Do Annual Geopotential Variations Affect IGS Products ?

1
Do Annual Geopotential Variations Affect IGS
Products ?

• Test effect of GRACE RL05 annual model fits from
  CSR
• consider terms (2,0), (2,1), (2,2), (3,1)
• Compare GPS results for two extreme weeks
• 1668 25 - 31 Dec 2011
• 1694 24 -30 Jun 2012
• Impacts at levels up to several mm
• Other ACs should test consider using in Repro2

J. Ray
NOAA/NGS
with major help from S. Bettadpur, J. Ries
U. Texas/CSR T.-S. Bae
Sejong
U. X. Collilieux
IGN/LAREG T. van Dam
U. Luxembourg K. Choi, J.
Griffiths
NOAA/NGS
IGS Workshop 2012, AC Splinter Meeting, Olsztyn,
Poland, 26 July 2012
2
Annual Geopotential Terms Considered
wk 1668
wk 1694

• Pick two extreme weeks 6 months apart for
  testing 1668 1694
• Difference NGS solutions WITH WITHOUT adding
  annual terms

3
Compare Test Orbits
WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668
dX dY dZ RX RY RZ SCL wRMS Medi
d 0 -0.2 0.6 -1.2 2 9 -2 0.02 4 4
d 1 0.1 -0.2 -0.9 -8 -6 -5 0.01 2 2
d 2 -0.6 -0.1 -1.0 -8 -4 7 0.01 3 3
d 3 0.1 -0.4 -0.6 -4 1 3 0.01 3 3
d 4 0.1 0.0 0.2 2 5 0 0.00 3 3
d 5 -0.2 0.3 -1.3 9 -7 9 0.01 3 3
d 6 -0.6 0.3 -0.5 3 -7 0 0.00 4 4
units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm units mm, mm, mm, µas, µas, µas, mm, mm
WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694
dX dY dZ RX RY RZ SCL wRMS Medi
d 0 0.2 -0.2 -0.7 2 10 0 0.00 2 2
d 1 0.4 -0.2 -0.9 -4 6 2 0.01 2 2
d 2 -0.1 0.0 -0.9 1 1 -7 0.00 2 2
d 3 -0.2 0.3 -1.0 -2 -1 -2 0.01 2 2
d 4 0.0 0.2 -0.2 0 0 -4 0.00 2 2
d 5 0.1 -0.1 -0.4 0 1 -1 0.01 2 2
d 6 0.0 0.2 -0.9 -2 7 2 0.00 3 3
units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm units mm, mm, mm, µas, µas, µas, ppb, mm, mm
4
Compare Test Terrestrial Frames
WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668 WITH ? WITHOUT Differences Wk 1668
dX dY dZ RX RY RZ SCL wRMS
offsets -0.61 -0.17 -1.32 5.2 2.2 -2.4 0.002 0.45
0.02 0.02 0.02 0.7 0.7 0.6 0.003
units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations units mm, mm, mm, µas, µas, µas, ppb, mm 228 stations
WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694 WITH ? WITHOUT Differences Wk 1694
dX dY dZ RX RY RZ SCL wRMS
offsets 0.24 0.10 -0.83 3.4 -3.4 0.5 0.007 0.29
0.01 0.01 0.01 0.4 0.5 0.4 0.002
units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations units mm, mm, mm, µas, µas, µas, ppb, mm 253 stations

• Orbit TRF frames both shift by about -1 mm in Z
  component
• probably due to N/S network asymmetry
• recall that current IGS Z bias wrt SLR origin is
  10 larger
• global WRMS impact on stations positions at level
  of 0.5 mm

5
Distribution of dU Shifts
Week 1668 (25-31 Dec 2011)
TASH
- (IGS-load)
6
IGS Repro1 Residuals (TASH Loads)

• TASH heights are too low each December
• annual geopotential effect might partially
  compensate ?

