Jovan Skuljan, Defence Technology Agency, John Kay, Defence Technology Agency, Auckland, New Zealand
Keywords: space situational awareness, wide-field imaging, automated astrometric analysis
Abstract:
An observational trial was conducted in the South Island of New Zealand from 24 to 28 February 2015, as a collaborative effort between the United Kingdom and New Zealand in the area of space situational awareness. The aim of the trial was to observe a number of satellites in low Earth orbit using wide-field imaging from two separate locations, in order to determine the space trajectory and compare the measurements with the predictions based on the standard two-line elements. This activity was an initial step in building a space situational awareness capability at the Defence Technology Agency of the New Zealand Defence Force. New Zealand has an important strategic position as the last land mass that many satellites selected for deorbiting pass before entering the Earth’s atmosphere over the dedicated disposal area in the South Pacific. A preliminary analysis of the trial data has demonstrated that relatively inexpensive equipment can be used to successfully detect satellites at moderate altitudes. A total of 60 satellite passes were observed over the five nights of observation and about 2600 images were collected. A combination of cooled CCD and standard DSLR cameras were used, with a selection of lenses between 17 mm and 50 mm in focal length, covering a relatively wide field of view of 25 to 60 degrees. The CCD cameras were equipped with custom-made GPS modules to record the time of exposure with a high accuracy of one millisecond, or better. Specialised software has been developed for automated astrometric analysis of the trial data. The astrometric solution is obtained as a two-dimensional least-squares polynomial fit to the measured pixel positions of a large number of stars (typically 1000) detected across the image. The star identification is fully automated and works well for all camera-lens combinations used in the trial. A moderate polynomial degree of 3 to 5 is selected to take into account any image distortions introduced by the lens. A typical RMS error of the least-squares fit is about 0.1 pixels, which corresponds to about 4 to 10 seconds of arc in the sky, depending on the pixel scale (field of view). This gives a typical uncertainty between 10 and 25 metres in measuring the position of a satellite at a characteristic range of 500 kilometres. The results of this trial have confirmed that wide-field measurements based on standard photographic equipment and using automated astrometric analysis techniques can be used to improve the current orbital models of satellites in low Earth orbit.
Date of Conference: September 20-23, 2016
Track: Instrumentation & Optical Surveillance