In the polar middle and upper atmosphere, Nitric Oxide (NO) is produced in large amounts by both solar EUV and X-ray radiation and energetic particle precipitation, and its chemical loss is driven by photodissociation. As a result, polar atmospheric NO has a clear seasonal variability and a solar cycle dependency which have been measured by satellite-based instruments. On shorter timescales, NO response to magnetospheric electron precipitation has been shown to take place on a day-to-day basis. Despite recent studies using observations and simulations, it remains challenging to understand NO daily distribution in the mesosphere-lower thermosphere during geomagnetic storms, and to separate contributions of electron forcing and atmospheric chemistry and dynamics. This is due to the uncertainties existing in the available electron flux observations, differences in representation of NO chemistry in models, and differences between NO observations from satellite instruments. In this paper, we use mesospheric-lower thermospheric NO column density data measured with a millimeter-wave spectroscopic radiometer at the Syowa station in Antarctica. In the period 2012–2017, we study both the long-term and short-term variability of NO. Comparisons are made with results from the Whole Atmosphere Community Climate Model to understand the shortcomings of current electron forcing in models and how the representation of the NO variability can be improved in simulations. We find that, qualitatively, the simulated year-to-year variability of NO is in agreement with the observations. On the other hand, there is up to a factor of two underestimation of the NO column density in wintertime, and the model captures only 27 % of the measured magnitude in the day-to-day variability. The observed day-to-day variability has a good correlation with three different geomagnetic indices, indicating the importance of electron forcing in atmospheric NO production. Using electron flux measurements from the Arase satellite, we demonstrate their potential in atmospheric research. Our results call for improved representation of electron forcing in simulations to capture the observed day-to-day variability.