Accelerometer-derived neutral mass density (NMD) is an important quantity describing the variability of the upper atmosphere. NMD is widely used to calibrate and validate some models used for satellite orbit determination and prediction. Quantifying the true NMD is nearly impossible due to, among others, the lack of simultaneous in-situ measurements for cross-validation and the incomplete characterization of the uncertainties of these NMD products. This study investigates the error distribution of three different accelerometer-derived NMD products from the CHAMP satellite mission during time periods of both high and low solar activity. Using a multimodel ensemble comprised of both physical and empirical models, the study characterizes the error variance of the NMD. The strategies employed here may be useful and applicable to other space missions spanning over longer time periods. The results show considerable differences among the three CHAMP data sets and also reveal a pronounced latitude dependence in their error distributions. The median error standard deviation of CHAMP NMD is smaller during time periods of high solar activity (11.0%) than during periods of low solar activity (13.1%). The results indicate that the method of processing the accelerometer data has a significant impact on the uncertainty estimates of the different CHAMP NMD products.
Accelerometer-derived neutral mass density (NMD) is an important measurement of the variability in upper atmosphere and one of the widely used measurements to calibrate and validate models used for satellite orbit determination and prediction. Providing precise estimates of the true uncertainty of these NMD products is a challenging task but essential for the space weather and geodetic communities. Using multiple data assimilation (DA) experiments and robust statistical techniques, we investigate the uncertainty distribution of three different accelerometer-derived NMD products from the CHAMP satellite mission. Here, in three different DA experiments, we use an ensemble Kalman filter to drive a physics-based model with CHAMP in-situ electron density and temperature data as well as neutral wind estimates from an empirical model. Using a multi-model ensemble comprised of both physical and empirical models, we characterize the error variances among the different NMD products. Our results indicate considerable differences among the CHAMP data sets and also show a pronounced latitudinal dependency for the estimated error distributions. On average, the error estimates for NMD vary in the range 6.5–15.6% of the signal. Our experiments demonstrate that DA considerably enhances the capability of the physical model. We note that the generic strategies applied here may be useful and applicable to other space missions spanning over longer time periods.
The paper presents experiments of driving a physics-based thermosphere model by assimilating electron density (Ne) and temperature (Tn) data using the ensemble adjustment Kalman filter (EAKF) technique. This study not only helps to gauge the accuracy of the assimilation, to explain the inherent model bias, and to understand the limitations of the framework, but it also establishes EAKF as a viable technique to forecast the highly dynamical thermosphere using realistic data assimilation scenarios. The results from the perfect model scenarios show that data assimilation changes and, more often than not, improves the model state. Data from Swarm-A, Swarm-C, CHAMP, and GRACE-A are used to validate the resulting analysis states. Independent validation results show that the Ne-guided thermosphere state does not outperform the model state without data assimilation along the considered satellite orbits. This may be due to the limited number of bonafide Ne profiles available for the thermosphere specification tasks in the experiments. More importantly, the results show that the Ne-guided thermosphere state does not deteriorate much in performance during geomagnetic storm time. The results reveal a few challenges of using Ne profiles in a hypothetical operational data assimilation exercise. In terms of estimating the mass density along the orbits of both CHAMP and GRACE-A satellites, the experiment with assimilating Tn shows more promise over Ne. The results show that the improvement gained in the overall forecasted thermosphere state is better during solar minimum compared to that of solar maximum. These results also provide insights into the biases inherent in the physics-based model. The systematic biases that the paper highlight could be an indication that the specification of plasma-neutral interactions in the model needs further adjustments.