Description: We present a partition-enhanced least-squares collocation (PE-LSC) which comprises several modifications to the classical LSC method. It is our goal to circumvent various problems of the practical application of LSC. While these investigations are focused on the modeling of the exterior gravity field the elaborated methods can also be used in other applications. One of the main drawbacks and current limitations of LSC is its high computational cost which grows cubically with the number of observation points. A common way to mitigate this problem is to tile the target area into sub-regions and solve each tile individually. This procedure assumes a certain locality of the LSC kernel functions which is generally not given and, therefore, results in fringe effects. To avoid this, it is proposed to localize the LSC kernels such that locality is preserved, and the estimated variances are not notably increased in comparison with the classical LSC method. Using global covariance models involves the calculation of a large number of Legendre polynomials which is usually a time-consuming task. Hence, to accelerate the creation of the covariance matrices, as an intermediate step we pre-calculate the covariance function on a two-dimensional grid of isotropic coordinates. Based on this grid, and under the assumption that the covariances are sufficiently smooth, the final covariance matrices are then obtained by a simple and fast interpolation algorithm. Applying the generalized multi-variate chain rule, also cross-covariance matrices among arbitrary linear spherical harmonic functionals can be obtained by this technique. Together with some further minor alterations these modifications are implemented in the PE-LSC method. The new PE-LSC is tested using selected data sets in Antarctica where altogether more than 800,000 observations are available for processing. In this case, PE-LSC yields a speed-up of computation time by a factor of about 55 (i.e., the computation needs only hours instead of weeks) in comparison with the classical unpartitioned LSC. Likewise, the memory requirement is reduced by a factor of about 360 (i.e., allocating memory in the order of GB instead of TB).

Description: Deutsche Forschungsgemeinschaft http://dx.doi.org/10.13039/501100001659

Description: Technische Universität München (1025)

Description: We evaluate the mean circulation patterns, water mass distributions, and tropical dynamics of the North and Equatorial Pacific Ocean based on a suite of global ocean-sea ice simulations driven by the CORE-II atmospheric forcing from 1963-2007. The first three moments (mean, standard deviation and skewness) of sea surface height and surface temperature variability are assessed against observations. Large discrepancies are found in the variance and skewness of sea surface height and in the skewness of sea surface temperature. Comparing with the observation, most models underestimate the Kuroshio transport in the Asian Marginal seas due to the missing influence of the unresolved western boundary current and meso-scale eddies. In terms of the Mixed Layer Depths (MLDs) in the North Pacific, the two observed maxima associated with Subtropical Mode Water and Central Mode Water formation coalesce into a large pool of deep MLDs in all participating models, but another local maximum associated with the formation of Eastern Subtropical Mode Water can be found in all models with different magnitudes. The main model bias of deep MLDs results from excessive Subtropical Mode Water formation due to inaccurate representation of the Kuroshio separation and of the associated excessively warm and salty Kuroshio water. Further water mass analysis shows that the North Pacific Intermediate Water can penetrate southward in most models, but its distribution greatly varies among models depending not only on grid resolution and vertical coordinate but also on the model dynamics. All simulations show overall similar large scale tropical current system, but with differences in the structures of the Equatorial Undercurrent. We also confirm the key role of the meridional gradient of the wind stress curl in driving the equatorial transport, leading to a generally weak North Equatorial Counter Current in all models due to inaccurate CORE-II equatorial wind fields. Most models show a larger interior transport of Pacific subtropical cells than the observation due to the overestimated transport in the Northern Hemisphere likely resulting from the deep pycnocline.

Description: Gravity surveying is challenging in Antarctica because of its hostile environment and inaccessibility. Nevertheless, many ground-based, airborne, and shipborne gravity campaigns have been completed by the geophysical and geodetic communities since the 1980s. We present the first modern Antarctic-wide gravity data compilation derived from 13 million data points covering an area of 10 million km2, which corresponds to 73% coverage of the continent. The remove-compute-restore technique was applied for gridding, which facilitated leveling of the different gravity data sets with respect to an Earth gravity model derived from satellite data alone. The resulting free-air and Bouguer gravity anomaly grids of 10 km resolution are publicly available. These grids will enable new high-resolution combined Earth gravity models to be derived and represent a major step forward toward solving the geodetic polar data gap problem. They provide a new tool to investigate continental-scale lithospheric structure and geological evolution of Antarctica.

