Week 26 of the Early Operations system optimization period
The week included a mix of dedicated on-sky engineering and pre-LSST observations.
Over the past several weeks, the team has been making detailed analyses of variations of optical aberrations across the field of view. These studies have shown that there are persistent higher-order spatial variations of the optical aberrations across the field of view that are not fully constrained by the four specialized wavefront sensors that are uniformly spaced around the periphery of the field of view. The higher-order variations of the optical aberrations across the field of view can be studied in detail by using full-array mode measurements that piston the entire science array to intra- and extra-focal positions. Analyses of these data suggest that improved, more uniform delivered image quality could be achieved across the field of view by using additional mirror bending mode degrees of freedom that are not currently actively controlled by the open loop and closed loop control systems of the AOS. As an initial step in this direction, the team used full-array mode measurements to derive corrections utilizing the full set of 50 degrees of freedom for a single telescope pointing, and tested application of those constant offsets during on-sky observations.
The team is examining the extent to which the higher-order variations of optical aberrations across the field of view are stable with respect to telescope pointing and other system telemetry (e.g., temperature gradients across the primary-tertiary mirror). On-sky engineering this past week included observations in full-array mode while rapidly alternating between two telescope elevations or two Camera rotator angles to directly measure differential changes in the optical state.
The team also deployed an updated weighting scheme for the optical degrees of freedom used to create the basis of non-degenerate modes employed by the AOS to apply corrections.
On 23 April, 6 additional dome louvers were opened for the first time during nighttime observations, for a total of 12 operational louvers.
During the pre-LSST observations this week, the typical per-night median instrument contribution to the delivered PSF FWHM was at the level of 0.4-0.45 arcseconds. Multiple analysis methods produce consistent estimates for the instrument contribution:
- optical aberrations measured at the corner wavefront sensors added in quadrature with laboratory measurements of the diffusion of photoelectrons in the LSST Camera detectors,
- variation of the PSF FWHM across the field of view, and
- comparison of delivered PSF FWHM to an estimate for the atmosphere seeing contribution derived from the corner wavefront sensors.
For the large majority of pre-LSST visits acquired during the week, the delivered PSF shapes are well described by their second-moments. Typical per-night median ellipticity was at the level 0.06-0.07. The delivered PSF has been comparatively stable, with a lower incidence of irregular PSF shapes, as identified from 3rd and 4th order moments or shapelet decomposition, relative to prior periods of the Early Operations on-sky campaign.
A Rubin Observatory record for delivered image quality was achieved on 18 April with PSF FWHM below ~0.6 arcseconds across the full field of view during an AOS closed-loop stability test at fixed telescope pointing.
Observing plans for the next week include both pre-LSST and dedicated on-sky engineering tests.
A filter swap to exchange u for y occurred on 21 April. The currently installed filter set is grizy.