Abstract. Detailed model calculations of auroral secondary and photoelectron distributions for varying conditions have been used to calculate the theoretical enhancement of incoherent scatter plasma lines. These calculations are compared with EISCAT UHF radar measurements of enhanced plasma lines from both the E and F regions, and published EISCAT VHF radar measurements. The agreement between the calculated and observed plasma line enhancements is good. The enhancement from the superthermal distribution can explain even the very strong enhancements observed in the auroral E region during aurora, as previously shown by Kirkwood et al. The model calculations are used to predict the range of conditions when enhanced plasma lines will be seen with the existing high-latitude incoherent scatter radars, including the new EISCAT Svalbard radar. It is found that the detailed structure, i.e. the gradients in the suprathermal distribution, are most important for the plasma line enhancement. The level of superthermal flux affects the enhancement only in the region of low phase energy where the number of thermal electrons is comparable to the number of suprathermal electrons and in the region of high phase energy where the suprathermal fluxes fall to such low levels that their effect becomes small compared to the collision term. To facilitate the use of the predictions for the different radars, the expected signal- to-noise ratios (SNRs) for typical plasma line enhancements have been calculated. It is found that the high-frequency radars (Søndre Strømfjord, EISCAT UHF) should observe the highest SNR, but only for rather high plasma frequencies. The VHF radars (EISCAT VHF and Svalbard) will detect enhanced plasma lines over a wider range of frequencies, but with lower SNR.
Explainable Models of Disease Progression in ALS: Learning from Longitudinal Clinical Data with Recurrent Neural Networks and Deep Model Explanation
