Re-entry predictions of potentially dangerous uncontrolled satellites: challenges and civil protection applications

Currently, nearly 70% of the re-entries of intact orbital objects are uncontrolled, corresponding to about 50% of the returning mass, i.e. approximately 100 metric tons per year. In 2015, 79% of the mass was concentrated in 40 upper stages and the remaining 21% mostly in about ten large spacecraft. The average mass of the sizable objects was around 2 metric tons. Predicting the re-entry time and location of an uncontrolled object remains a very tricky task, being affected by various sources of inevitable uncertainty. In spite of decades of efforts, mean relative errors of 20-30% often occur. This means that even predictions issued 3 h before re-entry may be affected by an along-track uncertainty of 40,000 km (corresponding to one full orbital path), possibly halved during the last hour if further tracking data is available. This kind of information is not much useful and manageable for civil protection applications, often resulting in confusion and misunderstandings regarding its precise meaning and relevance. Therefore, specific approaches and procedures were developed to provide understandable and unambiguous information useful for civil protection planning and applications, as shown in practice for recent re-entry prediction campaigns of significant satellites (UARS, ROSAT, Phobos-Grunt, GOCE, and Progress-M 27 M).