Contribution of Galileo observations to improve the quality of daily and sub-daily earth rotation

parameter estimates

Dzana Halilovic, Robert Weber

TU Wien, Department of Geodesy and Geoinformation, Higher Geodesy

EGU GENERAL ASSEMBLY 2020, 04 – 08 MAY 2020

Old vs. New Dataset – Station selection

2

• Stations: max 208 possible(approx. 191 per day)

• IGS: 174• MGEX: 34

• NNR constrain: 75• With ~16 Galileo + ~30 GPS satellites per

day, the contributional part of Galileo observations amounts max ~10%

• Stations: max 176 possible(approx. 165 per day)

• IGS: 176• MGEX: 104 (up to 55 per day)

• all 104 MGEX stations overlap with IGS list(if MGEX is not available for a day, IGS isused instead if available)

• NNR constrain: 75 (updated station list)• Contributional part of Galileo observations

amounts max ~24%

old IGSold MGEX

new IGSnew MGEX

04

.05

.202

0

Dza

na

Halil

ovic

Campaign Erpnet

Software Bernese v5.2Processing Period Jul - Dec 2017 (doy 184 – doy 363) → Jan 2017 – Jun 2018 (doy 002 2017 – 180 2018)Type of Solution 1-Day/3-Day Solution (1 hour time resolution)Satellite system GPS-Galileo/GPSObservations Phase and Code A priori Orbits and EOP ESOC Multi-GNSS Final ProductsStation Position and Station Velocities ITRF2014Absolute Antenna Model IGS14Station Network 191 Site observations daily (NNR → 75 stations) → 165 Site observations daily (NNR → 75)Processing Mode Double DifferencesAmbiguity Resolution QIF & WL/NLEarth‘s Gravity EGM2008_SMALL

Planetary Ephemerides DE405A priori Solar Radiation Pressure C061001 (Code Model COD9801, Springer et al. 98)Subdaily Pole Model IERS2010XY (based on Ray 1994, XY - values)Nutation Model IAU2000R06Solid Earth Tide Model TIDE2000 (IERS2000)Ocean Tides OT_FES2004 Site - specific Correction for Ocean Tidal Loading

FES2004

Solar Radiation Pressure model 9 Par. ECOM Model : 4 parameters constrained, 5 ECOM parameters fitted (D0, Y0, B0, Bc, Bs)

Old vs. New Dataset – Characteristics

04

.05

.202

0

Dza

na

Halil

ovic

3* Green rows indicate updates in the dataset: old setup written in grey, new setup written in green

04

.05

.202

0

Dza

na

Halil

ovic

4

Polar motion time series (X pole)

(Jan 2017 - Jun 2018) w.r.t. IERS2010

Figure: Xp – IERS2010

GNSS GPS/GAL

(mas)

GPS

(mas)

MAX (abs.) 0,70 1,31

STD (+/-) 0,20 0,27

MEAN (abs.) 0,16 0,21

MAX (abs.) 0,65 1,28

STD (+/-) 0,20 0,28

MEAN (abs.) 0,16 0,21

Statistics

OLD DATASET

NEW DATASET

Some remaining signalsin series

04

.05

.202

0

Dza

na

Halil

ovic

5

Polar motion time series (Y pole)

(Jan 2017 - Jun 2018) w.r.t. IERS2010

Figure: Yp – IERS2010

GNSS GPS/GAL

(mas)

GPS

(mas)

MAX (abs.) 0,73 1,48

STD (+/-) 0,21 0,29

MEAN (abs.) 0,17 0,21

MAX (abs.) 0,74 2,55

STD (+/-) 0,21 0,53

MEAN (abs.) 0,17 0,35

OLD DATASET

Statistics

NEW DATASET

04

.05

.202

0

Dza

na

Halil

ovic

6

LOD time series

(Jan 2017 - Jun 2018) w.r.t. IERS2010

Figu

re:

LOD

–IE

RS2

01

0

GNSS GPS/GAL

(ms/day)

GPS

(ms/day)

MAX (abs.) 0,66 0,99

STD (+/-) 0,18 0,23

MEAN (abs.) 0,14 0,18

MAX (abs.) 0,57 0,97

STD (+/-) 0,17 0,23

MEAN (abs.) 0,13 0,18

Statistics

NEW DATASET

OLD DATASET

Figure: Raw LOD

• LOD timeseries show satisfying agreementw.r.t. IER2010 in both solutions

• Smaller noise shown in combined solution

04

.05

.202

0

Dza

na

Halil

ovic

7

Amplitude spectrum

Area ofinterest

Artefacts fromdata sampling

Figure: X,Y-Pole amplitude spectrumGPS (blue) vs. GPS/GAL (red)

