Atmospheric odd nitrogen response to electron forcing from a 6D magnetospheric hybrid-kinetic simulation

Abstract. Modelling the distribution of odd nitrogen (NOx) in the polar middle and upper atmosphere has proven to be a complex task. Firstly, its production by energetic electron precipitation is highly variable across a range of temporal scales from seconds to decades. Secondly, there are uncertainties in the measurement-based but simplified electron flux datasets that are currently used in atmosphere and climate models. The altitude distribution of NOx is also strongly affected by atmospheric dynamics on monthly timescales, particularly in the polar winter periods when the isolated air inside the polar vortex descends from the lower thermosphere to mesosphere and stratosphere. Recent comparisons between measurements and simulations have revealed strong differences in the NOx distribution, with questions remaining about the representation of both production and transport in models. Here we present for the first time a novel approach, where the electron atmospheric forcing in the auroral energy range (50 eV–50 keV) is derived from a magnetospheric hybrid-kinetic simulation with a detailed description of energy range and resolution, as well as spatial and diurnal distribution. These electron data are used as input in a global whole-atmosphere model to study the impact on polar NOx and ozone. We show that the magnetospheric electron data provide a realistic representation of the forcing, which leads to considerable impact in the lower thermosphere, mesosphere, and stratosphere. We find that during the polar winter the simulated auroral electron precipitation increases polar NOx concentrations up to 215 %, 59 %, and 7.8 % in the lower thermosphere, mesosphere, and upper stratosphere, respectively, when compared to no auroral electron forcing in the atmospheric model. These results demonstrate the potential of combining magnetospheric and atmospheric simulations for detailed studies of solar wind–atmosphere coupling.