IGS LEO CHAMP precision estimates

Table 1 below shows the pairwise RMS of SLR residuals as obtained from the separate orbit solutions. The solutions have been ordered according to their RMS of SLR residuals, which is reflected by the increase of the pairwise RMS from the left column to the right column. The grey values have been excluded from the analysis for various reasons, typically because the corresponding value of D falls outside a reasonable noise band. This suggests that the two involved orbit solutions are not sufficiently independent, namely, the pairwise RMS of SLR residuals is larger than what the RMS of orbit differences indicates. The most reasonable explanation for such a situation is that the two involved solutions sontrain a common orbit error signal, which will be removed from the difference signal between the two (i.e. the orbit comparison results) but not from the SLR residuals.

The ratios D show a reasonably constant value, the mean value from this table is 1.65 at a standard deviation of 0.15 (9%). Applying this constant value D to the separate solutions results in the absolute orbit error estimates of Table 2 below.

At present, all orbits with an estimated precision better than 10 cm RMS are considered adequate for the IGS LEO Pilot Project analysis. This value of 10 cm has been asopted only as a working value, the actual analysis must show if this is really precise enough (to compare, the typical accuracy of IGS tracking station coordinates is better than 1 cm, i.e. an order of magnitude more precise). As can be seen in Table 2, the two IGS Analysis Centers that had early access to the CHAMP data - GFZ and JPL - have reached this precision level. It is reasonable to expect that AIUB and ESOC may reach this precision about 1.5 year later, corresponding to the period between launch of the CHAMP satellite and the release of the GPS data.