• High peak in prograde diurnal band• Artefacts from data sampling visible• Improvement in noise level when using

GPS/GAL compared to GPS

Figure: LOD amplitude spectrumGPS (blue) vs. GPS/GAL (red)

• Combined spectrum shows smaller noise floor than GPS only

• Artefacts still remain

04

.05

.202

0

Dza

na

Halil

ovic

8

Estimation of Tidal Wave Corrections

(July 2017- Dec 2017) – old dataset

2Q1 σ1Q1

RO1

M1

O1

T01

χ1

π1P1

S1

K1

ϕ1ψ1TT1

J1SO1

OO1

ν1

2N2

μ2

N2

ν2M2

λ2

L2

R2

S2

T2

K2

0

5

10

15

20

25

30

35

40

45

29

,07

28

,01

27

,85

26

,87

26

,72

25

,82

25

,8

24

,97

24

,85

24

,83

24

,13

24

,07

23

,94

23

,93

23

,87

23

,21

23

,1

22

,32

22

,3

21

,58

13

,13

12

,87

12

,68

12

,66

12

,63

12

,45

12

,42

12

,4

12

,19

12

,19

12

11

,97

11

,97

11

,75

11

,55

Resid

ual

am

plitu

de (

µas)

Period (h)

X Amplitude values

3D GPS/GAL - IERS

3D GPS - IERS

2Q1 σ1Q1

RO1

O1

T01

M1

χ1P1

S1π1

K1

ψ1ϕ1

TT1

J1SO1

OO1

ν1

2N2

ν2

μ2

N2M2

L2

λ2T2

S2

K2

R2

0

5

10

15

20

25

30

35

40

45

29

,07

28

,01

27

,85

26

,87

26

,72

25

,82

25

,8

24

,97

24

,85

24

,83

24

,13

24

,07

23

,94

23

,93

23

,87

23

,21

23

,1

22

,32

22

,3

21

,58

13

,13

12

,87

12

,68

12

,66

12

,63

12

,45

12

,42

12

,4

12

,19

12

,19

12

11

,97

11

,97

11

,75

11

,55

Resid

ual

am

plitu

de (

µas)

Period (h)

Y Amplitude values

3D GPS/GAL - IERS

3D GPS - IERS

X,Y – pole residual amplitudesCorrections w.r.t. IERS up to:• 30μas from GPS/GAL series• 40μas from GPS seriesMajor constituents in thediurnal band: 2Q1, O1, M1 , OO1, ν1Major constituents in thesemidiurnal band: 2N2, μ2, L2, S2

LOD residual amplitudesCorrections w.r.t. IERS up to:• 25μs/day from GPS/GAL• 30 μs/day from GPSMajor constituents in the diurnal band: M1 , OO1Major constituents in the semidiurnal band : M2, S2, K2

04

.05

.202

0

Dza

na

Halil

ovic

9

Conclusions

• New dataset (Jan 2017 – Jun 2018)• GPS/GAL combined data processing has clearly improved the estimated ERPs, i.e. overall

better agreement with IERS2010 model compared to GPS only

• Old dataset (Jul 2017 – Dec 2017)• Improvement of tidal wave amplitude corrections determination using GPS/GAL

• New dataset vs. Old dataset• Similar ERP statistics when comparing the two GPS/GAL solutions• Visible improvement in the GPS only solution when comparing the ERP statistics of the

new and old dataset

Recommendation: Processing of longer ERP time series based on combined GPS/GAL data

Future task: estimation of tidal wave corrections of the new dataset in order to analyze thepossibilities of an improvement when using longer time series and an enlarged number of MGEXstations compared to the old dataset

04

.05

.202

0

Dza

na

Halil

ovic

10

References

• Petit, G. & Luzum, B. 2010, IERS Conventions 2010, ed. G. Petit & B. Luzum (IERS Technical Note No. 36)

• Rolf Dach, Simon Lutz, Peter Walser, Pierre Fridez: Bernese GPS Software Version 5.2, November 2015

• D. Horozovic, R. Weber, "Bestimmung von hochfrequenten Erdrotationsparametern unter Verwendung von GPS und Galileo Beobachtungsdaten"; Talk: Geodätische Woche 2018, Frankfurt; 2018-10-16 - 2018-10-18; in: "Abstract Book Geodätische Woche 2018", Abstract Book Geodätische Woche 2018, (2018)

Network station RINEX data: • Igs (International GNSS Service):

ftp://igs.ensg.ign.fr/pub/igs/data/campaign/mgex/daily/rinex3/• CDDIS (Crustal Dynamics Data Information System):

ftp://cddis.gsfc.nasa.gov/pub/gps/data/daily/• ESOC MulltiGNSS Orbits and EOPs:

http://navigation-office.esa.int/products/gnss-products/