7
Distribution of dU Shifts
Week 1694 (24-30 Jun 2012)
Sometimes regions of good correlation
- (IGS-load)
8
Distribution of dN Shifts
Week 1668 (25-31 Dec 2011)
- (IGS-load)
9
Distribution of dN Shifts
Week 1694 (24-30 Jun 2012)
- (IGS-load)
10
Distribution of dE Shifts
Week 1668 (25-31 Dec 2011)
- (IGS-load)
11
Distribution of dE Shifts
Week 1694 (24-30 Jun 2012)
But also sometimes areas of poor correlation
- (IGS-load)
12
Compare Test ERPs
WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668 WITH wrt WITHOUT Annual Terms Wk 1668
Xpole Ypole Xprate Yprate LOD
d 0 2.9 -0.3 42.5 -21.7 -4.14
d 1 -3.0 -6.1 -6.8 34.7 -2.59
d 2 -2.2 -6.2 -3.3 -13.0 -2.22
d 3 -3.3 -1.3 18.3 -9.7 -0.56
d 4 6.4 -3.6 -33.0 -23.3 -1.60
d 5 -12.0 10.0 51.6 -32.8 -3.21
d 6 -10.7 3.5 -18.4 -0.1 -0.50
units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs
WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694 WITH wrt WITHOUT Annual Terms Wk 1694
Xpole Ypole Xprate Yprate LOD
d 0 11.7 1.7 4.4 -2.1 -1.14
d 1 6.6 -10.0 5.3 0.3 -3.14
d 2 -0.6 3.3 -64.3 14.2 -3.46
d 3 -2.3 -3.2 -47.9 -44.2 -0.60
d 4 -0.2 -0.2 -13.7 19.6 -1.31
d 5 5.1 0.9 -17.5 1.8 -2.29
d 6 8.6 0.5 6.5 -25.7 -1.39
units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs units µas, µas, µas/d, µas/d, µs
13
Conclusions Recommendations

• Annual geopotential variations have small but
  non-negligible impacts for IGS products
• DZ component of orbit terrestrial frames
  shifted by 1 mm
• LOD is biased by few µs
• subdaily orbit residuals differ up to 4 mm WRMS
• station positions shift by up to 0.7 mm
  horizontal, 3 mm vertical, probably seasonally
• systematic geographic shifts may significantly
  alias inferred GPS load signatures
• however, annual geopotential effect generally
  appears to be smaller than annual (GPS load)
  residuals, esp for dN dE
• Recommend further testing by other ACs
• need longer spans of results further
  comparisons
• Recommend possible adoption for Repro2
• if preliminary NGS results confirmed, IGS should
  consider adopting a conventional model for annual
  geopotential variations for Repro2
• must coordinate with GRACE, SLR, IERS groups
• Srinivas Bettadpur working on GRACE fit to degree
  15

14
Models Used from S. Bettadpur J. Ries (1/2)

• Subject Estimates of non-tidal degree-2 annual
  geopotential variability
• Author Srinivas Bettadpur
• Date June 27, 2012
• Version v 0.0
• The total variability at the annual frequency is
  a sum of many processes. Not all of these are
  included in the estimates here.
• Total_Annual 3rd Body Pert (relevant only for
  orbits) ltlt-- This is NOT included below
• All tides (solid, ocean,
  solidocean pole tide) ltlt-- This is NOT included
  below
• Atmosphere non-tidal oceans
  (AOD1B contents) ltlt-- This is included below
• Everything else left over (GSM
  contents) ltlt-- This is included below
• The estimates for "Everything else left over"
  depends on what was modeled for the parts labeled
  "NOT included below". This list is included
  below
• 3rd Body Pert DE405 for luni-solar positions
• Solid Tide Eq. 6.xx from IERS2010, with
  anelastic earth klm
• Ocean Tide Self-consistent equilibrium
• Solid Earth pole tide IERS C04 pole series
  with an-elastic earth klm
• Ocean pole tide IERS C04 pole series with
  self-consistent equilibrium model of Desai

15
Models Used from S. Bettadpur J. Ries (2/2)

• Table below gives the values of the annual
  amplitudes for all the degree-2 harmonics. The
  GRACEGAC values are labeled as "ANNUAL". For the
  (2,0) harmonic, the SLRGAC based estimates are
  also provided.
• name N M CBAR_cos CBAR_sin SBAR_cos
  SBAR_sin
• 
  
• ANNUAL 2 0 0.1103E-09 0.8033E-10
  0.0000E00 0.0000E00
• SLRGAC 2 0 9.9868E-10 1.1105E-10
  0.0000E00 0.0000E00
• ANNUAL 2 1 0.7377E-11 -.2024E-10
  0.7651E-10 -.2273E-10
• ANNUAL 2 2 -.1394E-10 -.7749E-11
  0.5471E-10 -.4229E-10
• --------------------------------------------------
  -----------------------------------------------
• Subject Re degree-2 annual coefficients
• Date Wed, 27 Jun 2012 154036 -0500
• From John C. Ries ltries_at_csr.utexas.edugt
• Hi Jim,
• I imagine that degree 2 is the 'tall pole' for
  GPS, but I'm curious about the effect of an
  odd-degree order 1 term. I think it will be too
  small for GPS, but it has shown to be important
  for lower satellites. A quick fit to RL05 gets,
  in the same convention as Srinivas
• name N M CBAR_cos CBAR_sin SBAR_cos
  SBAR_sin