Calibration of Herschel SPIRE FTS observations at different spectral resolutions

TL;DR

This paper investigates the calibration discrepancies between high and low spectral resolution modes of the Herschel SPIRE FTS, identifying the cause as temperature-induced response variations and proposing an empirical correction method.

Contribution

It reveals the origin of calibration differences between spectral resolutions and provides an empirical correction to improve data consistency.

Findings

Systematic continuum level discrepancy between HR and LR modes.

Variation in telescope and instrument response functions causes calibration differences.

An empirical correction effectively mitigates the response variation effects.

Abstract

The SPIRE Fourier Transform Spectrometer on board the Herschel Space Observatory had two standard spectral resolution modes for science observations: high resolution (HR) and low resolution (LR), which could also be performed in sequence (H+LR). A comparison of the HR and LR resolution spectra taken in this sequential mode, revealed a systematic discrepancy in the continuum level. Analysing the data at different stages during standard pipeline processing, demonstrates the telescope and instrument emission affect HR and H+LR observations in a systematically different way. The origin of this difference is found to lie in the variation of both the telescope and instrument response functions, while it is triggered by fast variation of the instrument temperatures. As it is not possible to trace the evolution of the response functions through auxiliary housekeeping parameters, the calibration cannot be corrected analytically. Therefore an empirical correction for LR spectra has been developed, which removes the systematic noise introduced by the variation of the response functions.