Conference PaperPDF Available

LDCmgm90: an improved multiple-data based monthly geopotential model set

Authors:
Conference Paper

LDCmgm90: an improved multiple-data based monthly geopotential model set

Abstract

The paper was published in Scientific Data (https://www.nature.com/articles/s41597-019-0239-7) with the data link https://doi.org/10.6084/m9.figshare.7874384.v1. LDCmgm90 is a monthly geopotential model set complete from degree and order 2 to 90, derived from various GRACE/SLR (Satellite Laser Ranging) monthly geopotential + Various GCM (General Circulation Model) outputs + ERP (Earth Rotational Parameter) measurement constraints using the Least Difference Combination (LDC) method. Therefore, LDCmgm90 has the same functions of the GRACE monthly geopotential model sets as released by different institutes, such as Center for Space Research (CSR), Deutsches GeoForschungsZentrum (GFZ), Jet Propulsion Laboratory (JPL) and Graz University of Technology (TUG), but with a number of defects or limitations in GRACE data removed. For a full description about how LDCmgm90 overcomes the defects in common GRACE data, one can refer to this presentation given at the TibXS2019 meeting.
1. School of G eodesy a nd
Geom at ics, Wu han University,
Wuhan, China
2. NOAA, Silver Spring,
Ma ryland, USA (Retir ed)
3. China Univer sity of
Geosicences, Wuha n, China
Wei Chen1*, J iesi Luo1, Hanwei Zhang1, J im Ray2, Nan Yu3, J ianCheng Li1
Email: wchen@sgg.whu.edu.cn
Introduction to LDCmgm90
What is LDCmgm90? -- Monthly geopotential model
set complete from degree and order 2 to 90
Various GRACE/SLR monthly geopotential + Various GCM outputs + ERP measument constraints
Data Sets Assimilated by LDCmgm90
The GCM outputs are combined to obtain the LDCgam global
angular momenta for atmoshere, oceans and land water (freely
available at https://doi.org/10.13140/RG.2.2.28698.49604) in
advance
Geopotential
GCM outputs
ERP measurements
CSR GRACE RL05 GSM /GAC/GAD
CSR GRACE RL06 GSM/GAC/GAD
NCEP/NCAR reanalyses AAM
ITRF2014 polar motion
CSR GRACE Mascon RL05
CSR GRACE Mascon RL06
ECCO OAM
IERS 14 C04 polar motion and length-of-
day variations
CSR SLR RL05
CSR SLR RL06
GSFC SLR RL06
ECMWF AAM (ESMGFZ ver.;
ERAinterim ver.)
IGS (final) polar motion and length-of-day
variations
GFZ GRACE RL05 GSM/GAA/GAB/GAC/GAD
GFZ GRACE RL06 GSM/GAA/GAB/GAC/GAD
MPIOM OAM (ESMGFZ ver.)
JPL GRACE RL05 GSM/GAA/GAB/GAC/GAD
JPL GRACE RL06 GSM/GAA/GAB/GAC/GAD
OMCT OAM (ERAinterim ver.)
JPL GRACE Mascon RL05M (Ver. 2)
JPL GRACE Mascon RL06
LSDM HAM (ESMGFZ ver.;
ERAinterim ver.)
TUG ITSG-Grace2018 RL06
LSDM SLAM (ESMGFZ ver.)
Why Correcting GRACE Data
Limitations and corrections in GRACE data
Problematic C20 --> Replaced with C20 derived from LDCgam
global angular momenta for atmoshere, oceans and land water
Strong longitudinal stripe-pattern errors --> Using corrected
MASCON grids and numerical integrating them to obtain Stokes
coefficients
Significant power losses near their Nyquist frequency -->
Replaced with refined components from meteorological models
(currently limited to degree-2 terms because we only got
NCEP/NCAR gridded data while those for European climate
models are not freely available)
Excess ±2.3 cpy signals --> Removed by fitting them
Tips on origin(s) of the ±2.3 cpy or 161-day signals
S2 alias
GPS orbital resonance near the S2 period
A significant but unknown excitation that is dependent on the beta-
angle of the GRACE orbit plane with respect to the Sun
161-day variations in atmospheric pressure and winds
Why Correcting GRACE Data (cont.)
Limitations and corrections in GRACE data (cont.)
Two jumps at Jan. 2006 and Jan. 2010 due to updates of
ECMWF model (all RL05 GAA and GRACE/SLR GSM data are
affected) --> Using GSM + GAA + GAB instead of GSM only,
then seperate them into GSM, GAA and GAB again using the LDC
method
Significant errors in RL05 GAB due to mismodelling in the OMCT
model --> Same as above
Long-period pole tides due to the IERS2010 nonlinear mean pole
model (for all RL05 GRACE and SLR data) --> Applying the long-
period pole tide corrections as proposed by Wahr et al. (2015)
Atomospheric, oceanic and hydrolopgical model outputs (used to
generate the RL06 AOD1B) violate the conservation of global
mass, and a sea level mass term is introduced to conserve the
mass; however, only the ocean model outputs are used to
produce the RL06 GAB (namely part of oceanic contributions are
not modeled) --> See above items
Differences between data released by JPL, CSR and GFZ are
notable --> Applying LDC weighting and combination of the
three MASCON solutions
Validations I
LDCmgm90 C20 series is close to SLR C20, but slightly
better in bands >~2cpy due to the assimilation of
variations from AAM, OAM, HAM and SLAM (if any)
GSM on ly
GSM + GAA + GAB
Validations II
C21, C21 and S21 from LDCmgm90, Geophysical
