Euclid Near Infrared Spectro-Photometer: spatial considerations on H2RG detectors interpixel capacitance and IPC corrected conversion gain from on-ground characterization

TL;DR

This paper presents a detailed on-ground characterization of Euclid's NISP H2RG detectors, focusing on spatial variations in interpixel capacitance and conversion gain, and introduces new correction methods for improved performance analysis.

Contribution

It introduces a novel IPC correction method and analyzes spatial variations in IPC and conversion gain for Euclid's H2RG detectors.

Findings

Per pixel IPC coefficient maps derived from SPR measurements.

Validated new IPC correction method.

Identified spatial variations in conversion gain across detectors.

Abstract

Euclid is a major ESA mission scheduled for launch in 2023-2024 to map the geometry of the dark Universe using two primary probes, weak gravitational lensing and galaxy clustering. \Euclid's instruments, a visible imager (VIS) and an infrared spectrometer and photometer (NISP) have both been designed and built by Euclid Consortium teams. The NISP instrument will hold a large focal plane array of 16 near-infrared H2RG detectors, which are key elements to the performance of the NISP, and therefore to the science return of the mission. Euclid NISP H2RG flight detectors have been individually and thoroughly characterized at Centre de Physique des Particules de Marseille (CPPM) during a whole year with a view to producing a reference database of performance pixel maps. Analyses have been ongoing and have shown the relevance of taking into account spatial variations in deriving performance parameters. This paper will concentrate on interpixel capacitance (IPC) and conversion gain. First, per pixel IPC coefficient maps will be derived thanks to single pixel reset (SPR) measurements and a new IPC correction method will be defined and validated. Then, the paper will look into correlation effects of IPC and their impact on the derivation of per super-pixel IPC-free conversion gain maps. Eventually, several conversion gain values will be defined over clearly distinguishable regions.