Flechtner F.M., Gruber Th., G?ntner A., Mandea M., Rothacher M., Sch?ne T., Wickert J. (Eds.) System Earth via Geodetic-Geophysical Space Techniques

Подождите немного. Документ загружается.

A 15-Year Reconstruction of Sea Level Anomalies 365

Using global positioning system-derived crustal velocities to estimate rates of absolute sea

level change from North American tide gauge records. J. Geophys. Res. 112, B04409, doi:

10.1029/2006JB004606.

Woodworth PL, Player R (2003) The permanent service for mean sea level: an update to the 21st

century. J. Coastal Res. 19, 287–295.

Part IV

SEAVAR

Continental Water Storage Variations

from GRACE Time-Variable Gravity Data

Andreas Güntner

1 Surface Mass Variations from GRACE

Since 2002, the Gravity Recovery and Climate Experiment (GRACE) satellite

mission monitors the Earth’s gravity ﬁeld. Given the unprecedented accuracy of

GRACE, mass redistributions caused by geophysical and hydro-climatological pro-

cesses above, at and below the Earth surface can be resolved by time-variable gravity

signals of GRACE (Tapley et al., 2004). The mission sensitivity to processes in

the continental water cycle (e.g. Schmidt et al., 2008a), post glacial rebound (e.g.

Tamisiea et al., 2007), changes in the polar ice sheets (e.g. Velicogna and Wahr,

2006) and mass transport in the oceans (e.g. Chambers et al., 2004) has clearly been

demonstrated during the last years. Also, interactions between these subsystems

have been analyzed with the help of GRACE data, such as the effect of continental

water storage changes on sea level rise (e.g. Chambers et al., 2007; Ramillien et al.,

2008a). GRACE thus provides a unique data s et to help understanding mass distri-

butions and mass transport processes in the Earth system at continental to global

scales.

The particular nature of GRACE is that the satellite gravity data integrate over all

mass contributions in the Earth system. Thus, before adequate data on mass varia-

tions in a certain Earth system component are obtained, a key point is to separate the

signal of interest from the other contributions included in the GRACE data. In the

standard GRACE monthly gravity ﬁelds provided by the three processing centres of

the GRACE mission (GFZ, CSR and JPL) (see Schmidt et al., 2008a and chapter

“Improved GRACE Level-1 and Level-2 Products and Their Validation by Ocean

Bottom Pressure” by Flechtner, this issue, for an overview on GRACE processing),

several time-variable gravity effects such as tides of the solid Earth, oceans and the

atmosphere and non-tidal mass variations due to atmospheric and oceanic dynamics

are already reduced. Thus, mainly mass signals from Earth surface processes, in par-

ticular water storage changes in the continental areas including ice caps and glaciers,

A. Güntner (B)

Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences,

Department 5: Earth Surface Processes, Telegrafenberg, 14473 Potsdam, Germany

e-mail: guentner@gfz-potsdam.de

369

F. Flechtner et al. (eds.), System Earth via Geodetic-Geophysical Space Techniques,

Advanced Technologies in Earth Sciences, DOI 10.1007/978-3-642-10228-8_31,



Springer-Verlag Berlin Heidelberg 2010

370 A. Güntner

can be expected to be left in the time-variable gravity ﬁelds. Other mass variation

components that are still contained in these reduced gravity ﬁelds are effects from

post-glacial rebound and from seismic deformations of the Earth crust due to earth-

quakes. Any errors committed in the separation process propagate into errors of the

signal of interest, s uch as continental water storage changes. These errors originate

in particular from uncertainties in the geophysical reduction models (e.g., atmo-

spheric and ocean circulation models) in terms of input data, model structure and

parameters.

Other factors that contribute to the error budget of GRACE data on Earth sur-

face mass variations originate from measurement errors of satellite instruments and

orbit determination, and from leakage errors caused by mass signals outside the

area of interest. The latter error term is again due to the integrative nature of satel-

lite gravity data that capture all masses around the satellites, with the contribution

of individual mass elements decreasing as a function of the squared distance to

the satellites. Usually, gravity ﬁeld models of the Earth are represented in the fre-

quency domain by truncated global spherical harmonics series. A limitation of this

type of representation is that the signal cannot be precisely located in the space

domain, i.e., spatially detailed information is smeared to some extent over the whole

globe.

