Page:SATCON2 Observations Report.pdf/23



The accuracy in satellite forecasting codes depends on the accuracy of the general perturbation TLEs and the quality of the software used to calculate the satellite's future position and trajectory. Currently the accuracy of Starlink and OneWeb TLEs translates to an accuracy in position on the night sky of ≦ 30 arcminutes (see Figures 2 and 3). This is not adequate for the needs of the astronomical community and satellite forecasting software. Additionally, recent survey observations of OneWeb satellites by the CLEOsat group show that 1 in 40 observations resulted in a negative detection. This is due to an orbital maneuver made by the satellite after the public release of the TLE data. Therefore, we recommend operators maintain the frequency of the TLEs being released (every eight hours), but in addition release a new TLE after a satellite maneuver, allowing forecasting software to update the satellite’s position and trajectory.

The introduction of error bars (uncertainties) with all orbital solution data is also essential to improve accuracy. This would protect critical optical observations, where a satellite trail could ruin an entire image, as well as radio astronomy, where the satellite radio beam can damage sensitive equipment.

New satellites launch regularly and many satellites in constellations change orbits very frequently. As a result, operators should begin sharing data as soon as they initially contact a newly launched satellite and indicate whenever a future maneuver is planned in advance.

3.1.1. Two case studies: validating orbital solutions through observations

Measuring the accuracy in the ephemerides derived from a TLE requires a large field of view and relatively short exposure time (a few seconds) to capture the start and end points of the satellite trail. With this type of observation, the position of the satellite as a function of time can be extracted from the image by integrating the angular velocity over the length of the satellite trail. Factoring in the telescope pointing error and correctly propagating uncertainties, it is possible to measure the accuracy of a TLE. An example is shown in Figure 2, where an observation of OneWeb-0210 obtained with the 0.6-meter telescope at Chungbuk National University Observatory, South Korea (courtesy of the CLEOsat group) provides a single measurement of the ephemeris-derived TLE accuracy to ~ 15 arcminutes.

Rh