Rie2009d
Airborne-based
Geophysical Investigation
In
Dronning Maud Land,
Antarctica
Dissertation
zur Erlangung des Grades Dr. rer. nat.
vorgelegt dem Fachbereich Geowissenschaften
der Universität Bremen
von
SVEN RIEDEL
Alfred Wegener Institut für Polar- und Meeresforschung
Bremerhaven
August 28, 2008
Sven Riedel
Alfred-Wegener-Institut für Polar- und Meeresforschung
Columbusstrasse
D-27568 Bremerhaven
gegenwärtige Anschrift:
Christian Albrechts Universität zu Kiel
Institut für Geowissenschaften, Abt. Geophysik
Otto Hahn Platz 1
D-24118 Kiel
Die vorliegende Arbeit ist die inhaltlich unveränderte Fassung der Dissertationsschrift, die 2008 dem Fachbereich Geowissenschaften der Universität Bremen
vorgelegt wurde.
Gutachter:
Prof. Dr. H. Miller (Universität Bremen/AWI)
Prof. Dr. H. Villinger (Universität Bremen)
Prüfer:
Prof. Dr. C. Spiegel (Universität Bremen)
Prof. Dr. R. Tiedemann (Universität Bremen/AWI)
Promotionskolloquium:
am 04.Februar 2009
2
This is about the times, the places,
the people, that have shaped me.
Kurzfassung
Die Antarktis steht bei der Rekonstruktion der Großkontinente Rodinia und
Gondwana jeweils im Mittelpunkt erdgeschichtlicher Untersuchungen. Geologische
Großstrukturen zeugen von der Bildung und dem Zerfall dieser Superkontinente, jedoch ist deren direkte Erschließung, bedingt durch die Eisbedeckung Antarktikas,
nicht möglich und begrenzt sich nur auf spärliche, geologische Aufschlüsse. Daher
ermöglichen nur satelliten- und flugzeuggestützte, also indirekte Messungen, eine
flächenhafte Erfassung geophysikalischer Daten in polaren Gebieten.
Geologische Untersuchungen manifestieren drei tektonische Großereignisse im Gebiet um Dronning Maud Land (DML): die Grenvillische Orogenese (1.1 Ga) führte
zur Bildung Rodinias, die Pan-Afrikanisch-Antarktische Orogenese (EAAO, 500 Ma),
gipfelte in der Entstehung von Gondwana sowie dessen Zerfall, um 180 Ma, als letztes
tektonisches Großereignis der Region.
Im Rahmen der vorliegenden Arbeit, die Teil des VISA Projektes ist (Validierung,
Verdichtung und Interpretation von Satellitendaten zur Bestimmung von Magnetfeld, Schwerefeld und Eismassenhaushalt sowie Krustenstruktur in der Antarktis, gestützt durch flugzeugbasierende und terrestrische Messungen) wurden
flugzeuggestützte Daten über einen Zeitraum von 4 Jahren (2001-2005) erhoben,
bearbeitet, zusammengefaßt und strukturell analysiert. Die Methodik der Messungen wird vorgestellt, wobei der umfangreichen, rechentechnischen und zeitaufwändigen Bearbeitung der Topographie-, Magnetik- und Schweredaten ein Großteil der
Arbeit gewidmet wird.
Die Arbeit resultiert in der flächenhaften Darstellung des Magnet- und Schwerefeldes,
sowie der Topographie und bietet erstmals eine einheitliche Datenbasis für die Region
um Dronning Maud Land, zwischen 14◦ W-20◦ E und 70◦ S-78.5◦ S. Weiterführend
wurden Verfahren angewandt, wie Wellenlängenfilterungen, Tiefenbestimmungen
geologischer Störkörper, Betrachtungen zur Isostasie und Attributbestimmungen, um
die geologische Situation umfassend zu analysieren.
i
Das präsentierte Kartenmaterial zeigt den Verlauf geologisch-tektonischer Großstrukturen, die zwar in der Literatur vermutet und eingehend diskutiert, jedoch in ihrer
Lage, Form und Ausprägung nicht vollständig bekannt waren. In Bezug zu bestehenden Daten aus aerogeophysikalischen Befliegungen im Gebiet um Dronning
Maud Land, umfassen nahezu 85% der Schwere- und 65% der Magnetikdaten, welche
in dieser Arbeit vorgestellt werden, bisher nicht erforschte Regionen. Dies stellt einen
großen Beitrag für die geologische Forschung in der Antarktis dar.
Die lithospherische Grenze zwischen dem Archaischen Craton, der Grunehogna Provinz, und dem Proterozoischen- / Früh-Paleozoischen mobilen Gürtel, der Maudheim Provinz, konnte aufgrund der gewonnenen Datenbasis sowie weiterführenden
Betrachtungen zur Isostasie und der Curvature-Analyse, interpretiert werden. Die
detaillierte Kartierung von Störungszonen weist auf Verlauf und Ausrichtung tektonischer Großereignisse hin. Diese Beobachtungen bilden ein kombiniertes geologisches Modell, welches bestehende Vorstellungen erweitert und verfeinert sowie
lokale Aussagen seismologischer, seismischer und geologischer Arbeiten bestätigt.
Abstract
Antarctica represents a key component in the investigation of the geological history
and reconstruction of the supercontinents Rodinia and Gondwana. Remnants of the
formation and disintegration of these former land masses can be found, although
great uncertainties remain in the location of tectonic boundaries beneath the ice
sheet of Antarctica due to general lack of outcrops and the limited amount of geological data. Space and airborne measurements are the only possibility to obtain
comprehensive spatial data coverage of geophysical data over the extensive large
polar areas.
Common knowledge of the geological framework displays three major tectonic events
which formed Dronning Maud Land (DML): the Grenvillian Orogen (1.1 Ga) build
up Rodinia, the Pan-African-Antarctic Orogen (EAAO, 500 Ma) rose in the supercontinent Gondwana and finally the breakup of Gondwanaland, at around 180 Ma.
During this work, as part of the VISA project (Validation, densification and interpretation of satellite data for the determination of magnetic field, gravity field, ice mass
balance and structure of the Earth crust in Antarctica using airborne and terrestrial
measurements), four years of investigated airborne based data (2001-2005) are processed, compiled and interpreted. The methods of measurements are explained in the
methodology chapter, focused on the complex computational and time-consuming
processing, to handle topographic-, magnetic-, and gravity data.
Finally, the thesis displays a compilation of a homogeneous database for the DML
region from 14◦ W to 20◦ E and from 70◦ S to 78.5◦ S. Furthermore, comprehensive studies and techniques, such as wavelength-filtering, depth estimation routines,
isostatic analysis and Curvature discussions, are applied for final geological interpretation.
iii
The presented maps display detailed boundaries of geologic and tectonic structures,
which already have been suggested or discussed in recent literature, but have never
been known to full extent, concerning detail, locations, boundaries and structures.
With respect to earlier conducted geophysical investigations in DML, up to 85% of
the gravity data and 65% of the magnetic data, presented in this thesis, cover unexplored regions and contribute therefore a large amount of new data to the Antarctic
geological research.
Old lithospheric boundaries between the Archaen Craton, the Grunehogna Province,
and a Proterozoic to Early Paleozoic mobile belt, the Maudheim Province, were be
interpreted on the basis of the new database and the use of isostatic and curvature analysis. Detailed mapping of thrust faults show the strike of major tectonic
events. All of these observations constitute an integrated geological model, which is
confirmed by recent seismologic-, seismic,- and geologic results.
Contents
1 INTRODUCTION
1
1.1
Antarctica . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1.2
The VISA project . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
1.3
Previous geological investigations . . . . . . . . . . . . . . . . . . . .
5
1.3.1
Supercontinent Cycle . . . . . . . . . . . . . . . . . . . . . . .
5
1.3.2
Gondwana assembly and breakup . . . . . . . . . . . . . . . .
5
1.3.3
Connection between Antarctica and South Eastern Africa . .
5
1.3.4
The EAAO . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
2 METHODOLOGY
11
2.1
Aircraft handling and design . . . . . . . . . . . . . . . . . . . . . . .
11
2.2
Radio Echo Sounding System . . . . . . . . . . . . . . . . . . . . . .
13
2.3
GPS System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
16
2.3.1
Satellite Constellation . . . . . . . . . . . . . . . . . . . . . .
16
2.3.2
Control Segment . . . . . . . . . . . . . . . . . . . . . . . . .
18
2.3.3
User Segment . . . . . . . . . . . . . . . . . . . . . . . . . . .
18
2.3.4
GPS Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . .
20
2.3.5
GPS Processing . . . . . . . . . . . . . . . . . . . . . . . . . .
21
Aerogravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
2.4.1
Eötvös correction . . . . . . . . . . . . . . . . . . . . . . . . .
25
2.4.2
Vertical Accelerations . . . . . . . . . . . . . . . . . . . . . .
25
2.4.3
The scalar ZLS Ultrasys gravity meter system . . . . . . . . .
26
2.4.4
Aerogravimetry Data Processing . . . . . . . . . . . . . . . .
30
2.4.5
Gravity data corrections . . . . . . . . . . . . . . . . . . . . .
30
Aeromagnetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
32
2.5.1
33
2.4
2.5
Processing of Airborne Magnetic Data . . . . . . . . . . . . .
vii
2.6
Data Visualisation . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36
2.6.1
Basic Interpretation of Magnetic Anomalies . . . . . . . . . .
37
2.6.2
Total Field Shaded Relief Map . . . . . . . . . . . . . . . . .
37
2.6.3
Derivative Based Filters . . . . . . . . . . . . . . . . . . . . .
38
2.6.4
Curvature attributes . . . . . . . . . . . . . . . . . . . . . . .
38
2.6.5
Analytic Signal . . . . . . . . . . . . . . . . . . . . . . . . . .
41
2.6.6
Tilt Derivative . . . . . . . . . . . . . . . . . . . . . . . . . .
42
2.6.7
Depth Estimation
. . . . . . . . . . . . . . . . . . . . . . . .
44
2.6.8
Isostasy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
47
3 SURVEYS and DATABASE
49
3.1
Campaigns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
49
3.2
Airborne RES Data . . . . . . . . . . . . . . . . . . . . . . . . . . . .
51
3.2.1
Topography VISA I . . . . . . . . . . . . . . . . . . . . . . .
52
3.2.2
Topography VISA II . . . . . . . . . . . . . . . . . . . . . . .
54
3.2.3
Topography VISA III . . . . . . . . . . . . . . . . . . . . . .
56
3.2.4
Topography VISA IV
. . . . . . . . . . . . . . . . . . . . . .
58
Airborne Gravity Data . . . . . . . . . . . . . . . . . . . . . . . . . .
61
3.3.1
Free-air Anomaly VISA I . . . . . . . . . . . . . . . . . . . .
61
3.3.2
Free-air Anomaly VISA II . . . . . . . . . . . . . . . . . . . .
64
3.3.3
Free-air Anomaly VISA III . . . . . . . . . . . . . . . . . . .
66
3.3.4
Free-air Anomaly VISA IV . . . . . . . . . . . . . . . . . . .
68
Airborne Magnetic Data . . . . . . . . . . . . . . . . . . . . . . . . .
70
3.4.1
TMI VISA I . . . . . . . . . . . . . . . . . . . . . . . . . . . .
70
3.4.2
TMI VISA II . . . . . . . . . . . . . . . . . . . . . . . . . . .
72
3.4.3
TMI VISA III . . . . . . . . . . . . . . . . . . . . . . . . . . .
74
3.4.4
TMI VISA IV . . . . . . . . . . . . . . . . . . . . . . . . . . .
76
Advices for future airborne operations . . . . . . . . . . . . . . . . .
78
3.3
3.4
3.5
4 COMPILATION and INTERPRETATION
79
4.1
Topography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
79
4.2
Gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
82
4.2.1
Free-air Anomaly Map . . . . . . . . . . . . . . . . . . . . . .
82
4.2.2
Complete Bouguer Anomaly Map . . . . . . . . . . . . . . . .
84
4.2.3
Isostasy Map . . . . . . . . . . . . . . . . . . . . . . . . . . .
87
4.3
4.4
4.5
Magnetic
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
89
4.3.1
Total Magnetic Intensity Map . . . . . . . . . . . . . . . . . .
89
4.3.2
Analytic Signal Map . . . . . . . . . . . . . . . . . . . . . . .
92
4.3.3
Tilt Derivative Map . . . . . . . . . . . . . . . . . . . . . . .
93
Areas In Detail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
95
4.4.1
Grunehogna Unit . . . . . . . . . . . . . . . . . . . . . . . . .
95
4.4.2
Maudheim Province . . . . . . . . . . . . . . . . . . . . . . . 101
Geologic model suggestions . . . . . . . . . . . . . . . . . . . . . . . 109
5 SUMMARY
111
6 OUTLOOK
115
ACKNOWLEDGEMENTS
117
REFERENCES
119
LIST OF TABLES
125
LIST OF FIGURES
127
A DGPS SETTINGS
131
A.1 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
B GRAVITY READINGS and TYING-PROCESS
133
B.1 Tying VISA I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
B.2 Tying VISA II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
B.3 Tying VISA III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
B.4 Tying VISA IV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
C DEPTH ESTIMATION SOLUTIONS
141
C.1 Source depths, VISA I . . . . . . . . . . . . . . . . . . . . . . . . . . 142
C.2 Source depths, VISA II . . . . . . . . . . . . . . . . . . . . . . . . . . 144
C.3 Source depths, VISA III . . . . . . . . . . . . . . . . . . . . . . . . . 146
C.4 Source depths, VISA IV . . . . . . . . . . . . . . . . . . . . . . . . . 148
D CURVATURE DISCUSSION
151
D.1 Regional and local observations . . . . . . . . . . . . . . . . . . . . . 151
Chapter 1
INTRODUCTION
1.1
Antarctica
The Antarctic continent plays a keyrole in the development and study of plate tectonics. The study of the lithosphere and identification of lateral discontinuities in
Antarctica and the surrounding areas are essential to an understanding of the geodynamic evolution of the continent.
Airborne acquisition of potential field and other remote-sensed data is essential due
to the extensive ice cover of the Antarctic continent. This large ice sheet hinders
direct observations of the surface geology.
Potential field anomaly maps are particularly helpful in connection with geological
mapping of the sparse outcrop, enabling the detection of major structures and interpretation of the composition of Antarctica. Magnetic and gravity data provide
regional maps of the structural grain in the basement, suture zones between basement terranes, the basement terranes themselves, and the nature of intra continental
rifts and the extent of major faults.
The relative timing and kinematics of the evolution of the Antarctic continent can be
derived from the detailed geophysical mapping of rock units. Distinctive magnetic
signatures provide important markers for extracting both the orientation of regional
faulting, and its relative timing.
Information on the isostatic state of the lithosphere is contained in long wavelength
anomalies, providing significant insights into the thermal structure of the Antarctic
lithosphere.
From reconstructions of Gondwana, it should be possible to identify the shapes of
old oceans and therefore to infer the pathways of currents, that would have played a
role in past climates.
Thus, receiving information on the crustal makeup of this remote and poorly understood region is essential. Furthermore, such projects will provided improved global
potential field modelling and assist with the verification and calibration of observations obtained from polar orbiting satellite missions.
1
INTRODUCTION
1.2. THE VISA PROJECT
A significant amount of airborne data have been collected during the past 40 years
and have contributed toward a better understanding of regional tectonic provinces
within Antarctica. For example the Russian regional aeromagnetic surveys undertaken by VNIIOkaengeologia in the 80s have imaged magnetic pattern and tectonics
from the Weddell Sea to the boundary of Dronning Maud Land (DML). More high
detailed investigations, but much smaller in extent were flown by British Antarctic
Survey (BAS) in the Jutulstraumen region and by the Bundesanstalt für Geowissenschaften und Rohstoffe (BGR) in the Schirmacher oasis and adjacent areas.
Previous investigations by the Alfred Wegener Institute (AWI) within the EMAGE
project were focused on the break-up history of Gondwana and the opening of the
Weddel Sea. The interpretation of this extensive magnetic anomaly dataset in the
eastern Weddell Sea constrained the ages and orientations of sea floor spreading
anomalies and were used to calculate a new set of model parameters for the opening
of the Weddell Sea. These parameters were combined with constraints on the South
America-Africa and Africa-Antarctica spreading systems, based on a compilation of
magnetic anomaly and fracture zone data, to determine a new model for Gondwana
break-up between 176Ma and 83Ma (König, 2005).
Within the new VISA project, as described in the following chapters, the efforts were
made to conduct purely continental investigations, including systematic airborne
coverage over an area of more than 1.2 Million km2 , from 2001 to 2005.
1.2
The VISA project
The VISA project (Validation, densification and Interpretation of Satellite
data for the determination of magnetic field, gravity field, ice mass balance and structure of the Earth crust in Antarctica using airborne and
terrestrial measurements) is an interdisciplinary project undertaken in collaboration between the Technical University Dresden (TUD) and the Alfred Wegener
Institute for Polar- and Marine Research, Bremerhaven (AWI).
The primary goal of this long-term project is to estimate the mass balance of Antarctica’s continental ice sheet based on temporal changes of the gravity field of the Earth.
All processes which result in a growth or shrinkage of an ice sheet also cause temporal
changes of the gravity field, as, for instance, mass is redistributed between the ice
sheets and the world’s oceans. Such temporal variations of the global mass distribution should be determined by new satellites with mission periods of up to 5 years
measuring the Earth’s gravity field. A very precise identification and quantification
of all signals influencing the gravity field is necessary in order to determine mass
induced effects and thus minimize errors in the calculation of the mass balance of
the ice sheet.
Apart from these, there is a huge need to understand the subglacial geology, of which
observations are impossible due to the ice cover.
2
INTRODUCTION
1.2. THE VISA PROJECT
Figure 1.1: The Antarctic continent and the related area of investigation within the VISA project.
Finally, other complementary indirect methods are utilized in this multidisciplinary
project, and are listed as follows:
• Validation and densification of gravity satellite mission data (CHAMP, GRACE
and GOCE) in Antarctica, using airborne gravity.
• Detailed investigations on ice, including height, gravity and mass changes including glaciological and geodetic field campaigns.
• Investigations of vertical motions due to lithospheric processes, using geodetic determinations of height and gravity changes on bedrock, estimation of
seismicity, and determination of focal mechanisms.
• Detailed investigations into the structures of the gravity field, magnetic field,
ice thicknesses, surface and bedrock topography, from airborne measurements.
3
INTRODUCTION
1.2. THE VISA PROJECT
This thesis deals mainly with the processing and interpretation of airborne data, in
detail: The evolution of the South Atlantic region, including the Weddell Sea and its
adjacent areas, is of crucial importance for understanding the processes responsible
for the structure and tectonics of the Antarctic lithosphere and its relation to geodynamic processes, especially to the timing and geometry of the initial stages of the
Mesozoic break-up between Africa, Antarctica and South America. The sub-glacial
geology is of utmost importance for unraveling the geological evolution of Antarctica prior to the break-up of Gondwana. Understanding the sub-ice geology allows
reconstruction of ancient mountain chains (collision zones) across continents, which
are separated by large ocean basins in the present world.
Since only the peaks of the Dronning Maud Land mountains can be geologically sampled, geophysical methods are required to uncover the geological structure beneath
the ice. Therefore, extensive airborne surveys were conducted across DML between
2001 and 2005 to close data gaps and to improve existing data sets.
The compilation of previous regional Russian investigations (magnetic and gravity)
gave a first impression on the sub-glacial geology in the Weddel Sea and adjacent
areas (Aleshkova et al., 2000; Golynsky et al., 2000), but the data are of insufficient
accuracy for direct comparisons with satellite missions. In addition, the variable
spacing of flight lines will cause in problems with their interpretation.
The extent of the new survey areas from 14◦ W to 20◦ E and from 70◦ S to 78.5◦ S, is
large enough to fully recognize long wavelength regional anomalies. However, smaller
features could also be mapped, owing to an average line spacing of about 10 km.
In detail, the new compilation of airborne magnetic and gravity data across DML is
focused on:
East Antarctic Craton / Maud boundary: The Grunehogna Craton is only
sparsely exposed at a small number of nunataks. Detailed airborne-geophysical investigations will define the Cratonic units exactly, as well as their regional extension.
The continent-ocean boundary is of utmost interest.
Grenvillian mountain chains: The southern extension of the Grunehogna Craton
is the Maudheim Province, which has an Grenvillian age of 1.1 Ga. The mutual
boundary is marked by strong magnetic anomalies and may correlate with structures
in Southern Africa. The exact boundary of the Grunehogna Unit (GU) is not well
developed.
Pan African overprinted crust: The Heimefrontfjella, Kirnvanveggen and Sverdrupfjella ridges mark the boundary of the influence of the East Antarctic African
Orogen (EAAO, 500 Ma), which is the result of the collision of East and West Gondwana. The existing, limited, datasets permits no detailed interpretation.
4
INTRODUCTION
1.3
1.3. PREVIOUS GEOLOGICAL INVESTIGATIONS
Previous geological investigations
In the distant past, Earth was a very different planet than the one we know today.
If you could travel through time to arrive at the Earth of a billion years ago, you
would have a hard time navigating. A strange giant continent and a single planetary
ocean would replace the familiar continents and oceans of todays world. 1
1.3.1
Supercontinent Cycle
During the Neoproterozoic, a supercontinent, referred as Rodinia, formed at ca.
1100 Ma and broke apart at around 800-700 Ma. Rodinia is thought to have included
Laurentia, Australia, Antarctica, Greater India and Amazonia, but their exact relative positions and the timing on break-up are the subject of ongoing debate in a number of papers dealing with palaeomagnetic and geological data. Grenvillian/Kibaran
aged deformation is exposed along margins of Laurentia, East Gondwana, Amazonia and Baltica. Geologic records indicate that Neoproterozoic and early Paleozoic
rift margins surrounded Laurentia, while similar-aged collisional belts crosscut Gondwana. Hence, the breakup of one supercontinent was followed rapidly by the assembly
of another, smaller, one: Gondwana.
1.3.2
Gondwana assembly and breakup
The reorganization of cratonic blocks following by the Neoproterozoic breakup of Rodinia gave birth to Gondwana. This reassembly of various continental fragments into
a new supercontinent started between 800-700 Ma and continued up to 500 Ma, with
the occurrence of collisional tectonics along orogenic belts (Pan-African/Brasiliano).
Traditionally, Gondwana can divided into three separate blocks. West Gondwana,
consisted of the Amazonia Craton of South America and the West African Craton.
Central Gondwana comprised the Congo and Kalahari cratons of Africa and the
Grunehogna Province of East Antarctica. While East Gondwana includes eastern
Madagascar, India, Sri Lanka, the East Antarctic shield and Australia.
1.3.3
Connection between Antarctica and South Eastern Africa
Geological correlations suggest that the crustal evolution of DML and SE Africa were
quite similar from Archea until Mesozoic times. The assembly of DML took place
before 1000 Ma as both the Grunehogna and Maudheim provinces are transected by
orogenic belts. The lithostratigraphic, metamorphism and deformation of the late
Archean to Mesoproterozoic granite-greenstone rocks and the volcanic-sedimentary
sequences of the Kaapval-Zimbabwe cratonic provinces are closely comparable to
those of the Grunehogna Province. Western DML consists of numerous Precambrian
elements, together referred to as the Grunehogna Craton, a fragment of the KalahariKaapval-Craton (KKC). These cratons parted during Gondwana breakup (Jurassic
1
Burke Museum of Natural History and Culture, University of Washington
5
INTRODUCTION
1.3. PREVIOUS GEOLOGICAL INVESTIGATIONS
times). Correlation with the KKC is on the bases of Archean and Mesoproterozoic
sediments with similar age.
The rocks of southern DML are grouped together as the Maudheim Province (MP),
which consists of high grade metamorphic rocks of Grenvillian age. The MP can be
seen as an eastward continuation of the Namaqua-Natal-belt in South Africa. Limited
outcrops exist and are confined to the nunataks at Sverdrupfjella, Kirvanveggen and
Heimefrontfjella.
The Heimefrontfjella (HF), a 140 km SW to NE oriented mountain range, is divided
into several blocks (Trottanfjella, Sivorgfjella, XU-Fjella and the Kottas-Mountain).
The range divides the EAAO into a western and an eastern, more overprinted part
(Jacobs, 1996). A prominent shear zone is recognized within the HF.
Central DML was strongly influenced during the assembly of Gondwana. The crystalline basement melted and the Grenvillian structures of the EAAO were covered.
An element of crustal extension is recognized within the continent-continent collision
settings, and has been suggested as possible related to the replacement of the orogenic root during delamination. Elsewhere, such processes are known to have caused
topographic uplift and subsequent orogenic collapse.
The youngest tectonic event in DML occurred during Jurassic times, the break-up
of Gondwana. A volcanic rifted continent margin formed and the Weddell Sea and
Lazarev Sea were established by rifting processes. Permo-triassic shear zones where
reactivated and forced strike-slip movements between Africa and Antarctica (Cox,
1992).
1.3.4
The EAAO
Due to the strong influence of the East African Antarctic Orogen (EAAO) on the
geological history of DML, some detailed notes are needed to help in understanding
the complexity.
The EAAO is one of the largest orogenic belts on the planet. It resulted from the
collision of various parts of proto–East and West Gondwana during late Neoproterozoic to early Paleozoic times (between 650 and 500 Ma). The orogen extents for over
nearly 8000 km, from Arabia in the north (Arabian–Nubian shield) along the East
African margin (Mozambique belt) into East Antarctica (e.g., Muhongo and Lenoir,
1994; Stern, 1994; Jacobs and Thomas, 2002). In most places along its length, the
orogen is more then 1000 km wide (see figure 1.2, left).
