Date: Thu, 17 Jun 2004 From: "John C. Ries" Subject: updated coordinates for BELB, CROB, JIUB and others Hi Everyone, This is just a summary of what I have been looking. A few sites look like they might need improvement (both new and some older ones), but the agreement on the update is not always good (or the update doesn't perform better than before). I mapped Pascal's estimates to 1997.0, using the ITRF2000 velocities (or the plate velocities given below) and tested them with Topex (using GGM01S as well as JGM3), comparing them to the original coordinates announced in the DorisMails. Note that this test is only based on T/P and a few cycles. Still, if a coordinate is a significant improvement, it should show up as such. If the change is not significant, then there may be no reason to change the nominal estimates (yet). Note that since I used the same velocity to map back to 1997 and then forward to April 2004, the test is insensitive to whatever velocities I actually used. JIUB cannot be tested with TOPEX...no data yet. I haven't seen any WETB data either. BELB is improved but the original estimates had a large uncertainty. The rest do not seem to perform better for T/P than the originals. I'm speculating that these solutions, with the amount of data available, probably have a true error of 2-3 cm (as compared to the formal errors quoted). Many of the adjustments were of the same size, and I don't see anything to suggest that they are sufficiently better to justify a change (except for BELB) from the original announced coordinates. I include below a few sites that appear to need adjustment, and the results of various attempts to do so. The last one in any sequence is the one I'm actually using at the moment. # # "General Belgrano" Argentine base in Antarctica (2/6/2004) DOMES # 66018S001 1997.0 BELB 4296 1106057.90 -763816.80 -6214233.61 13 cm uncertainty 1997.0 BELB 4296 1106057.9096 -763816.8156 -6214233.3913 Willis (2004) 1997.0 BELB 4296 1106057.8750 -763816.8299 -6214233.4970 Lemoine 2004 (1 cm uncertainty) Site STAVEL4296 21.3 -0.2 -31.7 ITRF2000 (from BELG) Frank Lemoine and Pascal agree to 1-3 cm in Y and X but differ in Z by 10 cm. Due to the latitude of this site, T/P cannot can it tell if one is better than the other. # # Yaragadee, Australia (new antenna Nov. 27, 2003) DOMES # 50107S011 1997.0 YASB 4292 -2389016.44 5043346.16 -3078493.70 Geosciences Australia (1 cm uncertainty) 1997.0 YASB 4292 -2389016.45 5043346.30 -3078493.73 Lemoine (2004) 1997.0 YASB 4292 -2389016.46 5043346.20 -3078493.68 Willis (2004) 1997.0 YASB 4292 -2389016.46 5043346.20 -3078493.74 Ries (2004) Site STAVEL4292 -46.98 7.87 48.84 ITRF2000 The agreement in X is good, but the scatter in Y and Z is large. Maybe there's something off with this station that makes various estimates disagree so much. I can't really say which of these is better based on the Topex fits. # # Sal, Cape Verde (new station replacing Dakar December 12, 2002) DOMES # 39601S002 1997.0 SALB 4286 5626877.20 -2380936.32 1824490.40 IGN (5 cm uncertainty) 1997.0 SALB 4286 5626877.22 -2380936.38 1824490.34 Ries (2003) 1997.0 SALB 4286 5626877.24 -2380936.40 1824490.36 Lemoine (2004) 1997.0 SALB 4286 5626877.24 -2380936.39 1824490.35 Pascal Willis (2003) Africa STAVEL4286 3.81 21.21 15.81 NUVEL NNR Here the agreement is excellent and any one of the updates should be good. # # Mahe, main island of the Republic of Seychelles DOMES # 39801S005 1997.0 MAHB 4274 3597814.60 5240899.67 -516791.42 IGN (2 cm uncertainty) 1997.0 MAHB 4274 3597814.60 5240899.58 -516791.39 Willis (2004) 1997.0 MAHB 4274 3597814.59 5240899.62 -516791.43 Lemoine (2004) 1997.0 MAHB 4274 3597814.54 5240899.52 -516791.41 Ries (2003) 1997.0 STAVEL4274 -20.60 24.80 11.8 ITRF2000 My estimate based on Topex-only unsurprisingly has the best Topex fit, but it has the largest deviation in X and Y. Maybe Topex wants to adjust the height more than it should. velocities estimates were provided by Herve Fagard in the DORISMails #306 (BELB), and #283 (JIUB) but not for CROB. Of recent sites, GAVB and CROB had no velocity, so I adopted the plate velocity (velocity is in mm/yr) 1997.00 GAVB 4291 4783636.527 2140697.655 3623255.283 2 cm uncertainty Eurasian STAVEL4291 -1.0 -2.2 2.6 NUVEL NNR 1997.0 CROB 4294 2719992.090 3463401.87 -4598608.19 3 cm uncertainty Antarctic STAVEL4294 -5.28 6.60 1.85 NUVEL NNR Time series will not depend on the apriori velocity, but a velocity is needed for POD purposes, so I propose these as nominal velocities for these sites. They should be adequate until the velocities can be accurately determined from the tracking. These seem to be performing reasonably well for Topex. Regards to all, JR