SBIR-STTR Award

The problem of developing computer-efficient thermospheric density models is addressed, and design investigations for improving the accuracy and speed of satellite orbital predictions in low earth orbit are proposed. Early computer models of the atmosphere were derived using quasi-static diffusion models, based primarily on satellite orbital decay data. Recent models include data from composition measurements, and more elaborate first-principles codes such as TIGCM, however without satisfactory improvement in decay prediction capability. The latest models provide thermospheric winds, a first order effect on satellite drag which has generally been ignored in orbit propagation software. Extensive and correlated satellite orbital histories are available, and may be used to evaluate these models for accuracy and design. Both research and software engineering development approaches are proposed in order to establish benchmarks, improve code efficiency, develop diagnostic tools, and identify deficiencies thru testing and thru theoretical considerations. This is proposed in Phase I.

Various empirical and theoretical atmospheric density models have been developed for application to satellite tracking data to make orbital predictions. Over the years, researchers have incorporated more observations into empirical models and added sophistication to theoretical models. However, none of these models has decreased the one-sigma standard error for density to less than 15%. In fact, NORAD prefers the 1970 version of the Jacchia density model for its best orbital predictions. One barrier to improving density models is the lack of sufficient information about geophysical parameters which influence temporal variations of the atmosphere. Currently, most density models use the 10.7cm solar flux and a geomagnetic index as indicators of the relevant geophysical parameters. SPACECOM has conducted a limited study which seems to indicate that their satellite tracking data, given sufficient observations and careful processing, yields high resolution density information. In fact, RADEX proposes using this density information as a near-real-time input to a density model to correct for temporal atmospheric variations beyond the resolution of present corrections. The density model will be updated continuously with recent density values to provide an improved basis for orbital predictions.