Thus, to extract mass variations for a particular region of interest such as a river

basin, either alternative GRACE processing techniques that result in regional grav-

ity solutions (e.g., Rowlands et al., 2005) or ﬁlter functions applied to the global

gravity data (e.g., Swenson and Wahr, 2002) are used. These ﬁlters have to be opti-

mized in a way such that they minimize at the same time (1) the contamination

of the river basin signal by disturbing contribution from outside the region (leakage

error) and (2) GRACE measurement errors which increase markedly with the spatial

detail that is to be extracted from the gravity ﬁelds. As a consequence, ﬁlters cannot

follow exactly the boundary of the region of interest as this would involve unac-

ceptably high GRACE errors, but they are gradual in space around the region and

thereby incorporate some signal from the surroundings into the ﬁnal basin-average

estimates of water mass anomalies. A variety of ﬁlter techniques has been devel-

oped during the last years (see e.g. Kusche, 2007 for an overview). The efﬁciency

of a particular ﬁlter type, i.e., high signal-to-noise ratio and low bias of signal ampli-

tudes and phases, varies with the location, shape and with the signal properties in

and around the region of interest. Thus, the optimum ﬁlter type to derive water stor-

age variations for the scale of large river basins varies between the basins (Werth

et al., 2009).

Near-surface mass variations on the continents are not only reﬂected in the

Earth’s gravity ﬁeld as monitored by GRACE but also in solid Earth deformations.

In particular, load variations such as caused by water storage changes induce an

elastic deformation of the Earth surface which can be observed by geodetic GPS

observations. Thus, in principle, an independent observation system exists that could

be compared and combined with GRACE data to estimate large-scale continental

mass variations (e.g., Davis et al., 2004; Van Dam et al., 2007).

Continental Water Storage Variations from GRACE Time-Variable Gravity Data 371

2 Surface Mass Processes: Continental Hydrology

Continental water storage is a key component of the global hydrological cycle. By

balancing precipitation, evapotranspiration and runoff, water s torage has an impor-

tant control over water, energy and biogeochemical ﬂuxes and thus plays a major

role in the Earth’s climate system. In spite of its importance, spatial and temporal

variations of continental water storage were largely unknown at regional and global

scales prior to the GRACE mission because of the lack of adequate monitoring

systems (Lettenmaier and Famiglietti, 2006). While disperse ground-based hydro-

logical monitoring networks exist, point measurements of, e.g., soil water content

or groundwater levels give only local estimates of water storage. It is practically

impossible to upscale these observations to larger scales because of the high vari-

ability of environmental conditions, e.g., in terms of climate, geology, soils and land

cover, which may cause very different water storage dynamics than what is observed

at speciﬁc points. Monitoring from satellites would in principle allow for a large

spatial coverage. However, remote sensing of soil water storage was limited to the

surface soil layer of few centimetres in depth and to areas free of a dense vegetation

cover (e.g., Wagner et al., 2003). For surface waters bodies, satellite altimetry and

imaging techniques could be used to derive variations of water levels and the extent

of inundated areas (e.g., Alsdorf and Lettenmaier, 2003) and water volumes by a

combination of both data types (Frappart et al., 2008). A common limitation of all

ground- and satellite-based observation techniques was that they only gave access

to individual water storage components. Total continental water storage, however,

is composed of storage in the form of snow, ice, soil water, groundwater and sur-

face water in rivers, lakes, wetlands and inundation areas. The contribution of single

storage components to total storage varies considerably between the Earth’s climate

zones (Güntner et al., 2007). Total water storage variations could not be monitored

in an integrative way before GRACE.

Besides water storage, also other components of the continental water cycle are

only sparsely known from observations at large spatial scales. While a global net-

work of gauging stations for river discharge exists (Global Runoff Data Centre.

Koblenz, Germany), many continental areas remain ungauged or no access to recent

data is available due to the decline in station density. Measurements of evapotranspi-

ration from the land surface are even more difﬁcult to obtain. Thus, the description

of water storage and other water balance components at continental scales primarily

came from global hydrological or land-surface models prior to the GRACE mission.

