NATIONAL REPORT OF THE FEDERAL REPUBLIC OF ... - IAG Office

66 Commission 2 – Gravity Field
gravity field solutions. The methodology to correct the
atmospheric non-tidal mass variations was revisited by
PETERS (2007). Non-tidal oceanic mass variations of the
latest releases of AOD1B are derived from output of the
baroclinic ocean model OMCT (Ocean Model for Circulation
and Tides) of the Technical University of Dresden
which is based on meteorological ocean surface forcing and
precipitation and evaporation data. DOBSLAW and THOMAS
(2006) showed that the impact of river run-off on global
ocean mass redistribution (un-modelled in OMCT) can be
neglected.
Monitoring the Continental Hydrological Cycle
Since during the processing of the GRACE mission data
to monthly gravity field solutions known tidal as well as
all short-term atmospheric and oceanic mass variations are
taken into account, time-variable gravity signals derived
from time series of GRACE-only gravity models mainly
reflect mass redistribution at the Earth’s surface due the
continental hydrological cycle. This was verified in studies
on global scales (RAMILLIEN et al., 2005a, 2005b; Schmidt
et al., 2006a, 2006b; GÜNTNER et al., 2007) as well as on
regional scales such as the monitoring of time variations
of regional evapotranspiration rates (RAMILLIEN et al.,
2006). In this way, GRACE-derived changes in surface
mass anomalies can be expected to contribute to the quantification
of the total water budget, which is an obviously
underestimated quantity as indicated by the GRACEderived
amplitudes of annual and semi-annual signals being
larger than predicted by global hydrological models. More
recently RAMILLIEN et al. (2007) have analysed seasonal
but also interannual change in land water storage over 27
large river basins from GRACE data and found significant
negative trends for some of the largest basins indicating
water mass loss over the investigated time period. NEU-
MAYER et al. (2006) showed a high correlation when
combining temporal gravity variations resulting from
superconducting gravimeter recordings, GRACE monthly
gravity field solutions and global hydrology.
To extract hydrological (and other geophysical) mass
variability from monthly GRACE gravity field solutions
special smoothing techniques have to be applied to the nonphysical
meridional-oriented striping in the GRACE geoids
and to avoid leakage from neighbouring basins or from the
ocean. To this end, MARTINEC et al. (2007) performed a
statistical analysis of the temporal variability of the GRACE
Stokes potential coefficients and Schmidt et al. (2007) made
an accuracy assessment for GRACE derived time variable
gravity field solutions. KUSCHE (2007) suggested an
approximate decorrelation and non-iso-tropic smoothing
of time-variable GRACE-type gravity field models. HOR-
WATH and DIETRICH (2006) estimated errors of regional
mass variations inferred from monthly GRACE gravity field
solutions. As an alternative to the concept based on spherical
harmonics FENGLER et al. (2005, 2007) and SCHMIDT
et al. (2006) calculated regional high-resolution temporal
GRACE gravity models using spherical wavelets. SNEEUW
et al. (2003) investigated the space-wise, time-wise, torus
and Rosborough representation in gravity modelling. In
SASGEN et al. (2007) a method based on Wiener filtering
applied for an optimized estimation of secular trends over
Antarctica.
GRACE Oceanic Applications
It has been shown by various authors that GRACE gravity
field time series also trace mass-induced gravity variations
over the oceans. For example, KANZOW et al. (2005) have
intercompared global patterns of ocean mass signals based
on early GRACE-only gravity field series provided by GFZ
and CSR with in-situ ocean bottom pressure data from a
ground truth site in the tropical northwest Atlantic Ocean
and the ECCO ocean model. The study indicated a general
agreement between these independent data sources but also
showed remaining deficiencies in the GRACE data processing
and suggested, among others, the substitution of the
non-tidal barotropic ocean model by a baroclinic one. On
a regional scale FENOGLIO-MARC et al. (2006) calculated
mass variations in the Mediterranean Sea from analysis of
hydrology corrected GRACE data and found reasonable
agreement with altimetry-based estimates corrected for the
steric part. LOMBARD et al. (2006) estimated steric sea level
variations from a combined GRACE and Jason data analysis
and found an overall good agreement. The net effect of the
land water contribution to sea level change was estimated
to be 0.19 ± 0.06 mm/yr which is comparable to the ice
sheet contribution. VINOGRADOVA et al. (2007) investigated
the relation between sea level and ocean bottom pressure
and the vertical dependence of oceanic variability.
Post Glacial Rebound and Ice Mass Loss
Since 2003, absolute gravity measurements have been
performed regularly by the Institute für Erdmessung
Hannover in the Fennoscandian land uplift network covering
Norway, Sweden, Finland and Denmark (TIMMEN et
al. 2005, 2006). In cooperation with the national agencies
and research institutions of the Nordic countries and BKG
in Frankfurt, terrestrial absolute gravimetry is applied to
observe the postglacial land uplift due to the isostatic
adjustment of the crust. Nearly all absolute stations are colocated
with continuously observing GPS stations. From
the comparisons between the participating instruments, an
overall accuracy of ±30 nm/s2 is indicated for a single
absolute gravimeter and a single station determination.
Thus, the gravity change due to the land uplift may be
observed with an accuracy of ±10 to 20 nm/s² for a 5-year
period. One purpose of these terrestrial in-situ observations
is to validate the GRACE results (ground-truth) and first
promising results have been presented in MÜLLER et al.
(2003, 2005, 2007a, 2007b). In the same context, WIEHL
et al. (2006) showed how the Baltic Sea water mass variations
mask the postglacial rebound signal in CHAMP and
GRACE gravity field solutions.
Predicted changes in the geoid about Greenland to be used
for GRACE validation have been described by FLEMING
et al. (2005). SASGEN et al. (2005) described signatures of
glacial changes in Antarctica, namely rates of geoid height
change and radial displacement due to present and past ice
mass variations. These are more or less due to changing ice
mass balance and ice dynamics and shall be detectable by