In-flight radiometric calibration of the ExoMars TGO Colour and Stereo Surface Imaging System

Pommerol, Antoine; Thomas, Nicolas; Miguel, Almeida; Read, M.; Becerra, P.; Cesar, C.; Valantinas, A.; Simioni, E.; McEwen, A.S.; Perry, J.; Marriner, C.; Munaretto, G.; Pajola, M.; Tornabene, L.; Mège, D.; Da Deppo, V.; Re, C.; Cremonese, G. (2022). In-flight radiometric calibration of the ExoMars TGO Colour and Stereo Surface Imaging System. Planetary and space science, 223, p. 105580. Elsevier 10.1016/j.pss.2022.105580

1-s2.0-S0032063322001660-main.pdf - Published Version
Available under License Creative Commons: Attribution (CC-BY).

Download (6MB) | Preview

The Colour and Stereo Surface Science Imaging System (CaSSIS) of the ExoMars Trace Gas Orbiter returns on
average twenty images per day of the Martian surface, most of them in 3 or 4 colours and some of them in stereo.
CaSSIS uses a push-frame approach to acquire colour images, with four bandpass filters deposited directly above
the sensor and an imaging cadence synchronized with the ground track velocity to cover the imaged area with
tens of small, partially overlapping images. These “framelets” are later map-projected and mosaicked to build the
final image. This approach offers both advantages and challenges in terms of radiometric calibration. While the
collection of dark and flatfield frames is considerably enhanced by the frequent and fast acquisition of tens of
successive images, mosaics assembled from the adjacent framelets highlight the straylight and changes in the bias
of the detector. Both issues have been identified on CaSSIS images, with low intensities overall (up to a few %),
but sufficient to generate prominent artefacts on the final assembled colour images. We have therefore developed
methods to correct these artefacts that are now included into the radiometric calibration pipeline. We detail here
the different steps of the calibration procedure and the generation of the products used for calibration, and discuss
the efficacy of the corrections. The relative uncertainties on the bias and flatfield frames are low, of the order of
0.2 and 0.1%, respectively. The uncertainty on the absolute radiometric calibration is of 3%, which is quite low
for such an instrument. The straylight adds an estimated ~1% error to the absolute calibration. The residuals after
corrections of the straylight and bias offsets are of the order of a few DNs to tens of DNs. As CaSSIS can observe the
Martian surface in challenging illumination conditions to provide unique views of the surface at early and late
local solar time, residuals from the straylight correction can become noticeable when the absolute signal is very
low. As they appear at the level of the noise in very low illumination images, these residuals do not limit the
scientific exploitation of the data. For most of the dataset, as the signal in well-exposed images reaches 8000 DNs
in the panchromatic filter and thousands of DNs in the colour filters, the residuals are negligible and CaSSIS
provides the best colour images available over many areas covered

Item Type:

Journal Article (Original Article)


08 Faculty of Science > Physics Institute > Space Research and Planetary Sciences
08 Faculty of Science > Physics Institute
08 Faculty of Science > Physics Institute > NCCR PlanetS

UniBE Contributor:

Pommerol, Antoine, Thomas, Nicolas, Miguel, Almeida, Read, Matthew James, Becerra Valdes, Patricio, Cesar, Camila, Valantinas, Adomas


500 Science > 530 Physics
500 Science > 520 Astronomy
600 Technology > 620 Engineering








Dora Ursula Zimmerer

Date Deposited:

16 Mar 2023 09:51

Last Modified:

16 Mar 2023 23:27

Publisher DOI:





Actions (login required)

Edit item Edit item
Provide Feedback