Photospheric electric fields, estimated from sequences of vector magnetic field and Doppler measurements, can be used to estimate the flux of magnetic energy (the Poynting flux) into the corona and as time-dependent boundary conditions for dynamic models of the coronal magnetic field. We have modified and extended an existing method to estimate photospheric electric fields that combines a poloidal-toroidal (PTD) decomposition of the evolving magnetic field vector with Doppler and horizontal plasma velocities. Our current, more comprehensive method, which we dub the {bf P}TD-{bf D}oppler-{bf F}LCT {bf I}deal (PDFI) technique, can now incorporate Doppler velocities from non-normal viewing angles. It uses the texttt{FISHPACK} software package to solve several two-dimensional Poisson equations, a faster and more robust approach than our previous implementations. Here, we describe systematic, quantitative tests of the accuracy and robustness of the PDFI technique using synthetic data from anelastic MHD (texttt{ANMHD}) simulations, which have been used in similar tests in the past. We find that the PDFI method has less than $1%$ error in the total Poynting flux and a $10%$ error in the helicity flux rate at a normal viewing angle $(theta=0$) and less than $25%$ and $10%$ errors respectively at large viewing angles ($theta<60^circ$). We compare our results with other inversion methods at zero viewing angle, and find that our methods estimates of the fluxes of magnetic energy and helicity are comparable to or more accurate than other methods. We also discuss the limitations of the PDFI method and its uncertainties.
Download