Euclid VI. NISP-P optical ghosts

TL;DR

This paper models and provides masking strategies for optical ghosts caused by the dichroic beamsplitter and filters in Euclid's NISP-P instrument, ensuring cleaner photometric data for the mission.

Contribution

It introduces detailed models and masking radii for optical ghosts in Euclid's NISP-P data, improving data quality for the mission's photometric analysis.

Findings

Ghost offsets are well modeled by 2D polynomials.

Masking radii depend on star brightness and wavelength.

Models are implemented in the NIR data pipeline.

Abstract

The Near-Infrared Spectrometer and Photometer (NISP) onboard Euclid includes several optical elements in its path, which introduce artefacts into the data from non-nominal light paths. To ensure uncontaminated source photometry, these artefacts must be accurately accounted for. This paper focuses on two specific optical features in NISP's photometric data (NISP-P): ghosts caused by the telescope's dichroic beamsplitter, and the bandpass filters within the NISP fore-optics. Both ghost types exhibit a characteristic morphology and are offset from the originating stars. The offsets are well modelled using 2D polynomials, with only stars brighter than approximately 10 magnitudes in each filter producing significant ghost contributions. The masking radii for these ghosts depend on both the source-star brightness and the filter wavelength, ranging from 20 to 40 pixels. We present the final relations and models used in the near-infrared (NIR) data pipeline to mask these ghosts for Euclid's Quick Data Release (Q1).