The focal plane of ODI will be composed of 8 x 8 Orthogonal Transfer Arrays (OTA). By allowing real-time controlled motion of the charges at a pixel level during integration, these detectors provide several different observing modes, some not found on other wide-field astronomical imagers. The general idea is improving the global image quality by moving the charges during an exposure. The following five modes have been defined in the Science Requirements:

A) Static Mode

In this mode, the camera is used as a conventional CCD imager. Optionally (i.e., this is the rule) four guide star cells will be configured to deliver guide star videos. Their centroid signal will be used to adjust telescope tracking and adjust the instrument rotator. No image corrections (i.e. no charge motion) will be applied on the detector.

Anticipated use for static mode:

The static mode without telescope guiding feedback will find applications for observing standard stars and to obtain twilight flat fields. Also, short exposure/snapshot programs might want to use this mode. The static mode with telescope guiding feedback will find applications for all science applications not depending on the very best possible seeing, requiring very low observational overheads, or wanting to avoid all calibration complications that OTA modes can bring with them.

B) Coherent Correction Mode

In the coherent correction mode the corner guide star information will also be used to apply a fast (~20Hz) motion correction on the detectors. The low frequency part (~1Hz) will be used to guide the telescope and the instrument rotator. Higher frequencies are used to correct windshake and other errors related to tracking. The OTA-correction will be the same for the entire focal plane, i.e., there will be no local variation in the astrometric reference frame.

Anticipated use for coherent correction mode:

This mode will probably be the most commonly used for ODI. We expect that both large surveys and small PI programs will make use of this mode. The wide field surveys due to the reduced observing preparation and overheads, and the small PI programs because all calibration complications associated to a rubber focal plane can be avoided.

C) Local Correction Mode

In the local correction mode, a multitude of guide stars will be defined, and the image motion detected by those stars will be corrected for all the OTAs on a distance-weighted scheme. This mode, the more complex with ODI, will provide the optimum correction for guiding, windshake, and atmospheric turbulence. Guide stars to determine the local correction have to be located on the same detector where the corrections will be applied. A guide star must be no more than 2’ away from a cell receiving local tip/tilt information. For detectors/cells where no local guide signal is available (e.g., due to the lack of guide stars), either no correction or correction as in the coherent correction mode is applied. The low frequency signal of the guide stars will again be used for telescope & instrument rotator guiding.

The best image quality offered by the local correction mode might not necessarily be needed over the entire field of view, but only for designated targets of interest across the field. In this case, only selected areas (“patches”) with enough guide stars would operate in the local correction mode. The remainder of the focal plane array would operate either in static or coherent correction mode, using four corner guide stars.

Anticipated use for local correction mode:

Programs depending on the best possible image quality, but at the same time not requiring a stable point spread function over the entire field of view, will be best suited for this mode.

D) Non-sidereal tracking guide mode

Non-sidereal tracking is an addition to the static & coherent correction mode. In addition to compensate for the earth’s rotation, the telescope will also correct for a high proper motion of solar system objects such as Kuiper belt objects or comets. Such a mode has been demonstrated with OPTIC with on-chip OT tracking; however, due to the small size of OTA cells this mode is impractical for ODI since significant sky coverage would be lost. Instead, the guide windows for the four guide stars will continuously be shifted on the detector without telling the telescope tracking system. As a consequence, the telescope tracking system will receive a constant push to follow the objects motion.

Anticipated use for non-sidereal tracking guide mode:

This mode is highly specialized and will be used for solar system science programs. Increased requirements for setup and quality control might force this mode to be offered only in visiting mode at the beginning.

E) Targeted Shutterless Photometry Mode

Targeted shutterless photometry utilizes the guide star video / telemetry stream for scientific purpose. The goal is to measure fast luminosity variations that are caused, e.g., by planetary transits, oscillation of stars, of accretion processes in double star systems. Typical video exposure times will range from 20ms to 40s. The shorted bound is set by limited photon flux and shutterless readout time, while the upper limit is chosen here since observing efficiency might break even with conventional exposures at 40s. Total exposure times (i.e, shutter open time) of targeted photometry observations can range from minutes to several hours.

In targeted photometry mode there will be no OT-correction applied on the detectors. It is anticipated that eventually a guide core with more enhanced photometry (compared to standard OT-corrected operations) will be used.

Anticipated use for targeted shutterless photometry mode:

This is a highly specialized mode that might find only limited applications.