The southern part of the EAAO reaches from Kenya to East Antarctica. In East
Antarctica, rocks of late Neoproterozoic-early Paleozoic age are exposed between
western Dronning Maud Land and the Lützow-Holm Bay area. Although only exposed in nunataks, the dominant orogenic structures can be traced under the ice by
aeromagnetic anomalies (e.g., Golynsky and Jacobs, 2001). The core of this part of
the orogen is exposed in central and eastern Dronning Maud Land (e.g., Jacobs et
al., 1998, 2003; Shiraishi et al., 1994), see figure 1.2, right.
6
INTRODUCTION
1.3. PREVIOUS GEOLOGICAL INVESTIGATIONS
Figure 1.2: The East African Antarctic Orogen and escape tectonics in DML. left: The East
African Antarctic Orogen (EAAO) is interpreted as the main collisional orogen along which parts
of proto-East and West Gondwana collided to form Gondwana. right: Escape tectonics model
for the southern termination of the EAAO. Abbreviations: ANS-Arabian-Nubian shield; C-Coats
Land; Da-Damara belt; DML-Dronning Maud Land; EF-European fragments; EH-Ellsworth-Haag;
F-Filchner block; FI-Falkland Islands; G-Grunehogna; H-Heimefrontfjella; K-Kirvanveggen; LHLützow-Holm Bay; M-Madagascar; Na–Na-Namaqua-Natal; SR-Shackleton Range; Z-Zambesi belt,
(after Jacobs, 2004).
Large volumes of high-temperature post tectonic granitoids occur in central Dronning
Maud Land (>50 percent outcrop area). Their petrology and geochemistry indicate
that they are crustal derivatives, probably the consequence of asthenospheric upwelling following delamination of the orogenic root, and subsequent orogenic collapse
(Jacobs et al., 2003).
In western Dronning Maud Land the 20 km-wide Heimefront transpression zone (Jacobs and Thomas, 2002) separates Mesoproterozoic rocks with a strong Pan-African
tectonothermal overprint to the east from unaffected rocks to the west. Consequently,
this shear zone has been interpreted as the western front of the orogen (Golynsky
and Jacobs, 2001). Further evidence that the Heimefront transpression zone is a
major crustal discontinuity is provided by geophysical data, which show that the
high-amplitude, elongate magnetic anomalies that characterize the eastern half of
the ca. 1.1 Ga Namaqua-Natal-Maud belt (southeast Africa and East Antarctica
juxtaposed in Gondwana) terminate sharply against it.
7
INTRODUCTION
1.3. PREVIOUS GEOLOGICAL INVESTIGATIONS
The Pan-African collision history of the southern part of the orogen in Antarctica
and Mozambique can be separated into three major phases:
• An earliest stage, recorded in the Schirmacher Oasis at ca. 620 Ma (HenjesKunst, 2004), followed by anorthosite magmatism in the main mountain range
of DML at ca. 600 Ma.
• The main deformation and medium- to high-grade metamorphism in the main
mountain range of DML and the Nampula Province of Northern Mozambique
is bracketed in age by metamorphic zircon rims to between ca. 590 and 550 Ma
and is interpreted to represent the collision phase (Jacobs et al., 2003).
• A late Pan-African stage is associated with extension, tectonic exhumation and
south-directed extrusion between ca. 530 and 500 Ma, exposing mid- to lower
crustal levels (e.g. Jacobs and Thomas, 2004). This period is accompanied
by syn-tectonic and late- to post-tectonic intrusions. The volume of igneous
rocks seems to drastically increase toward the end of the extensional period,
culminating in voluminous and extensive granitoid-charnockoid magmatism.
Late to post-tectonic granitoids with dominantly charnockitic mineralogy cover
an area of at least 15000 km2 in East Antarctica (where they can be traced
under the ice by geophysical means) and NE Mozambique, see figure 1.2.
Figure 1.3: Detailed geological sample work, including mapped intrusions and shear zones, after
Jacobs [pers. comm.].
8
INTRODUCTION
1.3. PREVIOUS GEOLOGICAL INVESTIGATIONS
Summarizing this, a widespread late-to post-tectonic Cambrian (Pan-African) magmatic province is recognized in the southern part of the East African-Antarctic Orogen in NE Mozambique and Central Dronning Maud Land, two areas that were
thought to be contiguous within Gondwana. It covers an area of at least 15000 km2 ,
and would have stretched from the northern margin of the Nampula Province (the
Lurio Belt) in Mozambique, through to central Dronning Maud Land, decreasing
gradually westwards in volume to the eastern Sverdrupfjella, where the magmatism
stops, close to the frontal zone of the orogen in that region. New SHRIMP dates
from Dronning Maud Land reveal that the intrusion of the granitoids is tightly
constrained to almost exactly 500 Ma, preceded and/or accompanied by extensional
shearing dated at c. 510 Ma. The intrusions are interpreted to have crystallized at
mid-crustal levels after collapse and extension of the orogen, possibly accompanied
by delamination of the lithosphere root. Hot asthenosphere, rising to the lower crust
above the subsiding orogenic root would have provided the heat source for the magmatism which is typically anhydrous, high temperature and charnockitic (Jacobs,
2007).
9
Chapter 2
METHODOLOGY
2.1
Aircraft handling and design
1 The
Alfred Wegener Institute for Polar and Marine Research, Bremerhaven (AWI),
owns the aircraft POLAR2 (D-CAWI) and POLAR4 (D-CICE). The planes are platforms for geophysical, meteorological, air chemistry and remote sensing research.
They provide logistical support for expeditions. Aircraft management is presently
commissioned to The Deutsche Forschungsanstalt für Luft- und Raumfahrt (DLR,
German Aerospace Research Establishment) by the Alfred Wegener Institute. The
DLR subcontracted OPTIMARE, Sensorsysteme AG, for the operational use of the
aircraft. OPTIMARE is responsible for the scientific instrumentation and data acquisition.
Figure 2.1: Polar 2 aircraft, operated by AWI, during field activities.
Airborne Polar missions need long-term planning, a task which is performed by the
Aircraft Planning Group of the Alfred Wegener Institute. It prepares an aircraft operation plan based on applications from the research group. The plan is presented to
the User Advisory Council, consisting of scientists from German research institutions
and universities. The board advises the director of the Alfred Wegener Institute on
1
taken from users handbook for the POLAR 2 and POLAR 4 research aircraft
11
METHODOLOGY
2.1. AIRCRAFT HANDLING AND DESIGN
the missions of the aircraft. The membership covers different disciplines interested
in aircraft based research in polar regions.
Figure 2.2: Cross-section of Dornier 228-200 Polar 2 aircraft showing the geophysical configuration
on board, adopted from Boebel (2000).
Figure 2.2 displays the aircraft used, Polar 2 or Polar 4, with the geophysical configuration on board, which is used within the VISA project. The combined system
consists of:
• two GPS-Trimble 4000ssi receivers for the estimation of the flightpath trajectory; inclusive geodetic GPS antenna onboard the aircraft roof
• one Honeywell LaserNav II inertial navigation system for estimation of the
aircraft orientation; aircraft rear-end
• one ADMB OPTECH- or Riegl LD 90 Laseraltimeter for estimation of the
altitude; roller door
• one ZLS Ultrasys gravity meter based on a LaCoste&Romberg AirSea S56
gravimeter for the acquisition of scalar gravity; near aircraft balance point
• two Geometrics CS-2 magnetometer measuring the total magnetic field; wing
tips
• one tri-axial Bellington fluxgate magnetometer, measures three components of
the magnetic field with respect to the movement of the plane for compensation
purposes; in the tail of the aircraft
• one 150 MHz Radio Echo Sounding system to map ice thicknesses and internal
layering of glaciers using Short Backfire Antenna; below each wing
12
METHODOLOGY
2.2. RADIO ECHO SOUNDING SYSTEM
Apart from these sensors, a data acquisition system were installed and mounted in
a rack to store the acquired data in a common database.
2.2
Radio Echo Sounding System
The Radio Echo Sounding System is used to map ice thicknesses and the internal
layering of glaciers, ice sheets and ice shelf areas. Installed on board the aircraft, it is
capable of penetrating ice thicknesses of up to 4 km. It uses radio pulses transmitted
downwards into the ice that are partly reflected at layers with contrasting electrical
properties, as well as by the boundary between ice and sea water.
The system was designed and built in cooperation between AWI, TUHH (Technische Hochschule Hamburg-Harburg, Institut für Hochfrequenztechnik), and Aerodata Flugmesstechnik GmbH, Braunschweig.
The system uses two antennae of the short backfire type mounted underneath the
wings, see figure 2.1. A burst of 150 MHz with a duration of 60 ns or 600 ns is
transmitted by the left-wing antenna and the reflected signal are received by the
right-wing antenna. The amplified and filtered signal is sent into a logarithmic
detector which produces a low frequency envelope signal from the bursts. This
output signal is passed onto the data acquisition system for printout and storage.
The transmitter and the preamplifier are mounted in the wings in order to reduce
noise and cable ringing.
Figure 2.3: Polar 2 Radio Echo Sounding instrumentation and principles, adopted from Steinhage
(2001).
13
METHODOLOGY
2.2. RADIO ECHO SOUNDING SYSTEM
A compression (stacking) of the signal is calculated, with the well calculated loss
of phase information. The advantage of this is faster processing of the data set,
with signal repetition frequency of 20 Hz; 1000 pulses are consecutively digitized
and stacked. At a speed over the ground of 130 knots (240 kmh−1 ), the horizontal
resolution is 3.25 m. The vertical resolution depends on the pulse-length setting,
the duration and the shape of the pulse, and the physical properties of the ice.
Theoretically, a pulse of 60 ns provides a resolution of 5 m, a pulse of 600 ns about
50 m. By using the RES-system in toggle mode (after each registration the pulse
changes between 600 ns and 60 ns) it is possible to arrive a signal that combines the
highest vertical resolution and maximum power (penetration depth). A corollary is
the loss of horizontal resolution by a factor of 2 (3.25 m with a single pulse to 6.5 m
in toggle mode), but this is compensated for by the increased vertical information
content.
The received signals are converted into digital format at a rate of 75 MHz and 24 bit
resolution. This data stream is stored in a database and post-processed to SEGYformat. The hard disks can hold up 7.5 hours of airborne ice radar survey. The
RES system can be entirely computer supervised except for setting the transmitting
frequency (between 100 and 200 MHz) and the signal filter.
When a wave encounters an interface between materials of different properties, the
wave may be refracted, reflected, or both. Snell’s Law describes the reaction of
light to a boundary between materials of different dielectric contrasts (or refractive
index), based on the angle at which a ray (perpendicular to the wave front) hits the
interface. The angle of the incoming ray (Angle of Incidence αi ) is equal to the Angle
of Reflection αr . The Angle of Refraction αR is determined by the ratio of the sines
of the Angle of Incidence to the Angle of Refraction and the ratio of the dielectric
constants for the upper and lower layers (1 and 2 ).
There is a point where the Angle of Incidence is large enough, close to horizontal,
that there is no refraction. This is called the Angle of Critical Refraction where all
the incoming waves are either reflected or refracted along the interface. Any angles
larger than the Angle of Critical Refraction result in only reflection.
The receiver records the amount of time between the arrival of the transmitted wave
and any reflected waves as well as the strength of the waves. The radio waves travel
at different speeds through different materials. For example, the velocities of radio
waves are known with 3 x 108 ms−1 through air and a little less than half as much
in ice at 1.69 x 108 ms−1 . Where a glacier bed echo is recorded, the time delay is
used to calculate the ice thickness.
An example of a 500 km length RES profile (cross section over Jutulstraumen) is
shown in figure 2.4. This profile is measured with the 60 ns pulse, is 50 times stacked,
and has been band-pass filtered, amplitude scaled and corrected to a constant flight
level.
14
METHODOLOGY
2.2. RADIO ECHO SOUNDING SYSTEM
Dark gray represents the travel path through air. The first reflection represents the
surface-topography shown with the dark black line. A strong surface multiple reflection (result of reverberation within the ice) is clearly seen as is the internal layering
of the ice sheet. The subglacial topography is marked as second main reflection horizon. The loss of signal is clearly recognized over locations with strong gradients in
surface topography.
Figure 2.4: Cross section of RES sounding profile. Data are measured with 60 ns pulse, 50 times
stacked, band-pass filtered, amplitude scaled and corrected to a constant flight level [Steinhage
(pers. comm.)].
15
METHODOLOGY
2.3
2.3. GPS SYSTEM
GPS System
Precise navigation of the aircraft is of utmost importance for the later interpretation
of scientific data. In particular the gravity field measurements, are sensitive to positioning accuracy and accelerations. Due to this, it is necessary to understand GPS
techniques more then just a basic tool.2
The Global Positioning System (GPS) is a network of 24 Navstar satellites orbiting
Earth at 21000 km, originally established by the U.S. Department of Defense (DOD).
GPS provides specially-coded satellite signals that can be stored or processed in a
GPS receiver, enabling the receiver to compute position, velocity and time. Using
the stored data to enable post-processing options yields advantages from having more
precise algorithms and additional information.
For normal code processing, the
signals of four GPS satellite have
to be used to compute positions in
three dimensions and the time offset in the receiver clock. Higher
accuracy can be archived with
phase observations of the GPS signal. Here, an additional unknown
variable has to be determined,
so that a minimum of five GPS
satellite observations is needed.
The Global Positioning System
is comprised of three segments: Figure 2.5: The Global Positioning System; Measurements of code-phase arrival times to estimate position
satellite constellation, ground con- and time (after Dana, P.H.).
trol/monitoring network and user
receiving equipment.
The formal GPS Joint Program Office (JPO) programmatic terms for these components are the space, operational control and user equipment segments, respectively.
2.3.1
Satellite Constellation
The satellite constellation comprises satellites in the orbit that provide the ranging signals and data messages to the user equipment. A GPS satellite transmits
two microwave carrier signals. The L1 frequency (1575.42 MHz) carries the navigation message and the single point positioning code signals. The L2 frequency
(1227.60 MHz) is used to measure the ionospheric delay by pulse per second (PPS)
generator equipped receivers. Three binary codes shift the L1 and/or L2 carrier
phase.
2
most illustrations and suggestions are taken from P. H. Dana, The Geographer’s Craft Project,
Department of Geography, The University of Colorado at Boulder,
http : //www.colorado.edu/geography/gcraf t/notes/gps/gpsf .html
16
METHODOLOGY
2.3. GPS SYSTEM
Figure 2.6: GPS satellite signals, adopted from Dana, P.H.
The C/A code (coarse acquisition) modulates the L1 carrier phase. This C/A signal is
a repeating 1 MHz pseudo random noise (PRN) code. This noise-like code modulates
the L1 carrier signal, "spreading" the spectrum over a 1 MHz bandwidth.
The C/A code repeats every 1023 bits (one millisecond). There is a different C/A
code PRN for each satellite. Their PRN number, the unique identifier for each
PRN code, often identifies GPS satellites. The C/A code that modulates the L1
carrier is the basis for the civil single point positioning (SPS). The P-code (precise
code) modulates both the L1 and L2 carrier phases. The P-Code is a very long
(seven days) 10 MHz PRN code. In the anti-spoofing (AS) mode of operation, the
P-code is encrypted into the Y-code. The encrypted Y-code requires a classified
anti-spoofing module for each receiver channel and is for use only by authorized users
with cryptographic keys. The P (Y)-Code is the basis for the PPS. The navigation
message also modulates the L1 -C/A code signal. The navigation message is a 50 Hz
signal consisting of data bits that describe the GPS satellite orbits, clock corrections,
and other system parameters.
The nominal GPS operational constellation consists of 24 satellites that orbit the
earth in 12 hours (see figure 2.7). There are often more than 24 operational satellites
as new ones are launched to replace older satellites. The satellite orbits repeat almost
the same ground track (as the earth turns beneath them) once each day. The orbit
altitude is such that the satellites repeat the same track and configuration over any
point approximately each 24 hours (4 minutes earlier each day). There are six orbital
planes (with nominally four satellites in each), equally spaced (60 degrees apart), and
inclined at about 55 degrees with respect to the equatorial plane. This constellation
provides the user with between five and eight satellites visible from any point on the
earth.
17
METHODOLOGY
2.3.2
2.3. GPS SYSTEM
Control Segment
The Control Segment (OCS) has responsibility for maintaining the satellites and
their proper functioning. This includes maintaining the satellites in their proper
orbital positions (called station keeping) and monitoring satellite subsystem health
and status. The OCS also monitors the satellite solar arrays, battery power levels, and propellant levels used for manoeuvres, and activates spare satellites. The
overall structure of the operational ground/control segment is as follows: Remote
monitor stations constantly track and gather C/A and P(Y) code from the satellites
and transmit this data to the Master Control Station, which is located at Falcon
Air Force Base, Colorado Springs. There is also the ground uplink antenna facility, which provides the means of commanding and controlling the satellites and
uploading the navigation messages and other data. The unmanned ground monitor
stations are located in Hawaii, Kwajalein in the Pacific Ocean, Diego Garcia in the
Indian Ocean, Ascension Island in the Atlantic and Colorado Springs, United States.
Ground antennas are also located in these areas. These locations have been selected
to maximize satellite coverage.
Figure 2.7: GPS nominal constellation: 24 satellites in 6 orbital planes, 4 satellites in each plane,
20.200 km altitudes, 55 degree inclination, after Dana, P.H..
2.3.3
User Segment
The GPS user segment consists of the GPS receivers and the user community. As
mentioned before, four satellites are required to compute the four dimensions of x,
y, z (position) and time. GPS receivers are used for navigation, positioning, time
dissemination, and other purposes. Navigation in three dimensions is the primary
function of GPS.
Carrier-phase tracking of GPS signals has resulted in a revolution in land surveying.
A line of sight along the ground is no longer necessary for precise positioning. This
use of GPS requires specially equipped carrier tracking receivers. The L1 and/or L2
carrier signals are used in carrier phase surveying. L1 carrier cycles have a wavelength
18
METHODOLOGY
2.3. GPS SYSTEM
of 19.029 cm, while L2 has 24.421 cm. If tracked and measured, these carrier signals
can provide ranging measurements with relative accuracies of millimeters under special circumstances. Tracking carrier phase signals provides no time of transmission
information. The carrier signals, while modulated with time tagged binary codes,
carry no time-tags that distinguish one cycle from another. Therefore, to archive the
range between satellites and receivers, the ambiguity of phases has to be solved.
Figure 2.8: Carrier phase tracking: range from SV to remote has changed by 7 cycles if no cycle
slips have occurred, after Dana, P.H..
The measurements used in carrier-phase tracking are differences in carrier-phase
cycles and fractions of cycles over time. At least two receivers track carrier signals
at the same time. Ionospheric delay differences at the two receivers must be small
enough to insure that carrier phase cycles are properly accounted for. This usually
requires that the two receivers are within about 30 km of each other. Carrier phase
is tracked at both receivers and the changes in tracked phase are recorded over time
in both receivers.
All carrier-phase tracking is differential, requiring both a reference and remote receiver tracking carrier phases at the same time, see figures 2.8 and 2.9. Unless the
reference and remote receivers use L1 -L2 differences to measure the ionospheric delay, they must be close enough to ensure that the ionospheric delay difference is less
than a carrier wavelength. Using L1 − L2 ionospheric measurements and long measurement averaging periods, relative positions of fixed sites can be determined over
baselines of hundreds of kilometers. Phase difference changes in the two receivers
are reduced using software to differences in three position dimensions between the
reference station and the remote receiver. High-accuracy range difference measurements with sub-centimeter accuracy are possible. Problems result from the difficulty
of tracking carrier signals in noise or while the receiver moves.
Two receivers and one satellite over time result in single differences. Two receivers
and two satellites over time provide double differences. Post-processed static carrierphase surveying can provide 1-5 cm relative positioning within 30 km of the reference
receiver given a measurement time of 15 minutes for short baselines (10 km), or one
hour for long baselines (larger than 30 km). Rapid static or fast static surveying can
provide 4-10 cm accuracies with 1-kilometer baselines and 15 minutes of recording
time. Real time kinematic (RTK) surveying techniques can provide centimeter measurements in real time over 10 km baselines tracking five or more satellites and real
time radio links between the reference and remote receivers.
19
METHODOLOGY
2.3. GPS SYSTEM
Figure 2.9: Carrier phase positioning, adopted from Meyer, U..
2.3.4
GPS Errors
GPS errors are a combination of noise, bias and blunders. Noise errors are the
combined effect of PRN code noise (around 1 meter) and noise within the receiver
(around 1 meter). The potential accuracy of the C/A code, of around 100 meters,
is reduced to 30 meters (two standard deviations). A is the intentional degradation
of the SPS signals by a time varying bias. The SA bias on each satellite signal is
different, and so the resulting position solution is a function of the combined SA
bias from each satellite vehicle used in the navigation solution. Because SA is a
changing bias with low frequency terms in excess of a few hours, position solutions
or individual SV pseudo-ranges cannot be effectively averaged over periods shorter
than a few hours. Differential corrections must be updated at a rate less than the
correlation time of SA (and other bias errors).
SV clock errors uncorrected by the Control Segment can result in one-meter errors.
Ephemeris data errors are about 1 meter, tropospheric delay errors around 1 meter.
The troposphere is the lower part of the atmosphere from ground level up to around
8 to 13 km, that experiences the changes in temperature, pressure, and humidity associated with weather changes. These effects are independent for both frequencies,
and so cannot be eliminated by L1 /L2 processing. Models of tropospheric delay are
complex and require estimates or measurements of these parameters. Un-modeled
tropospheric delay errors are in the range of 10 meters. The ionosphere is the layer
of the atmosphere from 50 to 500 km that consists of ionized air. The transmitted
model can only remove about half of the possible 70 ns of delay leaving a ten meter un-modeled residual. The ionospheric delay can indeed be calculated by L1 /L2
observations, but for low satellite constellation, the error increases.
Multi-path errors are about 0.5 meters. Multi-path is caused by reflected signals
from surfaces near the receiver that can either interfere with, or be mistaken for, the
signal that follows the straight-line path from the satellite. Multi-path is difficult to
detect and sometimes hard to avoid.
20
METHODOLOGY
2.3. GPS SYSTEM
Table 2.1: Precise error model, C/A code. This is the statistical ranging error (one-sigma) that
represents the total of all contributing sources. The dominant error is usually that arising from the
ionosphere.
error sources
Ephemeris data
Satellite clock
Ionosphere
Troposphere
Multipath
Receiver measurement
User equivalent range
error (UERE), rms
filtered (UERE), rms
2.3.5
bias
2.1
2.0
1.0
0.5
1.0
0.5
one-sigma error in meter
random
0.0
0.7
0.5
0.5
1.0
0.2
total
2.1
2.1
1.2
0.7
1.4
0.5
DGPS
0.0
0.0
0.1
0.1
1.4
0.5
3.3
3.3
1.5
0.4
3.6
3.3
1.5
1.4
GPS Processing
As shown before, each satellite transmits signals on two sinusoidal carrier waves, L1
and L2 . Modulated onto L1 are two pseudo random noise codes, in RINEX (receiver
independent exchange format) notation called C1 and P1 . A second P-code, P2 , is
modulated onto the L2 frequency. Assuming the clocks in the satellite and in the
receiver are synchronized, the travel time signal can be determined by measuring
the shift between the internal and the incoming versions of the code in the receiver.
The pseudo random noise codes are designed to have a low autocorrelation, allowing
the shift to be measured precisely and without ambiguity. Multiplication of the
transmission time by the speed of light gives the range between the satellite and
the receiver. However, in practice the time synchronizations between receivers and
satellites are not perfect. Because of this effect and the influence of other error
sources these ranges are called pseudo-ranges. It is possible to measure the carrier
beat phase more precisely than the codes themselves, however integer ambiguities
must be resolved for centimeter and better positioning.
2.3.5.1
DGPS Processing
The idea behind all differential positioning is to correct bias errors at one location
with measured bias errors at a known position. A reference receiver, or base station,
computes corrections for each satellite signal. Because individual pseudo-ranges must
be corrected prior to the formation of a navigation solution, DGPS implementations
require software in the reference receiver that can track all SVs in view and form
individual pseudo-range corrections for each SV. These corrections are passed to the
remote, or rover, receiver which must be capable of applying these individual pseudorange corrections to each SV used in the navigation solution. Applying a simple
position correction from the reference receiver to the remote receiver has limited
21
METHODOLOGY
2.3. GPS SYSTEM
effect at useful ranges because both receivers would have to be using the same set of
SVs in their navigation solutions and have identical Geometric Dilution of Precision
terms (GDOP, not possible at different locations) to be identically affected by bias
errors.
Differential corrections may be used in real-time or later, with post-processing techniques. Therefore, in this work, a kinematic DGPS program, Trimble Geomatic
Office (TGO) in post-processing option was used.
The first crucial step in DGPS processing is the so-called static positioning process
of the base stations, which are normally grounded on surface rock. Due to limited
available outcrops it was also necessary to use ice grounded GPS stations. Because
the ice sheet flows, the static GPS solution has limited value (only during flight time)
or a model of ice sheet movement has to be calculated, see Kirchner (2002).
Static GPS processing was done twice within the VISA-project. First, calculations
are done by the geodetic group of TU Dresden, using Berner software. Static ground
stations are combined with a known station, i.e. SANAE 4, in the International
Terrestrial Reference Frame (ITRF). A second calculation, within this work, uses
the TGO software. Direct comparisons of both static solutions show similar results.
For example the Weigel-Nunatak station (2002/2003), grounded on surface rock,
displays differences in a range of ∆ x=0.033 m, ∆ y=0.012 m, ∆ z=0.019 m.
