The optical performance and stability of JWST is described in this article. The performance exceeds expectations, with observatory-level wavefront error (WFE) at NIRCam usually between 60 and 80 nm rms.

The OTE was required to be diffraction limited for wavelengths λ > 2 μm, with predictions for the full observatory wavefront error level (telescope plus instrument plus dynamics) being better than 100 nm rms WFE at NIRCam. See Lightsey et al. (2014) for details of pre-flight modeling, requirements, and predictions.

In fact, the optical performance measured at the end of commissioning and into science operations exceeds expectations, with, for example, the OTE plus NIRCam SW showing diffraction-limited image quality at ∼1.1 μm, and observatory level WFE at NIRCam usually between 60 and 80 nm rms. A complete discussion of the telescope’s optical performance against requirements at the end of commissioning is reported in Knight & Lightsey (2022), Rigby et al. (2022), and McElwain et al. (2023) 

For a discussion of the active optics operations concept for JWST, see JWST Sensing and Control.

Routine wavefront sensing and control

The OTE has been in routine wavefront sensing and control (i.e., "maintenance") mode since the end of OTE commissioning. In this mode, the wavefront is measured every 2 days using NIRCam weak lenses, and mirror alignment corrections are made as needed. This 2-day sensing cadence has a loose tolerance; the goal is to schedule wavefront sensing observations so as to accommodate any time-critical observations and to not disrupt long mosaic or time-series observations.

Wavefront corrections via mirror control have proven relatively infrequent, averaging a correction every 29 days prior to March 2024. After the correction in March 2024, there was a 186 day stretch without corrections, raising the average to 36 days. (The requirement was no more frequently than once every 2 weeks.) Wavefront sensing and control observations, necessary to maintain image quality, take < 1.5% of the total observatory time, which is accounted as part of the observatory calibration overheads. In December 2024, the cadence of sensing measurements was increased from 2 days to 4 days. This cadence could be changed again in future, based on the stability of the telescope and its level of performance.

Note that the intent of the mirror corrections (i.e., wavefront control) is to maintain the telescope alignment, not to specifically alter it. That is, the goal of corrections is to bring the OTE back to the nominal aligned state that had been achieved at the end of the commissioning period, and ensure it continually remains near that state. There is no plan for "campaign" style observations in which the OTE would be temporarily optimized for one instrument over another. Nor is there any need for observers to request scheduling their observations with any particular timing constraints relative to wavefront sensing. However, to mitigate any possible impacts of thermal changes to the point spread function during certain  high-contrast imaging observations, it is recommended that the target and reference stars be observed sequentially. Users should consult documentation on the specific instrument and mode they wish to use for best observing practices.

The wavefront sensing image data from NIRCam and the derived wavefront maps are delivered to the MAST Archive, similar to other calibration program data, and are accessible to the community. Two examples of trending the observatory optical performance, illustrated below, provide views into the wavefront error (WFE) as a function of time since the start of the Cycle 1 (mid-July 2022). This and other trending capabilities are offered as part of the WebbPSF Python software. See, e.g. https://webbpsf.readthedocs.io/en/stable/jwst_measured_opds.html.

Performance & Stability

In the figure above, one sees the behavior comprising two main components:

• A low-level slow and largely continuous rise in the overall wavefront error between corrections:
  These are believed to result from misalignments in the primary mirror segments that accumulate over time as the composite back plane undergoes small temperature changes from varying thermal loads. They can be seen here to have generally displayed steeper trends earlier in the mission than present.
  
  
• Punctuated abrupt and significant jumps in the wavefront state: 
  These are colloquially and somewhat misleadingly referred to as "tilt events." This term has come to mean any uncommanded change in the position of one or more primary mirror segments that occurs abruptly and contributes to a significant increase in the overall OTE WFE. Distinct from the mechanism behind the more evolutionary gradual misalignments, tilt events are thought to result from a stick/slip sort of sudden release of stored mechanical loads that have been fed into various parts of the OTE structure, likely at the time of observatory cooldown in space vacuum. These so-called tilt events have been readily corrected as part of the nominal routine maintenance program.

One can see something of a resurgence of significant "tilt events" in early 2024. With the exception of one, the WFE from these remained within the 130 nm requirement, and were nonetheless corrected entirely routinely. The outlying event (seen in the plot as off-scale in late February) had WFE that was the largest seen in the science era. This event was expeditiously analyzed and corrected, and its effects documented for the community in Perrin et al. (2024). After this event, and the rash of others within a relatively short time, the telescope has subsequently exhibited an unprecedented level of stability, which can be seen in the right-most set of data points following the March 31st mirror correction in the figure's top pane. 

Unlike these mirror tilts and misalignments that STScI corrects, the effects of micrometeoroid impacts have been infinitesimal since the start of science. They can only be detected at low levels by specialized high frequency phase retrieval and do not manifest in a measurable way in science data, analyses, or interpretation. Nor do they alter the positions of the mirror segments, so do not produce "tilt events", which is a persistent misconception. This characterization of the effects of micrometeoroids excludes, of course, the large impact that occurred in May 2022, which STScI is now understood from additional experience to most likely be an exceptional outlier. Despite this, and out of an abundance of caution, the JWST mission nonetheless implemented the micrometeoroid avoidance zone (MAZ) to reduce the amount of time that JWST's orbital velocity (which is similar to typical micrometeoroids) will maximally add to the kinetic energy of potential impacts. A more detailed discussion of the MAZ and references are found at JWST Micrometeoroid Avoidance Zone.

By providing an updated time baseline for the wavefront behavior, this section also serves to supplement the treatment given by Lajoie et al. (2023)

References

Knight, J.S. & Lightsey, P.A. 2022, SPIE 12180, 121800V

Lajoie, C. -P., et. al. 2023, JWST-STScI-008497
A Year of Wavefront Sensing with JWST in Flight: Cycle 1 Telescope Monitoring & Maintenance Summary

Lightsey, P. A. et al. 2014, SPIE 9143, 914304-1 
Status of the Optical Performance for the James Webb Space Telescope  

McElwain, M. W. et al. 2023, PASP, 135, 058001
The James Webb Space Telescope Mission: Optical Telescope Element Design, Development, and Performance

Perrin, M.D. et al. 2024, JWST-STScI-008650
OTE Science Performance Memo 3, JWST-STScI-008650 Detection and Correction of a Uniquely Large Mirror Shift during Cycle 2

Rigby, J. et al. 2023, PASP, 135, 048001
The Science Performance of JWST as Characterized in Commissioning

Telfer, R. et al., 2024, DOI:10.1117/12.3020273
Empirical Characterization of JWST Wavefront Error Variations