Node Definitions

Probably a bit late for this question, but here goes:

How you define the event associated with certain nodes can significantly impact the location of those nodes. For instance, the ID node after NS station-keeping appears to be placed roughly where the “inclination range” has been exceeded, but based on my analysis of certain objects, that range limit does not appear to be defined by the most recent peak, or the first peak, or the average peak, or the max peak. This choice causes variation of the node location well beyond the 6-step tolerance.

Are definitions of what constitutes the event for each node type available? I didn’t see that in the paper on the dataset.


Hi Jeff,

NS ID node placement is highly dependent on the preceding SK behavior, but the majority of cases are either driven by an EW ID node, in which case the NS ID node is placed at the same timestamp, or determined by when the NS SK pattern is broken. If the NS SK is time-based (i.e. maneuvers every 14 days) instead of triggered by an inclination threshold, then the ID node wouldn’t necessarily align with a particular inclination and should be placed at the timestamp when the next inclination correction would have been observed had the SK pattern continued. This convention reflects the earliest time that a behavior change could be detected if this analysis were to be performed in real time.

Hope this helps!

1 Like