Description & Release Notes for hmi.ME_720s_fd10 Data
Definitive, science-grade, HMI vector magnetic field data are now (May, 2012) available via JSOC under the series name hmi.ME_720s_fd10.
The polarimetric filtergram data are processed with the Milne-Eddington inversion code VFISV (Borrero et al. 2010) to produce full-disk vector magnetic field maps every 12 minutes (or 720 seconds). Data products are available, on average, a week after observation. We are retroactively processing data from 2010 onwards. Current processed database includes, at least, 2 full-disk images per day, with a more exhaustive coverage of particular time intervals (click here for link to coverage map: http://jsoc.stanford.edu/doc/data/hmi/coverage_maps/ then select the fd10/ folder). The inversion code is operating with a fixed magnetic filling factor of unity, and the azimuth has not been disambiguated (it ranges between 0 and 180 degrees). Recent updates to the inversion code include the addition of a regularization term to the merit function in order to bias the solution towards lower eta0 values. The aim of this modification is to reduce spatial discontinuities due to the presence of double minima in the chi-square surface. Other updates focus on the convergence criteria and speed optimization.
Be aware that some magnetic field values in strong umbrae and complex regions are erroneous in HMI vector data. The dynamic range of the HMI measurements goes up to ~3500 Gauss. Strong magnetic fields leads to a complete split of the spectral line (Zeeman saturation). In our case, stronger values than 3500 Gauss extends the spectral line outside of the observed HMI spectral range. High Doppler velocities can also shift the spectral line partially outside of the HMI spectral range. When either one of these facts or their combination happens, the inversion code does not have enough information to constrain the field strength, and occasionally it hits the preset upper limit of 5000GThe dynamic range of the HMI measurements goes up to ~3500 Gauss. Only a very, very small fraction of the pixels exhibits such problems, which are easily seen 'by eye'. These and other minor issues will be described in detail at ReleaseNotes3 page.