Neutral Density and Crosswind Determination from Arbitrarily Oriented Multiaxis Accelerometers on Satellites

An iterative algorithm for determining density and crosswind from multiaxis accelerometer measurements on satellites is presented, which works independently of the orientation of the instrument in space. The performance of the algorithm is compared with previously published algorithms using simulated data for the challenging minisatellite payload. Without external error sources, the algorithm reduces rms density errors from 0.7 to 0.03% and rms wind errors from 38 to 1 m is in this test. However, the effects of the errors in the instrument calibration and the external models that are used in the density and wind retrieval are dominant for the challenging minisatellite payload. These lead to mostly systematic density errors of the order of similar to 10-15%. The accuracy of the wind results when using the new algorithm is almost fully determined by the sensitivity of the cross-track acceleration component to the calibration and radiation pressure modeling errors. The applicability of the iterative algorithm and the accuracy of its results are demonstrated by presenting challenging minisatellite payload data from a period in which the satellite was commanded to fly sideways and by comparing the density and wind results with those from adjacent days for which the satellite was in its nominal attitude mode. These investigations result in recommendations for the design of future satellite accelerometer missions for thermosphere research.