Solar prominence diagnostics and their associated estimated errors from 1D NLTE Mg II h&k modelling

Aims.

We present further development of the rolling root mean square (rRMS) algorithm.

These improvements consist of an increase in computational speed and an estimation of the uncertainty on the recovered diagnostics.

This improvement is named the cross root mean square (xRMS) algorithm.

Methods.

We used the quantile method to recover the statistics of the line profiles in order to study the evolution of the prominence observed by IRIS on 1 October 2019.

We then introduced the improvements to rRMS.

These improvements greatly increased the computational speed, and this increase in speed allowed us to use a large model grid.

Thus, we utilised a grid of 23 940 models to recover the thermodynamic diagnostics.

We used the 'good' (but not 'best') fitting models to recover an estimate of the uncertainty on the recovered diagnostics.

Results.

The maximum line-of-sight (LOS) velocities were found to be 70 km/s.

The line widths were mostly 0.4 {\AA} with the asymmetries of most pixels around zero.

The central temperature of the prominence was found to range from 10 kK to 20 kK, with uncertainties of approximately +/-5 to +/-15 kK.

The central pressure was around 0.2 dyn/cm2, with uncertainties of +/-0.2 to +/-0.3 dyn/cm2.

The ionisation degree ranged from 1 to 1000, with uncertainties mostly in the range +/-10 to +/-100.

The electron density was mostly 10^10 /cm3, with uncertainties of mostly +/-10^9.

Conclusions.

The new xRMS algorithm finds an estimation of the errors of the recovered thermodynamic properties.

To our knowledge, this is the first attempt at systematically determining the errors from forward modelling.

The large range of errors found may hint at the degeneracies present when using a single ion and/or species from forward modelling.

In the future, co-aligned observations of more than one ion and/or species should be used to attempt to constrain this problem.

;Comment: 9 pages, 12 figures