How does the quality of lower atmospheric drivers help predict thermospheric weather?

Abstract:

The development of whole atmosphere models (ground to exobase) in the last decade along with the availability of high altitude atmospheric data assimilation has brought the geospace community closer to important understandings of the prediction of short-term variability (≤ 10 days) in the lower thermosphere: a large fraction of this thermospheric variability is driven by lower atmosphere weather from below; and, because of the wind-dynamo coupling, such neutral variability is mapped into the ionospheric behavior. Thus, thermosphere-ionosphere predictions have become gradually a whole atmosphere problem in the last decade, highlighting that the quality of short-term thermospheric forecasts is dependent on how well we can predict lower atmospheric weather. However, a quantification of such connection between lower and upper atmospheric forecasts, and an assessment of the ability of the lower atmosphere to predict the dynamical and ionospheric variability out to few days has never been attempted. To achieve this goal, in this study we use the state-of-the-art Whole Atmosphere Community Climate Model eXtended (WACCM-X) coupled to the NRL ionosphere SAMI3 model via the Navy Highly Integrated Thermosphere-Ionosphere Demonstration System (Navy-HITIDES) coupling layer. In order to represent the weather of the day, the coupled thermosphere-ionosphere system is nudged below 90 km toward the atmospheric state provided by the prototype Navy Global Environmental Model for High-Altitude (NAVGEM-HA). Ensembles of 5-day forecasts issued by NAVGEM-HA are used to constrain the coupled thermosphere-ionosphere system to the predicted weather of the day and study the thermosphere-ionosphere response. The resulting simulations illustrate how the quality of thermosphere-ionosphere predictions depend on the quality of the lower atmosphere forecasts. Specifically, we discuss the representation of planetary-scale waves (normal modes, migrating and non-migrating solar tides, lunar tides), their wave-mean flow interactions, as well as the nature of wave-to-wave non-linear interactions that occur in such a highly non-linear system. Ultimately, differences in the prediction of thermospheric variability, and the quality of, is reflected on the ionospheric behavior.