Description: SummaryRecently, a variety of gravity observations in Antarctica has become available through extensive e orts of airbornesurveys. Aircrafts serving as multi-instrumentation platforms provide measurements on gravity, bedrocktopography, ice surface topography and ice thickness. Collected datasets are valuable in terms of resolution andhomogeneity, which make them suitable for studying regional geoid determination in selected Antarctic regions.Within this context the German joint project VISA provided an excellent database for improving the regionalgeoid by combining gravity and topographic data from aerogeophysical observations with long-wavelength informationfrom global gravity eld models. Using the remove-compute-restore technique in conjunction withleast-squares collocation a regional geoid for Dronning Maud Land, East Antarctica, will be presented. A signalthreshold of up to 6 m added to the global model that was used as a basis can be expected. The accuracy ofthe regional geoid will be estimated to be at the level of 15 cm.Citation: J. Muller, S. Riedel, M. Scheinert, M. Horwath, R. Dietrich, D. Steinhage, H. Anschutz, W. Jokat(2007), RegionalGeoid and Gravity Field from a Combination of Airborne and Satellite Data in Dronning Maud Land, East Antarctica { OnlineProceedings of the 10th ISAES, edited by A.K. Cooper and C.R. Raymond et al., USGS Open-File Report 2007-xxx, ExtendedAbstract yyy, 1-4.IntroductionThe new datasets provided by the satellite missions CHAMP, GRACE and GOCE (to be launched by theend of 2007) enable a homogeneous determination of the gravity eld. Furthermore, in the polar regions icesurface heights could be determined in a similar quality by ICESat. These new satellite data shall be validatedand densi ed by the German joint project VISA (Validation, Densi cation and Interpretation of Satellite Datafor the Determination of Magnetic Field, Gravity Field, Ice Mass Balance and Structure of the Earth Crust inAntarctica, uitilizing Airborne and Terrestrial Measurements) of TU Dresden and AWI Bremerhaven.For this purpose western and central Dronning Maud Land (DML), East Antarctica, were chosen as areaof investigation. Airborne as well as terrestrial observation campaigns were carried out to provide appropriatedatasets on height and height changes, gravity and gravity changes, magnetics, glaciology and seismology. Incombination with the satellite data these measurements will be applied to yield more detailed models of thegravity eld and the regional geoid, of the crustal structure and litosphere dynamics and of the dynamics andmass balance of the Antarctic ice sheet in the working area.Observation campaignsBetween 2001 and 2005 four airborne observation campaigns and two terrestrial observation campaigns werecarried out in western and central DML in order to conduct geodetic and geophysical measurements (Fig. 1,left). The scienti c program of the aerogeophysical campaigns for the observation of the gravity eld, magnetic eld, ice surface height and ice thickness (Radio Echo Sounding (RES)) contains more than 350 ight-hourswith a line-spacing between 10 and 20 kilometers. The terrestrial eld work took place at two di erent areas,during the season 2003/04 at Schirmacher Oasis - Potsdam Glacier - Wohlthat Mountains and one year later(season 2004/05) at Heimefrontfjella - Kirwanveggen. GPS and seismometer stations on bedrock were installed,kinematic GPS pro les, relative gravimetry on ice and ground penetrating radar (GPR) measurements werecarried out as well as samplings of rn cores and snow pits (Anschutz et al., 2007; Anschutz et al., 2006;Scheinert et al., 2005; Nixdorf et al., 2004).Regional Geoid ImprovementCombining satellite observations from CHAMP and GRACE with terrestrial data, high-resolution models ofthe Earth gravity eld have been obtained. Latest examples of these combination models are EIGEN-CG03C, EIGEN-GL04C (Forste et al., 2005; Forste et al., 2006) and GGM02C (Tapley et al., 2005). In Antarctica, thedetermination of the global gravity eld is problematic becausen due to the remoteness (often inaccessibility)and harsh conditions the terrestrial gravity data coverage features very large gaps. Only for a few smallerregions ground-based or airborne measured gravity was included into the combination. In order to improve theterrestrial gravity coverage and to determine the Antarctic geoid, the IAG Commission Project 2.4 "AntarcticGeoid" (chaired by M. Scheinert) was set into action, which is closely linked to SCAR Expert Group on GeodeticInfrastructure in Antarctica (GIANT) project 3 "Physical Geodesy". An overview on the situation is given in(Scheinert, 2005), and the strategy of regional geoid improvement is discussed in (Scheinert et al., 2007b) for thePrince Charles Mountains region, East Antarctica (PCMEGA), as well as for Palmer Land, Antarctic Peninsula(Scheinert et al., 2007a).Within this context, the VISA observation campaigns de-Figure 2: Free-air Anomalies (preliminary resultswith a spatial resolution of 14 kilometers)scribed above provide an excellent database for the validationof the gravity eld and, more importantly, for the determinationand improvement of the regional geoid. Fig. 2 showspreliminary results for the free-air anomalies derived from airbornemeasurements over the western and central DML witha resolution of 14 kilometer (Riedel and Jokat, 2007). Comparedwith the subglacial topography (Fig. 3, left panel) thestrong correlation between these two datasets is clearly visible.The right panel of Fig. 3 shows the ice surface heightin the area of investigation. The datasets of Fig. 3 a ord toderive the ice-thickness, which will be needed in addition tothe subglacial topography for the computation of an improvedgeoid. The high resolution of these datasets make them muchmore suitable than BEDMAP data (Lythe et al., 2000), whichwere a valuable source of information prior to the VISA radarobservations in DML.Especially in Antarctica problems occur when satellite observationsfrom CHAMP and GRACE up to a certain spherical harmonic degree (typically 120) should be combined with terrestrial data. Geophysically extrapolated gravity anomalies do not necessaily reect the actualgravity eld in Antarctica, though they are inevitable to provide a globally complete data coverage neededfor the solution of the closed surface integrals. For this reason, shorter wavelength information (higher thanspherical harmonic degree 120) is unreliable for most Antarctic areas (Fig. 1, right). This evinces when comparingthe gravity anomalies from EIGEN-GL04C for a harmonic window (degrees 121 to 360) (Fig. 1, right)with the free-air anomalies derived from VISA airborne measurements (Fig. 2). While a higher correlation canbe seen near the coastline, it diminishes in the southern part of DML.For the calculation of the regional geoid the remove-compute-restore technique (RCRT) was applied, whichis discussed in detail e.g. in (Forsberg and Tscherning, 1997) and (Sjoberg, 2005) and which was also usedin the PCMEGA case (Scheinert et al., 2007b). In the remove step, a long-wavelength part (predicted by aglobal gravity eld model) and a short-wavelength part (predicted by topography) are removed from the originalgravity data. In the compute step, the obtained band-pass ltered gravity anomalies are transformed into geoidheights, using least-squares collocation in this study. Least-squares collocation o ers the advantage of providingerror estimates for the resulting geoid. After having carried out the compute step, the long-wavelength part andthe short-wavelength part are restored in the geoid. For the computations, we could make use of the programpackage GRAVSOFT (Forsberg et al., 2003; Tscherning, 1974), which o ers a variety of tools for the geodeticgravity eld modelling.ConclusionCombining gravity and topographic data from VISA aerogeophysical campaigns with a global gravity eldmodel a regional geoid for Dronning Maud Land, East Antarctica, will be presented. Studies in other regionsof Antarctica (Scheinert et al., 2007a; Scheinert et al., 2007b) have shown that a signal threshold of up to 6 mto the global gravity eld model that was used as a basis can be expected when comparing the improved geoidwith the global model up to spherical harmonic degree 120. The accuracy of the regional geoid is estimated tobe at the level of 15 cm. Considering the current data situation in Antarctica, the accuracy level of 1 dm is arealistic and appropriate goal for this area of the world. The data coverage in Antarctica will most likely besubject to major improvements when further airborne surveys are carried out. The International Polar Year2007/ 2008 provides a reasonable framework for international and interdisciplinary cooperation in that eld.SCAR-GIANT project 3 "Physical Geodesy" and IAG Commission Project 2.4 "Antarctic Geoid" work towardsthe goal of closing the gaps in the gravity data coverage and at improving the geoid in Antarctica.