fluid models & GRACE/SLR, respectively
Differences between the observed polar motion excitations and
geophysical excitations estimated from geop hysical fluid
models and LDCmgm90 (denoted as LDCgsm in the figure)
LDCmgm90 agrees better both with EOP measurements and GCM outputs
Validations III
Differences between the geopotential maps for Nov
2010 and Oct 2010
LDCmgm90 (Neither smoothing nor destriping is applied)
CSR GRACE RL06 (Neither smoothing nor destriping is applied)
The maps for CSR, GFZ,
JPL and TUG (only the map
for CSR RL06 is provided
here) have strong stripe-
pattern noises, which
overwhelm any geophysical
signal of interest!!!
TVG Perturbations to Satellite Orbits I
Currently, the GPS (and perhaps other satellites)
orbital perturbation models have not incuded the
effects of time-variable gravity (TVG)
Accurate GPS orbits are of great significance for
positioning
navigation
orbital determinations for other satellites
Recognizing the significance of TVG perturbations,
the IERS and GPS community are calling for a
standard TVG model
TVG Perturbations to Satellite Orbits II
For TVG pertubation with angular frequency σ
Acceleration caused by Rn m
Rn m would cause periodic radial displacement of
the orbit, with amplitude maxima
Lower-degree terms (such as C20) dominate the perturbations!
It implies high-degree TVG
terms would have little (or
no) impacts on orbital
perturbation!!!
Numerical Test for Real GPS Orbits
Orbital difference (unit: km)
Will be serveral times larger
if using the standard way
that the IGS compares
orbits as did by J. Ray
(2012)!
Orbital differences would
be much larger for lower-
orbit satellites, such as
GRACE, Etalon-1 and -2!
Conclusions
LDCmgm90 is reliable and better than individual
GRACE/SLR models
Reliable C20 derived from LDCgam
No longitudinal stripe-pattern errors
No power losses near their Nyquist frequency
Excess ±2.3 cpy signals removed
No jumps at Jan. 2006 and Jan. 2010
Consistent with the new IERS linear mean pole model
TVG has non-negligible impacts (a few mm) on GPS
orbits
The TVG perturbations to lower-orbit satellites
would be much larger and should be considered
Least Difference Combination (LDC)
CHARACTERS OF LDC
A frequency-domain combination method proposed by Chen et al.
(2013b) and extended by Chen et al. (2017)
Can handle both the magnitude and phase of each frequency
component simultaneously
Can handle the trends and very-low frequency components of
series (we need not to remove data trends)
Article
Accurate determination of the Chandler wobble (CW) period (Tcw) and quality factor (Qcw) is of great significance to our understanding of the Earth's dynamic figure parameters, elasticity, rheology, and energy dissipation. Tcw and Qcw were typically determined in the time domain using the digital filter designed by Wilson (1985); however, we developed an alternative method to estimate Tcw in the frequency domain. We adopted the frequency domain expression solving the Liouville equation for polar motion (equation (3) in the following) rather than the time domain to separate the free-damping CW and excited parts. Next, we substituted various excitation functions derived from the outputs of several general circulation models and selected monthly gravity models into the above frequency domain expression; hence we estimate Tcw. The preferred Tcw value using this method and the least difference combination mgm90 model is 430.4 ± 2.0 mean solar days. Our results correlate with previous studies within the error range, providing an independent way of estimating Tcw.
Article
We carry out a comprehensive analysis and assessment of degree-2 gravitational changes ΔC21, and ΔS21, estimated using the Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-On (GFO), satellite laser ranging (SLR), Earth Orientation Parameters (EOP), and geophysical models over the period April 2002–February 2020. The four independent estimates of ΔC21 and ΔS21 variations agree well over a broad band of frequencies. The GRACE/GFO Release 6 (RL06) solutions show major improvements over the previous RL05 solutions at both seasonal and intra-seasonal time scales, when compared with EOP and SLR estimates. Among the four independent estimates, highest correlation coefficients and smallest RMS residuals are found between GRACE/GFO and EOP estimates of ΔC21 and ΔS21 variations. GRACE/GFO and EOP ΔC21 and ΔS21 estimates exhibit slightly different trends, which are related to the implementation and interpretation of the pole tide correction in GRACE/GFO data processing. This study provides an important early validation of GFO ΔC21 and ΔS21 solutions, especially the new pole tide correction applied in GRACE/GFO RL06 solutions using independent estimates.
ResearchGate has not been able to resolve any references for this publication.