In-Orbit updates for Observers

These include technical news, in-orbit performance, data systematics, and tips from the CHEOPS Science Team.

1) Technical news

Science images during the South Atlantic Anomaly

As of 2 March 2025, science images taken within the South Atlantic Anomaly (SAA) are now downlinked and included in all CHEOPS data products. While these images are often heavily affected by cosmic rays, they do not compromise the quality of data points acquired outside the SAA. These images are excluded when computing the observation efficiency of a visit and should therefore be considered a "bonus." The affected data points are flagged in the DRP products (see details in the DRP reports) and may, in some cases, be useful for data analysis—particularly when observing bright targets.

Improved Sun Exclusion Angle

The Sun Exclusion Angle (SEA) is the minimum angle allowed between the Sun position and CHEOPS' line of sight. It is defined to prevent sunlight from shining on the radiators (passive cooling system) of the instrument. Initially 120° at launch, the nominal SEA allowed CHEOPS to cover 71.3% of the sky. With the aim of increasing CHEOPS’ sky coverage (new targets and longer observability), the SEA constraint was gradually relaxed over the year 2025 (see table below). Since the end of October 2025, the SEA is 115°, extending the sky coverage of CHEOPS to 77.6%.

* Observing at Sun angles smaller than 115.75° can yield partial blinding of the star trackers required for pointing stability. Such events are brief, occurring only up to 3% of each CHEOPS orbit and resulting in blurred images, which in turn might degrade photometric precision. If programme managers wish to avoid such short events, they should constrain observations to always have Sun angles ≥ 115.75°. This can be enforced in the Observation Requests using the Earliest Start Date and Latest End Date parameters (estimates of these dates are provided by the CHEOPS visibility tool).

Default window location

To optimize the limited telemetry of CHEOPS, only a portion of the full frame images, known as "window images", are downlinked for nominal science observations. These window images are 200x200 pixels in size and feature the target star at the centre. Additionally, to further reduce data usage, the spacecraft's onboard software crops the corners of these window images to create circular images of a diameter of 200 pixels that are then transmitted to the ground. The position of the target, and therefore the window in the CCD, is evaluated weekly as part of the Monitoring and Characterization Program. The goal is to minimize noise caused by bad pixels by moving the target to different regions of the CCD if necessary. The table below shows the log of the default window locations over time. The coordinates correspond to the CCD-pixel where the bottom-left corner of a 200x200 window is located.

* See section Window shape below for details.

Change in the read-out mode allocation

The instrument team decided to change the readout mode for long exposures (Texp = 11.27 s to 60 s) from FAINT to FAINT FAST. This change was implemented to be less sensitive to hot pixels in the covered section. The FAINT read-out mode reads the full-frame image and the window image (which is the product downlinked to the ground) is cropped afterwards by the In-Flight Software. The ULTRABRIGHT, BRIGHT and FAINT FAST read-out modes, on the other hand, only read out the stripe of the CCD containing the window, avoiding any traces of hot pixels in the cover section placed above Y=200 px. Both read-out modes, FAINT FAST and FAINT, have the same read-out frequency (100 KHz) and can be used interchangeably when the exposure time is longer than 11.27 seconds. It was therefore decided to replace the FAINT read-out mode with FAINT FAST in the science visits.

To summarise, from launch until 29.01.2023 any Observation Request (OR) submitted using the PHT-2 was allocated readout modes as listed in the following table:

As of 30.01.2023, any new OR submitted using the PHT-2 is allocated the following read-out modes as a function of the exposure time:

Window shape

As the CCD ages, bad pixels become more of a concern when evaluating the quality of the data. In order to monitor bad pixels, especially hot pixels and RTS pixels, with regular science visits, we need to be able to detect them throughout the CCD subframe. If the PSF of the target star is always in the centre, it is not possible to characterise these pixels as they are constantly illuminated. We, therefore, developed a new strategy for positioning the target star in different pixels.
We chose three alternative positions in the CCD, close to the centre, so that on successive visits, the target would be located in different pixels. However, for the sake of repeatability and scientific performance, a given target will always have the same position when it is observed multiple times.
Since 04/05/2024, the coordinates of the alternating window positions are [389, 382], [373, 440] and [448, 453]. Note that these are the x and y coordinates (in px) of the lower left corner of the 200x200 px window.
We have also switched to square windows instead of the so far default circular ones. This allows us to monitor the pixels in the corners and better characterise the CTI trails.

2) In-orbit performance

The CHEOPS Mission Consortium was responsible for the in-orbit commissioning of the instrument and is now responsible for an on-going monitoring and characterisation programme. A full analysis of the first three and a half years of operations can be found at:

• CHEOPS in-flight performance: a comprehensive look at the first 3.5 years of operations (Fortier et al. 2024)

Details of the performance that has been measured over the course of the mission are summarised in the presentations listed below:

• Measured photometric performance - overview (updated January 2022)
• Observing outside the nominal magnitude range of CHEOPS (May 2021)
• Illustration of photometric performance, including light curves (January 2022)

3) Data systematics

All CHEOPS science observations are processed at the Science Operations Centre at Geneva Observatory, using the CHEOPS data reduction pipeline (DRP). The DRP (Hoyer et al., 2020) is designed to calibrate and correct CHEOPS observations, and to extract photometric light curves from the corrected image series. The DRP does not remove time-dependent trends in the data which are related to, for example, instrument temperature or the position of the target on the detector. These systematic effects are removed from the data by decorrelation or detrending. Identifying and correcting for systematics has been one of the major tasks of the two technical working groups within the CHEOPS Science Team.

• Presentation summarising systematics that have been identified in CHEOPS data (updated in January 2022)
• pycheops - a python package developed by members of the CHEOPS Science Team for the analysis of CHEOPS light curves (Maxted et al., 2021)

In the list below are papers in which the treatment of specific systematics seen in CHEOPS data has been described.

• Bonfanti et al. (2021) "CHEOPS observations of the HD 108236 planetary system: a fifth planet, improved ephemerides, and planetary radii" (https://doi.org/10.1051/0004-6361/202039608) - detailed elaboration of the roll angle, its impact on CHEOPS data and how to decorrelate for it; discussion on smearing patterns, due to bright stars in the field, and an example of internal reflection and its manifestation in CHEOPS images.
• Wilson et al. (2022) "A pair of sub-Neptunes transiting the bright K-dwarf TOI-1064 characterized with CHEOPS" (https://doi.org/10.1093/mnras/stab3799) - includes treatment of the "ramp" and various roll angle-related issues, and also describes the PSF PCA (Principal Component Analysis) method that has been developed to remove these systematics.

4) Tips from the CHEOPS Science Team

The presentations below capture the understanding and experience of CHEOPS and its data that has been built up by technical working groups in the Science Team.

• Tips and guidelines for determining the duration of individual visits (Phase 1) and tips on how to define your individual observation requests (Phase 2) at this link (December 2022)
• Heading off-piste - how well multi-epoch observations can be combined, e.g. for stellar monitoring, and the impact of transit stacking (February 2022)