****************************************************************************** dorisreport 30-Sep-2013 16:41:46 Message No 3453 ****************************************************************************** Author: "Lemoine, Frank G. (GSFC-6980)" Subject: Delivery of gscwd20 SINEX series (1993 - 2013.0) to IDS Data SUMMARY OF GSC DORIS SUBMISSION for 1992-2013 wd20 SINEX files September 30, 2013 We have delivered 1070 SINEX files to the IDS data centers at the NASA CDDIS and the IGN. These SINEX files represent a new series (wd20) spanning the weeks from 1993/003 to 2012/365. This series is a new baseline to support the development of the IDS contribution to ITRF2013. This series contains the following changes w.r.t. the previous series (wd18). 1. Explicit modelling of the pole according to the IERS2010 standards (Table 7.7, pp 115, IERS Conventions (2010), Petit and Luzum. 2. C21, S21 are modelled in the long term according to equation 6.5 of the above document. 3. A new DORIS station set was adopted (DPOD2008_1_12). This updates some of the stations, and adds new stations towards the end of the series. Consequently, compared to wd18, the newer DORIS stations not included in wd18 are now part of the wd20 processing. 4. In wd18, we identified stations which had less than 250 observations and removed them at the normal equation stage - however they had remained as part of the POD processing. In wd20, we explicitly deleted stations from the POD analyses that had less than 250 observations in any given week. In our experience, stations with too few observations were invariably associated with weak coordinate solutions. 5. A timing bias model was added for the DORIS processing of TOPEX, derived from the SLR-DORIS analyses (wd18). 6. Due to a discrepancy in the attitude law for Cryosat-2 in the previous version of GEODYN we now use the quaternions supplied by E.J.O. Schrama (TU Delft). We continue to use the CNES 7plate macromodel for Cryosat-2. 7. For SPOT-5, from 2006.0 to 2013.0, the SAA-corrected data were used, instead of the standard DORIS data. No deweighting scheme was applied to the SPOT5/SAA stations, however Cacheoira (CACB, CADB) was explicitly deleted from the SPOT-5 POD processing, since this was the station whose data were most severely disrupted by the SAA. 8. More frequent Cd (drag) adjustments were applied for certain time periods, as per Laurent Soudarin from the DORIS AWG Meeting Toulouse (April 2013). This affected SPOT-2, SPOT-3, SPOT-4, SPOT-5, Envisat and Cryosat-2. 9. It was ascertained that the solar array pitch of SPOT-5 was modified on almost a monthly basis after March 2012. These modifications were explicitly incorporated into the SPOT-5 processing after March 2012. 10. The DORIS data arcs were editted, truncated or recast in places to eliminate large data gaps at the beginning, middle, and end of the arcs. 11. GOCO2S is now applied only to 150x150 instead of 180x180. 12. For the wd18 series (and all previous series) the DORIS data in GEODYN used a POD sigma of 2 mm/s. At the normal equation level all satellite contributions in any week had unit weight. This resulted in overly pessemistic sigmas for the weekly station adjustments. Here in the wd20 series, we reweight the normal equations based on 1.25 * the average RMS of fit (computed over all the arcs). So in wd20, the normal equation scale factors by satellite are: TOPEX 9.6898 SPOT-2 11.5398 SPOT-3 10.2811 SPOT-4 11.8906 SPOT-5 13.5913 ENVISAT 9.9986 CRYOSAT2 12.755 JASON-2 17.6356 With respect to wd18, we note that the following features remain part of wd20. 1. As with wd18: GEODYN now explicitly accounts for the DORIS beacon frequency changes at the DORIS sites from the nominal value reported in the DORIS2.2 files. 2. The macromodels for the DORIS satellites are as described for the wd18 series, and presented at the DORIS AWG meeting (Toulouse, April 2013) in the GSC Analysis Center Report, and have not been changed. 3. We use the DORIS CoM/Offset corrections supplied on the DORIS2.2 data. 4. The Troposphere refraction Corrections rely on the GPT model for a priori meteorological data, and the GMF mapping function. Only a correction for the wet delay is adjusted on a pass-by-pass basis. 5. With the exception of C21, S21. The field to 4x4 is represented as a series of terms including secular rates, annual, and semiannual terms. These are derived from a fit to an SLR-DORIS weekly solutions for the time-variable gravity field of the Earth (1993-2012). (cf. Lemoine et al., 2011). 6. The annual variations in the geopotential (other than 4x4) to 20x20 are modelled from a multi-year average of solutions to GRACE data. --- Special Note: We have expedited the production of this series (wd20) to support the DORIS AWG meeting in Washington (October 2013), and the preparations for ITRF2013. If further analysis detects any glitches in any particular weeks, we will resubmit and update the SINEX's in question. --- Contact information for questions Frank Lemoine, GSFC/NASA Greenbelt, MD USA frank.g.lemoine@nasa.gov Douglas Chinn, SGT Inc., Greenbelt MD USA douglas.s.chinn@nasa.gov Please do not reply directly to this message, but send comments and suggestions to IDS.central.bureau@cls.fr