Speaker
Description
The increasing numbers of satellites in the night sky pose risks to ground-based astronomy in the optical regime. The Vera C. Rubin Observatory has found that most images from the Legacy Survey of Space and Time (LSST) will have contamination from satellite streaks, ranging from localised pixel effects due to faint objects, to entire observations being unusable due to crosstalk from bright satellites. Mitigation attempts are underway to minimise these effects but will struggle if proposed mega-constellations such as SpaceX’s filing for one million space datacentres or Reflect Orbital’s thousands of space mirrors go ahead. Motivated to protect Earth-based astronomy, the International Astronomical Union’s (IAU) Centre for the Protection of the Dark and Quiet Sky from Satellite Constellation Interference (CPS) recommends satellites be no brighter than seventh magnitude for orbits lower than 550km altitude - the majority of LEO. This requirement is also proposed in the EU Space Act to come into force in 2030.
Several satellite operators are voluntarily bringing in measures to reduce their optical impact, such as dark pigments and mirror-like finishes. For satellite operators to be able to know their impact on astronomy, and show their mitigation efforts are effective before launch, requires physical testing of the satellite in the lab to determine their bi-directional reflectance distribution function (BRDF) and creating a digital twin to simulate their orbit. This is surface reflectance, used extensively in terrestrial cases such as pigment classifications and Earth albedo measurements. Adapting terrestrial methods to satellites is non-trivial, due to the size of satellite parts, their irregular surfaces, commercial sensitivity, and cost.
In a UK Space Agency study led by the University of Edinburgh, 3S Northumbria researched existing brightness prediction methods for satellites. It was found that there were several digital twin capabilities, both open- and close-sourced and of various levels of complexities, but a lack of public BRDF data to use as input. This motivated 3S Northumbria to do this further research into existing BRDF acquisition capabilities, to determine the current landscape of satellite brightness predictions.
It was found that there are two main methodologies for acquiring reflections: firstly, using scatterometers. They generally take small (approx. 5cm) samples and run fast image-based measurements. There are some that can take samples up to 30cm, but cost rises with sample size capabilities. As scatterometers take small samples, testing actual satellite panels is not possible for most cases so accurate BRDFs are not possible. The advantage of this method is the speed and low cost - appealing to commercial actors in space. The second is using gonioreflectometers, larger apparatus that have arms holding a light source and detector with which accurate incident and reflection angles can be recorded. This method is time-consuming and data dense – exploring the hemisphere above a sample in 0.1$^\circ$ increments results in over 3$\times$10$^6$ data points. Applying either of these methods to pre-launch requirements would require time and effort from operators and certification bodies, and the market would require time to adapt to provide reflectometry services en masse.