With respect to the calculations of disturbing accelerations in gravity processing,
differences in absolute positioning of the so called static GPS stations will not affect
the result of the "relative position" of the aircraft.
Two kinematic datasets are available for the positioning of the aircraft, namely, those
from the front and rear antennae solutions. Both receiver solutions are calculated
and quality checked, in that we expected their spacing with respect to each other
to be fixed at 4.5 m: the onboard spacing of the receivers. Irregularities are noted
and, if necessary, the decision for one receiver solution was made. Furthermore, the
acceptance of one receiver solution can be supported by a kind of statistical analysis,
done by the TGO software.
The "variance of the probability distribution" can be defined as the relationship of
the variance to the best and second-best solutions. A high value can be interpreted
for the correct, or best, calculation. If the calculated variance is below the setting
parameter, the output solution is set to false and the float solution is used.
The handling of the qualitative "variance of the reference" is not well documented,
but seems to have characteristics of a-posteriori to a-priori analysis. A low value
indicates a good comparison with the parameter settings and the model, but gives,
in reality, no quality check, because a single spike can arise in a high value. Parameter
settings for the TGO software, depending on 1 s data frame (kinematic) and 30 s data
(static) are displayed in the Appendix A.
22
METHODOLOGY
2.4
2.4. AEROGRAVITY
Aerogravity
Detailed theoretical descriptions of airborne gravity measurements are given by several authors, like LaCoste (1967; 1982; 1988), Harlan (1968), Childers (1999), LaFehr
(1967; 1991), Nettleton (1960), Swain (1996), Vailliant (1991; 1976; 1992) and others. Here, just a summary of the main points is given to brief readers on the sensor
system, refer to Boebel (2000) and Meyer (2004).
For all geodetic and geophysical tasks of airborne gravity, Newton’s Gravitational
Law is still valid. In rotating Earth’s system the gravity potential is the sum of
the potential of the gravitation and the potential of the centrifugal force. From the
scalar field of the gravity potential, Φ(x), the gravitational acceleration, gi , will be
calculated by gradient operations:
−∂Φ
∂xi
(2.1)
ρ(x )
ωearth 2 2
d
0 dV +
|x − x |
2
(2.2)
gi = −∇i Φ =
with
Φ(x) = −G
Z
V
0
Here G = 6.67 10−11 [m3 kg−1 s−2 ], d is the orthogonal distance to the rotation
axis, and ωearth is the angular velocity of the Earth. The force f , which works
on the test mass (gravimeter) is the product of its own mass and the gravity ac0
0
celeration due to the density ρ(x ) or mass distribution µ(x ). Gravity unit is
[10−5 ms−2 ] = 103 Gal = 1 mGal.
Scalar gravity measurements will be recognized along the lot-line, as shown in figure
2.10. Measurement of the gravitational attraction on a moving platform is influenced
by disturbances which for a given altitude, can be solved as follows:
δg(ϕ, λ, h, t) = δgsensor (ϕ, λ, h, t) + ∆innererror + ∆outererror + ∆corr + ∆redu (2.3)
∆innererror
=
sensor − dependent
∆outererror
=
g 2
2 (x,initial
∆corr
=
r̈z (t) + ∆Etv (ϕ, α, ṙaircraf t )
∆redu
=
γ(ϕ, λ, hnormal ) + ∆g(ϕ, λ, hnormal , t)tidal .
+ 2y,initial + r̈x (t)sinx (t) + r̈y (t)siny (t)
23
METHODOLOGY
2.4. AEROGRAVITY
ϕ, λ, h are set for geographic longitude, latitude and height above sea level, defined
on a reference ellipsoid, with the definition of the normal gravity γ(ϕ, λ, hnormal ).
The flight height, h, is given above the ellipsoid. hnormal is the surface of the ellipsoid
and t represents the time, related to the GPS system (most above WGS 84 and UT
(World Geodetic System 1984 and Universal Time)). α is the flight direction, ṙaircraf t
the aircraft velocity in the given flight height.
x and y are the orthogonal errors of leveling (tilt correction) for the longitudinal
and transverse axes of the aircraft. r̈x , r̈y , r̈z , are the corresponding orthogonal
components of aircraft accelerations. g, the static gravity measurement on ground
before take off, will represent the gravity level within the area of investigation.
The ∆outererror are representing the leveling mismatch (only if a scalar system is
used). Errors due to vibrations are not taken into account as they will be eliminated
using damping and filters within the instrument. Another assumption is that the
time varying data are well synchronized. The measured gravity disturbance at flight
level can be reduced to gravitational disturbances (reference on ellipsoid) and Free-air
Anomaly (reference geoid).
The main task are pointed out as follows and yielding in the correction for Eötvöseffect, Vertical accelerations and Tilt correction.
Figure 2.10: Principle of scalar gravimeter systems, adopted from Meyer, 2004.
24
METHODOLOGY
2.4.1
2.4. AEROGRAVITY
Eötvös correction
In general, all types of gravity meters or accelerometers, when based on a moving
platform, are affected by their motion over a curved, rotating Earth. The motion of
a unit of mass in a rotating coordinate system is expressed by the vector equation
[Harlan, 1968]:
a = r̈ + 2ωearth ṙ + ω̇earth × r + ωearth × ωearth × r
(2.4)
with ωearth = angular velocity, r = radius vector. This formula is the precise expression for the Eötvös correction and differs from the Eötvös’ original expression in
that is takes the aircraft speed into account.
Using the ellipsoidal coordinate system, the formula can be rewritten to correct
for the Eötvös effect on airborne measurements, refer to velocities over ground and
velocities in flight height, the effect due to measurements on a platform moving with
respect to the Earth, Harlan [1968]:
∆Eotv =
ṙ2
h
h
(1 − − (1 − cos2 φ(3 − 2sin2 α))) + 2ṙωearth cosφsinα(1 + ) (2.5)
a
a
a
∆Eotv =
ṙ2
h
(1 − − (1 − cos2 φ(3 − 2sin2 α))) + 2ṙωearth cosφsinα
a
a
(2.6)
Errors in navigation have a large impact in the Eötvös correction. Consequently, only
the best possible navigation solution should be used for airborne gravity correction.
2.4.2
Vertical Accelerations
No gravity sensor can distinguish between gravity and platform acceleration. Therefore, any raw, relative scalar "gravity" measurement on a moving platform is actually
the addition of the vertical acceleration and the change in gravity. Consequently, the
accelerations have to be calculated using the GPS solution.
gsens = g + z̈
(2.7)
Furthermore, during airborne operations, the instantaneous vertical accelerations
are generally 10000 to 100000 times greater, and of much higher frequency, than
the expected variation of the gravity signal. Solutions become less stable for high
frequencies in vertical accelerations. It is therefore mandatory to filter the data.
The amount of filtering is highly dependent on the quality of the gravity meter, the
quality of the platform, and the flight conditions. Due to the problems described,
25
METHODOLOGY
2.4. AEROGRAVITY
Figure 2.11: Effects on moving platform, adopted from Meyer, 2004.
the gravity meter response and the filter response must both be linear and stable
in phase to avoid artifacts in the filtered and Eötvös-corrected data. Various filters
are available, for example a Butterworth filter with a cut-off wavelength of 200 s and
RC-filter (3x20 s), depending on data quality.
Filtering the data has an influence on the averaged gravity measurements as well
as on the gradient in the vertical acceleration. Due to the fact that g and the
platform are both sensor-dependent, the vertical accelerations must be calculated
due to a non-inertial mass-system with the focus on non-implementation of mass
distributions.
2.4.3
The scalar ZLS Ultrasys gravity meter system
The ZLS Ultrasys gravity meter system is based on the older LaCoste & Romberg
S56 air/sea gravity meter. The sensor consists of a highly damped, zero-spring type
gravity sensor mounted on a gyro stabilized platform with associated electronics.
The sensor incorporates a hinged beam supported by a zero-length spring. A zerolength spring is a spring whose equilibrium length, with a test mass attached, is zero
(see figure 2.12). Damping of the large vertical accelerations due to the aircraft’s
motion is achieved through the use of internal air dampers.
26
METHODOLOGY
2.4. AEROGRAVITY
Nevertheless, the vertical accelerations of the aircraft make it impossible to keep the
beam constantly nulled. Therefore, it is necessary to read the gravity sensor whilst
the beam is in motion. A mathematical analysis of the spring type gravity sensor
shows that this is possible through observations of the beam position, the beam
velocity, and the beam acceleration at any given time. If the beam motion is highly
damped, the beam acceleration term can be neglected. If the gravity sensor has a
very high sensitivity over a high range, the beam position can be neglected as well.
The ZLS Ultrasys gravity meter fulfills both requirements. Accordingly, it can be
read without nulling by measurement of the beam position parallel to the adjusted
spring tension.
Utilizing the zero-length spring principle in a particular geometry results in a vertical
suspension that can have infinite periods [LaCoste et al., 1988]. When the period
is infinite and the torque exerted by the spring exactly balances the torque exerted
by gravity, the beam will remain stationary at any position. When this position is
achieved, the smallest change in gravity will cause the beam to rotate to one stop
or the other. Thus, infinite period corresponds to infinite sensitivity [Valliant et al.,
1992]. If the period is less than infinite and the beam is displaced from its equilibrium
position, a restoring torque will return it back to the equilibrium position - this is
the case for land gravity meters.
Figure 2.12: Simplified gravimeter (left) and sensor (right): The mass M is attached to the movable
beam OB that is free to rotate about O. The beam is supported by a zero-length spring attached
at the points A and B. In practice, the beam’s total travel distance between top stop and bottom
stop is some mm in the gravity meter.
27
METHODOLOGY
2.4. AEROGRAVITY
So finally, for the gravity meter, the basic equation to gain the relative gravity at a
given time, and thus at a given location, is:
δgsens (φ, λ, h, t) = ST + k Ḃ + CC
(2.8)
The movement of the beam depends on ST (spring tension)- and is needed for the
linear expression of worldwide measurements. The velocity of the beam, Ḃ, itself
is defined by the gravitational attraction, while the correction term, CC, named as
Cross Coupling will be used for the mathematical expression of the sensor-mechanics.
k is a constant, which is a function of the average beam sensitivity and the damping
system.
For best performance and accuracy of the airborne gravity measurements it is imperative to keep the gravity meter system as close as possible to horizontal. For this
task, a platform with two accelerometer and two gyroscopes is implemented.
The platform itself is controlled manually and levelled when the gravity meter is in
an undisturbed environment.
The gyroscopes itself only measures the angular rates of the platform, which can
keep the platform in the stable orientation, but delivers no information on its own
about orientation. For this, the accelerometer input is needed.
The output signal of the accelerometer varies linearly with the tilt angle of the platform. The accelerometer signal is sent to the gyro processor in order that the signal
is appropriately shaped for gyroscopes input.
The combined signals are filtered and sent to the servomotor to correct actual deviations of the platform from the horizontal. This Proportional-Integral Feedback
algorithm maintains a stable platform that performs like a damped pendulum. The
reaction time of this negative feedback loop is close to immediate but it has a limited
"memory" due to the gyroscope drift. The memory time used with the filter is about
4 minutes for airborne application.
Figure 2.13: The scalar ZLS Ultrasys S56 Air/Sea gravity meter.
28
METHODOLOGY
2.4. AEROGRAVITY
Table 2.2: Overview of gravity measuring systems, in use worldwide. In center of the table, the
aircraft system used is shown. For VISA, this system will be the connection between satellite- and
ground-based measurements as well and will be used to validate satellite gravity data, overview
after Meyer (2004).
satellite missions
system
CHAMP
GRACE
GOCE
principle:
trajectory from GPS
correction nonconservative forces
by accelerometer
trajectory from GPS
distance via microwave
correction non-conservative
forces by accelerometer
trajectory from GPS
3D accelerometer
(gradiometer)
height:
450-300 km, decreasing
490-300 km, decreasing
250-200 km
velocity:
−1
2700 km
◦
, I=87
−1
2700 km
◦
, I=89
3000 km−1 , I=96.5◦
error:
0.1-10 mGal, decreasing
0.1-5 mGal, decreasing
1 mGal
resolution:
650 km until Earth revolution
450 km until Earth revolution
100 km until Earth revolution
application:
global gravity and geoid
global gravity and geoid
global gravity and geoid
system
long range
middle range
short range
principle:
scalar and vector gravity meter
velocity and accelerations
from GPS
position by INS/IMU
scalar and vector gravity meter
velocity and accelerations
from GPS
position by INS/IMU
scalar and vector gravity meter
and gradiometer
velocity and accelerations
from GPS
position by INS/IMU
height:
5-15 km
0.5-5 km
0.1-3 km
aircraft missions
−1
−1
velocity:
750-1000 km
200-300 km
50-200 km−1
error:
1-5 mGal
1-5 mGal
1-5 mGal
resolution:
20-8000 km,
continental scale
5-500 km,
regional scale
1-250 km,
regional to local scale
application:
mountains, fracture zones
deserts, ocean structures
mountains, fracture zones
deserts, coastal structures
mountains, fracture zones
marshes, deltas
surface missions
system
ship and submarine
airborne-point
surface
principle:
scalar gravity meter
velocity and accelerations
from GPS
position by INS/IMU
scalar gravity meter
one point measurement
observatory studies
scalar gravity meter
and gradiometer
height:
sea surface
topography
topography
−1
velocity:
10-30 km
-
-
error:
1-5 mGal
0.1-2 mGal
0.1-2 mGal
resolution:
0.1-1000 km
0.1-300 km
0.1-300 km
application:
oceanic structures
regional to local structures
regional to local structures
29
METHODOLOGY
2.4.4
2.4. AEROGRAVITY
Aerogravimetry Data Processing
Airborne-based gravity measurements are the difference of two data records: gravity
signal and accelerations from flight trajectory.
The kinematic differential GPS data are computed in the initial phase of data processing. The second primary data input are the raw gravity readings from the gravity
meter.
This subtraction requires precise synchronization of both datasets. To ensure this,
the symmetry of the cross correlation between gravity signal and vertical acceleration
is used (Olesen and Fosberg, 1997).
After synchronization is ensured, the vertical aircraft accelerations are computed
from the GPS heights, as well as the horizontal accelerations from the positions in
the X and Y directions, yielding the Eötvös correction, Tilt correction, and Free-air
reduction. All these computations are applied within the unfiltered, common 1 Hz
data frame. Once all corrections and reductions are applied, the data are low-pass
filtered, with filter choice, depending on data quality and flight behavior. Optional
low-pass filters are 3x20sec RC, Infinite Impulse Response (IIR), Butterworth and
Finite Response (FIR). The final result is further reduced for the normal gravity field
and related to sea level, thus reaching the Free-Air Anomaly, having a mean spatial
resolution of about 6.5 km.
2.4.5
Gravity data corrections
To compare measurements made at different sites, several corrections must be applied
to the datasets:
δg = δgm − δgt − δgl − δgd − δgf a − δgb − δgice − δgw − δgtopo
δgm
δgt
δgl
δgd
δgf a
(2.9)
value determined in the field,
correction for tides, neglected for airborne operations,
correction for geographic latitude,
correction for instrumental drift,
Free-air correction.
The first four corrections result in the Free-air Anomaly by subtracting the
latitude correction (theoretical gravity) from the absolute gravity and adding a
correction for the station elevation.
30
METHODOLOGY
2.4. AEROGRAVITY
The Bouguer Anomaly is the result of correcting the Free-air anomaly for the
mass of material that exists between the station elevation and the spheroid. In the
Antarctic there are four typical cases for correction for different media: rock only;
ice/rock; ice/water/rock; water/rock, such that:
δgba = δgf a − 0.0419088[ρhs + (ρw − ρ)hw + (ρi − ρw )hi ] + δgcurv
(2.10)
in which
δgba
δgf a
ρ
ρw
ρi
hs
hw
hi
δgcurv
Bouguer anomaly [mGal],
Free-air anomaly [mGal],
Bouguer density of rock [gcm−3 ],
Bouguer density of water [gcm−3 ],
Bouguer density of ice [gcm−3 ],
station elevation [m],
water depth [m] including ice,
ice thickness [m], and
curvature correction.
The additional curvature correction converts the geometry for the Bouguer correction
from a infinite flat slab to a spherical cap whose thickness is the elevation of the
station and whose radius from the station is 166.735 km. The formula from LaFehr
(1991) was applied.
The Complete Bouguer Anomaly is the result of correcting the Bouguer anomaly
for irregularities in the Earth due to terrain in the vicinity of the observation points.
δgcba = δgba + δgtc
(2.11)
The terrain corrections are calculated using a combination of the methods described
by Nagy (1966) and Kane (1962), which calculate the regional terrain correction from
a coarse regional Digital Elevation Model (DEM) draped over a more finely sampled
local DEM that covers the survey area. For this purpose, a more regional DEM from
the BEDMAP database and, for the local DEM, VISA’s own calculated DEMs was
used. This yields in a regional correction grid that represents the terrain correction
beyond a local correction distance.
31
METHODOLOGY
2.5
2.5. AEROMAGNETIC
Aeromagnetic
The success with which the magnetic method can be applied depends on the contrast in magnetic properties of the rock types concerned. These magnetic properties
arise from the presence of magnetic minerals such as magnetite, pyrrhotite, ilmenite,
franklinite and specular hematite in the rock. By far the most common of there
minerals is magnetite. The magnetic effect of rocks is almost entirely due to their
magnetite content.
Crustal rock material becomes magnetized within the core field, the inducing field on
the magnetic minerals in the crustal material. Magnetic domains within the minerals
align themselves in the direction of the inducing field and thereby generate their own
magnetic field, which is superposed on the core field. This causes anomalies in the
smoothly varying core field.
The main magnetic field generated in the Earth’s core and that induced in the crust
are both vector quantities, which interact with each other. At Earth’s surface, only
the magnitude of this resultant force is measured by the magnetometer. A further
complication is that the field present in the crustal rocks consists of two parts, a
remanent or permanent field and the temporarily induced field describes above. The
strength of the induced field is proportional to the core field and parallel to it.
Remanent magnetization represents a remnant of the magnetic field imprinted at an
earlier stage in the Earth history.
The magnetization is a measure of the magnetic polarization M. The magnetic field
is proportional to the magnetizing field H :
M=kH
(2.12)
Since M and H are both measured in [Am], the susceptibility k is dimensionless in
the SI system. The magnetic induction, B (unit: [nT]) is the total field, including
the effect of magnetization:
B = µ0 (M + H)
(2.13)
= µ0 (1 + k) H
(2.14)
= µ µ0 H .
(2.15)
Magnetized matter contains a distribution of microscopic magnetic moments. Unpaired electron spins are the most important sources of magnetic moment. Magnetization, M, is defined as the magnetic dipole moment per unit volume of the material. Induced magnetization, Mind , is the component of magnetization produced in
response to an applied field. The induced magnetization varies in proportion with
changes in the applied field and vanishes when the field is removed. Remanent magnetization or remanence, Mrem , is the permanent magnetization that remains when
the applied field is removed, and is essentially unaffected by weak fields.
32
METHODOLOGY
2.5. AEROMAGNETIC
The total magnetization is the vector sum of the induced and remanent magnetizations:
M = Mind + Mrem
(2.16)
For sufficiently weak fields, such as the geomagnetic field, the induced magnetization
is approximately proportional to the applied field.
For most rock, the induced magnetization is essentially parallel to the applied field,
irrespective of the field direction. In this case the susceptibility is a scalar quantity,
i.e. it is characterized simply by its magnitude and is isotropic. The Koenigsberger
ratio, Q, is a convenient parameter for expressing the relative importance of remanent
and induced magnetizations. It is given by:
Q=
Mrem
Mrem
=
Mind
kH
(2.17)
Thus, Q > 1 indicates that remanence dominates induced magnetization, whereas
Q < 1 implies that the induced magnetization is dominant.
2.5.1
Processing of Airborne Magnetic Data
The processing of aeromagnetic flight line data to a grid of values ready for the
application of enhancement techniques and interpretation involves the following sequential process: editing, correction for diurnal effects, the levelling of all data to a
common base and removal of the core magnetic field and, finally, the application of
a gridding algorithm.
Thus, the overall processing of aeromagnetic data involves the following major steps
in two phases:
Phase 1-Pre-processing:
• Verifying and editing the raw data
• Merging the raw magnetic data with the flight path coordinates via GPS time
Phase 2-Processing:
• Removing diurnals
• Removing the component attributable to the core field (IGRF correction)
• Levelling the data
• Gridding and contouring
33
METHODOLOGY
2.5. AEROMAGNETIC
Despite this, sources of errors still exist and are listed as follows:
Magnetometers: Modern magnetometers give absolute measurements with high
sensitivity and virtually no drift and, to all intents and purposes, can be regarded as
giving an exact reading. The noise envelope of the Geometrics CS-2 Magnetometer
used is 0.2 nT including all orientation errors.
Aircraft effects: The magnetic signature of the aircraft consists of three components, namely those due to its permanent magnetization, the magnetization induced
by the motion of the aircraft through the Earth’s magnetic field, and that due to
the flow of electrical currents within the aircraft. The permanent magnetization of
the aircraft leads to the familiar heading errors caused by the vector addition of the
aeroplane’s induced field to Earth’s field. Higher frequency errors are introduced by
aircraft movements and is called manoeuvre noise. The general method for removing
these effects is called compensation, and involves measuring the pitch, yaw and roll
of the aircraft. The use of feedback compensators (flying periodic manoevers in each
direction) will detect these additional field components and allow to remove them in
real time.
Navigational effects: The availability of the Global Positioning System (GPS)
improved the quality of navigational data, in this case to positional accuracies of
better then 1.5 m for the x- and y directions.
IGRF: The main component of the measured magnetic field originates from the
magnetic dynamo in the earth outer core. This field is preliminary bipolar, with
amplitudes of 50000 nT, but spherical harmonic terms up to about order 13 are
significant. Since the core field is much larger than that due to crustal magnetization,
and since it has a significant gradient in many parts of the world, it is desirable to
remove a model of the global field from the data before further processing. The model
most widely used today is the International Geomagnetic Reference Field (IGRF).
The IGRF is modified every 5 years and includes coefficients for predicting the core
field into the near future. The contour values of 47000 nT in the unprocessed data
show that the regional (non-geologic) field component still persists in the magnetic
data. So it was necessary to remove the regional field as approximated by the IGRF
for the corresponding epoch of the survey from the total magnetic field (observed
value) to obtain the anomalous field due to crustal sources according to normal
convention.
Time variation in the magnetic field (diurnal): Earth’s magnetic field varies
with time. The variations can be random or cyclic, varying from effects of the
11 year sunspot cycle (secular variation) down to geomagnetic pulsations with periods
of the order of seconds. To remove these effects a stationary magnetometer that
simultaneously measures the time varying magnetic field for later subtraction from
the survey data is necessary. There is still considerable debate on how many such
base stations are needed to adequately sample the spatial variations of the external
field for large surveys or when the survey area is at a considerable distance from the
base of operations.
34
METHODOLOGY
2.5. AEROMAGNETIC
Ground clearance and altitude variation/common datum: The amplitude
of local magnetic anomalies varies with distance from the recording instrument, i.e.
with respect to the ground clearance of the aircraft. The rate of change increases
as the wavelength of the anomaly decreases. Earth’s magnetic field thus varies with
height above the ellipsoid. Typically, the rate of change with height is 0.025 nT m−1 .
Because the surveys were carried out at different flying heights, ranging from 1500
to 4500 m, all the line data and grids needed to be reduced to a common datum. A
datum of 3500 m was chosen because the majority of the data under consideration
were flown at that heights. Difficulties arise with data of VISA III-campaign, for
which a few lines were flown at 4500 m, and downward continuation produced poor
results (see chapter Surveys and Database).
Levelling: Some flight path errors were evident in the initial grids as narrow elongate
anomalies along the flight lines. This noise is often prominent in that it interrupts
the real anomaly pattern of the images, making it difficult to interpret real anomalies. Thus, it was necessary to minimize this effect. Levelling using tie lines was
originally developed as an alternative method to the use of base stations. Nowadays
it is a standard step after base station corrections. The purpose today is to minimize residual differences in level between adjacent lines, and the long-wavelength
errors along lines that inevitably remain after compensation and correction for external field variations by base station subtraction. The differences in field value at
the intersections of lines and tie lines are calculated, and corrections are applied to
minimize these differences. The most common method is to calculate a constant correction for all lines by least squares fitting, sometimes using a low order polynomial.
Another method is to treat the tie lines as fixed and to adjust only the survey lines.
Note that all these procedures are empirical.
Gridding: The x y z data were gridded separately for each data set using the Minimum Curvature Interpolation Method. As the spacing of the flight lines is approximately 10 km, except in parts of the VISA III-campaign where it is 20 km, and the
track spacing reaches 66 m, a grid cell size of 3 km was chosen to reconstruct the
crustal anomalies.
35
METHODOLOGY
2.6
2.6. DATA VISUALISATION
Data Visualisation
Gridded data were used to produce images and maps for interpretation at suitable
scales for the complete region as well as interesting fragments, using the Lambert
Conformal projection (2sp).
Maps were interpreted to identify regional features like tectonic boundaries between
cratons and mobile belts, major faults and shear zones, dykes etc. Qualitative interpretation involved zoning of a map by outlining zones with distinct characteristic
anomaly patterns. Thus, zoning helps in mapping the subsurface extents of geological units. Total Magnetic Intensity (TMI) maps were used for zoning, Tilt derivative (TDR) maps were used to delineating linear features, and the Analytic Signal
(AS) for simplification of complex magnetic anomalies. Furthermore, calculation of
Complete Bouguer Anomaly, Isostasy- and Curvature discussions are used to define
structural boundaries as well.
