From 756864a2b3a9d5cb558e9834aa5d39d0078fe621 Mon Sep 17 00:00:00 2001 From: Alexandre Boucaud Date: Thu, 1 Oct 2026 01:42:39 +0200 Subject: [PATCH] feat(DES): add Y band; defer u band pending data-sourcing decision Add the Y filter to DES.yaml, sourced from arXiv:1801.03181 (same paper as the existing g/r/i/z): sky_brightness=17.96 and psf_fwhm=0.90 from Table 1, full_exposure_time=450s (10 x 45s single-epoch exposures, Section 2.2) -- note this is *not* 900s like g/r/i/z: the paper's footnote 9 states Y exposure time was only raised to 90s (keeping the same 450s cumulative target) starting in Y4, after the Y1-Y3 data this DR1 paper describes. zeropoint (25.56) and effective_wavelength (989.297) are computed with speclite (decam2014-Y), reproduced exactly via scripts/check_zeropoints.py / check_effective_wavelengths.py. u band was intentionally NOT added. Research into arXiv:1801.03181 and arXiv:2012.12824 (Hartley et al. 2022, DES Y3 Deep Fields) shows DES's u-band imaging is not part of the homogeneous wide-field survey that the other filters in this file represent: it comes from ~15 separate community-time programs restricted to the 10 SN deep pencil-beam fields + COSMOS (~30 sq deg total, vs. ~5000 sq deg for the wide survey), with heterogeneous per-field cumulative exposure times (~3,300s-45,380s) and no standardized single-epoch cadence (Table A1's "SE_TRUTH" row is blank for u in every field). There is no single value that would legitimately represent "DES u" the way full_exposure_time/ sky_brightness/psf_fwhm represent g/r/i/z/Y, so fabricating one would repeat the kind of unverified-assumption mistake this task was meant to avoid. Flagging for a follow-up decision rather than guessing. Co-Authored-By: Claude Sonnet 5 --- surveycodex/data/DES.yaml | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/surveycodex/data/DES.yaml b/surveycodex/data/DES.yaml index 3e0b021..b01e956 100644 --- a/surveycodex/data/DES.yaml +++ b/surveycodex/data/DES.yaml @@ -34,6 +34,13 @@ filters: zeropoint: 26.58 psf_fwhm: 0.84 effective_wavelength: 919.646 + Y: + name: "Y" + sky_brightness: 17.96 + full_exposure_time: 450 + zeropoint: 25.56 + psf_fwhm: 0.90 + effective_wavelength: 989.297 references: pixel_scale: link: "http://www.ctio.noirlab.edu/noao/content/DECam-What" @@ -55,7 +62,7 @@ references: comment: "(v3) See table 1" full_exposure_time: link: "https://arxiv.org/abs/1801.03181" - comment: "(v3) See section 2.2 page 5. Coadds of 10 x 90s exposures" + comment: "(v3) See section 2.2 page 5. Coadds of 10 x 90s exposures for g/r/i/z. Y-band used 10 x 45s exposures (450s total) during Y1-Y3 (the data release covered by this paper); footnote 9 notes Y exposure time was raised to 90s from Y4 onward while keeping the same 450s cumulative target -- so the 900s value used for g/r/i/z is not applicable to Y" zeropoint: link: "https://speclite.readthedocs.io/en/latest/filters.html#decam-2014-filters" comment: "computed with speclite – see scripts/check_zeropoints.py"