Jonathan T. Wojack asks >Why does SatEvo always have a lower magin of error (+/-) than SpaceCom? >Is it just how the program SatEvo works, or is it than SatEvo has >historically been more precise than SpaceCom? My estimate of uncertainty over the last 10-15 days is usually based on a very simple formula which is of the form L = remaining lifetime (from epoch of latest elset) Uncertainty = (0.2 * L) + (0.01 * L * L) For some objects (eg highly eccentric decayers or others for which the elsets appear uncertain) or at some times (eg during unsettled geomagnetic activity), I may use my judgement to increase the "0.2" to "0.3" or more. Also, for estimates very close to decay (last few orbits) I include factors such as "is this more likely to decay close to perigee, to an equator crossing or anywhere around its orbit?". The "uncertainty" is based on my analysis of almost 2,000 published SatEvo decay estimates, and corresponds to almost twice the standard deviation in the observed accuracy of the predictions. I hope that the actual decay to fall within the "window" more than 80% of the time, at least when the window is more than one or two orbits wide. I cannot say how SpaceCom sets its own uncertainty figures in its Decay Warnings, though IMO these SpaceCom predictions are no more inaccurate than my SatEvo ones, and are often better. Unfortunately, this cannot be said of the decay predictions in the "60-Day Decay Forecast Report" issued by NASA's GSFC and available via OIG. Surprisingly, I believe the latter originates at SpaceCom, too; someone could correct me if this is wrong. Alan -- Alan Pickup | COSPAR 2707: 55d53m48.7s N 3d11m51.2s W 156m asl Edinburgh | Tel: +44 (0)131 477 9144 Fax: +44 (0)870 0520750 Scotland | SatEvo page: http://www.wingar.demon.co.uk/satevo/ ----------------------------------------------------------------- Unsubscribe from SeeSat-L by sending a message with 'unsubscribe' in the SUBJECT to SeeSat-L-request@lists.satellite.eu.org http://www2.satellite.eu.org/seesat/seesatindex.html
This archive was generated by hypermail 2b29 : Thu Jun 29 2000 - 15:40:02 PDT