Presently, two procedures exist for the interpretation of potential field data, namely
the solution of the so-called direct and indirect cases. The indirect problem deals
with theoretically-calculated anomalies based on a postulated model which is altered
and compared with the observed anomaly until a reasonably "good fit" is obtained.
When the body parameters are calculated directly from the observed magnetic field
a direct interpretation approach is used. Considering this, it must be borne in mind
that the accuracy of any quantitative analysis is reduced by three main factors:
• Imperfect source body geometry, e.g. deviation from flat top and parallel sides
and infinite depth extension,
• Heterogeneity of i) the magnetic susceptibility, e.g. chilled margins will be finer
grained and thus of a lower susceptibility, ii) density contrasts,
• remanent magnetism.
Due to the fact that geological control on Antarctic anomalies is lacking or limited
to nunataks, the huge extent of the compiled surveys and the more regional aspect
of the project lead to the decision only to apply direct methods and interpretation
techniques. Magnetic interpretation of terrains in particular is often difficult, due to
their complex geological and tectonic evolution, resulting in uneven distributions of
magnetic minerals, polyphase deformation and associated metamorphism and variability in remanent magnetization.
The use of different Fourier filtering techniques will enable numerical analysis and
interpretation of potential field data. All the filter outputs are based on the measured
field and are limited by the data quality, the quality of filter techniques and the
observer’s experience. Sharpening filters, such as high-pass, downward continuation,
vertical and horizontal derivatives are useful to enhance short wavelength features.
The opposite effect can be realized with smoothing filters, like low-pass-filter and
upward continuation, to enhance longer wavelength features. A third class of task
includes transformations to convert data from one phase to another, for example by
Reduction to the Pole.
36
METHODOLOGY
2.6.1
2.6. DATA VISUALISATION
Basic Interpretation of Magnetic Anomalies
Magnetic interpretation is somewhat complicated by the fact that the magnitude and
shape of the anomaly is not only related to the magnetic mineral content, depth and
attitude of the causative body but also to its attitude or orientation to the direction
of the inducing field. It may be further complicated by the presence of remanent
magnetism often in a different direction to that of the present magnetism induced
field component.
The amplitude is determined by the depth, the magnetic susceptibility of the body,
the magnitude of the inducing field and to a lesser extent by the attitude of the body.
Surprisingly though it may seem, amplitude is of least interest in interpretation. This
is because of the large ranges of susceptibilities of apparent similar rock types.
The shape of the anomaly is of prime importance. From the shape it is possible to
determine the depth below the surface, the dip and to get some idea of the dimensions
of the body.
As already known by potential field data its often not possible to distinguished
between different types of the body geometry, like sphere and horizontal cylinder or
vertical cylinder, vertical thin dyke and thick dyke structures. The thin dyke is a
very common form of magnetic body and beside true intrusive dykes many other
geologic bodies take on the form or appearance of a dyke. A thin dyke is defined
as a sheet like body extending in strike to infinity and extending from surface or
subsurface to infinite depth and having a width of less then one-third of the depth
to the top of the body.
As mentioned, the anomaly shape stays in relation to dip, strike, depth and magnetic
inclination. For a deeper interpretation they are several catalogs and tables available,
which display these relationships, refer to body geometry. These help was used within
the detailed interpretation.
2.6.2
Total Field Shaded Relief Map
Directional sun-shading of the data can be done with varying inclination (elevation)
and declination (azimuth) angles of the illumination source. Shaded relief images
prove to be useful in determining geological strike and delineating linear features
like faults, shear zones etc. as they enhance the visibility of features in a desired
direction and suppress those in other directions. In essence, a shaded relief image
represents the first horizontal derivative in a given direction. As such, the near surface
features that are not well resolved in a simple color raster map tend to appear more
prominently in shaded relief maps.
37
METHODOLOGY
2.6.3
2.6. DATA VISUALISATION
Derivative Based Filters
The first vertical derivative is theoretically equivalent to observing the vertical gradient directly with a magnetic gradiometer. A vertical derivative map enhances the response from shallow sources, suppressing deeper ones by enhancing high-wavenumber
components of the spectrum. Thus, closely-spaced sources can be better differentiated on derivative maps. The (first) vertical derivative, n = 1, sharpens the anomaly
amplitudes, thus helping in the identification of more geological features.
L(ω) = ω n
(2.18)
and uses ω as angular wavenumber in radians per ground units as well as n as order
of differentiation.
The use of horizontal derivatives can be useful for the identification of geological
boundaries:
L(ω) = (ωi)n
2.6.4
(2.19)
Curvature attributes
The aim of the curvature analysis is to improve visualization of potential data and
hence optimize interpretation. The analysis of horizon attributes is not new at all,
but already have a prowen track record in many disciplines, e.g. medical brain
scanners, Optometry and Terrain Analysis. The application of curvature attributes
to potential field data is described by Kollersberger (2005).
Curvature attributes are related to surface attributes. The term surface is taken to
mean any surface, which is either flat, interpreted or used to control a window from
which a volume attribute can be extracted (Roberts, 2001).These surface-related
attributes can be grouped in: surface-associated attributes are those which use a
surface to extract values from a secondary data source, e.g. seismic amplitude,
volume attributes. Surface-derived attributes are computed directly from the surface
itself (dip, edge and azimuth). Within this category, curvature falls into a separate
group, the second derivative attributes, which includes Laplace-based attributes.
The third category involves surface-rendered attributes, like shaded relief and 3D
visualization.
Curvature is a two-dimensional property of a curve and describes how bent a curve is
at a particular point on the curve, i.e. how much the curve deviates from a straight
line at this point (K = dw/ds rate of change of angle dw with respect to the arc
length dS). For one point P on a curve the curvature can be defined by means of
the radius of curvature R of the osculated circle, i.e. possesses that circle of the one
common tangent T with this curve. Then one gets the simple connection that the
curvature K is the reciprocal of the radius of curvature R: K = 2π/2πR = 1/R.
38
METHODOLOGY
2.6. DATA VISUALISATION
Figure 2.14: Definition of curvature: For a particular point P on a curve, the curvature can be
defined in terms of the radius of curvature, R of the osculating circle. This circle possesses a
common tangent T with the curve. N is the vector normal to the curve at point P, which defines
the local dip angle θ. The curvature at point P is defined as the reciprocal of the radius of curvature.
The smaller the radius of curvature, the more strong the curvature of the curve is
defined. Over the entire range of a circle the curvature is constant. The curvature
K is related to the 2nd derivative of a curve.
For interpretation purposes, surface anticlines will yield in positive curvatures, synclinal surfaces will yield in negative curvatures and saddles will yield both. Ridges
will yield positive curvature in the direction across the ridge (zero curvature in the
direction along). Troughs will yield negative curvature in the direction across and
zero curvature along the trough line. More than 40 attributes are available and only
the applied techniques are briefly described.
Dip angle: extracts the curvature in the direction of maximum dip as a measure of
the rate of change of dip in the maximum dip direction, magnitude and direction of
faults is preserved, method of first derivative.
Azimuth: indicates from point of view the angle distance between point which are
on horizontal circle and the north pole, indicates changes in properties, method of
first derivative.
Minimum curvature: smallest curvature values are computed and illustrated, if the
minimum values are very small or zero, then their surface is evolved. if the minimum
values are very large, then minimum curvature shows where it folds or the break
came, this attribute is suitable more for bodies with small density/susceptibility
contrast.
Maximum curvature: the maximum curvature stands perpendicularly to the minimum. is suitable for distinguishing disturbances in geometry, moreover, the curvature
defines the orientation of these disturbances, positive curvature values indicate the
ascending side and the negative values show the sloping side, distinction between disturbances and other lineaments is possible, however, due to enormous information
39
METHODOLOGY
2.6. DATA VISUALISATION
contained in the data at times it could be confusing.
Most negative/positive curvature: is suitable only for seeking out lineaments, here
one can recognize nearly each individual lineament within a surface, it is possible
to filter the size of the lineaments and get generally better resolution in identifying
lineaments.
Dip curvature: attribute is called also "profiles curvature" and computes the idea
curvature, this is the process of extracting curvature in the direction of maximum
breaking and is a measure of the rate of the change of the angle of incidence toward
the maximum angle of incidence, size and direction of the disturbances are conserved,
method bends and exaggerates each local relief contained within the surface and can
be used to increase the differential connectivity characteristics.
Strike curvature: gives an extraction of the curvature in a direction, which runs
in the angles to the "Dip Curvature"(along the impact), separates the surface into
ranges of descending forms and edges.
Figure 2.15: Sign convention for curvature attributes: Arrows represent vectors, which are normal
to the surface. Where these vectors are parallel on flat or planar dipping surfaces, the curvature is
zero. Where the vectors diverge over anticlines, the curvature is defined as positive and where they
convergence over synclines, the curvature is defined as negative.
The analysis of curvature attributes has been done using topography, gravity and
magnetic data and with the help of algorithm as describes by Robertson (2001). The
curvature-code was written by Dr. S. Schmidt, Geophysics Division, at Christian
Albrechts University Kiel.
Each of the attributes gives slightly individual insight into the mapped surface depending on the step and size of the analyzing window used. It may be possible to
differentiate between local (short wavelength and related to near surface anomalies)
and more regional/global (long wavelength) anomalies. Regional (50 km analyzing
window) and local (10 km analyzing window) characteristics of the data are presented
in the APPENDIX D.
No surface attribute should be used in isolation and must be interpreted with reference to the origin. Within this thesis, the focus lies not within a detailed analysis
of curvature attributes.. The displayed attributes do not belong to specific physical
properties. The aim of the selected comparison (local and regional) is to identify
40
METHODOLOGY
2.6. DATA VISUALISATION
similarities to outline geologic structures (more quantifying then qualifying). Thus,
in combination with the more classical interpretation methods (see chapter Data Visualisation), the Curvature analysis will lead to the combined terrain model of DML
lithosphere.
2.6.5
Analytic Signal
The interpretation of observed magnetic anomalies is often complicated by their horizontal displacements with respect to their sources. This displacement, or skewness,
results from the fact that the directions of the geomagnetic field and induced magnetization are, in general, not vertical. Repositioning of anomalies by reduction to
the pole can be complicated, when magnetization inclination is low. Most methods
assume knowledge of the orientation of the present day magnetic field and of the
source body magnetization. The inclination and declination of the present day magnetic field are well known. In the absence of oriented magnetic samples, one often
assumes that the source body magnetization is purely induced; an assumption that
is very often not justified. The Analytic Signal is given by:
s
A(x, y) =
(
∂T 2
∂T 2
∂T 2
) +(
) +(
)
∂x
∂y
∂z
(2.20)
where T is the observed field at x and y. While this function is not a measurable parameter, it is extremely interesting in the context of interpretation, as it is completely
independent of the direction of magnetization and the direction of the inducing field.
This means that all bodies with the same geometry have the same analytic signal.
Furthermore, the peaks of analytic signal functions are symmetrical and occur directly over the edges of wide bodies and directly over the centers of narrow bodies.
Under the assumption that the anomalies are caused by vertical contacts, the analytic signal can be used to estimate source depth using a simple amplitude half-width
rule (accuracies in depth determination are in the order of 30%). This avoids the
difficulties that are often faced in the conventional process of reduction to pole for
δT , when the effects of natural remanent magnetization on the source magnetization
distribution are usually unknown. The implementation of the AS calculation has
three steps.
• low-pass filtering of δT
• processing to obtain the gradients of δT with respect to the x, y and z directions
• calculation of the AS
The calculation of the AS is illustrate in figure 2.16, which shows how it results in
the determination of source characteristics without making assumptions about the
direction of source body magnetization.
41
METHODOLOGY
2.6. DATA VISUALISATION
Figure 2.16: Schematic outline of the Analytic Signal, after Roest (1992).
However, the data should nevertheless be interpreted with care for the following
reasons:
• The amplitude of the analytic signal varies with the effect of magnetization,
and therefore remains a function of the ambient magnetic field parameters.
• The analytic signal over magnetization contrasts that are closely spaced or
dipping are more complicated than the assumed bell-shape function found over
a single contrast.
• The analytic signal over structures that intersect at an acute angle is complicated because of the nonlinear combination of signals.
• The calculation of the analytic signal is based on derivatives of the magnetic
anomalies, so that gridding artifacts, errors like track corrugations, and noise
in general, are all enhanced.
Processing was realized with Geosoft Oasis montaj software.
2.6.6
Tilt Derivative
Derivatives of potential field data can help define and estimate the physical properties
of the source structure causing the anomaly. The tilt derivative is highly suitable
for mapping shallow basement structures and has distinct advantages over many
conventional derivatives.
The problems to overcome in determining the shape and edges of magnetic source
structures are to identify and map:
42
METHODOLOGY
2.6. DATA VISUALISATION
• subtle anomalies attenuated in dynamic range due to the presence of high
amplitude magnetic anomalies
• continuity of individual bodies that feature lateral changes in susceptibility and
/or depth of burial and
• edges of structures by adequately accounting for the nature of the rock magnetization. Rock magnetization is a vector quantity that can consist of both
remanent and geomagnetically induced components. The remanent component can affect the shape of the magnetic field response and result in spurious
derivatives
The complex Analytical Signal for 2 D structures is
A(x, z) = |A|jθ
(2.21)
with the known Analytical Signal:
s
|A| =
(
∂T 2 ∂T 2
) +
)
∂x
∂y
(2.22)
where T is the magnitude of the total magnetic intensity (TMI) and θ =
∂T
tan−1 [ ∂T
∂z / ∂x ] is the local phase.
The tilt derivative is similar to the local phase, but uses the absolute value of the
horizontal derivative in the denominator:
T DR = tan−1 [
V DR
]
T HDR
(2.23)
with VDR and THDR as the first vertical and total horizontal derivatives, respectively, of the TMI.
The important points to note about the tilt derivative are:
• the AS is invariant for all inclinations
• it normalizes a magnetic field image and discriminate between signal and noise
Processing was realized with Geosoft Oasis montaj software.
43
METHODOLOGY
2.6.7
2.6. DATA VISUALISATION
Depth Estimation
Potential field data have, by their nature, a very broad band of information in a single
measurement that includes the contributions due to all physical sources (geology).
The resolution of different sources is dependent on the noise levels of the measuring
system and on the ability to resolve overlapping signals. Roughly qualitative information is given by a spectral analysis of gridded data. The energy spectrum is a 2D
function of the energy relative to wavenumber and direction. The radially averaged
energy spectrum is a function of wavenumber alone and is calculated by averaging
the energy in all directions for the same wavenumber.
When considering a grid that is large enough to include many sources, the use of the
log spectrum of this data is useful to determine the statistical depth to the tops of
an ’ensemble of sources’, using the relationship (see Spector and Grant, 1970):
logE(k) = 4πhk
(2.24)
where h is the depth in ground units and k is the wavenumber in cycles/ground
units.
The Nyquist wavenumber, N , is the largest wavenumber that has been sampled by
the grid, and is defined as one over twice the grid cell size.
N = 1/(2 ∗ cellsize)
(2.25)
If the gridded cell size is 3 km, the Nyquist wavenumber is 0.16 km. Furthermore,
the smallest detectable depths are defined using the size of the grid cell by:
hmin = 0.4δx
(2.26)
The grid cell size of VISA data is 3 km, which leads to an estimate of the smallest
detectable source at 1.2 km depth.
It is possible to determine the depth of the source ensemble by measuring the slope
of the energy spectrum and dividing by 4π. A typical spectrum may exhibit three
parts: a deep source component, a shallow source component and a noise component.
However, in conjunction with the finite detectable wavelength (wavenumber), defined
by the Nyquist frequency (see above), it is difficult to define a full ensemble of sources.
Due to this, a 5 point average slope of the energy spectrum is used, and will be
illustrated in the solutions of the approximate depth calculations (see APPENDIX
C).
Note that the estimation of linear trends for the ensemble of sources is strongly
subjective. For an a exponential value (energy), the estimation of the depth will be
calculated using linear approximation of trends. The qualitative aspect of this trend
44
METHODOLOGY
2.6. DATA VISUALISATION
of interpretation is that, for the increasing depth solution, the error also increases.
For low frequencies, only a few points are recognized in contrast to high frequency
anomalies. This is due to the transformation process from the space domain (grid)
to the wavenumber domain. Consequently, the solutions from this analysis can only
be used as a rough guide in qualitative interpretation.
2.6.7.1
3D Euler Deconvolution
The objective of the 3D Euler process is to produce a map that will show the locations
and corresponding depths of the geologic sources observed in a two dimensional grid.
The Standard 3D Euler method is based on Eulers’ homogeneity equation, that relates the potential field and its gradient components to the location of its sources,
by the degree of homogeneity N , which may be interpreted as a structural index
[Thompson, 1982]. The structural index is a measure of the rate of change with
distance of a field.
The calculation uses a least squares method to solve Euler’s equation simultaneously
for each grid position within a sub grid (a square grid that is moved along each grid
row). At each grid point, there will be 10 grid window equations, from which the
four unknowns (x,y,z as location and the background value) and their uncertainties
are obtained for the specific structural index. A solution is recorded if the depth
uncertainty of the calculated depth is less than a specific tolerance and the solution
is within a limiting distance to the center of the data window.
Any 3-dimensional function F (x, y, z) is to be homogeneous at the degree n if the
function obeys the expression:
F (tx, ty, tz) = tn F (x, y, z)
(2.27)
From this, the Euler’s equation can also be satisfied:
x
∂F
∂F
∂F
+y
+z
= nF
∂x
∂y
∂z
(2.28)
An anomaly over an idealized symmetrical source can be written as:
F (x, y, z) =
K
rN
(2.29)
with r2 = (x − x0 )2 + (y − y0 )2 + (z − z0 )2 , (x0 , y0 , z0 ) is the position of a source whose
field f is measured, K a constant and N a real number, which depends on source
geometry, the measure of the fall-off rate of the field, and which may be interpreted
as the structural index (SI) and thus equivalent to −n in Euler’s equation. The
equation is homogeneous of grade n = −N .
45
METHODOLOGY
2.6. DATA VISUALISATION
For z = 0, the observation plane, Euler’s equation can be re-stated as:
(x − x0 )
∂F
∂F
∂F
+ (y − y0 )
+ (z0 )
= −N F (x, y, 0)
∂x
∂y
∂z
(2.30)
Thus, Euler deconvolution provides an excellent means of gaining a broad indication
of the depths and locations of various sources in a given area, provided appropriate
dimensions are selected for parameters like the grid cell size, window size and structural index. Euler deconvolution is a faster method of covering the whole area for
depth and boundary estimation of sources than modeling individual anomalies. It
also helps in delineating linear features more precisely, as focused solutions cluster
along these features.
The significant advantage of Euler deconvolution for magnetic data is that it is
insensitive to the effects of magnetic inclination, declination and remanence.
The following table summarizes the structural index for simple models:
geologic model
sphere
pipe
horizontal cylinder
dyke
sill
contact
number of infinite dimensions
0
1(z)
1(x-y)
2(z and x-y)
2(x and y)
3(x,y,z)
magnetic SI
3
2
2
1
1
0
gravity SI
2
1
1
0
0
NA
The overall processing sequence for Euler deconvolution consists of the following
steps:
• Preparation and gridding the potential field data with respect to sampling
interval and line spacing,
• Applying of FFT and convolution grid enhancement,
• Processing to calculate derivatives,
• Analyze grids (Standard or Located Euler Deconvolution) for each structural
index,
• Plot results,
• Repeat until acceptable.
The unusual aspect of using Euler Deconvolution is that one must have some initial
estimate of the sources types in order to select a structural index. Accordingly, there
is a need for additional information, for example from drill sites. In the case of the
DML region, additional information comes from field geologists (structural geologist,
J.Jacobs, University of Bergen), who has investigated DML in detail.
46
METHODOLOGY
2.6. DATA VISUALISATION
In regional interpretations, one is interested in identifying contacts and faults, so that
an index of between 0 and 1 should be used. The maximum distance for acceptable
solutions was set to 20 km, taking into account the average spectra analysis. The
processing window size is 45 km, which is large enough to include variations within
the data, but small enough not to include effects from multiple sources. As the
deconvolution is a statistical process, with associated uncertainties, the maximum
depth tolerance was set to 15 %.
The results are displayed in ordinary maps, combining the location and the depth
solution. Additionally, the clustering of a given solution given the choice of an
appropriate structural index, can be used as an interpretation tool. For example, a
dyke structure would be displayed with a linear trend of solutions while vertical pipe
would be shown as a point solution.
The results of the Standard Euler Deconvolution is displayed in the Appendix C.
Due to the huge amount of information inside the Standard Euler Deconvolution,
which includes uncertainties, which may be over-interpreted, the Located Euler Deconvolution was applied too.
The Located Euler Deconvolution uses the Analytic Signal to find and recognize
peaks in the anomaly pattern, so that solutions are only estimated over recognized
anomalies. After these peaks are localized, their locations are used for the deconvolution process. This process combines a window size that is varied according to the
observed anomalies. Finally, the Located Euler Deconvolution produces far fewer
solutions than the standard method, which are consequently simpler to handle.
In the interpretation chapter, the solutions of the Located Euler Deconvolution are
displayed using the AS map with overlain depth solution.
Processing was realized with Geosoft Oasis montaj software, accordingly to Reid
(1990) and Thompson (1982).
2.6.8
Isostasy
In order to study density variations within the upper crust, the thickness of the
crust, mass distributions and isostatic adjustment have been computed. Important
lithospheric units have been identified and the resulting structural information have
been used in the interpretation of tectonic provinces. In view of isostasy, two classical
and contrary conceptual models exist:
Pratt isostasy (Pratt, 1855): The density of the crust varies inversely with the
height of the topography and the depth of compensation is at the base of the horizontal crust-mantle boundary. This model is generally accepted within the range of
continental transition zones at passive continent-ocean boundaries.
Airy isostasy (Airy, 1855): The crustal density is constant beneath both the
elevated topography and the level region. The roots extend beneath the elevated
topography and the depth of compensation is at the base of the crust where the
pressure is constant.
47
METHODOLOGY
2.6. DATA VISUALISATION
Both models assume the local reconciliation of the topographic loads and it is well
known, that part of the loads is regionally compensated (see Barrell, Vening Meinesz,
Gunn). The principle of regional compensation led to advancement of the iso-static
models.
Figure 2.17: Isostasy: local isostasy (Airy) and regional isostasy (Vening Meinesz)
The computation of the Isostasy after Vening Meinesz (1939) is based on the principle of Airy’s model. In addition, the model incorporates flexural rigidity of the
lithosphere which partly supports the topographic load and takes into consideration
the regional iso-static reconciliation. The regional compensation at the crust-mantle
boundary can be considered as low-pass filter.
The calculation was realized using algorithm from Parker (1972) and Banks (1977).
The following parameters have been used: crustal-density of 2670 kgm−3 , density
contrast of 400 kgm−3 across the Moho accommodated the mass effects of the compensated terrain, with averaged crustal-thickness of 34 km.
48
Chapter 3
SURVEYS and DATABASE
3.1
Campaigns
The campaign/processing chapters will give a short summary of the main processing
steps and the individual characteristics and similarities of the different datasets and
campaigns. All calculations are done by the author, except the determination of the
onset of the RES-data.
The extent of the survey areas, from 14◦ W to 20◦ E and from 70◦ S to 78.5◦ S, is large
enough to fully recognize long wavelength regional anomalies. However, even smaller
features could be mapped, owing to the average line spacing of about 10 km. The
VISA project was subdivided into four austral-summer campaigns:
Figure 3.1: Overview of the study area and campaigns.
49
SURVEYS and DATABASE
3.1. CAMPAIGNS
VISA I campaign
VISA I lasted from December 2001 to February 2002 and lasted 100 flight-hours
during which 27700 profile km were flown. The survey area stretches from 4◦ W
to 10◦ E and from 70◦ S to 75◦ S with Neumayer Station (Germany) and E-Base
(South Africa) as operating base stations. A line spacing of 10 km was chosen.
Three GPS reference stations were established at various locations in DML: Kohnen
Station, Bleskamin Ice Rise and Soerasen. Additionally, magnetic base stations were
established at E-Base, Kohnen and Neumayer Station.
VISA II campaign
The VISA II campaign (December 2002 to January 2003) amounted to 54.5 hours
flying time, covering 13300 profile km. As gravity measurements do not allow
frequent flight level changes, a level of 3960 m a.s.l. was adopted for most flights in
view of the surface topography and cloud level. The spacing of the parallel profiles is
10 km. For post-processing two GPS reference stations were established, at Kohnen
on Weigel Nunatak near Kottas Camp, and close by the Watzmann seismic array
at Halvfarryggen. In addition, magnetic base stations were established at Kohnen
and Kottas Camp. Furthermore, GPS data and magnetic data were collected at the
geophysical observatory of Neumayer.
VISA III campaign
VISA III (December 2003 to February 2004) was completed with 133 hours of
flying time and consists of 31150 profile km. The survey flights were carried out
from Novolazarevskaya (Novo) runway and SANAE IV. While the profiles flown
from Novo runway had a spacing of 20 km, a flight level around of 4000 m, and
a north-south orientation (6◦ W to 20◦ W and 71.5◦ S to 76.5◦ S), the flights from
SANAE IV had a spacing of 10 km and maximum flight level of 1500 m toward
the north (1◦ W to 16◦ E and 70◦ S to 71.5◦ S). Four GPS reference stations were
established at DML25, next to Kohnen, at Weigel Nunatak near Kottas Camp,
FOR1 at Schirmacher Oasis and Novo Runway. A magnetic base station was set up
at Novo Runway. Furthermore, GPS data and magnetic data were obtained at the
geophysical observatories at Neumayer and SANAE IV.
