Spectroscopic Diagnostics of the Non-Maxwellian κ-distributions Using SDO/EVE Observations of the 2012 March 7 X-class Flare

Abstract

Spectroscopic observations made by the Extreme Ultraviolet Variability Experiment (EVE) on board the Solar Dynamics Observatory (SDO) during the 2012 March 7 X5.4-class flare (SOL2012-03-07T00:07) are analyzed for signatures of the non-Maxwellian κ-distributions. Observed spectra were averaged over 1 minute to increase photon statistics in weaker lines and the pre-flare spectrum was subtracted. Synthetic line intensities for the κ-distributions are calculated using the KAPPA database. We find strong departures (κ 2) during the early and impulsive phases of the flare, with subsequent thermalization of the flare plasma during the gradual phase. If the temperatures are diagnosed from a single line ratio, the results are strongly dependent on the value of κ. For κ = 2, we find temperatures about a factor of two higher than the commonly used Maxwellian ones. The non-Maxwellian effects could also cause the temperatures diagnosed from line ratios and from the ratio of GOES X-ray channels to be different. Multithermal analysis reveals the plasma to be strongly multithermal at all times with flat DEMs. For lower κ, the DEM κ are shifted towards higher temperatures. The only parameter that is nearly independent of κ is electron density, where we find log(n e [cm −3 ]) ≈ 11.5 almost independently of time. We conclude that the non-Maxwellian effects are important and should be taken into account when analyzing solar flare observations, including spectroscopic and imaging ones.

Plasma Diagnostics from Active Region and Quiet-Sun Spectra Observed by Hinode/EIS: Quantifying the Departures from a Maxwellian Distribution

“…The N e,0 is dependent on κ only slightly, with the DEMweighted ratios for κ = 2 yielding densities about 0.1-0.2 dex lower than the corresponding Maxwellian curves. This behavior of the density-sensitive coronal lines ratios with κ is well-known (e.g., Dzifčáková & Kulinová 2010;Dudík et al 2014Dudík et al , 2015Dzifčáková et al 2018) and occurs only for low κ → 2 values. The resulting uncertainty in density is small enough to permit an initial estimate of the κ value using the ratio-ratio technique.…”

Effects of electron distribution anisotropy in spectroscopic diagnostics of solar flares

“…The temperature diagnostics for the κ-distributions show similar behavior to bi-Maxwellian ones (Dzifčáková et al 2018). However, differences in the diagnosed temperatures are much p = 1/5 (blue dashed lines), 1/10 (green dashed lines), 1/20 (red dashed lines), and p = 5 (blue full lines), 10 (green full lines), and 20 (red full lines).…”

“…We assume that the bi-Maxwellian electron distributions are formed in the flaring plasma in the reconnection plasma outflow or in collapsing magnetic traps, where the anisotropic acceleration can be effective. The electron densities in such a plasma can be high, up to several times 10 11 cm −3 (e.g., Milligan et al 2012;Del Zanna & Woods 2013;Dzifčáková et al 2018). Therefore we take the plasma density at locations with the bi-Maxwellian distributions as 5 × 10 10 cm −3 .…”

“…The equilibrium EUV line synthetic spectra of iron for the bi-Maxwellian distributions were calculated for p = 1/5, 1/10, and 1/20 and for p = 5, 10, and 20, the spectral bands of SDO/EVE spectrograph 70-600 Å the temperature range log(T/K) = 6.9−7.5, and electron density N e = 5 × 10 10 cm −3 . The spectral, temperature, and density ranges were adopted according to Dzifčáková et al (2018) to compare effects of the bi-Maxwellian distribution on the flare spectra with the effects of κ-distribution.…”

“…We note that in many acceleration mechanisms a description of the distribution anisotropy by the bi-Maxwellian distribution only can be considered as a first approximation. Finally, we compare the results of the spectroscopic diagnostics using these bi-Maxwellian distributions with those using the κ distributions (Dzifčáková et al 2018). .…”