Technical description of solution bkg00003.eops 1. Purpose of solution: Session EOP 2 - Analysis Center: Federal Agency for Cartography and Geodesy (BKG Leipzig) and Geodetic Institute of the University of Bonn 3. Short narrative description of solution: All available Mark III/IV sessions were analysed for this solution except one baseline and small network sessions which are not suitable for reliable EOP determination. Celestial reference frame was fixed to ICRF-Ext1. Terrestrial reference frame ITRF2000 with the associated velocity field is used for station coordinates, reference epoche is 1997.0. 4. Estimated parameters: a. celestial frame: No b. terrestrial frame: No c. Earth orientation: X, Xdot, Y, Ydot, UT1-UTC, UT1dot, dPsi, dEps d. zenith troposphere: 1h piece-wise linear functions, rate constraint generally 50 ps/hour e. troposphere gradient: east and north offset, offset constraint 0.5 mm, rate constraint 0.5 mm/day f. station clocks: 1h piece-wise linear functions, rate constraint generally 5.E-14 g. baseline clocks: set in initial analysis - usually used h. other: No 5. Celestial reference frame: a. a priori source positions: ICRF-Ext.1 b. source positions adjusted in solution: No 6 - Terrestrial reference frame: a. a priori station positions: ITRF2000 b. a priori station velocities: ITRF2000 c. reference epoch: 1997.0 7. Earth orientation: a. a priori precession model: IERS 1996 b. a priori nutation model: IERS 1996 c. Short-period tidal variations in x, y, UT1: Model used according to: John Gipson, VLBI determination of neglected tidal terms in high frequency Earth Orientation Variation. Journal of Geophysical Research, Vol. 101, No. B12, p. 28051-28064 Dec 10, 1996 d. EOP estimation: X, Xdot, Y, Ydot, UT1-UTC, UT1dot, dPsi, dEps each session 8. A priori geophysical models: a. troposphere: hydrostatic component with NMF mapping function b. solid Earth tide: IERS Conventions c. ocean loading: IERS Conventions d. atmosphere loading: none 9. Data type: Group delays 10. Data editing: Editing of outliers during adjustment when necessary. 11. Data weighting: Observations are weighted using std reported in observational files; re-weighting iteration for each session to achieve the chi-square unity. 12. Standard errors reported: Reported formal errors are derived from least-squares estimation propagated from data uncertainties and weighted as discussed in #11. 13 Software: Calc 9.12, SOLVE release: 2001.12.21 14. Other information: Field 18 contains the session code Field 19 contains the duration of experiments