Theses models, however, suffer from important uncertainties in terms of climate

forcing data, model structures and parameters, and the observed river discharge is

usually the only observable to evaluate simulation results (e.g., Hunger and Döll,

2008). Having only this single constraint does not guarantee that the hydrological

processes are correctly represented in the model and that water ﬂuxes and storage

are adequately partitioned between the different components of the water cycle.

This can only be evaluated and improved by additional observation data on other

hydrological state variables and ﬂuxes.

372 A. Güntner

From the absence of a global and comprehensive hydrological monitoring system

and the limitations in quantifying water balance components by observations and

modelling as summarized above, the extraordinary value of GRACE for the ﬁeld

of continental hydrology becomes apparent. GRACE is an unprecedented source of

information to quantify spatio-temporal variations of the continental hydrological

cycle as, for the ﬁrst time ever, water storage changes for large continental areas are

monitored with GRACE.

The ﬁrst about 7 years of the GRACE mission have clearly demonstrated its

capacity to resolve water storage variations (see Schmidt et al., 2008a and Ramillien

et al., 2008a for reviews). With monthly to 10-days time-steps, maps of water stor-

age patterns for the continental areas and time series of area-average water storage

for selected regions and large river basins worldwide have been derived. The reso-

lution and accuracy of water storage change data is in the range of 400–500 km and

some ten mm water equivalent. Errors tend to increase for smaller areas and depend

on various factors as explained above. GRACE water storage data were used in

resolving the water balance equations for other components of the water cycle, such

as evapotranspiration (e.g., Rodell et al., 2004) and river discharge (e.g., Syed et al.,

2005). With the help of ancillary data, also mass variations in individual water stor-

age compartments have been isolated from GRACE, such as snow (e.g. Frappart

et al., 2006) or groundwater (e.g. Rodell et al., 2007).

The focus of many studies using GRACE data for continental hydrology was on

seasonal variations of water storage as the dominant signal in the time series. In a

few studies it could additionally be shown t hat GRACE is able to detect longer-term

variations and extremes in continental water storage, for example related to large

ﬂoods and droughts in Central Europe and Australia (Andersen et al., 2005; Seitz

et al., 2008; Schmidt et al., 2008b). The potential of relating water storage anomalies

from GRACE to large-scale inter-annual modiﬁcations of global circulation patterns

such as El-Niño-Southern Oscillation (ENSO) has for the ﬁrst time been shown by

Morishita and Heki (2008). Ramillien et al. (2008b) analyzed trends in continental

water storage and their contribution to sea level variations. Nevertheless, such ana-

lyzes of longer-term variations are still considerably limited due to the short period

with available GRACE data.

Given the scarcity of other observation data at large scales, water storage varia-

tions from GRACE have a large potential to validate, calibrate and improve global

hydrological and land surface models by (1) demonstrating deﬁciencies of the model

output relative to GRACE data, (2) identifying sources of these model deﬁcien-

cies in terms of model input data, model structure or model parameters, and by (3)

modifying hydrological models and evaluating the modiﬁcations against GRACE

observations (see Güntner, 2008 for a review). However, while qualitative com-

parisons of model results to GRACE data have frequently been done, only f ew

attempts have been made so far towards a full integration of GRACE data into the

modelling process. The particular nature of GRACE data requires adequate pro-

cessing and modelling steps both on the hydrological and geodetic side to assure

the comprehensive and consistent improvement of large-scale hydrological mod-

els. Such prerequisites include, for instance, the consistency of GRACE data and

Continental Water Storage Variations from GRACE Time-Variable Gravity Data 373

model variables in terms of ﬁltering, reliable error estimates, advanced methods

of multi-objective model calibration or data assimilation and the full evaluation of

the water balance at the river basin scale after model adjustments with GRACE

(Güntner, 2008).

