Estimating Neutral Atmosphere Drivers using a Physical Model

Abstract : In the initial phase of this study, research focused on implementing a physical model into an ensemble Kalman filter to estimate the location and magnitude of the upper atmospheric heating. Since the neutral atmosphere is strongly externally driven during geomagnetic storms, specifying the upper atmospheric heating is necessary when describing the time-dependent evolution of the neutral density. Challenges arise when only a portion of the upper atmosphere is observable. Because the upper atmospheric dynamics is nonlinear and incompletely observed, determining the heating location from the observed atmospheric response is not straight forward, as shown in the example Figure 1. The nonlinear system of equations can provide numerous possible solutions, only one of which is correct. Although the response of the neutral density to the heating is initially linear at storm onset, the system becomes increasingly nonlinear with time as illustrated in Figure 1. The true heating source (blue circles) becomes increasingly more difficult to predict (red crosses) as the storm progresses.