Acceleration effects in MDI magnetogram data

Abstract

Acceleration effects are found in the Michelson Doppler Imager (MDI) magnetogram data because changes in the line profiles during the 30-s measurement are introduced by underlying p-mode velocity variations. This imparts an oscillatory component to the magnetic flux signal. Simulated profiles using Maltby M and Harvard Smithsonian Reference Atmospheres (HSRA) are shifted in accordance with a given velocity amplitude and period and the MDI algorithm for data measurement is applied. The simulated oscillatory component to the magnetic flux density always has a phase difference with respect to the underlying velocity of −90◦. The maximum introduced RMS amplitude is a function of velocity amplitude and field strength, but realistic errors are on the order of 5/2000 G, or 0.25% of field strength. Comparison of simulations with observations shows RMS amplitudes of MDI flux density are much greater than predicted by this effect. A 2-component HSRA model, tested to determine if stronger fields with smaller fill factors could fit the data, still can not reproduce the observations. It is concluded that oscillatory amplitudes of magnetic flux density measured with MDI are not due to acceleration effects, although the effect could contribute up to 25% of the measured amplitude. Attempts to remove acceleration effects from the magnetic flux signal are not successful. Also, we confirm that velocities measured in linearly polarized light in the vicinity of a strong magnetic field contain larger errors than velocities measured in circularly polarized light (Yang and Norton, 2001).