3 The TIVAGAM Project

Research within the project TIVAGAM (Time-Variable Gravity and Surface Mass

Processes: Validation, Processing and First Application of New Satellite Gravity

Data) was devoted to topics outlined above, from the adequate processing of

GRACE time-variable gravity ﬁeld data to their application for improved modelling

and understanding of Earth surface processes, here for the example of continental

hydrology. Important goals were (1) to develop strategies for signal separation from

integral satellite gravity observations with emphasis on identifying typical spatio-

temporal characteristics of hydrological mass signals, (2) to evaluate GRACE-based

mass variations in comparison to geophysical model data and independent obser-

vations and (3) to develop methods to make use of GRACE data towards an

improvement of large-scale hydrological models.

Regarding the ﬁrst goal, Petrovic et al. (chapter “Surface Mass Variability from

GRACE and Hydrological Models: Characteristic Periods and the Reconstruction

of Signiﬁcant Signals” by this issue) propose an analysis method to extract the

characteristic spatio-temporal morphology of surface mass anomalies, in particu-

lar their periodic components, from time series of GRACE gravity models in the

space domain. The method works as a ﬁlter to enhance the signal-to-noise ratio

and was applied to reconstruct basin-average water storage signals from GRACE

which were subsequently used for the calibration of a hydrological model (chap-

ter “Calibration of a Global Hydrological Model with GRACE Data” by Werth

and Güntner, this issue). With the similar aim to enhance the signal properties of

GRACE data in particular for regional analyses of surface mass processes, Freeden

et al. (chapter “Time-Space Multiscale Analysis and Its Application to GRACE and

Hydrology Data” by this issue) introduce methods based on the theory of spher-

ical wavelets to allow for comparing GRACE and hydrology data in a consistent

way in time and space and with good localization in the space domain. The multi-

resolution approach can be applied for evaluation of different hydrological data sets,

i.e., towards identifying improved hydrological models.

Horwath et al. (chapter “Mass Variation Signals in GRACE Products and in

Crustal Deformations from GPS: A Comparison”, this issue) analyze the comple-

mentary value of GRACE gravity data and GPS surface deformation data to describe

surface mass variations. They illustrate the challenges in terms of homogeneous data

processing and reference systems to make the data useful for identiﬁcation of errors,

e.g., in hydrological or other geophysical background models to be used for GRACE

processing. In addition to the atmospheric and oceanic models that are routinely

applied with high temporal resolution in GRACE processing for signal separa-

tion, daily continental water storage variations are simulated with the WaterGAP

374 A. Güntner

Global Hydrology Model (WGHM) by Fiedler and Döll (chapter “Monthly and

Daily Variations of Continental Water Storage and Flows”, this issue) and are tested

for de-aliasing disturbing high frequency mass signals in the GRACE observations.

Finally, concluding the TIVAGAM research framework from GRACE satel-

lite data to hydrological applications, Werth and Güntner (chapter “Calibration of

a Global Hydrological Model with GRACE Data”, this issue) propose a multi-

objective calibration framework to incorporate time series of measured river

discharge and GRACE water storage variations directly into the parameter tuning

process of the global hydrological model WGHM. They make use of the adapted

GRACE signal separation methods mentioned above and ensure consistency by

applying identical processing strategies for the water storage ﬁelds of the hydrolog-

ical model. Fiedler and Döll (chapter “Monthly and Daily Variations of Continental

Water Storage and Flows”, this issue) in a complementary study improve the process

representation of water storage and transport in WGHM by adjusting ﬂow routing

and reservoir schemes in the model. In combination, both studies highlight the ben-

eﬁcial nature of GRACE data as a constraint for improving hydrological models and

towards best estimates of continental water storage variations at large scales.

References

Alsdorf DE, Lettenmaier DP (2003) Tracking fresh water from space. Science 301,

1491–1494.

Andersen OB, Seneviratne SI, Hinderer J, Viterbo P (2005) GRACE-derived terrestrial water

storage depletion associated with the 2003 European heat wave. Geophys. Res. Lett., doi:

10.1029/2005GL023574.

Chambers DP, Wahr J, Nerem RS (2004) Preliminary observations of global ocean mass variations

with GRACE. Geophys. Res. Lett. 31, L13310.