VISA IV campaign
VISA IV flights were conducted in Western Dronning Maud Land with Neumayer
and Kottas Camp as their main bases from December 2004 to January 2005. The
survey area was located from 14◦ W to 0◦ and from 74◦ S to 78.5◦ S. The spacing
of the parallel profiles was 10 km, with flight altitudes of 3200 m. GPS reference
stations and magnetic base stations were established at Amundsenisen and Kottas
Camp. Additional GPS and magnetic data were obtained from Neumayer Station
and SANAE IV.
50
SURVEYS and DATABASE
3.2
3.2. AIRBORNE RES DATA
Airborne RES Data
The aim of processing RES data was to generate maps showing the ice surface and
underlaying bedrock topographies. These are of crucial importance in understanding
the geology, especially in the tectonic content. Several topographic models already
exist, resulting from satellite investigations and small scale RES soundings, and are
available from the international continent-scale BEDMAP database. Within the
VISA project, the RES technique was used to generate a more local topographic
model. Furthermore, in conjunction with airborne gravity data, these more highly
detailed models are needed to calculate the Complete Bouguer Anomaly.
The input data sets are the kinematic GPS solution and the relative thicknesses
of different media (air and ice) from the RES data. After merging these data by
timecode we receive absolute values of surface and bedrock topography by simple
subtraction from the well-known aircraft position.
The data are leveled using the tie lines as reference and archive blockshift fit. After
this, the Kriging gridding routine was applied, with a cell spacing of 3 km. Ordinary
Kriging is a geo-statistical interpolation method, which determines the interpolated
values as a function of the distance to neighboring data points and the associated
variance and relies on the spatial correlation structure to calculate the weighting
values during interpolation.
The gridded ice surface topography data show a smooth surface without any noise or
artifacts. Comparisons to other ice topographic datasets and models (i.e. ICESat;
Bamber-model, 1997; BEDMAP-data) show that the results are good. Only in
regions with strong gradients are misfits, of up to 400 m recognized after comparing
VISA data to the older products (see Wesche, (submitted)). These large errors
can be neglected, because each dataset is limited by its own errors. For example,
BEDMAP data show strong differences along the coast lines (these data are mostly
not direct measured but instead are digitized from other datasets), while ICESat
data show differences at locations with strong gradients in topography, especially on
the outcropping nunataks. Additionally, the ICESat data are limited due to cloud
level conditions and this may result in differences of up to 50 m.
To level the ice topography data, it has to be taken into account that the electromagnetic waves of the RES have variable penetration depth with respect to snow
and firn conditions. Ground GPS data are thus needed for ground truthing. These
additional GPS data are provided along traverse profiles made using snow vehicles
on their way to support the Kohnen Station or were done by static measurements at
different camp sites within the field. Differences of up to 20 m in height between the
GPS measurements and RES results are recognized, and the mean average correction
was applied to the ice topography dataset.
With the exception of the VISA IV campaign which is purely continental, three
topographic provinces can be recognized within each survey: The ice sheet, which
has its origin in the hinterland and is continuously "fed", the mountain ranges with
local nunataks and channelized ice flow, and the grounding line region which marks
the transition from grounded ice sheets to floating ice shelves.
51
SURVEYS and DATABASE
3.2.1
3.2. AIRBORNE RES DATA
Topography VISA I
Ice Topography
(1) The coastal area is characterized by flat-lying ice shelves and sea ice with an
average height of 40 m above sea level. This area flattens toward the north and is
only disturbed where ice sheets and shelves are grounded, which yields in differences
in topography of several tens to hundreds of meters at separated locations. These
locations are well known, namely Soerasen and Halvfarryggen. Another interesting
feature is the ice flow of the Jutulstraumen glacier, which is focused over a narrow
sub-ice trough and widens out to join the sea ice over the shelf.
(2) From the continental coast line, a general increase in ice topography is recognized
until the mountain range, especially in the eastern part. The western extension
displays a more disturbed pattern due to the presence of mountain chains, which are
separated into two linear and NNE striking lineaments, near the coast.
In the central part over the ice sheet, a smoothing effect of thick ice, without significant undulations, is recognized. Consequently, the sheet flows by following the
subice topography.
(3) The hinterland is a more or less stable region, except for a central trough, with
topographic heights starting ranging from 2000 m to 3400 m.
From the resulting ice-topography, the bedrock topography is calculated by simple
subtraction of the ice thickness.
Bedrock Topography
(4) The offshore region, which is marked by the coastline (thick line) shows topographic features below grounded ice both at and below sea level. A pattern of troughs
and basins is evident. Remarkable features are the Halvfarryggen, situated at the
most north-western extent of the survey, and around E-Base, and the northern extension of the Jutulstraumen graben system, which is characterized by a trench-like
lineament reaching toward the north. The exact topography north of the grounding
line can not investigated with the RES technique as the ice-water interface beneath
the floating ice shelves constitutes a barrier to further propagation of radar signals.
A continuous increase of bedrock topography can be recognized within a band of
50 km from sea level to the mountain range. Two thirds of this area is dominated by
average elevations of 1000 m and more. Apart from this, strong segmentation due to
tectonic processes is recognized.
(5) Deep-seated valleys and rift structures separate the mountainous subice topography, namely by the Jutulstraumen-Penksökket-Graben system. These, suggested
to be parts of a failed rift system, are subdivided into three known parts and extend
further toward the SE, beyond the edge of figure 3.2. The main branch trends NESW, reaching lowest topographic heights of 1500 m below sea level and ends at the
shelve break. The real topography of this trench can not be traced further towards
the north. A further trench, parallel to this is situated nearly 80 km towards the
East. Both trenches continue in the more NE to SW oriented Penksökket graben
(not shown on this map, but seen in the compilation chapter). The third trench
strikes NS over a distance of more then 200 km.
52
SURVEYS and DATABASE
3.2. AIRBORNE RES DATA
The most southern extent of this survey displays a more or less stable region with
average heights of 500 m above sea level, but is disturbed by EW and NE-SW trending
highlands or mountain chains with average heights of 800 m, and intervening valleys
which can reach depth of 200 m below sea level.
Figure 3.2: RES Results from the VISA I campaign. top: ice topography, overlain by 200 m contour
interval, bottom: bedrock topography.
53
SURVEYS and DATABASE
3.2.2
3.2. AIRBORNE RES DATA
Topography VISA II
Ice Topography
(1) The northern area is characterized by the ice shelves, which display a smooth
and flat surface structure, starting at 40 m a.s.l.. The Ekstroemisen bay is flanked
by tongue-like topographic features, namely Halvfarryggen and Soerasen, striking
towards the north. The grounded Halvfarryggen and Soerasen display heights of up
to 700 m and 400 m a.s.l..
(2) A stable and continuous rise in ice topography toward the mountains in the south
is recognized. Only the center-east region is characterized by stronger gradients, due
to a spur of the mountains, which reaches the 2000 m level, toward the coast.
South of 73◦ S, the topography decreases slightly down to 1800 m over the Penksökket
graben system, which strikes SW-NE beneath the ice sheet.
(3) Over the mountain chains, topography again reaches the 2000 m level. Several
outcropping nunatak groups are visible. Further south, the topography becomes
plateau-like at around 2800 m and more.
Bedrock Topography
(4) The sub-ice topography can be subdivided into four distinctive parts. The northern part combines both the shelf region, with a lowest depth of around 600 m below
sea level and is divided by the Halvfarryggen and Soerasen. A SW-NE striking basin
marks the boundary toward the continent.
(5) At 1000 m above sea level, the first prominent feature inland of the coastline
toward the south is a massive block with an area of 22000 km2 that strikes SW-NE.
The SW-NE trending continuation of the Jutulstraumen-Penksökket graben system
divides the area. The Penksökket graben itself is characterized by topographic depth
of up to 500 m below sea level and increases up to 100 m a.s.l toward its eastern end.
No direct continuation to the Jutulstraumen part is recognized due to the flank of a
nearly 20 km-wide rock massif.
(6) The southern part is dominated by the SW-NE striking mountain range, with
average heights of 2000 m and a width of 100 km.
The most southerly extension becomes less clear, with fragments of trenches and
valleys, reaching down to 200 m below sea level, whose strikes vary wildly. Distinctive
rock massifs 30 km in width are situated in the SE, reaching heights of around 400 m
above sea level.
54
SURVEYS and DATABASE
3.2. AIRBORNE RES DATA
Figure 3.3: RES Results from the VISA II campaign. top: ice topography, overlain by 200 m contour
interval, bottom: bedrock topography.
55
SURVEYS and DATABASE
3.2.3
3.2. AIRBORNE RES DATA
Topography VISA III
Ice Topography
The displayed topography can be divided into three parts. (1) The northern part
consists of the ice shelf area with flat topographic behavior. Ice rises disturb the
region, namely Vigridisen and Nivilisen, as do the north-trending pin points between
8-9◦ W.
(2) From the coastline southward, a more or less constant increase in height can
be recognized until the mountain chains, with elevations around 2000 m. These
mountainous regions strike E-W parallel to the coast. A wide band of nunataks is
recognizable.
(3) From this linear feature, 50 to 100 km southward, the 3000 m contour marks
the boundary to the south, a more stable region, very flat and plateau-like, tilted
slightly upwards toward the SE. Maximum topographic heights reach the 3550 m
level.
Bedrock Topography
The survey can also be divided into three major parts: the flat region, situated in
the furthest north, the mountain chain in the center, and the hinterland plateau,
which is internally segmented.
(4) The northern region contains the ocean transition zone with topographic heights
below sea level. A well-defined basin structure extends over several hundreds of
kilometers at mostly 500 m b.s.l.. This basin strikes EW, parallel to the coast.
Several N-S orientated ridges reach north from the mountains, giving rise to a slight
increase in topography over a width of 50 km or more. Apart from these trends, the
area around the Russian base Novo and adjacent Schirmacher Oasis is characterized
by a more E-W trend over an area of 2300 km2 , with elevations around 300 m.
(5) The mountainous region also trends E-W, parallel to the coastline, and is 130 km
wide with maximum heights of more than 2500 m above sea level.
(6) South of the mountains, the 1000 m contour marks the boundary to the hinterland region, with maximum topographic heights between 500 and 1000 m a.s.l..
This area is characterized by prominent branches, segments, lineaments and basins
up to 40 km wide and below sea level. These features display a rough and disturbed
pattern without any preferred orientation.
56
SURVEYS and DATABASE
3.2. AIRBORNE RES DATA
Figure 3.4: RES Results from the VISA III campaign. top: ice topography, overlain by 200 m
contour interval, bottom: bedrock topography.
57
SURVEYS and DATABASE
3.2.4
3.2. AIRBORNE RES DATA
Topography VISA IV
Ice Topography
As the survey covered a purely continental area, the topography is monotonous and
shows only minor signatures induced from bedrock and glacier movement.
(1) In the northernmost part, a SW-NE-trending mountain group is situated and
marks a region with strong topographic gradients, climbing from 900 m a.s.l. to
2000 m over a distance of 30 km.
(2) Reaching the 2000 m level, where limited outcrops are situated, the hinterland
topography becomes very stable again. A slight SW-NE trend can be recognized
in elevation, which results in an average elevation of 2400 m above sea level in the
further southeast.
Bedrock Topography
(3) The northern region is dominated by a small-deep valley, reaching down to 600 m
b.s.l., flanked to the SE by the mountainous region with average heights of 2000 m.
The strike direction is more or less SW-NE, but the upland displays several internal
structures which are more N-S oriented.
A broad ridge trends SE from the mountains, and is flanked by basin structures
20 km wide and with depth down to 500 m b.s.l.. This feature marks the northern
boundary of a huge basin structure that dominates the south of the survey area.
(4) The full basin is not covered by the survey, but the imaged area is more then
70400 km2 . This basin depth decreases gently to around 1000 m b.s.l.. Step sided
circular features are recognized in the SE and SW corners, and reaching elevations
of 500 m above sea level.
58
SURVEYS and DATABASE
3.2. AIRBORNE RES DATA
Figure 3.5: RES Results from the VISA IV campaign. top: ice topography, overlain by 200 m
contour interval, bottom: bedrock topography.
59
SURVEYS and DATABASE
3.2. AIRBORNE RES DATA
60
SURVEYS and DATABASE
3.3
3.3. AIRBORNE GRAVITY DATA
Airborne Gravity Data
For accurate levelling of the airborne gravity data it is necessary to have precise
knowledge of the absolute gravity at the airport and of its daily variation as well
as the detection of drift and disturbance induced effects for the airborne measuring
system.
To connect the local measurements to the absolute gravity net, a tie to the last
absolute gravity reference point (South Africa) was made. This first crucial step
must be done due to the relative absence of absolute gravity readings in Antarctica.
3.3.1
Free-air Anomaly VISA I
The tying process to the IGSN-71 net started in Cape Town, the nearest absolute
gravity observatory point to the survey. At Cape Town, the following measurements
were made at:
station
absolute gravity
value [mGal]
measured relative
gravity value [mGal]
979657.90
3159.52
Poller 94
The following gravity values were obtained for the three stations in Dronning Maud
Land, Boreas/Passat-nunataks, Neumayer Station and E-Base:
station
Boreas/Passat
Neumayer Station
E-Base
absolute gravity
value [mGal]
measured relative
gravity value [mGal]
982648.49
982748.67
982723.41
6150.11
6250.28
6225.03
Gravity values at Neumayer Station and E-Base are used for the tie process, because
these where the base stations. The quality of the calculated absolute gravity values
are difficult to assess, because both stations are situated on the ice shelf, which is
moving due to glaciological processes as well as influenced by tidal movement.
A second important step involves making daily base station readings, normally before
and after flight, to compare the ground truth data with the airborne gravity meter.
This will give a main level base value and detect sensor disturbances due to the
sensor’s mechanical system. Also, drift parameters are recognized and these are part
of quality control. A table of base station, ground-truth and airborne-sensor readings
is given in the Appendix B.
Summarizing the observed drift by base station and stage of operation- the sensor
system was demounted during the campaign; the survey had a medium quality. The
first part of the campaign, based on Neumayer, display a 3.5 mGal drift (9 days).
61
SURVEYS and DATABASE
3.3. AIRBORNE GRAVITY DATA
Based on E-Base, a drift of 6.7 mGal (17 days) is recognized and may have been
influenced by a hard landing procedure. The last stage of the campaign, based on
Neumayer station again, displays 0.7 mGal drift over 15 days. The real drift will
decrease due to the influence of tidal movement of up to 0.5 m of the ice shelf, which
is not recognized. The raw data indeed will highlight jumps in gravity readings and
drift processes.
The levelling routine minimizes these uncertainties, but we have to bear in mind,
that the application of levelling, although a numerical routine, is influenced by the
operator’s technique. Levelling was difficult due to the limited number of tie lines, so
that every single line was levelled separately on different intersection points, meaning
not even every tie line (1 full tie line and 2 half tie lines) was generally used as
reference-level.
Other difficulties arise during the process of calculating the Free-air anomaly itself.
Due to in flight-disturbances, or survey design limitations, changes in the flight path
trajectory gave rise to additionally forces on the gravity meter’s sensor. As a result it
was necessary to separate each line due to the external disturbances and this results
in a strong segmentation. The levelling (only zero order trend applied) minimized
the intersection error to 4.4 mGal, with an standard deviation of 5.7 mGal.
The free-air gravity map is characterized by long wavelength anomalies that represent
mostly deep-seated crustal structures. As expected, there is a strong correlation
between free-air gravity and sub-ice elevation.
A prominent SW to NE trending gravity low can be interpreted as the expression
of the Jutulstraumen-Penksökket graben system (2). The measured values on this
sharply defined gravity low is -113 mGal, which is the most negative gravity anomaly
in this survey.
Positive gravity anomalies are situated over the flanks of the Jutulstraumen rift system and the sub-ice mountain range, mostly induced due to the topography. These
long wavelength anomalies reach a maximum of 170 mGal.
A gravity low orientated parallel to and NW of the Jutulstraumen terminates at an
E-W orientated gravity anomaly low (1). Orientated along the coast this long wavelength and gravity anomaly low is interpreted as a basin structure near the transition
zone between ocean and continent. Further north, positive high-amplitude gravity
anomalies with average values of 100 mGal represent seaward dipping reflectors and
a shallow crust-mantle boundary (Jokat et al., 2003, 2004).
A NW-SE trending low branches off from the center of the Jutulstraumen rift system
and continues well to the south, although here it is much broader with a minimum
value of -80 mGal (3).
Influenced by strong gradients in topography the free-air anomaly displays a disturbed pattern with extreme values ranging from 170 mgal down to -110 mGal.
62
SURVEYS and DATABASE
3.3. AIRBORNE GRAVITY DATA
Comparison with the detailed BAS survey from 2001, situated in the center of the
area around Jutulstraumen, published by Ferracioli et al., (2005), shows no significant
differences.
Figure 3.6: Free-air anomaly of VISA I campaign.
63
SURVEYS and DATABASE
3.3.2
3.3. AIRBORNE GRAVITY DATA
Free-air Anomaly VISA II
The tying process was less accurate due to the absence of any gravity reading at
a known station in the IGSN 71 net before surveying. The only way of using the
investigated survey data was the use of a dummy value from the year before readings
as well as the post survey tying process. This, of course, involves questionable
assumptions, but there are no other information available.
Using this "virtual" Cape Town station with readings of 3159.50 mGal (relative),
979657.90 mGal (absolute) with the single reading at Neumayer Station yields the
following absolute gravity value for the VISA II campaign:
station
Neumayer Station
absolute gravity
value [mGal]
measured relative
gravity value [mGal]
982738.83
6240.42
In fact, it is not possible to make any statement about the quality of gravity measurements. Long term experience, combined with the knowledge of gravity measurements
of the upcoming seasons (2003/2004) makes these reading/calculations repeatable.
A yearly drift of 10 mGal for the Neumayer station seems likely in view of the slow
movement of the ice shelf by nearly 160 m per year.
Summarizing, the drift of 3.9 mGal over 16 days for the aero-gravimeter during surveying and the tying uncertainties together mean that the survey has bad input
quality, but is internally very stable.
The levelling (only zero order trend applied) minimized the intersection error to
3.6 mGal, with an standard deviation of 3.9 mGal.
The strong correlations with sub-ice topography are clearly recognized within the
free-air anomaly map. Compared to the topography, the free-air anomaly is much
smoother and is mainly influenced by the masses of rock material above the reference
as well as separates due to deep-seated crustal structures. This is especially seen in
the most northern part, where positive long wavelength anomalies with values up to
150 mGal, can be interpreted as seaward dipping basalt flows, in conjunction with
reduced crustal thickness in the continent-ocean transition zone (1).
A prominent gravity low, with measured values of -100 mGal, trends SW-NE, with a
wavelength of 80 km and characterizes the boundary to the massive SW-NE trending mountainous region (2) seen in the sub-ice topography, whose gravity reaches
100 mGal. The southern region is divided by the Penksökket graben, which displays
a gravity low with values of -50 to 10 mGal.
The E-W trending mountain range (3) is characterized by a chain of positive gravity
anomalies, 80 km wide, with partially high amplitudes of 170 mGal and a wavelength
along the chain of 50 km. Two additional features are recognized from the main
mountain chain, which strikes mostly parallel to the coastline. In the south-eastern
and the south-western parts, the free-air anomaly field is characterized by isolated
anomalies, divided by a distinctive 20 km wide low of 0 mGal. Summarizing this,
64
SURVEYS and DATABASE
3.3. AIRBORNE GRAVITY DATA
the free-air anomaly displays a more or less undisturbed pattern with respect to the
topographic behavior.
Figure 3.7: Free-air anomaly of VISA II campaign.
65
SURVEYS and DATABASE
3.3.3
3.3. AIRBORNE GRAVITY DATA
Free-air Anomaly VISA III
The procedure for tying the local gravity measurements to the IGSN 71 network was
difficult because no gravity measurements were made before survey. Only the post
survey reading at Cape Town (South Africa) was used as a real measured value for
this purpose. Under the assumption that no disturbing effect was encountered during
the survey, the following measurements and calculations are used:
station
Cape Town (UTC)
Novo (airfield)
Sanae (airfield)
absolute gravity
value [mGal]
measured relativ
gravity value [mGal]
979616.80
982467.58
982457.72
3254.19
6104.98
6095.12
During this austral summer campaign, an absolute gravity measuring project carried
out by J. Mäkinen. His absolute values, also measured at the Novo airfield and Sanae
station, much later confirmed our tying process within an accuracy of 1.0 mGal.
The drift process is referred to the airborne gravimeter, with the corresponding base
stations displaying normal sensor operations. A drift of 4.28 mGal over 9 days at
Novo might be quite high, but is within the acceptable range. The observed drift at
Sanae station, of 0.2 mGal over 4 days, displays normal operation.
The levelling (only zero order trend applied) minimized the intersection error to
5.5 mGal, with an standard deviation of 6.9 mGal.
With respect to the survey and topographic behavior, the interpretation of the freeair anomaly can be made with reference to four different areas.
The northern extent, flown with E-W profiles, displays the continent-ocean transition. Strong field gradients are recognized, beginning from 150 mGal offshore and
dominantly striking parallel to the coast. As elsewhere, these can be interpreted as
structures and seaward dipping basalt flow sequences (1).
The positive anomalies are followed inland by an extensive gravitational low. This
-100 mGal low dominates the foreland of more than 400 km E-W and 150 km N-S
extent. There is a good correlation with the bedrock topography, which displays a
basin structure (2).
The central part of the survey is dominated by the mountain region, with values of
around 150 mGal. Apart from the bedrock topography, segmentation in N-S direction
disturb this mostly homogeneous part of the field (3).
The southern reaches show a smooth field, with levels of about -20 to 20 mGal and
corresponding to topographic features, which are characterized by individual segmentation (4).
66
SURVEYS and DATABASE
3.3. AIRBORNE GRAVITY DATA
Figure 3.8: Free-air anomaly of the VISA III campaign.
67
SURVEYS and DATABASE
3.3.4
3.3. AIRBORNE GRAVITY DATA
Free-air Anomaly VISA IV
The tie process to connect the local gravity measurements to the IGSN 71 network
was realized by colleagues of TU-Dresden, who undertook local GPS measurements
within the field campaign. As reference station in the IGSN 71 network, they used
the South African UCT NEW station, situated at Cape Town University.
station
Neumayer (construction)
Neumayer (airfield)
Kottas (campsite)
Kottas (airfield)
absolute gravity
value [mGal]
measured relative
gravity value [mGal]
982733.48
982729.61
982311.87
982310.57
6372.29
6368.43
-
The observed airborne gravity sensor drift of 2.9 mGal over 11 days is consistent with
normal operation during the campaign.
The levelling (only zero order trend applied) minimized the intersection error to
3.8 mGal, with an standard deviation of 4.3 mGal.
Compared to the bedrock topography, the map shows a much smoother field. The
influencing changes in topography, with a dominant N-S trend, are clearly recognized.
The most northerly extent is characterized by a gravitational low, while the mountain
range itself displays high positive values of around 150 mGal, and trending SW-NE
(1). Within this anomaly complex, local E-W and N-S trends are recognizable. A
smooth change in free-air anomaly to a level of 20 mGal describes the mountainous
hinterland region, with N-S trends.
Further south, values of around 0 mGal are measured, trending SW-NE. The most
southern area is dominated by a gravitational low of down to -80 mGal, striking E-W,
with a N-S striking branch in the furthest southeast (2). This pattern is disturbed
by small-scale and local positive anomalies of 40 mGal, which are entirely contained
within the survey.
68
SURVEYS and DATABASE
3.3. AIRBORNE GRAVITY DATA
Figure 3.9: Free-air anomaly of the VISA IV campaign.
69
SURVEYS and DATABASE
3.4
3.4. AIRBORNE MAGNETIC DATA
Airborne Magnetic Data
A sequence of processes, see section Aeromagnetic, is applied to produce maps of the
TMI. The quality of the resulting map depends on the quality of the dataset and
on the use of different, operator based (subjective) techniques. Manual editing will
remove disturbances or spikes of noise induced by current flow inside the aircraft.
The correction for diurnal effects is mainly based on the ground recording sample
interval and distance between the magnetic base station and the area of operation.
3.4.1
TMI VISA I
Problems with magnetic base station recordings (malfunction results in recording
stop) resulted in strong segmentation of these datasets. To close existing base station
data gaps during flight times, magnetic data from the Neumayer Station are also used
to correct for diurnal variations. This involves on the other hand long baselines of
up to 700 km, which may result in a phase shift of the base station record. Other
problems occur due to strong daily variations in the magnetic field, and especially
influence the tie lines and night time flights.
Due to limited number of tie lines in the origin flight level, it was necessary to use
profiles from draped flown previous campaigns for levelling after upward continuation. Following continuation to a common datum, at 3500 m a.s.l., the errors at
intersection points are reduced to 5.1 nT after levelling, with a standard deviation of
4.7 nT.
The map shows strong variations in magnetic intensity, suggesting a wide variety
of different magnetic properties. The survey marks a boundary between magnetic
provinces, and probably also includes several magnetic subunits.
E-W striking magnetic anomalies mark the northern extent of this survey (1). With
high amplitudes of up to 150 nT, these coast-parallel anomalies can be related to
seaward dipping basalt layers [Jokat et al., 2003, 2004]. A magnetic low, with amplitudes of -200 nT, is recognized parallel to the south of these anomalies, and can
also be related to these basalts.
