Effects of Uncertainties in the Atmospheric Density on the Probability of Collision Between Space Objects

Abstract

The rapid increase of the number of objects in orbit around the Earth poses a serious threat to operational spacecraft and astronauts. In order to effectively avoid collisions, mission operators need to assess the risk of collision between the satellite and any other object whose orbit is likely to approach its trajectory. Several algorithms predict the probability of collision but have limitations that impair the accuracy of the prediction. An important limitation is that uncertainties in the atmospheric density are usually not taken into account in the propagation of the covariance matrix from current epoch to closest approach time. The atmosphere between 100 km and 700 km is strongly driven by solar and magnetospheric activity. Therefore, uncertainties in the drivers directly relate to uncertainties in the neutral density, hence in the drag acceleration. This results in important considerations for the prediction of Low Earth Orbits, especially for the determination of the probability of collision. This study shows how uncertainties in the atmospheric density can cause significant differences in the probability of collision and presents an algorithm that takes these uncertainties into account to more accurately assess the risk of collision. As an example, the effects of a geomagnetic storm on the probability of collision are illustrated.

Satellite Orbital Drag During Magnetic Storms

“…The results of this work are consistent with the results found by Bussy‐Virat et al. (), who computed probabilities of collision between objects in space affected by thermospheric density uncertainties, as the ones introduced by magnetic storms. They found that during a severe storm (Table ), the collision probability increased by 50%.…”

The Effect of Neutral Density Estimation Errors on Satellite Conjunction Serious Event Rates

“…However, the expectation is that the sample size used in this study—nine multiday periods of screenings against an entire satellite catalogue—is large enough that differences introduced by this simplification will tend to level out statistically: in some cases, the multiplicative alteration of the ballistic coefficient will result in conjunctions that are more risky than would have been produced by a nuanced altering of the density field, and in other cases it will result in conjunctions that are less risky, so it is believed that the overall result would be similar under either approach. A study performed recently by Bussy‐Virat et al (), which was similar in intent to the present analysis but instead performed a detailed investigation of a small number of conjunction examples, did in fact account for density field variations, and their findings showed similar variations in conjunction severity to those of the present analysis (which will be presented in section 7). A fully definitive adjudication of this general question would in principle begin with a comprehensive statistical investigation of the forecast errors present in space weather indices (the Bussy‐Virat et al study makes a good beginning at this but investigates only two years' worth of data and limits itself to F10 and Ap) and, using the statistical models generated from such an investigation, generate statistically consistent space weather index errors that could then be applied to the input indices.…”

Space Weather Modeling Capabilities Assessment: Neutral Density for Orbit Determination at low Earth orbit

“…Consequently, thermosphere model performance depends strongly on the choice of solar and geomagnetic activity drivers, and in case of density prediction, on the accuracy of the forecast of the drivers. The error due to activity forecast is out of the scope of this paper but has recently been addressed by Bussy‐Virat et al () and Hejduk and Snow ().…”