Description: In Dronning-Maud-Land, Ostantarktis, wurden sowohl Flugzeugmessungen (2001-2005) als auch terrestrische Beobachtungskampagnen (2003/04 und 2004/05) durchgeführt. Die dabei gewonnenen Daten betreffen Höhen und Höhenänderungen, Schwere und Schwereänderungen, Magnetfeld, Seismizität und Glaziologie.Kombiniert man diese mit Satellitendaten, so lassen sich sowohl Vergleiche durchführen als auch Modellverbesserungen bzw. -verdichtungen erreichen. So wurden detaillierte Modelle des regionalen Schwerefelds bzw. Geoids, der Krustenstruktur und Lithosphärendynamik sowie der Dynamik und Massenbilanz des antarktischenEisschildes im Untersuchungsgebiet abgeleitet. Die Bestimmung der raum-zeitlichenMuster von Eismassenänderungen liefert wertvolle Information, um die mittels GRACE abgeleiteten zeitvariablen Schwerefeldinformationenin Hinblick auf antarktische Massenänderungen zu validieren. Hierbei ist es u.a. unabdingbar, zwischen Höhen- und Dichteänderungen des Eiskörpers zu unterscheiden, so daß nur durch eine Kombination der verschiedenen Beobachtungstechniken ein komplettes Bild zu erreichen ist.

Description: We found evidence for highly periodic dunes in a near-coastal area of Dronning Maud Land, East Antarctica.Analysis of accumulation patterns, derived from ground-penetrating radar (GPR) internal layering, and GPS-based characteristics of surface topography, by means of auto- and cross-covariance, indicate quasi-harmonic oscillations of surface undulations, surface slope, and accumulation.The dunes occur at an elevation range of 1000-1350 m with the dominant wavelength of 5 km and undulations of about 10 m.Accumulation and slope are anticorrelated at zero lag, with a mean annual accumulation of some 140 kgm-2a-1 and variations on the order ofsome 100 kgm-2a-1.Our findings have implications for studies aiming at the mass balance of coastal areas in Antarctica.