Euclid data release 1 (DR1) - SIR spectroscopic calibration

The 30-second take

• Dedicated SIR calibration pipelines use special pointings and fields because the science exposures are too entangled to self-calibrate.
• The focal plane is modeled as one system so positions and wavelengths stay continuous across the field of view.
• Survey users of Euclid DR1 inherit more than 170 million validated red- and blue-grism spectra whose accuracy tracks this calibration.

What the paper actually did

The Euclid mission maps the large-scale structure of the Universe to study dark energy and dark matter. The spectra that feed that map come from the Near-Infrared Spectrometer and Photometer and are reduced by the SIR processing function, the spectral near-infrared pipeline. Slitless spectra are too intrinsically tangled for the science pipeline to calibrate them from the science exposures themselves. Separate SIR calibration pipelines use specific pointing sequences and observation fields. Their products are what keep the reduced spectra inside requirements for both wavelength and flux. This paper describes the calibration pipelines that produced the products used by the science pipelines for Data Release 1. A central difficulty is that the entire focal plane has to be modeled as one entity. That global model is what keeps astrometry and spectroscopy continuous across the field of view. DR1 contains more than 170 million validated spectra in both the red-grism and the blue-grism modes. The abstract says the calibration products and the resulting science spectra meet mission requirements, and that specific areas for later optimization have already been identified.

What makes this disruptive

A slitless spectrum overlaps neighbors, so the usual hope that the survey calibrates itself does not hold. Euclid's answer is a dedicated calibration program and a focal-plane model that is global rather than detector-by-detector. Continuity of position and wavelength across the field is the difference between 170 million spectra that can be stacked into a map and 170 million spectra with a seam every detector. The paper's job is to document the pipelines that make the release usable.

Why it matters (outside the lab)

Cosmology from Euclid is only as good as the wavelength and flux scales on those spectra. Anyone using DR1 to trace large-scale structure is using this calibration whether they open the calibration paper or not. The statement that current outputs meet mission requirements is the operational claim. The list of optimizations left for future releases is the warning that the scale is not finished.

Limitations & open questions

The abstract says requirements are met and does not quote the residual wavelength or flux errors. Areas for optimization are identified and not enumerated here. This paper describes calibration products. It does not present a dark-energy or dark-matter measurement. 'More than 170 million' is the count of validated spectra, not a count of galaxies with finished redshifts.