

The bulk of the DES papers containing the main cosmological results from the first year of observations (Y1) were published in 2017. The next major step will be the publication of the results of the cosmological analyses of the first three years of observations (Y3), now expected in 2020. In 2019, DES published two analyses led by IFAE researchers, one using Y1 data and another in preparation of the Y3 analyses.

In 2019, the group at IFAE led a DES paper involving cross-correlations between DES and CMB measurements and another concerning cosmological constraints from the moments of the (mostly dark) matter distribution.

Two PhD students and a post-doctoral researcher at IFAE led the analysis of the correlation between the position of DES super-voids in Y1 and the lensing convergence field measured in the Planck maps of the CMB. The imprint of the DES super-voids in the Planck convergence map is detected at the 5-σ level, a first. This analysis was motivated by an earlier finding by the same IFAE post-doc that the imprint of DES super-voids in the CMB temperature was not well described by simulations based on the prevailing ΛCDM cosmological model. In contrast, their imprint in the convergence field is well reproduced by ΛCDM, hence adding to the mystery. The results are reported in P. Vielzeuf, A. Kovács, U. Demirbozan et al. (DES Collaboration) 2019, “Dark Energy Survey Year 1 Results: the Lensing Imprint of Cosmic Voids on the Cosmic Microwave Background”, arXiv:1911.02951 [astro-ph.CO]. Figure 3 shows the stacked lensing maps in and around the DES super-voids (right panel) compared to the ΛCDM expectations (center panel).
Figure 2: Stacked CMB lensing convergence (κ) maps in and around super-voids in the MICE simulation (left without κ noise and center with κ noise) and in the observed Planck-DES-Y1 sample (right). Dashed circles mark the (re-scaled) void radius. Taken from P. Vielzeuf, A. Kovács, U. Demirbozan et al. (DES Collaboration) 2019, “Dark Energy Survey Year 1 Results: the Lensing Imprint of Cosmic Voids on the Cosmic Microwave Background,” arXiv:1911.02951 arXiv:1911.02951 [astro-ph.CO]. In this case, there is good agreement between data and simulations.

Another PhD student at IFAE led the preparatory phase for the cosmological analysis of the moments of the matter distribution in Y3, something that was not attempted in Y1. The resulting paper (M. Gatti et al. (DES Collaboration) 2019, “Dark Energy Survey Year 3 Results: Cosmology with Moments of Weak Lensing Mass Maps – Validation on Simulations”, arXiv:1911.05568 []) presents the analysis strategy, the computation of the theoretical predictions, the validation of the method in simulations, and the science reach, which turns out to be very competitive with respect to the standard analysis based on the 2-point correlation function. Furthermore, the moments analysis is able to capture non-gaussianities in the matter distribution that the 2-point function is insensitive to.