To extract the information that the Mg ii NUV spectra ( observed by the Interface Region Imaging Spectrograph ; IRIS ) , carries about the chromosphere during solar flares , and to validate models of energy transport via model-data comparison , forward modelling is required .
The assumption of statistical equilibrium is typically used to obtain the atomic level populations from snapshots of flare atmospheres , due to computational necessity .
However it is possible that relying on statistical equilibrium could lead to spurious results .
We compare solving the atomic level populations via statistical equilibrium versus a non-equilibrium time-dependent approach .
This was achieved using flare simulations from RADYN alongside the minority species version , MS_RADYN , from which the time-dependent Mg ii atomic level populations and radiation transfer were computed in complete frequency redistribution .
The impacts on the emergent profiles , lightcurves , line ratios , and formation heights are discussed .
In summary we note that non-equilibrium effects during flares are typically important only in the initial stages and for a short period following the cessation of the energy injection .
An analysis of the timescales of ionisation equilibrium reveals that for most of the duration of the flare , when the temperatures and densities are sufficiently enhanced , the relaxation timescales are short ( \tau _ { \mathrm { relax } } < 0.1 s ) , so that the equilibrium solution is an adequate approximation .
These effects vary with the size of the flare , however .
In weaker flares effects can be more pronounced .
We recommend that non-equilibrium effects be considered when possible , but that statistical equilibrium is sufficient at most stages of the flare .