Further southeast lies a prominent band of positive magnetic anomalies (2), with
wavelengths of up to 30 km, striking SW-NE. These anomalies peak with amplitudes
of 1200 nT, with an average of around 170-300 nT. This feature can be interpreted
as the border between two magnetic units.
To the north-west (3), past a subdued magnetic low of around -150 nT, the Jutulstraumen rift is characterized by positive anomalies that are more isolated, and
separated from one another at moderate wavelengths of 5-30 km with intensities of
100 nT.
The central and eastern region is dominated by singular high amplitude anomalies
of limited extent, with 5-10 km wavelength, flanked by a magnetic low of -150 nT.
The maximum amplitudes of these anomalies can reach values of 400 nT.
Eastward of the SW-NE striking anomaly complex, individual spot-like anomalies
70
SURVEYS and DATABASE
3.4. AIRBORNE MAGNETIC DATA
(4) are recognized, which are high in amplitude (200 nT) and of limited extent up
to 25 km. These are aligned along the SW-NE direction, and can be interpreted as
forming over intrusive igneous bodies.
The southern region of the survey is characterized by a more parallel orientated
ensemble of magnetic anomalies (5), striking WSW-ENE, consisting of well separated
anomalies with wavelengths of 10 km and intensities of 200 nT. Further south, the
magnetic pattern becomes less organized, with amplitudes up to 50 nT.
Figure 3.10: Total Magnetic Intensity map of VISA I campaign.
71
SURVEYS and DATABASE
3.4.2
3.4. AIRBORNE MAGNETIC DATA
TMI VISA II
The data quality can be described as good to excellent. The analyzed ground-truth
base station recordings display normal diurnal variations and these datasets, recorded
at Neumayer Station and Kohnen Station, could be used without any problems.
Base lengths, depending on base stations, are moderate, with maximum distances of
around 300 km.
Levelling was undertaken after continuing the data to a common datum, at 3900 m
a.s.l., using a statistical block-shift method, which yields a mean error at intersection
points of 7.3 nT with a standard deviation of 5.9 nT.
The resultant TMI-map shows a more or less quiet area with short wavelength anomalies, separated only by a few discrete point-like and linear anomalies in the north of
the surveyed area. These linear features occur within the continent-ocean transition
zone, strike E-W and parallel to the coast and are more then 100 km in length, with
wavelengths of 25 km and amplitudes of 300 nT (1).
A distinctive low of -250 nT, also E-W striking, lies to the south.
A positive anomaly complex, 175 km long, with a discrete boundary of 20 km in
wavelength and amplitudes of 170 nT, is situated further south. Two discrete elliptical anomalies are situated on the eastern flanks, each 40 km in length and 18 km
wide, with amplitudes of 220 nT (2).
Further south, in the central part of the survey an extensive low amplitude low with
values around -70 nT is situated (3). Short wavelength anomalies are recognized,
superimposed on this feature. Single circular spot anomalies, as well as bands of
lineaments up to 12 km in wavelength, and with intensities of 100 nT mark a change
in the magnetic pattern.
The most prominent anomaly in this area is the Penksökket Anomaly (4), which
strikes E-W and is more then 200 km in length, with maximum amplitudes of 230330 nT.
A magnetic anomaly low parallel to and south of the Penksökket Anomaly forms a
discrete boundary to two distinctive lows with -200 to -400 nT, which are followed
by two elongate anomalies, nearly 50 km in extent and up to 20 km in wavelength
with intensity of 150 nT. The anomalies show different strike directions, E-W and
SW-NE, and are surrounded by discrete and short wavelength anomalies on their
flanks (5).
72
SURVEYS and DATABASE
3.4. AIRBORNE MAGNETIC DATA
Figure 3.11: Total Magnetic Intensity map of VISA II campaign.
73
SURVEYS and DATABASE
3.4.3
3.4. AIRBORNE MAGNETIC DATA
TMI VISA III
The survey had to be divided into two operative parts, see below, each of which
required a different data processing sequence. The N-S flight lines, with a spacing
of 20 km had their operational base station at Novolasarewskaja and the second, the
W-E orientated flight pattern with 10 km spacing was flown from Sanae station.
Data handling, especially removing diurnal variations, levelling to a common datum
and the statistical levelling process, was influenced by the operator and strongly
subjective, due to:
• strongly spiked base station recordings at Novo station,
• long baseline induced phase shift of diurnal variations (Sanae station as magnetic base station),
• different flight altitudes,
• the absence of any tie lines for the E-W orientated flights.
For the N-S pattern, the magnetic base station recordings at Novo are strongly influenced by disruptive cultural noise. Despiking of the dataset shows partially useful
solutions (time windows), but unfortunately not for all flights. Transformation to
a common flight datum failed for levels between 3500 and 4600 m. This means a
common datum (upward continued) might be around 4600 m, which results in a
significant loss of information. A trial downward continuation to 3800 m displays
data which are strongly influenced by filtering. Comparison of the different solutions
(upward, downward, without) led finally to the decision to level the line-data individually, for their best solution, and not to attempt any transformation to any the
common datum for the whole campaign.
The E-W-pattern was flown at heights of between 800 to 3500 m. This survey was
levelled to a common datum at 3500 m. The use of magnetic base station data from
Sanae station, situated up to 650 km distant resulted in a phase shift of borderline
usefulness, but shows good primary results due to the absence of any other groundtruth data as well as the nonexistent tie lines for levelling.
The first attempt to level the data used the intersection points with the VISA I
dataset at the most western extent of VISA III. This yielded a bad solution, as only
the most western part (i.e. 1/5 of the complete flight line length) was levelled under
control, while the most important parts of the profiles are "free" of any levelling
control, which result in uncontrolled shifts in intensities.
Consequently a new levelling strategy was developed, to make the dataset viable.
This strategy takes the following steps:
• use of the VISA I campaign data in the furthest east and level base
• use of the GEOMAUD data (BGR Hannover) in the furthest northwest as
second data base
74
SURVEYS and DATABASE
3.4. AIRBORNE MAGNETIC DATA
• different trials of forward-backward levelling-procedures (statistical, individual,
automatic) to combine these two datasets with the VISA III data.
Finally, the displayed TMI map shows, except in the most south-eastern part (marked
with (x)), good results from this individual levelling procedure. In the marked area,
the influences of the different base levels of the flight lines are visible at around 73.5◦ S.
Direct conclusions about the internal quality with respect to the error at intersection
points are not useful, but can be given at 15.3 nT with a standard deviation of
12.1 nT.
This most eastern region of the investigated parts of DML shows, except for the
continent-ocean transition zone, low amplitudes and short wavelength anomalies with
little or no continuity.
The most northern part of the survey, over the continent-ocean transition, is characterized by strong coast parallel orientated anomalies (1), with high amplitudes
between 200 and 480 nT and wavelengths of up to 25 km.
Figure 3.12: Total Magnetic Intensity map of VISA III campaign.
75
SURVEYS and DATABASE
3.4. AIRBORNE MAGNETIC DATA
An extensive low (2), of averaging -250 nT, is displayed between the coast and the
mountain chains. Small changes in intensity occur with this low. These well located
anomalies disturb a magnetically quiet area over a region of nearly 50000 km2 .
South of the mountain chains (i.e. Wohlthat-Massive), the magnetic structure displays several anomalies and can be described as irregular, with short wavelength
variations of hundreds of nT (3). E-W and possible NW-SE trends can be interpreted but, due to the campaign uncertainties, it would be inappropriate to interpret
the southern area in more detail.
3.4.4
TMI VISA IV
The quality of the airborne based data can be defined as good. No significant additional noise was observed during the flights. Recordings of the diurnal variations
were problematic when it came to the correction of three flight lines. Additional
ground truth data from Neumayer Station, Sanae or Kohnen could not fix the problem. Consequently, two flights were unsuitable for levelling and do not contribute
to the gridded data. The levelling (only zero order trend applied) minimized the
intersection error to 7.5 nT, with an standard deviation of 6.6 nT.
Two magnetic trends are visible in this map: an E-W and a N-S segmentation.
The northern part, where the SW-NE trending mountain chain group is situated, is
dominated by a magnetic low of -200 nT, which separates into east-west and southeast trending branches (1). The branches are flanked by magnetic highs of several
hundreds of nT.
Beginning at 75◦ S the magnetic field adopts a background level of around -50 nT,
with long wavelengths. Positive, around 100-150 nT anomalies, 40 km in length and
striking SW-NE, with wavelengths of 15 km are superimposed on this field. A sharp
boundary to the west is recognized (2).
Southward, the dominant strike direction changes more to SW-NE, in an area that is
separated from the northern part by a magnetic low of -250 nT, followed by 110 km
in long and 35 km wide positive anomaly complex with amplitudes of 300 nT (3).
76
SURVEYS and DATABASE
3.4. AIRBORNE MAGNETIC DATA
Figure 3.13: Total Magnetic Intensity map of VISA IV campaign.
77
SURVEYS and DATABASE
3.5
3.5. ADVICES FOR FUTURE AIRBORNE OPERATIONS
Advices for future airborne operations
The investigations within the VISA project, which took place between 2001 and 2005
are combined here with a deeper understanding of the acquisition of potential field
data in general as well as of solving detailed problems within data handling. This is
some advice for future investigations; partially this material has been covered.
Gravity: It is essential to improve the quality of the absolute gravity estimation as
a fundamental part of gravity investigations. This requires the correct estimation
of a base value within the area of investigation as well as the verification of the
airborne gravity sensor.
Furthermore, daily repeat readings at the base station before and after surveying is
essential for the same task of quality control.
Twenty four hour heating of the gravity sensor ensures its stable operation and
reduces the burden of pre-flight procedures on the operator.
For survey planning and acquisition, it is essential not to fly on pre-defined
waypoints, because every correction of the flight trajectory during the flight causes
in additional accelerations, which work on the gravity sensor and must be balanced
within the system, leading to data gaps being introduced during the levelling
procedure.
Magnetics: A correctly operating airborne environment is not the only essential
prerequisite for the success of magnetic investigations. The measured data are
mainly influenced by the later reduction techniques, for example the correction of
daily variations of magnetic field activities. This requires the perfect completion
of the base station recordings as well as their location with respect to the airborne
component.
Generally: The levelling process is the essential tool to improve data quality. This
is realized with the use of tie-lines, which must be flown in a rectangular pattern
and with an approximate ratio of lines to tie-lines of 5:1. To ensure the highest
quality of tie-lines (important, because these define the base level of the data) it
would be advantageous to check the space weather forecast in advance. This can
be easily done via internet access or satellite telephone, which is always available
within the field. Investigations during night times are of no use due the mostly
strong magnetic background activity. A detailed quality and quantity check of the
measured data after each flight would be useful. An update of common processing
software is strongly recommended.
78
Chapter 4
COMPILATION and
INTERPRETATION
In the following chapter, the compilation of each dataset from the four parts of VISA,
measured from 2001 until 2005 is described. Apart from these, and with respect to
the huge extent of the surveyed area, different regions are displayed in more detail.
4.1
Topography
After merging the data into one common database, a systematic adjustment procedure was used to reduce the crossover errors between the different campaigns. The
individual lines were adjusted by a first order polynomial fit, which reaches in a
mean error of 2.5 m and a standard deviation of 2.4 m for the ice-topography. The
bedrock-topography displays, due to stronger gradients and uncertainties, a mean
error of 15.2 m, with an standard deviation of 11.1 m after levelling procedure.
The topographic features can be subdivided into three distinctive regions.
The most northern extent is dominated by the ice shelf region, which is more or less
flat, with average height of 40 m above sea level. The seaward flow of ice streams
dominates the topography and is also influenced by irregularities in the bedrock
topography. A dominating feature is the tongue of the Jutulstraumen, which is up
to 40 m higher then the surrounding ice of the Fimbul Ice Shelf. Other topographic
features arise from bedrock topography, like ice rises (contact of ice with bedrock
topography in the ice shelf region) and the Soerasen and Halvfarryggen.
The second remarkable area is the foreland of the mountain chains, which marks the
mechanical and natural boundary of the ice. Here, from sea level to 2000 m above
it, topographic changes with an average gradient of 2 mkm−1 are recognized. This
area too is dominated by the Jutulstraumen, which follows a geologic trough that
is of structural origin: here the ice sheet drains through the mountain range. This
system can be subdivided into two branches.
79
COMPILATION and INTERPRETATION
4.1. TOPOGRAPHY
Figure 4.1: Compilation Ice Topography.
The outcropping mountains, at the 2000 m contour, mark the boundary with the
highland plateau to the south, a more or less flat region, which has its maximum at
3400 m a.s.l..
The bedrock topography is more complex and has to be subdivided into the northern
shelf, the central mountain chain and the hinterland, which displays deep basin
structures within a highland plateau.
Contouring at sea level, which differs from the continent boundary displayed in
widely available maps, the shelf region is characterized by huge basin structures,
which are mostly orientated parallel to the coast. The average depth of these basins
is around 500 m b.s.l.. Positive structures, albeit always below sea level, can be seen
at the Soerasen and Halvfarryggen in the northwest and the ice rises Jelbartisen and
Fimbulisen in the center-north. The Jutulstraumen- and Penksökket-trough-system
seems to terminate at 0◦ . This tounge is the signature of both the speculative extent
of the Jurassic trough system and the present mechanical erosion due to the glacier
flow at 1 kma−1 .
80
COMPILATION and INTERPRETATION
4.1. TOPOGRAPHY
Figure 4.2: Compilation Bedrock Topography.
The mountain group was built up during the collision of East and West Gondwana,
and can be described as an E-W striking band, 200 km in N-S extent, with maximum heights of 2000 m a.s.l.. This topographic massif is cut by the JutulstraumenPenksökket graben system. A N-S orientated branch of this trough, more than 10 km
wide, with average heights of 400 m b.s.l., separates the EW segments of the mountain chain.
In the north-western area lie fragments of the Kalahari-Kapvaal Craton, which broke
apart during Gondwana separation in Jurassic times. During the earlier collision with
the old Antarctic craton, a crustal boundary was initiated that was much late reactivated as the Jutulstraumen-Penksökket graben system (72.5◦ S, 5◦ E). The graben
system itself consists of four separate branches:
the E-W trending Penksökket, the NW-SE branch, west of Borgmassivet, parallel
to the main branch of the Jutulstraumen trough, which reaches maximum depths of
1500 m b.s.l., and a N-S cutting branch.
81
COMPILATION and INTERPRETATION
4.2. GRAVITY
The mountain-hinterland can be subdivided into two areas.
In the south western area, 200 km south of the Kottas mountains and Heimefrontfjella
shear zone (Penksökket graben), a deep basin structure dominates the area of 270 x
300 km. The elevation decreases southward and reaches a minimum of 1000 m b.s.l..
At the edges of the surveyed area, only suggestive changes in topography can be
recognized.
The hinterland of central DML (75◦ S, 10◦ W) displays average heights of 800 km
above sea level. But internal segmentation is clearly visible on the basis of troughs,
graben systems and highland plateaus. Numerous small trough-like features of maximum 20 km in width and below sea level may be highlighted as subglacial environments and may be worthy of more detailed investigation. A smoother plateau, well
separated from the rest, can be recognized in the central southern area, measuring
300 km E-W and 150 km N-S, and may indicate an area of broad uplift.
4.2
Gravity
The free-air gravity map is characterized by long wavelength anomalies which represent deep seated crustal as well as upper mantle structures. As expected, there is
a strong correlation between free-air gravity and sub ice elevation, as seen in the ice
penetrating radar data. Due to the influence of these mass effects, the interpretative
potential of the free-air anomaly is limited. The effect of the ice-rock density contrast
can be compensated with the calculation of the Complete Bouguer anomaly, which
is displayed in subsection 4.2.2.
4.2.1
Free-air Anomaly Map
After merging the data into one common database, a systematic adjustment procedure was used to reduce the crossover errors between the different campaigns.
After this correction, the mean error can be given with 4.3 mGal, with an standard
variation of 5.2 mGal.
The northern extent can be interpreted as typical of a stable continent-ocean transition. These positive anomalies of up to 100 mGal, which are oriented parallel to the
coast, are induced by a crust thickness of 10 km including seaward dipping basalt
segments (Jokat, 2003; 2004). These long wavelength anomalies portray in detail a
more or less two dimensional continent-ocean transition zone. Only in the center,
were the Jutulstraumen trough crosses onto the shelf, does the gravity field display
any disturbance when compared to the surroundings (1).
The continent-ocean transition zone is normally characterized by gradients of
5 mGalkm−1 . From positive values the FAA decreases southward into gravity lows
of around -70 mGal. Northern central DML displays a 200 km width gravity low
(2), while the western part shows a 100 km wide low between the shelf and the
mountainous foreland (3), caused by the increasing topography.
The Jutulstraumen-Penksökket graben system, with its four branches (4), is still
82
COMPILATION and INTERPRETATION
4.2. GRAVITY
Figure 4.3: Compilation Free-air anomaly.
recognized as a set of gravity low structures with sharp flanking gradients. Here,
minimum values of -130 mGal (comparable with the topographic minima of 1500 m
b.s.l.) are measured.
The mountain chains display positive anomalies of 100 mGal and more, but show
also internal segmentation, while the hinterland is characterized by values of around
0 mGal. In the southwest, a -50 mGal gravity low, 100 km in EW and NS extent, and
with an additional N-S branch, is seen (5). The center-east shows short wavelength
anomalies (<50 km) strongly segmented, with both positive and negative values (6).
83
COMPILATION and INTERPRETATION
4.2.2
4.2. GRAVITY
Complete Bouguer Anomaly Map
The so-called Complete Bouguer anomaly reveals the internal crustal structure more
clearly then the FAA, but is strongly influenced by the different calculation techniques used to derive it, especially the terrain correction. The Complete Bouguer
Anomaly can be defined as the measured gravity field after subtraction of the gravity
effect of a simple model of the crust. This effective tool allows comparisons of gravity
anomalies worldwide.
Two main regional structures are visible:
The offshore region displays positive anomalies of around 100 mGal, striking parallel to the coastline and representing a crustal thickness of around 10 km and the
presence of seaward dipping basalt sequences (1). A southward decrease in gravity,
with a gradient of 1.5 mGal/km represents this passive and stable continent-ocean
transition, of a type known from numerous locations worldwide.
The second structure is entirely limited to the continental crust, but nonetheless
shows significant changes in gravity.
The area of the Grunehogna Unit (2) is displayed by a rough and disturbed pattern
of Bouguer Anomalies ranging from -100 mGal to 40 mGal. The geological history of
this fragment of the Kalahari-Kapvaal Craton, including the formation of a failed rift
system that appeared during Jurassic times is all recorded in this disturbed pattern.
Some additional observations:
• the Borgmassivet retwas a different CBA signal to those of the surrounding
topographic blocks,
• the J-P trough system returns positive anomalies with respect to the surrounding mountainous areas, consistent with the presence of deep seated crustal material of high density. This might be interpreted as underplated mafic rocks,
which can be in Jura related to a mantle driving mechanism for the failed rift,
• Receiver function analysis (Bayer, 2007) shows a crustal thickness of 38 km for
Sanae station, situated on a block structure (mountain), which is represented
in the CBA map by values of 40 mGal,
• the boundary to the Antarctic craton is marked by sharp gravity anomaly contrasts at the expected suture zone between the Grunehogna Unit and Maudheim Province.
The Antarctic craton itself displays various gravitational terranes:
In central DML, the Wohlthat Massif displays a remarkable gravity low (3) of up to
-160 mGal, which may be influenced by the orogenic root of the mountain chain and
which would be consistent with the crustal thickness of 48 km, from seismology. This
anomaly differs from those of other mountainous regions with similar topographic
elevations, which return gravity lows of around -80 to -100 mGal. Furthermore,
84
COMPILATION and INTERPRETATION
4.2. GRAVITY
Figure 4.4: Complete Bouguer anomaly.
this gravity low (-140 mGal) strikes SE, with small segmentations in amplitude to
-100 mGal and wavelengths of 15 km.
SW of the Wohlthat massif gravity low, a well defined pattern of stable gravity
values of around -100 mGal is recognized. Internal structures, reaching values up to
-40 mGal and with wavelengths around 20 km, with various orientations, are seen
(4).
A completely different gravity field is seen in the SW (5). The small mountain
group of the Heimefrontfjella is displayed by the expected values of -130 mGal, but its
southern extension displays a much higher level of -40 to -20 mGal. This fundamental
shift indicates a complete change in crustal behavior and may be induced by crustal
thinning or the presence of a layer of denser material. Given the calculated crustal
thickness of 53 km in the northern extent (Bayer, 2007), it is different to decide
on which explanation is more reasonable. Whatever its source, the anomaly marks
clearly a remarkable boundary.
85
COMPILATION and INTERPRETATION
4.2. GRAVITY
Figure 4.5: Results of filtering to separate gravity signals, left: 70 km high-pass, right: 70 km lowpass.
Wavelength filtering was applied in order to separate minor local inhomogeneities
from regional gravity information. The filter length applied was 90 km (effective
70 km) for both the high and low pass filters. The high-pass focused on local inhomogeneities within the upper crust. Most of the positive anomalies revealed belong to
causative sources that also give rise to magnetic anomalies. These anomalies might
therefore be reasonably interpreted as due to gabbroic intrusions. Furthermore,
granitic intrusions often poor in high susceptibility minerals. Gabbroic intrusions
(also denser, hence positive CBA at short wavelength) often give rise to strong magnetic anomalies (magnetite, ilmenite). Additionally, this map suggests the presence
of terrane structures (green level vs. dark blue) at the mountain chains and parts of
the Grunehogna Unit.
Because of their great size, these units are better recognized following low-pass filtering. A better segmentation due to the crustal behavior is seen:
the continent-ocean transition zone (high anomalies), segments of the Grunehogna
Unit, especially at the suture zone with the Maudheim Province (high anomalies),
the Maudheim Province itself (medium anomalies, green) with some fragments, and,
in the further SW a region of positive anomalies (high-orange), which may reveal a
subunit of the MP (due to some internal layering) or, more speculatively, parts of
the Coats Land Province (previous model suggestions would translate these units
further to the SW, refer to Golynsky, 2007).
86
COMPILATION and INTERPRETATION
4.2.3
4.2. GRAVITY
Isostasy Map
The isostatic models, which are based on topography that includes hard-rock material as well as load from the ice-sheet, display an idealized crustal behavior. The
calculation considers an average crustal-density of 2670 kgm−3 , density contrast at
the crustal-mantle boundary of 400 kgm−3 and an average crustal depth of 34 km.
The general trends of the isostatic anomalies are in good agreement with that of the
Bouguer Anomaly. Differences in amplitudes can be interpreted as follow:
• The continent-ocean boundary cannot be described within the model suggestions. Here the Pratt model is favored,
• The extensive lows indicate the extent of the mountain range where the Airy
model displays slightly better trend in gravity (not displayed here)
• Minor variations of Bouguer Anomaly cannot be realized within the isostasy
based on local discontinuities
• Isostasy represents an idealized model and the residuals are used for interpretation purposes
The isostatic residuals are calculated by subtracting the isostatic gravity from
Bouguer anomaly.
Figure 4.6: Isostasy. left: Regional Isostasy, right: Isostatic Residual.
87
COMPILATION and INTERPRETATION
4.2. GRAVITY
Interpretation of the residual due to lift processes of the mountains lead to the
assumption that negative amplitudes represent uplift structures. This is displayed
within the mountain range, in correlation with the main trend. A remarkable low is
situated around the Wohlthat massive (71.5◦ S, 12◦ E), including the foreland region.
Two scenarios will explain this case:
• A mass deficit: due to a sediment basin with lower density
• Existence of orogenic root; the region is not in an isostatic equilibrium and
the crust-mantle boundary layer is deeper than expected (root depth from
seismology near 50 km, Bayer, 2007)
Both ideas are discussed by Reitmayer, (2005), GeoMaud expedition, 1995/96, who
modeled the two cases (pers. comm.):
The mass deficit: Except the basin structure from RES data within the foreland
region (no infill informations!), there is no direct indications and no additional data
are available. Reitmayer modeled a basin with density contrast of 0.5 gcm3 and a
depth of 3 km.
The orogenic root: The theory of the orogenic root leads to the question as to why
the mountain range is not in equilibrium. An additional ice layer of 1.7-2.2 km would
compensate the deficit- and glacial observations indicate reduction in ice sheet during
the last thousands of years. The deficit are too high by a factor of 2-3 and not really
realistic.
Positive anomalies indicates downlift processes (only in view of tectonic processes
within the mountain range) or, if a correlation with the topography is given, the
existence of masses within the upper crust. The resulting gravity force of the masses
works in the opposite direction due to isostasy and is forced by horizontal tectonic
and up- or downlift processes.
These suggestions correlate very well with the positive isostatic residual gravity
within the Grunehogna Unit, which is characterized by rifting processes. All positive anomalies correlate with the main structures of the Jutulstraumen rift system
indicating mass deficits in topography and moreover the positive gravity anomalies
correlate with magnetic signatures as well.
Direct conclusions from isostatic residual gravity are difficult without the presence
of additional information. Furthermore, it is well known that most of the mechanism working within the crust and mantle are represented within the gravity field.
Crust to mantle processes includes oceanic subduction, continental subduction and
delamination. Processes, which works from mantle to crust include underplating,
intrusions and volcanism. All processes reported by geologists within this region.
88
COMPILATION and INTERPRETATION
4.3
4.3. MAGNETIC
Magnetic
As expected, the geologic structure of DML is rather complex and reflects several
tectonic events resulting in different magnetic units. In the magnetic pattern, the
western, central and eastern parts show similarities as well as their own characteristic
pattern. Additionally, the data always display an overprint by magnetic anomalies
that are associated with the fragments involved in Gondwana break-up.
