Originally Posted by
Propduffer
Although we don't know exactly what it was based on - because we were never told what radar had reported this, but in the first few days all the media was reporting that they had been told that the GS for QZ8501 was very low.
Now that we have the ADS-B information we can be sure this low groundspeed couldn't have come from secondary radar so it had to have come from a military primary radar. Or it might have been bogus information, but it has never been retracted afik.
Edit: After a bit of research I've come to the conclusion that the entire basis for the low speed assumptions for QZ8501 have been based on that radar plot which was released on day one of this event and had the numbers 363 which was taken to be the altitude and the number 353 which was taken to be the speed (IAS at FL 36.0).
That's not very solid evidence, that "radar plot" may have been composed for a news release quickly - it may have been mostly eye candy.
In modern systems the controller sees a 'track plot' that is made up from multiple surveillance inputs. Way back in the thread it was reported that secondary radar responses were lost first followed by ADS-B responses later. So the system was receiving both.
The speed displayed can be:
* Generated by the ATC system from the distance between successive track positions and the time. Most systems smooth this by taking the distance between for example the latest report and the one 5 reports ago. This is normal with SSR
* Replicated from the ground speed field of the ADS-B report
* Multi-Sensor Comparison with the speed from one source used but validated/corrected by the other source(s)
What will not happen is that the speed will be invented.
The problem with a snapshot is that you do not know the history of the ground speed displayed. Even if the previous update was 450 followed by 353 then that may not be a significant slow down, it may be an effect of going outside the bounds of the smoothing Kalman filter. What is needed is the raw SSR and ADS-B data received (and required to be retained for 28 days by ICAO). What you may be seeing is a replay of the time the incident involved using standard replay tools that are used for local investigations and training it may even be a replay of the data sent to the controller's display rather than reprocessing the input data.
I would not impugn the ATC by claiming they were producing made up 'eye candy'.