Events

Inversion methods are critical to inferring the physical properties of the solar atmosphere from spectropolarimetric measurements. A general approach is to iteratively solve a linear system of equations for corrections, to be added to a guess atmospheric model, such that the spectral output from the inverted model matches the observed spectrum. The current state-of-the-art SIR inversion method computes smooth solutions that best fit a given observed spectrum. These solutions are not optimally depth-resolved and are usually dominated by cross-talk errors from other variables. In the state-of-the-art observations from the Daniel K. Inouye Solar Telescope, great effort has been put into improving the horizontal spatial resolution of the data. To use the telescope to its full potential, it is equally important to determine and invert at the vertical resolution limit and compute solutions that are optimally error-balanced.

In this talk, I will discuss the strengths (and limitations) of the iterative Optimally Localized Averages (OLA) inversion method (originally invented for geo- and helio-seismological applications) we have developed to infer the thermodynamic properties of the solar photosphere from spectroscopic measurements. The method linearly combines response functions to form highly localized averaging kernels to compute solutions that are optimally depth-resolved and have minimal cross-talk error between variables. Moreover, the width of these kernels estimates the vertical resolution of the inverted solution.

I will also discuss a quantitative metric (preliminary work) we have devised that can identify spectral lines best suited for a given scientific query and can assess their inversion capabilities. Such a tool could be used independently of the inversion scheme and would greatly help researchers with limited expertise in line selection when developing missions and observing proposals.

Watch the recording.