4.3.1
Total Magnetic Intensity Map
After merging the data into one common database, a systematic adjustment procedure was used to reduce the crossover errors between the different campaigns.
The data of VISA I/VISA III campaign are used as reference and data of VISA II
campaign (by 20 nT) and VISA IV campaign (by 30 nT) were shifted. After this
correction, the mean error can be given with 8.8 nT, with an standard variation of
7.3 nT.
Seaward dipping basalt sequences like those known from volcanic passive margins
worldwide exist along the coast of DML. They are located within the continentocean transition zone and are marked by high amplitude magnetic anomalies, with
strong gradients of up to 300 nT and with wavelengths of 25 km.
The western part of DML is dominated by the Grunehogna unit (1), a cratonic fragment, with discrete, high intensity, spot-like and linear short-wavelength anomalies.
A SW-NE, and coast parallel striking, linear anomaly of more than 150 km length
displays amplitudes of 170 nT, with a wavelength of 15 km. Around this are scattered
three dimensional anomalies with amplitudes of 220 nT and wavelengths of 15 km.
Magnetic subunits situated in the western part of the GU are partially coexistent
with topographic features, like the Borgmassivet, with wavelengths of up to 20 km.
In the central parts, prominent NE-SW striking anomalies in the H.U. Sverdrupfjella (2) with maximum amplitudes (in the south) of 1200 nT, and wavelengths
of 30 km can be seen along with E-W trending anomalies of 300 nT in amplitude
and 30 km width over the Penksökket. The prominent Jutulstraumen-Penksökket
anomaly marks significant changes in magnetic strike direction (3).
A WSW-ENE trending anomaly complex dominates the central part of the map (4).
The most eastern part of the complex consists of two individual anomalies, with
wavelengths of 20 km and amplitudes of 200 nT. Very short wavelength anomalies
are situated on the flanks. Further NE, the complex consists of a set of four parallel
anomalies with amplitudes of up to 300 nT and wavelengths of 20 km. The overall
length of this complex is nearly 350 km.
In the southeastern region, the magnetic anomaly pattern becomes less well organized
(5). Small circular and elliptical anomalies dominate the area, set on a background
level of 50 nT, having maximum amplitudes of 200 nT and wavelengths of between 5
and 15 km.
89
COMPILATION and INTERPRETATION
4.3. MAGNETIC
Figure 4.7: Compilation Total Magnetic Intensity.
The most easterly part of DML shows low amplitude, short wavelength anomalies
with no or little continuity. A prominent structure is the magnetic low zone (6),
where the field is only disturbed by a small number of short wavelength anomalies.
The south-western region is dominated by two different magnetic anomaly complexes.
The northern complex shows similarities to the central DML complex (4), like the
parallel orientation, but at lower amplitudes of around 100 nT. The extent of this
complex is also limited to an area of 50 km from E to W, as it dies out into a lower
magnetic intensity background to the east (7).
Further south a 150 km long, well defined anomaly complex (8), strikes N-S. The
anomaly shows two distinctive maxima with amplitudes from 200 to 400 nT, and has
a wavelength of 40 km. On its eastern flank a 40 km wide magnetic low is situated,
beyond which short wavelength anomalies with irregular structure occur.
90
COMPILATION and INTERPRETATION
4.3. MAGNETIC
Figure 4.8: Main magnetic units (upward continued) with results of detailed mapping trend analysis
overlaid.
To a first approximation, upward continuation of the TMI field will highlight the main
magnetic structures and units, whereas highs sensitive trend mapping analysis can be
undertake to display the complex pattern of the original. The overprinting character
of magnetic anomalies varies somewhat throughout the study area, when interpreted
from the different magnetic-tectonic trends. Various trend-regimes, ENE-WSW and
E-W, as well as completely irregular pattern can be recognized, as displayed in figure
4.8, and must be interpreted in terms of different geotectonic histories of the various
regions covered by the map.
91
COMPILATION and INTERPRETATION
4.3.2
4.3. MAGNETIC
Analytic Signal Map
The calculation of the Analytic Signal (AS) helps to reorientate magnetic anomalies
directly over their sources, and gives more detail compared to the TMI, which is
dominated by the superposition of neighboring anomalies. In comparison to the TMI
map, a stronger segmentation of magnetic sources can be observed, combined with
"zooning" (colourscale). The main structures, or deep-seated anomaly sources are
highlighted in the displayed AS map, shown in figure 4.9. The following description
is made with reference to the TMI map and so will highlight the observed differences:
The AS map is mainly characterized by more or less three dimensional anomalies:
the loss of complex linear structures is clear. Neverless, linear alignments of the
localized features are interpretable.
• The continent-ocean transition zone is now reduced to two main anomaly complexes, situated in the NE and the central north (1). Between these, a band of
circular anomalies is recognized. In the area of the Jutulstraumen ice tongue,
a large circular anomaly (2) is now more strongly delineated than it was in the
TMI map.
• The area of the Grunehogna unit displays 8 major anomalies (3), which are
irregularly orientated, while the rest of the unit shows rather incoherent magnetic signature.
• The H.U. Sverdrupfjella anomaly complex displays a new segmentation (4) and
may be reinterpreted (see section Areas In Detail). The Penksökket anomaly
can now be divided into a set of four single anomalies (5).
• The WSW-ENE striking central anomaly complex (6) consists now of a set of
well-separated elliptical anomalies arranged like in a chain. Similar chains but
much lower in amplitude, can be recognized parallel to this to the north and
south.
• In the most eastern part of DML (7), three complexes of anomalies are highlighted: two of them, in the north and center, with a preferred EW orientation,
and the third in the south centered region with a more NW-SE orientation.
• The northern part of the south-west displays a set of minor isolated anomalies,
while the southern part displays a more massive anomaly complex striking NS
(8).
92
COMPILATION and INTERPRETATION
4.3. MAGNETIC
Figure 4.9: Analytic Signal analysis.
4.3.3
Tilt Derivative Map
Applying the tilt derivative highlights the signature of shallow magnetic sources and
so gives some guidance on interpreting known lineaments and shear zones. In the
displayed map, a frequency cut off is applied to focus on anomalies with positive
contrasts in susceptibility. Red color belong to these positive anomalies, situated
directly over their sources.
The main focus of this analysis is to find boundary structures between magnetic
units, and so concentrate on the tectonic evolution. In general, the continent-ocean
transition zone displays parallel striking complexes (1) but is disturbed in the eastern
part, where it gives way to the GU. The GU can be identified by its circle-like
structure (2), displayed in the map. Internal segmentation of the unit is not strong.
The lineaments and shear zones (3) suggested by Golynsky, (2000), and the detailed
interpretation of the small but highly detailed survey by Ferracioli, (2005), can be
93
COMPILATION and INTERPRETATION
4.3. MAGNETIC
confirmed and continued. These features mark the boundary between the GU and
MP situated in the H.U. Sverdrupfjella and Penksökket.
The central anomaly complex (4) indicates the presence of WSW-ENE lineaments,
and confirms geologic observations (Jacobs) further north in the Wohlthat Massif
where they may change to be more EW orientated.
The south-western region becomes less clear (x), because the TDR process highlights
shifts between the flight lines at the expense of lithological signals. But, nevertheless,
the northern part displays a parallel EW orientation, which is divided to the south
by NS striking features.
Figure 4.10: Tilt Derivative filter, to highlight only sources with positive susceptibility, low values
are cut off.
94
COMPILATION and INTERPRETATION
4.4
4.4. AREAS IN DETAIL
Areas In Detail
The investigated area is to vast for a detailed description of it all to be made. Here,
interesting patterns are extracted and investigated in detail. The huge, but geologically self-consistent Grunehogna Craton is focused on. Furthermore, different parts
of the Maudheim Province are examined in detail.
4.4.1
Grunehogna Unit
The Grunehogna Unit corresponds to the Grunehogna (Province) Craton, a fragment
of the Kalahari-Kapvaal Craton, consisting of Archaen granite basement overlain
by undeformed Proterozoic sediments. Magnetically, the GU is characterized by
broad featureless lows, with a few short wavelength anomalies. The widespread low
amplitudes are situated in the center, while short wavelength spot-like highs and
magnetic subunits are situated more towards the borders. The boundary to the
MP can be followed along the E-W striking Penksökket Anomaly and the SW-NE
striking Sverdrupfjella Anomaly complex.
GU-coast and shelf region: the offshore area is mainly influenced by linear, coastparallel striking anomalies. These features, compared to purely continental anomalies, are huge in extent (more than 100 km in length), with average wavelengths of
up to 50 km and amplitudes of 300 nT. Magnetic anomaly profile analysis suggest a
slightly dip toward the south. These anomalies coincide with seaward dipping reflector sequences in seismic reflection data and mark the boundary between oceanic
and continental crust. They connect the Princess Martha Coast Magnetic Anomaly
(PMCMA) to the Explora Anomaly, both wide bands of coastal magnetic anomalies
to the east and west. The AS-analysis supports the mostly linear nature of these
anomalies, but also highlights amplitude maxima that can be interpreted as formed
over large basaltic intrusion complexes. The TDR highlights shallow structures,
which are connected to the wide, coast-parallel striking subcrop of basalts. Most
of the anomalies are situated offshore, and there is no correlation with topography.
The southern extension, toward the continent, is characterized by a magnetic low
with amplitudes of around -300 nT, 400 km in E-W extent, with a width of 45 km.
GU-A unit: here, an E-W striking anomaly, nearly 250 km long, with amplitudes of
200 nT and a wavelength of around 25 km is prominent. While the TMI displays a
more or less linear behavior, the AS separates this anomaly complex into a 100 km
linear, E-W striking feature, with a depth-solution of between 7 and 2 km. The shape
of magnetic anomaly suggests it may best be modeled as due to a thin dyke. Dipping
slightly towards the south the anomaly betrays a more or less vertical source. Eastwards, two high amplitude and more N-S oriented elliptical anomalies are observed.
Maximum amplitudes reach 500 nT for the southern and 900 nT for the northern
anomaly, with estimated source depths between 2 and 1 km. The TDR suggests
that more shallow sourced anomalies build the connection between these features as
well as the presence of a EW oriented shear zone to the south.
95
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
Figure 4.11: TMI, Grunehogna Province with its magnetic subunits, and the adjacent area of
Maudheim Province, situated to the south and east.
Additionally, the abrupt changes in strike direction can be taken as indications for
some lineaments. This corresponds well with the topography: the anomaly complex
forms over a kind of channel, with topographic heights around 500 m b.s.l. that
strikes dominantly parallel to the coast.
GU-B unit: this unit strikes NW-SE, for 75 km, with a maximum wavelength of
30 km and peak amplitudes of 100 nT. The response from AS displays several maxima, with diverse structures, while the TDR shows a network of EW and NS striking
shallow sourced anomalies. Estimated depths for causative bodies lie within a range
of 4 km at the flanks and 1 km over the center. The topographic background of this
offshore region displays EW and NS trends at about 400 m b.s.l..
The southern continuation is the 60 km long, 40 km wide, GU-C unit. Moderate amplitudes of 60 nT, with four maxima, are seen in the TMI. A large block as magnetic
source can be inferred from the shape of the magnetic anomaly. No special characteristics can be observed in the AS, but the TDR displays strong segmentation in a
preferred NW-SE orientation.
96
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
Figure 4.12: Analytic Signal, Grunehogna Province, AS overlain by Located Euler Deconvolution
solution.
Indications for shear zones are given on each flank of this subunit. The eastern
boundary in particular indicates a NW-SE striking shear zone, while the western
flank is characterized by a chain of 8 km wide localized highs orientated NW-SE.
Topographically, the area is characterized by moderate gradients and the presence
of the Ahlmannryggen nunatak group.
Southward, the GU-D unit, can be recognized, as a set of four distinctive highs, in
total nearly 100 km in length and with a maximum width of 40 km, and maximum
amplitudes of 75 nT. Strike changes between E-W and N-S and corresponds partially
with the topographic terrain of the Borgmassivet complex. Significantly, most of the
maxima correspond to strong topographic gradients, like the flanks of the trough
system. Overall, the anomaly shape suggests another large block as the causative
body. The AS displays no significant internal structures, while the TDR is characterized by slight NE-SW and EW trends. The Located Euler Deconvolution gives
limited solutions for approximate depth, at 4 and 1 km.
97
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
Figure 4.13: Tilt Derivative, Grunehogna Province, TDR to enhance shallow magnetic sources and
interpreted lineaments.
The eastern and southern boundaries of this complex are connected with a lineament
(shear zone). The boundaries are shallow sourced, linear SW-NE striking magnetic
anomalies that display strong correlation with the strikes of bedrock topographic
features.
Focusing on the strike of the boundary to the northeast, an abrupt change into western and eastern lineaments can be recognized. Additional subunits can be defined in
each of these arms.
GU-E unit: high in amplitude at 600 nT, with a wavelength of 12 km and striking
NS for 45 km, this anomaly coincides with the Straumsnutane nunatak group. A
more or less vertical dyke source can be interpreted from the shape function of the
magnetic anomaly. The AS marks a massive source with an approximate depth of
around 0.7 km, while the TDR displays two separate bodies. Geographically situated
on the eastern flank of the Jutulstraumen trough, this subunit covers a topographic
horst structure.
98
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
Figure 4.14: Bedrock Topography, Grunehogna Province, including some geographical names.
Two subunits bound in a kind of wedge, are seen further north: GU-F, NE of GU-E,
consists of three, dominantly E-W striking anomalies with 10 to 20 km wavelength
and moderate positive amplitudes of 130 nT. Only shallow responses from the TDR
are recognized within the preferred E-W orientation. The eastern extent is characterized by sharp gradients, suggesting tectonic lineaments or shear zones.
Eastward, GU-G consists of a set of large circular anomalies arranged along a NWSE line, 75 km long and up to 50 km wide. The TMI shows two maxima with amplitudes of 700 nT and 200 nT , whereas the AS and TDR show this complex as
consisting of three parts. The most southern feature dominates in intensity and
size. The source geometry can not be differentiated from the shape of the magnetic anomaly, but approximate depth of the causative bodies lies within the 4 - 3 km
depth range. Relationships to bedrock topography are not well defined, but the main
anomaly of this subunit is situated on the N-W flank of the Jutulstraumen trough
system, offshore, where the topography displays rugged relief.
99
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
GU-others unit: There are numerous other less-well defined subunits in the GU’s
magnetic expression. In the center, where the TMI displays an extensive low, with
a background level of -50 nT, there are several indications for localized magnetic
sources. Simple zooming of color scale will highlight these features, but the AS and
TDR analysis display several local anomalies with average wavelengths of 10 km.
These features are dominantly symmetrical and round, but some minor linear trends
can be recognized. These well separated features are all situated over strong gradients
in bedrock topography. The anomalies are more or less irregularly oriented except
in as much as the linear shallow sources are strongly correlated with subsurface
topographic lineaments and trends.
Table 4.1: Subunits within the Grunehogna Craton.
unit
trends
GU-coast
linear,
coast-parallel,
≈100 km
GU-A
E-W,
linear,
250 km,
local flanking highs,
strike change to N-S
amplitudes /
wavelengths
300 nT,
50 km wide,
approx.
depth
9 to4 km
topography
gravity
normal shelf,
≈500 m b.s.l.
strong gradients
200 nT,
25 km wide
variable,
7 to 2 km
channel structure,
500 m b.s.l
linear,
strong gradients
900/500 nT,
15 km wide
2 to 1 km
GU-B
linear,NW-SE,
75 km,
strongly segmented
100 nT,
30 km wide
4 to 1 km
no significant,
400 m b.s.l.
disturbed
GU-C
NW-SE, irregular,
60 km long,
weak, 3 maxima,
strong flanking gradients,
complex margins
100 nT,
40 km wide
no direct
solution
variable,
Ahlmannryggen,
up to 1k m a.s.l.
disturbed
GU-D
tabular,
E-W and N-S strike,
100 km long,
three maxima,
strong flanking gradients
100 nT,
40 km wide
4 to 1 km
strong gradients,
Borgmassivet,
up to 2k m a.s.l.
disturbed
GU-E
elliptical/tabular,
45 km NS trend
600 nT,
12 km wide
0.7 km
horst-flank
structure
disturbed
GU-F
E-W trend,
three sources,
flank,
strong flanking gradients
130 nT
10-20 km
1 km
rugged,
300 m b.s.l.
strong gradients
GU-G
NW-SE,
75 km,
three maxima
200-700 nT,
50 km
4 to 3 km
flank of rift,
rugged relief
strong gradients
GU-other
wide, broad, few,
localized small anomalies
<100 nT,
10 km
variable
weak,
smoothed
100
COMPILATION and INTERPRETATION
4.4.2
4.4. AREAS IN DETAIL
Maudheim Province
The suture zone between the Maudheim Province (MP) and Grunehogna Unit (GU)
is characterized by strong magnetic anomalies. The boundary is marked with the
Sverdrupfjella-Kirvanveggen Anomaly (SKA) and the Penksökket Anomaly, see figure 4.11.
The linear, E-W-trending Penksökket Anomaly, 220 km long and up to 50 km in
wide, is the most prominent feature. While the TMI, with amplitudes from 200 to
450 nT, suggests a linear character, the AS displays segmentation into four different
bodies with a SW-NE orientation. The dip direction, referred to a thin dyke model,
varies over the whole extent from vertical to a slightly southward direction. The
depth solution varies from 7 to 4 km. The sourced anomalies show the segmentation
enhanced and highlights indications that might be interpreted as due to thrust and
shear zones seen in outcrop. Geographically, the anomaly appears directly over the
Penksökket trough, with average elevations of -300 m b.s.l. [MP-Penk].
To the east, at Kirvanveggen, the E-W trend bends towards the NE becomes much
broader, and remains so until the region around Neumaerskarvet. There, a more
S-N trend is observed, similar to thrust and shear zones seen in outcrop, as well as
a set of shallow and irregular anomalies. Amplitudes of 40-80 nT and wavelengths
of 5 to 10 km are observed. Topographically, the area is mountainous with the same
S-N orientation.
The anomaly associated with the H.U. Sverdrupfjella dominates the eastern part of
the map, shown in figure 4.11 and 4.15. This anomaly strikes SW-NE for 300 km
and is 40 km wide with high amplitudes of up to 1200 nT. The anomaly consists of
three distinctive segments with internal structure [MP-H.U.S.].
East of Neumearskarved, the maximum amplitude of 1200 nT is observed where the
anomaly wavelength is 40 km. The AS highlights a wide tabular anomaly, which is
situated on the eastern flank of a S-N-trending trough in a kind of bay. The dip
direction of an intrusion interpreted from this anomaly is not clear from the shape of
the anomaly, and tends from vertical to all possible dip directions. Estimated depths
of causative sources vary between 5 and 0.5 km. TDR-highlighted shallow anomalies
show a continuation towards the north and across the basement trough into the
Sverdrupfjella mountain group.
At 72.5◦ S, the H.U.S. is offset by 20 km toward the east. The southern segment
130 km long with a wavelength of 35 to 40 km, and typical intensities of 250 nT are
observed. Deep sources, 7 km, characterize the central part of this segment, while
the flanks exhibit shallower-sourced anomalies, of 1 km. Maximum amplitudes are
found over SW-NE trending topographic lineaments on the Sverdrupfjella mountain
chain.
One main intrusion at Straumsvola and Tvora nunatak group, further NW, displays a
NW-SE orientation, like some of the subunits of the GU. With a wavelength of 20 km,
a length of 30 km and observed intensity of 270 nT, these anomalies are aligned along
the eastern flank of the Jutulstraumen trough, where they are situated on a horst
structure. A purely vertical source geometry is interpreted, and a depth solution gives
101
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
a value of 4 km. In additional, several localized anomalies (<15 km wavelength) can
be seen in the surrounding area.
The third and northernmost segment of the H.U.S. anomaly occurs after a further
eastwards offset of 20 km (71.5◦ S), and terminates at a coastline-parallel, directions
and intensities are observed.
4.4.2.1
Central DML
Various magnetic patterns are recognized in central DML, as shown in figure 4.15.
Figure 4.15: Maudheim Province, central DML, top: Total Magnetic Intensity, bottom: Bedrock
topography.
102
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
We focus on the SW-NE trending anomalies. These mostly positive magnetic anomalies, displaying complete different pattern and their origins are unknown due to the
lack of geological control.
MP-Central-1, situated in the most southwestern region, consists, with reference
to the TMI, of two large anomalies. The western part strikes E-W for 75 km, with a
wavelength of 35 km and amplitudes of 200 nT. The eastern anomaly strikes NE for
75 km with a width of 35 km and maximum amplitudes of 160 nT. The deep source
characteristics and individual shapes of the anomalies, however, are dissimilar.
Figure 4.16: Maudheim Province, central DML, top: Analytic Signal, overlain by Located Euler
depth solution, bottom: Tilt Derivative, overlain by interpreted lineaments.
103
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
The western unit consists of numerous well-separated features, with variable orientations and depth solutions of around 1 to 2 km. The source geometry is more or less
vertical. The northern part consists of a chain of linked anomaly highs, trending NE
and suggestive of shearing and related vertical thin dyke structures.
This part of MP-Central-1 is formed over a branch of the Kirvanveggen in the south
and a chain of mountainous lineaments in the north.
The eastern part of MP-Central-1 consists of a main anomaly body situated in a NESW orientated trough. A approximate source depth of 4 km was calculated and may
correspond to a large tectonic block. Shallow sources, characterized in the TDR, are
related to changes in topography. On its western flank, two small circular anomalies
are recognized with wavelengths of 15 km, intensity of 200 nT, and source depth of
0 km.
MP-Central-2 can be described as a system of linear parallel SW-NE trending
anomalies. This unit is nearly 400 km long and has an average width of 65 km. Maximum amplitudes of 460 nT are measured for the individual striking anomalies, but
averages are around 200 nT. Internally the anomalies displays N-S and E-W segmentation. The dominant source geometry is vertical. The AS highlights the northern
substructures, with depths solutions ranging from 4 to 0.5 km. Shallow sources are
mostly orientated parallel to the NE trend, consistent with a wide band of thrust and
shear zones, marked in the figure 4.16. This interpretation is based on extrapolation
of observations in the Wohlthat Massif.
The northern extent of MP-Central-2 displays several circular anomalies, with average wavelength of 30 km and intensities of up to 400 nT. The shapes of these magnetic
anomalies suggest a vertical geometry of the causative bodies.
MP-Central-3 is characterized by a broad and weak anomaly pattern, with variable orientations. The background level is positive, with maximum amplitudes of
50 nT, and depth solutions around 1 km. Three main anomaly complexes disturb
the background in the east, with wavelengths of 18 km and intensities of 150 nT.
Estimated depths of these causative bodies are 6 km, 1.5 km and 0.7 km. Sub-ice topography provides some indications for an uplift structure related to MP-Central-3,
both limited to the area of investigation.
104
COMPILATION and INTERPRETATION
4.4.2.2
4.4. AREAS IN DETAIL
Eastern DML
Figure 4.17 displays the most north-eastern part of the investigated area. Three
significant changes in magnetic pattern are recognized and can be confidently related
to the geology.
Figure 4.17: Maudheim Province, eastern DML, top: Total Magnetic Intensity, bottom: Bedrock
topography.
105
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
The shelf unit in the north, MP-East-1, is dominated by extensive anomalies similar
to those known from the other coastal regions of DML. Large in extent, at 70-130 km
length and with wavelengths up to 35 km, the unit strikes parallel to the coast-line.
Amplitudes of a few hundreds of nT, and a maximum of 500 nT, are observed. The
focusing of magnetic anomalies over their sources, as shown in the AS, displays a
more complex structure in orientation and extent. Segmentation from 10 to 40 km
wavelength, for deep seated causative bodies, with 8 to 4 km depth, is established.
Shallow anomaly complexes dominate parallel to the coast. The complete offshore
region is dominated by more or less flat topography, which does not correlate with
the magnetic anomaly pattern.
The MP-EAST-2-unit, is characterized by short wavelength low-amplitude magnetic anomalies superimposed on a negative background of -250 nT. Linear chains
trend mostly E-W. Circular anomalies with average wavelengths of 12 km are recognized with amplitudes of -40 nT. The AS-filter highlights some additional features that are not evident in the TMI-map. Estimated depths for these anomalies’
causative bodies range from 9 to 4 to 1 km, and they are mostly situated over topographic irregularities. Shallow and more linear sources are observed over the whole
area, whose background field is negative. The dominant orientation in the central
part is E-W. Orientations becomes less obvious in the western part, towards the
H.U. Sverdrupfjella, where a more NE-SW trend is recognized. The eastern extent
of this unit lies beyond the limits of the survey. An eastern complex consists of five
isolated anomalies with wavelengths of 11 km and intensities of 200 nT, but does not
define the boundary of this subunit. The estimated depth of these anomalies lies
in a range between 2 and 0.5 km. Topographically, the area is underlain by a huge
basin structure. Minor structures, especially south and south-east of the Russian
base Novolasarewskaja and around the Schirmacher Oasis disturb these deep seated
structures with average topographic heights of 500 m b.s.l for the basin structure and
160 m for the Schirmacher Oasis and outcrops.
A change in the magnetic pattern is recognized within the mountain chains, where
the MP-EAST-3-unit is situated. MP-East-3 continues beyond the southern edge
of this map.
A negative background of -80 nT characterizes the study area and is disturbed by
smaller local anomalies without any obvious geometric or geological signature. The
AS displays a number of prominent circular anomalies that partially correlate with
outcrops of A-type granitoides within and around the Wohlthat-Massif. Intensities
of 100 to 200 nT are recognized, with average wavelengths of 10 km. Various depth
solutions are calculated, varying from 4 to 1 km. The TDR signal reflects better correlation between geologically-defined granite localities and shallow magnetic sources.