Chambers D, Tamisiea ME, Nerem R, Ries J (2007) Effects of ice melting on GRACE observations

of ocean mass trends. Geophys. Res. Lett. 34, L05610.

Davis JL, Elosegui P, Mitrovica JX, Tamisiea ME (2004) Climate-driven deformation of the solid

Earth from GRACE and GPS. Geophys. R es. Lett., doi: 10.1029/2004GL021435.

Frappart F, Ramillien G, Biancamaria S, Mognard-Campbell N, Cazenave A (2006) Evolution of

highlatitude snow mass derived from the GRACE gravimetry mission (2002–2004). Geophys.

Res. Lett., doi: 10.1029/2005GL024778.

Frappart F, Papa F, Famiglietti JS, Prigent C, Rossow WB, Seyler F (2008) Interannual variations

of river water storage from a multiple satellite approach: a case study for the Rio Negro River

basin. J. Geophy. Res., doi: 10.1029/2007JD009438.

Güntner A, Stuck J, Werth S, Döll P, Verzano K, Merz B (2007) A global analysis of temporal and

spatial variations in continental water storage. Water Resour Res. 43(5), W05416.

Güntner A (2008) Improvement of global hydrological models using GRACE data. Surv. Geophys.,

doi: 10.1007/s10712-008-9038-y.

Hunger M, Döll P (2008) Value of river discharge data for global-scale hydrological modeling.

Hydrol. Earth Syst. Sci. 12, 841–861.

Kusche J (2007) Approximate decorrelation and non-isotropic smoothing of time-variable

GRACE-type gravity ﬁeld models. J. Geod. 81, 733–749.

Lettenmaier DP, Famiglietti JS (2006) Hydrology – water from on high. Nature 444(7119),

562–563.

Continental Water Storage Variations from GRACE Time-Variable Gravity Data 375

Morishita Y, Heki K (2008) Characteristic precipitation patterns of El Niño/La Niña in time-

variable gravity ﬁelds by GRACEM. Earth Plan. Sci. Lett., doi: 10.1016/j.epsl.2008.06.003.

Ramillien G, Bouhours S, Lombard A, Cazenave A, Flechtner F, Schmidt R (2008a) Land water

storage contribution to sea level from GRACE geoid data over 2003–2006. Glob. Planet Change

60(3–4), 381–392.

Ramillien G, Famiglietti JS, Wahr J (2008b) Detection of continental hydrology and glaciology

signals from GRACE: a review. Surv. Geophys., doi: 10.1007/s10712-008-9048–9.

Rodell M, Famiglietti JS, Chen J, Seneviratne SI, Viterbo P, Holl S, Wilson CR (2004) Basin scale

estimates of evapotranspiration using GRACE and other observation. Geophys. Res. Lett., doi:

10.1029/2004GL020873.

Rodell M, Chen J, Kato H, Famiglietti J, Nigro J, Wilson C (2007) Estimating ground water storage

changes in the Mississippi River basin (USA) using GRACE. Hydrogeol. J. 15, 159–166.

Rowlands DD, Luthcke SB, Klosko SM, Lemoine FG, Chinn DS, McCarthy JJ, Cox CM,

Anderson OB (2005) Resolving mass ﬂux at high spatial and temporal resolution using GRACE

intersatellite measurements. Geophys. Res. Lett. 32, L04310.

Schmidt R, Flechtner F, Meyer U, Neumayer KH, Dahle C, König R, Kusche J (2008a)

Hydrological signals observed by the GRACE satellite mission. Surv. Geophys., doi:

10.1007/s10712-008-9033-3.

Schmidt R, Petrovic S, Güntner A, Barthelmes F, Wünsch J (2008b) Periodic components of

water storage changes from GRACE and global hydrology models. J. Geophy. Res., doi:

10.1029/2007JB005363.

Seitz F, Schmidt M, Shum CK (2008) Signals of extreme weather conditions in Central Europe in

GRACE 4-D hydrological mass variations. Earth Plan. Sci. Lett. 268(1–2), 165–170.

Swenson S, Wahr J (2002) Methods for inferring regional surface-mass anomalies from Gravity

Recovery and Climate Experiment (GRACE) measurements of time-variable gravity. J.

