The Impact of OMNI Data Accuracy on Thermospheric Neutral Density Simulations at Grid Cell Resolutions

As physics-based global magnetosphere-ionosphere-thermosphere models enter operations for space weather forecasting, it is imperative that the sources of uncertainty in model outputs are thoroughly investigated. Global magnetosphere models rely on a set of input solar wind conditions to drive the model; such inputs are often propagated from upstream in the solar wind where the interplanetary magnetic field and plasma parameters can be constantly monitored. Generally, this data is provided by the OMNI data set, a set of observations of the solar wind typically collected near the (Formula presented.) Lagrange point and propagated to Earth's bow shock nose. The actual solar wind conditions outside the bow shock may differ from the OMNI data, leading to erroneous model inputs. Erroneous model inputs will lead to erroneous model outputs, and contribute to poor predictions of space weather phenomena. We present a study on the effects of variability in OMNI solar wind propagation accuracy on ionosphere-thermosphere model outputs at the individual grid cell level. Models are driven with OMNI and ARTEMIS (Acceleration, Reconnection, Turbulence & Electrodynamics of the Moon's Interaction with the Sun) solar wind time series for three events at varying levels of correlation between both inputs (Pearson Correlation Coefficient = 0.95, 0.82, and 0.58). Thermospheric neutral densities at three heights (200, 400, and 600 km) are compared for each simulation at grid cells over Kennedy Space Center, Florida, USA, and College Magnetometer Station, Alaska, USA.