Geographically, these intrusions are based within areas of strong topographic gradients.
Due to the incomplete success of the levelling process, any detailed interpretations
that aim to map geological boundaries would be inappropriate for the more southern
parts of this area. With respect to these, the AWI will fly more tie-lines in January
2008, which should lead to a better levelling solution for this survey and so help in
further interpretation.
106
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
Figure 4.18: Maudheim Province, eastern DML, top: Analytic Signal, overlain by Located Euler
depth solution, bottom: Tilt Derivative, overlain by interpreted lineaments.
107
COMPILATION and INTERPRETATION
4.4. AREAS IN DETAIL
Table 4.2: Subunits within the Maudheim Province.
unit
trends
MP-Penk
E-W, linear,
≈220 km,
segmented,
four units
amplitudes /
wavelengths
200-400 nT,
50 km wide,
approx.
depth
7 to 4 km
topography
gravity
trough,
300 m b.s.l.
disturbed
MP-HUS
SW-NE, linear,
300 km,
segmented,
three unit trends
ave.200 nT,
1200 nT max.,
40 km wide
variable,
7 to 0.5 km
trough system,
mountainous
disturbed
MP-Central-1
linear, 75 km,
EW and NE trend,
2 segments,
flanking circular intrusion,
strong flanking gradients
200 nT,
35 km wide
2 to 1 km
4 km,
strong gradients
on mountainous
flanks
disturbed
MP-Central-2
SW-NE trend,
400 km, parallel,
strong flanking gradients
200 nT,
460 nT max.
65 km wide
4 to 0.5 km
1k m a.s.l.,
mountainous
segmented,
broad
MP-Central-3
weak, broad,
various,
shallow sourced,
circular anomalies
50 nT,
18 km wide,
1 km
800 m a.s.l.,
SW-NE segmentation,
uplift structure
broad
150 nT
6 to 1 km
MP-East-1
70-130 km,
various,
coast parallel
200-500 nT,
10-40 km wide
8 to 4 km
offshore,
≈500 m b.s.l.
strong,
smoothed
MP-East-2
mostly E-W,
magnetic low zone,
several circular anomalies
-40 nT,
200 nT,
11 km wide
9 ,4 ,1 km
basin, 500 m b.s.l.,
topogr. irregularities,
160 m a.s.l.
broad, weak,
isostaticinterest
MP-East-3
undifferentiated,
several circular anomalies
200 nT,
10 km wide
4 to 1 km
strong gradients,
mountainous
broad
MP-others
wide, broad,
single circular anomalies
200 nT,
10-15 km wide
108
2 to 0.5 km
related
COMPILATION and INTERPRETATION
4.5
4.5. GEOLOGIC MODEL SUGGESTIONS
Geologic model suggestions
The suggestions for a combined geological model are based on a detailed mapping
strategy, as shown in the previous chapters, in combination with the Curvature
analysis, displayed in the Appendix D. The displayed models are consistent with
each other, including potential field data and the bedrock topography..
The Grunehogna Unit corresponds to the Grunehogna (Province) Craton, a fragment
of the Kalahari-Kapvaal-Craton and is characterized:
• topographically: by a well-separated units in combination with a complex
graben structure. Strike and direction of the mountainous region emphasizes
the collision with the East Antarctic Craton.
• gravitational: the lithospheric response is characterized by strong gradients
within the transpression zone, the central part displays a stable pattern.
• magnetically: widespread low amplitudes and spot-like highs in combination
with several subunits towards the border.
The southern extension of the Grunehogna Craton is the Maudheim Province, which
has Grenvillian age. The suture zone is characterized by strong anomalies. The
mobile belt system displays:
• topographically: three well separated main structures: mountain, highland
plateau and basin
• gravitational: various units which corresponds with the magnetic subunits
• magnetically: magnetic units are very variable and display various trends,
overprinting character is always visible
The suggested separation of the mobile belt, in Maudheim Province and the Central
Dronning Maud Land, Golynsky (2007), cannot be verified. An internal segmentation
due to terranes is clearly seen and corresponds with all three datasets, but the exact
definition in terms of geological provinces is limited with reference to the surveyed
area.
109
COMPILATION and INTERPRETATION
4.5. GEOLOGIC MODEL SUGGESTIONS
Figure 4.19: Geologic model suggestions, top: topographic units, middle: gravity units, bottom:
magnetic units.
110
Chapter 5
SUMMARY
Within the VISA project the continental lithosphere of DML was systematically
mapped and geophysically investigated in order to achieve a better understanding of
the geology of Gondwanaland. This investigation covers a very large region of more
than 1.2 Million km2 , which is about four times the size of Germany at 350000 km2 .
Within this long-term collaborative project between the TU-Dresden and AWI, which
includes a variety of multidisciplinary targets, this thesis concentrates on airborne
potential field data and the interpretation of the Antarctic lithosphere. Since only
the highest peaks of the DML mountain chains can be geologically sampled, indirect
geophysical methods are required. Given this and the limited time window for scientific campaigns during the austral summer, fast and highly detailed investigation
techniques are required.
Aero-magnetic and gravity surveys have been flown during the past 20 years. The
development of airborne techniques have contributed toward a better understanding
of the variety of regional tectonic provinces of the Antarctic continent. The extent
of the survey area, from 14◦ W to 20◦ E and from 70◦ S to 78.5◦ S, is large enough to
fully recognize long-wavelength anomalies. However, smaller features could also be
mapped, owing to an average line spacing of about 10 km. With this spacing, the
line spacing permits a good regional interpretation.
Old lithospheric boundaries between the Archean Craton, the Grunehogna Province,
and the Proterozoic to Early Paleozoic mobile belt, the Maudheim Province, could
be interpreted. Thrust faults have also been mapped, and their formation must be
related to tectonic events that occurred in both Grenvillian and Pan-African times.
The Archaen to Mid-Proterozoic Grunehogna Province consists of Archean granites, which crop out in the Annandandagstoppane region and is overlain by the
Mid-Proterozoic Ritscherflya supergroup of relatively undeformed sedimentary and
volcanogenic rocks of the Ahlmannryggen and Jutulstraumen groups. Intrusions, theoleiitic sills and dykes, are reported within the Ritscherflya supergroup. The Straumsnutane Formation, andesitic lava flows with pyroclastics and sediments, forms a further element of the Ritscherflya supergroup (e.g. Groenewald et al., 1991; Martin,
1986).
111
SUMMARY
The Grunehogna unit is associated with a broad and relatively featureless magnetic
low, with a few circular short-wavelength anomalies, whose low amplitudes and short
wavelengths suggesting a shallow magnetic basement. The central part is transected
by the PMCMA, and may delineate a structure of crustal weakness. This anomaly
pattern can be traced further west into the Weddell Sea shelf. To the east, two
circular magnetic anomalies, which may locate mafic intrusions, define the eastern
limit. Other magnetic units are recognized on the eastern extent, next to the Jutulstraumen. These units are mostly low in amplitude and display various trend
characteristics, most of which can be related topographically-defined horst structures. The gravity field displays an undisturbed pattern in the central part
Of special note is the Jutulstraumen area. This region may represent a branch of
a Jurassic-aged rift system. This system may be related either to active rifting
processes involving the mantle plume that sourced the Karoo-Ferrar large igneous
province in Africa and Antarctica (Cox, 1992), or may represent a passive rift structure.
A NE-SW oriented thrust fault crosses this region, and the topography may point
to complex deformation. This area is occupied by a major structural boundary between the Grunehogna- and Maudheim Provinces and may have been part of a major
transpressional fault system which was active during Pan-African deformation. The
non-magnetic zone over the proposed Jutulstraumen rift is difficult to explain, but
may reflect the presence of post-Jurassic sedimentary rock or sedimentary infill, or,
alternatively an amagnetic rifting process. Prominent trends could be mapped. The
NE-SW trend has already been recognized (Ferraccioli, 2005) and a highly perpendicular NW-SE trend can also be defined. Notably, these trends are found over the
entire transpression zone, i.e. in the H.U. Sverdrupfjella and in the adjacent subunits
of the GU. This leads to the suggestion that the rift was a region of active extension
at different times and in directions.
The boundary to the Maudheim Province is defined by continuous linear magnetic
highs: the Penksökket Anomaly in the south and the H.U. Sverdrupfjella Anomaly
complex in the east.
The Maudheim Province, and its Sverdrupfjella supergroup displays two lithostratigraphic ensembles: amphibolite facies calc-alkaline metavolcanic rocks in the west
and granulite facies para- and ortho-gneisses in the east, which are intruded by
Pan African-age granites. Outcrops at Kirvanveggen, Heimefrontfjella and Vestfjella
display continental flood-basalts that can be interpreted as remnants of the KarooFerrar large igneous province. At the H.U. Sverdrupfjella, Jurassic alkaline intrusions
are observed (Straumsvora and Tvora) and may relate to crustal extension and the
presence of a rift system .
The central area of DML hosts Grenvillian rocks that were entirely transformed during Pan-African reactivation. These events altered the metamorphic assemblages
as well as reactivating older tectonic structures. The reactivation and intrusion of
magma into the Grenvillian crust at 600 Ma was followed by tectonism, metamorphism and a late stage of magmatism at 500 Ma (e.g. Shiraishi, 1994, Jacobs, 1998).
112
SUMMARY
Different magnetic trends could be mapped, and these vary from NE-SW to E-W.
Irregular patterns are also present. Low amplitudes with weak linear trends, are
recognized. Only a few circular highs interfere with this magnetic pattern. On the
basis of changing gradients in the TMI, the MP can be subdivided into several
units: the continent-ocean transition zone, marked by a strong anomaly pattern,
followed southward by a distinctive magnetic low. The Central Dronning Maud
Land is characterized by a WSW-ENE linear anomaly complex and further south,
by a weak zone in terms of amplitudes. The southwestern region can be divided
into three distinctive pattern. Nearly all units of the entire region display numerous,
mostly circular, magnetic highs, which are well separated.
The overprinting character of magnetic anomalies is very variable within the area and
shows various trends separating subunits as well as many indications of shearing and
thrust faults. The sources of many anomalies are obvious, where they correspond
to known outcrop geology. Others, and these are the majority, relate to structures
concealed under ice-cover, and their origin is more speculative.
Often, mismatches are recognized between magnetic observations in the TMI field,
and well recognized bodies in outcrop geology. Problems like this can be addressed
in different ways. Observation techniques, line spacing and sample interval which
define the lateral resolution, were chosen to detect structure in the order to regional
studies. On the other hand, the geological observations will always take precedence
over the remote sensing techniques- and, of course, the Earth is not homogeneous.
Normally, granitoids are associated with high susceptibility values, and so they ought
to be detectable with magnetic methods. In field studies, however, such bodies often
occur with an anorthositic component which has a low susceptibility, and granitoids
are confined to its flanks (Piech et al., 2005). This arrangement may give rise to
variations in size and susceptibility that, when combined with the resolution limits
of a given survey, lead to only the larger magnetized bodies being recognized. The
applied TDR technique correlates well with known geological sample areas.
Attempts to match cratonic fragments on the basis of potential field methods alone
are always difficult in particular with low resolution data with line spacings of 10 km.
On the other hand, airborne-based investigations are a proper tool to provide first
insights of large-scale tectonic features. As such, further investigations are necessary, based on seismic and geological techniques and using more detailed airborne
investigations, based on closer line-spacings.
113
Chapter 6
OUTLOOK
Aircrafts serving as multi-instrumentation platforms provide measurements of magnetic intensity, gravity, bedrock- and ice topography, and result in a fast and effective
observation technique for the interpretation of lithospheric boundaries. The resulting data are only as valuable as their resolution and homogeneity allow, and this
depends mainly on the flight line spacing and used equipment. For a comprehensive
classification of tectonic units and subunits, full coverage is essential. This criterion
was not achieved within this first project, and leads naturally to some uncertainties.
The gaps will be closed with further investigations.
These future investigations may target also subglacial environments, a continental
phenomenon that occurs below ice sheets. These isolated environments are natural
separated microcosms and of interest understanding evolution of fauna and flora in
the antarctic region. For this purpose, the topographic maps compiled here define a
good data base.
Apart from the calculation of Free-air- and Bouguer anomalies for interpretation of
the crustal structure, the data are useful for the calculation of a new geoid-model
(see PhD thesis (in preparation), Jan Müller, TU-Dresden). The calculated free-air
anomaly provides an excellent database for improving the regional geoid by combining gravity and topographic data from aero-geophysical observations with long
wavelength information from global gravity models.
Furthermore, with respect to one of the aims within the VISA project, the new
datasets will contribute to the validation of new data provided by satellite missions
CHAMP, GRACE and GOCE.
115
ACKNOWLEDGEMENTS
I sincerely thank Prof. Dr. H. Miller who gave me constant support and the opportunity to carry out this work.
The project was initiated from Dr. Wilfried Jokat. I particularly thank him for the
support over many years of my work.
Special thanks go to Dr. Graeme Eagles for his continued and detailed corrections
which contributed to the success of this work.
I also like to thank my close friend Dr. Tobias Boebel and OPTIMARE Sensorsysteme AG, not only for the instrumentation support of the flight campaigns as well
as Tobias fundamental work with airborne gravity at the Alfred Wegener Institute.
Furthermore, thanks go to Dr. Oliver Ritzmann, Dr. Vera Schlindwein and Dr.
Matthias Koenig for suggestions and corrections of the final thesis.
Dr. Joachim Jacobs introduced me to the amazing geology of the Antarctic continent.
His work is strongly dedicated to Dronning Maud Land and provided substantial
models for understanding geological history.
Many thanks go to the Bundesanstalt für Geowissenschaften und Rohstoffe (BGR),
to Dr. Uwe Meyer, for his software support, to Dr. Gernot Reitmayer for his help
referring to gravity data, to Dr. Detlef Damaske for his discussion of magnetic data
interpretation and Felix Goldmann for technical assistance in data processing.
Colleagues from TU-Dresden, Planetary Geodesy Division, i.e. Dr. Mirko Scheinert,
Axel Rühlke and Jan Müller, are thanked for their teaching of processing GPS data.
Thanks to all the members of the working groups of geophysics and glaciology, particularly Dr. Alfons Eckstaller, Dr. Daniel Steinhage and Christine Wesche, for all
the advice and support during the normal daily work.
Preparation of this work was supported by the Deutsche Forschungsgemeinschaft
(DFG) through the VISA Project, founded under grants Di 473/17-1 and Jo 191/81.
117
References
Aleshkova, N., Golynsky, A., Kurinin, R., Mandrikov, V. (1998). Gravity mapping
in the southern Weddell Sea region, Abschlussbericht zum Forschungsprojekt
BMFT 03F08GUS, Teilprojekt 4.
Bamber, J.L. and Blindschadler, R.A. (1997). An improved elevation dataset for
climate and ice-sheet modelling: Validation with satellite imagery, Anals of
Glaciology 25: 439–444.
Bayer, B. (2007). Untersuchungen der Lithosphäre des Dronning Maud Landes,
Antarktis, mit geophysikalischen Methoden, PhD thesis, University Bremen,
Bremen, Germany.
Bell, R., Childers, V., Arko, R., Blankenship, D. and Brozena, J. (1999). Airborne
gravity and precise positioning for geological applications, Journal of Geophysical Research 104 (7): 15281–15292.
Boebel, T. (2000).
Flugzeuggestützte Topographie- und Schweremessungen:
Messsystem und Anwendung auf die Region Framstrasse, Spitsbergen und Nordostgrönland, PhD thesis, University of Bremen, Bremen, Germany.
Brozena, J.M. (1991). Airborne gravity, topographic and magnetic mapping of an
entire continent, in Colombo, O.L., From Mars to Greenland Charting Gravity with Space and Airborne Instruments, Band 110, IAG Symposium Series,
Springer, Berlin,: 203–214.
Brozena, J.M., and Peters, M.F. (1988). An airborne gravity study of eastern North
Carolina, Geophysics 52: 245–252.
Childers, V., Bell, R. and Brozena, J. (1999). Airborne gravimetry: An investigation
of filtering, Geophysics 64 (1): 61–69.
Cox, K.G. (1992). Karoo igneous activity, and the early stages of the break-up
of gondwanaland, in: Storey, B.C., Alabaster, T. and Pankhurst, R.J. (eds),
Magmatism and the Causes of Continental Break-up, Special Publication 68,
Geological Society London, pp. 137–148.
Damaske, G., Marcinkowski, V., Möller, H.D. (2005). Aeromagnetic survey in Central Dronning Maud Land, East Antarctica, during the 1995/1996 GeoMaud
119
expedition: Layout, execution and data processing, Geologisches Jahrbuch,
B97:pp. 53–83.
Ferraccioli, F., Jones, P., Curtis, M., Leat, P. and Riley, I. (2005). Tectonic and magmatic patterns in the Jutulstraumen rift (?) region, East Antarctica, as imaged
by high-resolution aeromagnetic data, Earth Planets and Space 57 (8): 767–780.
Ferraccioli, F., Jones, P.C., Curtis, M.L. and Leat, P.T. (2005). Subglacial imprints
of early Gondwana break-up as identified from high resolution aerogeophysical
data over Western Dronning Maud Land, East Antarctica, Terra Nova 17: 573–
779.
Golynsky, A. and Jacobs, J. (2001). Grenville-age versus Pan-African magnetic
anomaly imprints in Western Dronning Maud Land, East Antarctica, Journal
of Geology 109: 136–142.
Golynsky, A.V. (2007). Magnetic anomalies in East Antarctica: a window on major
tectonic provinces and their boundaries, in: Cooper, A.K. and Raymond, C.R.
et al. (eds), Proceedings of the 10th ISAES, USGS Open-File Report.
Golynsky, A.V. and Aleshkova, N.D. (2000). New aspects of crustal structure in the
Weddell Sea region from aeromagnetic studies, Polarforschung 67 (3): 133–141.
Golynsky, A.V., Chiappini, M., Damaske, D., Ferraccioli, F., Ferris, J., Finn, C.,
Ghidella, M., Isihara, T., Johnson, A., Kim, H.R., Kovacs, L., LaBrecque,
J.L., Masolov, V.N., Nogi, Y., Purucker, M., Taylor, P. and Torta, M. (2001).
ADMAP - Magnetic anomaly map of the Antarctic, 1:10000000 scale map, BAS
(Misc.)10, Morris, P. and von Frese, R. (eds), British Antarctic Survey.
Golynsky, A.V., Grikurov, G.E. and Kamenev, E.N. (2000). Geological significance
of regional magnetic anomalies in Coats Land and Western Dronning Maud
Land, Polarforschung 67 (3): 91–99.
Groenewald, P., Grantham, G. and Watkeys, M. (1991). Geological evidence for a
proterozoic to mesozoic link between southeastern Africa and Dronning Maud
Land, Antarctica, Journal of Geological Society, London 148: 1115–1123.
Harlan, R.B. (1968). Eotvos corrections for airborne gravimetry, Journal of Geophysical Research 73 (14): 4675–4679.
Jacobs, J. and Thomas, R.J. (2004). Himalayan-type indenter-escape tectonics model
for the southern part of the late neoproterozoic-early paleozoic East African
Antarctic Orogen, Geology 32 (8): 721–724.
Jacobs, J., Bauer, W., Fanning, C.M. (2003a). Timing of grenville-age vs. panafrican medium- to high grade metamorphism in Western Dronning Maud Land
(East Antarctica) and significance for correlations in Rodinia and Gondwana,
Precambrian Research 125: 1–20.
Jacobs, J., Bauer, W., Spaeth, G., Thomas, R. J. and Weber, K. (1996). Lithology
and structure of the Grenville-aged (? 1.1 Ga) basement of Heimefrontfjella
(East Antarctica), Geologische Rundschau 85 (4): 800–821.
Jacobs, J., Klemd, R., Fanning, C.M., Bauer, W., and Colombo, F. (2003b). Extensional collapse of the late neoproterozoic-early palaeozoic East African-Antarctic
Orogen in Central Dronning Maud Land, East Antarctica, Geological Society,
Special Publication pp. 271–287.
Jokat, W. (2004). East antarctic tectonics and sedimentation history from 40W to
60E, Marine Geophysical Researches 25 (3-4): 181–182.
Jokat, W., Boebel, T., König, M. and Meyer, U. (2003). Timing and geometry of
early Gondwana break-up, Journal of Geophysical Research 108 (B9): 2428.
Kane, M.F. (1962). A comprehensive system of terrain correction using a digital
computer, Geophysics 27 (4).
Kirchner, M. (2002). Kinematische Modellierung von GPS-Referenzstationen auf Eis
für flugzeuggestützte Messvorhaben, Diploma, Technical University Dresden,
Dresden, Germany.
König, M. (2005) Processing of shipborne magnetometer data and revision of the
timing and geometry of the Mesozoic break-up of Gondwana. PhD thesis, University Bremen, Bremen, Germany.
Kollersberger, T. (2005). Anwendung von Krümmungsattributen zur Visualisierung
und Interpretation von Potentialfeldern und Gradienten, Diploma, Freie Universität Berlin, Berlin, Germany.
LaCoste, L. (1967). Measurement of gravity at sea and in the air, Reviews of Geophysics 5 (4): 477–526.
LaCoste, L., Ford, J., Bowless, R., and Archer, K. (1982). Gravity measurements
in an airplane using state-of-the-art navigation and altimetry, Geophysics 47
(5): 832–838.
LaCoste, L.J.B. (1988). The zero-length spring gravity meter, Geophysics 7: 20–21.
LaFehr, T. and Nettleton, L. (1967). Quantitative evaluation of a stabilized platform
shipboard gravity meter, Geophysics 32: 110–118.
LaFehr, T.R. (1991). An exact solution for the gravity curvature (bullard b) correction, Geophysics 56: 1179–1184.
Lythe, M. B., Vaughan, D. G. (2001). BEDMAP: A new ice thickness and subglacial
topographic model of Antarctica, Journal of Geophysical Research 106(B6)
(11): 335–352.
Meyer, U., Boedecker, G., Pflug, H. (2003). Angel - airborne navigation and gravimetry ensembles and laboraty - introduction and first airborne tests, Scientific
Technical Report, GFZ-Potsdam.
Meyer, U., and Heyde, I. (2004).
Entwicklung und Anwendung von aerogravimetrischem Messsystemen, DGG Kolloqium Aerogeophysik, pp. 17–44.
Nagy, D. (1966). The gravitational attraction of a right rectangular prism, Geophysics 31(2).
Nettleton, L., LaCoste, L., and Harrison, C. (1960). Tests of an airborne gravity
meter, Geophysics 25 (1): 181–202.
Olesen, A.V. and Fosberg, R. (1997). Airborne gravimetry using Lacoste & Romberg
gravimeter- An error analysis, Geomatics and Navigation, Intern. Symp. on
Kinematic Systems in Geodesy, pp. 613–617.
Piech, H.J. (2005). Present Knowledge of the Geology of Central Dronning Maud
Land, East Antarctica: Main results of the 1995/1996 GeoMaud expedition,
Geologisches Jahrbuch B97: 341–408.
Piech, H.-J., Bauer, W., Piazolo, S., Jacobs, J., Markl, G. (2005). Comparison of the
Geology of Central Dronning Maud Land, East Antarctica, with other areas of
the Gondwana Supercontinent, Geologisches Jahrbuch B97: 309–340.
Reid, A.B., Allsop, J.M., Granser, H., Millet, A.J., Somerton, I.W. (1990). Magnetic
interpretation in the three dimensions using Euler Deconvolution, Geophysics
55: 80–91.
Reitmayr, G. (2005). Gravity survey in Central Dronning Maud Land, East
Antarctica, during the 1995/1996 GeoMaud expedition, Geologisches Jahrbuch
B97: 141–164.
Roberts, A. (2001). Curvature attributes and their application to 3d interpreted
horizons, First Break 19 (2): 285–100.
Roest, W.R., Verhoef, J., Pilkington, M. (1992). Magnetic interpretation using the
3d analytic signal, Geophysics 57 (1): 116–125.
Spector, A. and Grant, F.S. (1970). Statistical models for interpreting aeromagnetic
data, Geophysics 35 (2): 293–302.
Steinhage, D. (2001). Beiträge aus geophysikalischen Messungen in Dronning Maud
Land, Antarktis, zur Auffindung eines optimalen Bohrpunktes für eine Eiskerntiefbohrung, PhD thesis, University of Bremen, Bremen, Germany.
Swain, C. (1996). Horizontal acceleration corrections in airborne gravimetry, Geophysics 61 (1): 273–276.
Thompson, D.T. (1982). Euldph: A new techniquw for making computer-assisted
depth estimates from magnetic data, Geophysics 47: 31–37.
Torge, W. (1989). Gravimetry, Walter de Gruyter, Berlin, 465pp.
Torge, W. (1991). Geodesy, Walter de Gruyter, Berlin, 264pp.
Valliant, H. (1991). Gravity meter calibration at LaCoste & Romberg, Geophysics
56: 705–711.
Valliant, H. (1992). The LaCoste & Romberg Air/Sea Gravity Meter: An overview,
In Hydrocarbons, Band 1, CRC Press, Inc.
Valliant, H. and LaCoste, L. (1976). Theory and evaluation of the Lacoste & Romberg
three-axis inertial platform for marine gravimetry, Geophysics 41: 459–467.
Wesche, C., Riedel, S., Steinhage, D. (submitted). Precise surface topography of the
grounded ice tounge at the Ekstroemisen, Antarctica, based on several geophysical data sets, