Radiometric error and re-calibration of the MGS TES spectra

Abstract

Several sources of systematic error were identified in the spectra of the Thermal Emission Spectrometer (TES) onboard the Mars Global Surveyor (MGS) spacecraft during its mission. Some of these errors were corrected, some still remain and contaminate spectra. One of the most significant remaining errors is a time-variable systematic radiometric error. This error significantly affects nighttime and polar spectra, and spectra of the Mars’ limb. The existence of this error hampered analysis of roughly half of the data collected by TES spectrometer. The error arises due to a periodic sampling error of TES interferograms, which is a common problem in Fourier-transform interferometers. The error negatively affects calibrated TES spectra in two ways: it introduces an error into estimates of the Instrument Response Functions (IRF) and instrument’s radiances that are used to calibrate TES spectra, and it introduces an error into TES spectra themselves.This paper presents a new approach to calibrating TES spectra that enables removing the error from the calibration functions. The new approach utilizes long-term averages of uncalibrated TES spectra of deep space to estimate the true shape of the TES IRF and its dependence on instrument temperature. This, and parameterization of the radiometric error spectral shape, enables removing the error from calibration. Examples of re-calibrated spectra are presented. The largest improvement in the quality of the spectra is observed for nighttime and polar spectra, and spectra of the Mars’ limb. Re-calibration would significantly improve retrievals of aerosol abundances and surface temperatures from these spectra.