This page contains a description of an anomaly in the HMI optical system that will have an ongoing impact to data obtained near the south-west limb of the Sun. This page will be updates as learn more about the problem and take corrective actions that may be available.

On 2018.08.17 at 04:52UT, an artifact appeared in HMI filtergrams from both cameras in the form of a dark stripe across the upper left quadrant of the raw images. When images are rotated to solar north up, the artifact is in the solar south-west, lower right, quadrant. The artifact is visually the same in both normal observing mode and HMI calibration mode so the artifact is the result of some change in the filter section of the instrument. This artifact results in noticeable streaks in the intensity observables. The effect on velocity and magnetic field measurements is considerably smaller, though detectable with careful analysis. There is both an intensity reduction in the affected region and some scattered light in its vicinity.

While it is not known with certainty what is causing this artifact, the team believes data from the apparently unaffected 95% of the solar image in not degraded.

Normal data processing was stopped while the HMI team investigated the issue. We believe that some of the effect can be corrected with the use of updated flat-fields and are resuming processing on a provisional basis – we expect further improvements to our corrections as more calibration data become available. Users can tell whether or not their data are affected by this issue by checking the QUALITY keyword. For Level 0 the 0x2000 bit and for Level 1 the 0x20 bit will be set. For Level 1.5 (that is, all 45s and 720s intensity, Doppler, and magnetic field data) the 0x80 bit will be set, indicating that the data is provisional and may be reprocessed in the near future.

The affected area is about 5% of the solar disc, lying beyond a chord in the west south-west part of the image at a minimum central distance of about 0.8 of the solar image radius and between position angles of about 75 and 145 deg measured west from north.

The HMI team is examining the situation and does still not have a full understanding of the impact of the probably permanent change to the instrument.

It is very likely that all data from outside the small effected region is not impacted and data from those regions will be OK to use for analysis of the Sun.

The effect in the impacted disk region is a pair of brighter than normal stripes surrounding a region of reduced intensity. While solar features such as supergranulation appear normal we do see scattered light off the solar limb so the change is not simply reduced intensity.

The event started over a few minutes and reduced in amplitude over the next two days. It has settled to an amplitude in a few days and has been changing slowly as of 31 August.

JsocWiki: HMI2018AugustAnomaly (last edited 2018-09-01 04:36:53 by PhilScherrer)