Geophys. Res., doi: 10.1029/2001JB000576.

Syed T, Famiglietti J, Chen J, Rodell M, Seneviratne S, Viterbo P, Wilson C (2005) Total basin dis-

charge for the Amazon and the Mississippi River basins from GRACE and a land-atmosphere

water balance. Geophys. Res. Lett., doi: 10.1029/2005GL024851.

Tapley BD, Bettadpur S, Watkins M, Reigber C (2004) The gravity recovery and climate

experiment: mission overview and early results. Geophys. Res. Lett. 31, L09607.

Tamisiea ME, Mitrovica JX, Davis JL (2007) GRACE gravity data constrain ancient ice geometries

and continental dynamics over laurentia. Science 316, 881–883.

Van Dam T, Wahr J, Lavallée D (2007) A comparison of annual vertical crustal displacements from

GPS and Gravity Recovery and Climate Experiment (GRACE) over Europe. J. Geophys. Res.,

doi: 10.1029/2006JB004335.

Velicogna I, Wahr J (2006) Acceleration of Greenland ice mass loss in Spring 2004. Nature 443,

329–331.

Wagner W, Scipal K, Pathe C, Gerten D, Lucht W, Rudolf B (2003) Evaluation of the agree-

ment between ﬁrst global remotely sensed soil moisture data with model and precipitation

data. J. Geophys. Res., 108(D19), 4611, doi: 4610.1029/2003JD003663.

Werth S, Güntner A, Schmidt R, Kusche J (2009) Evaluation of GRACE ﬁlter tools from a

hydrological perspective. Geophys. J. Int., doi: 10.1111/j.1365-246X.2009.04355.x.

Surface Mass Variability from GRACE

and Hydrological Models: Characteristic

Periods and the Reconstruction

of Signiﬁcant Signals

Svetozar Petrovic, Roland Braun, Franz Barthelmes, Johann Wünsch,

Jürgen Kusche, and Rico Hengst

1 Introduction

Using data supplied by the Gravity Recovery and Climate Experiment (GRACE),

scientists from various disciplines have for the ﬁrst time been able to observe time-

variable gravity signals related to mass redistributions at, on, and below the Earth’s

surface caused by ongoing geophysical and climatological processes. The mission’s

sensitivity to processes such as continental water cycle (e.g. Schmidt et al., 2006),

post glacial rebound (e.g. Tamisiea et al., 2007), changes in the polar ice sheets

(e.g. Velicogna and Wahr, 2006) and mass transport in the oceans (e.g. Chambers

et al., 2004) has been clearly demonstrated in the recent past.

An important goal is the quantiﬁcation of the mass redistribution related to

the individual phenomena to improve the modelling and understanding of various

geodynamic processes. A key question is the signal separation of the individual

contributions contained in the integral satellite gravity observations. In order to

derive the characteristic spatio-temporal morphology of the surface mass anoma-

lies traceable in time series of GRACE gravity models we propose an analysis

method to derive such features from time series of surface mass anomalies in the

space domain with a focus on periodic components due to continental hydrology.

The method combines a decomposition of the spatio-temporal signal using well-

known Principal Components Analysis (PCA) with a speciﬁc frequency analysis

method which allows for the determination of harmonic (i.e. sine and cosine) waves

of arbitrary periods.

As a basic result we obtain characteristic spectra of periodic mass variations

contained in the GRACE data and in independent global hydrology models. In

this way it is possible to identify the hydrological components contained in the

GRACE data. Based on this, we derive ﬁltered GRACE signals reconstructed from

selected components (considered signiﬁcant) which can contribute to calibration

of hydrological models. The proposed method is not limited to the examination of

S. Petrovic (B)

Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences,

Department 1: Geodesy and Remote Sensing, Telegrafenberg, 14473 Potsdam, Germany

e-mail: sp@gfz-potsdam.de

377

F. Flechtner et al. (eds.), System Earth via Geodetic-Geophysical Space Techniques,

Advanced Technologies in Earth Sciences, DOI 10.1007/978-3-642-10228-8_32,



Springer-Verlag Berlin Heidelberg 2010