We've had quite a few e-mails from people asking why our ground track display is still showing the position of Tiangong 1 after the re-entry has been widely reported in the media. It seems many people have been confused by the words "real-time display". Our displays are not receiving real-time data on the position of Tiangong-1, because there is no source providing such data. The words "real-time" were intended to indicate that the displays update automatically, and the web page does not need to be refreshed by the user to show the current position.

Much of the time, satellites in low Earth orbit are not visible from any tracking station. Orbits are calculated on the basis of many measurements of speed and position over the previous hours and days when tracking data was available. A set of "two-line elements" are then published on web sites such as Space-Trackweb. We have a script running which checks for a new set of elements every hour.

Our display works by propagating the last published orbit into the future, using a complex model called SGP4 which was developed by NASA and accounts for many factors which perturb the orbit, such as the Sun, Moon, the shape of the Earth and the residual atmosphere at the height of the satellite. It is this last factor which is the most uncertain, because the density of the atmosphere several hundred kilometers above the surface varies widely and unpredictably. This makes predictions into the future become less and less accurate as time goes by, and this is especially true of satellites approaching re-entry when the air-drag becomes the dominant perturbation. This makes predicting the exact time and place of re-entry very difficult even a few hours before it happens. A further uncertainty is caused by the unknown orientation of the satellite - it make a big difference if the solar panels, for example, are edge-on or perpendicular to the direction of motion. These uncertainties are the reason new orbital elements need to be published regularly. Other web sites which provide tracking data will all work in a similar manner.

We did not remove the displays or update the status of Tiangong 1 earlier, for the simple reason that the re-entry happened in the middle of the night for us here in Germany and we were asleep!

We apologise to all those people who feel they have been misled - this was not our intention, and the tracking page showed the position as accurately as possible from the data available. We will make sure that the limitations of the display and how it works are made clearer when the next major re-entry is imminent.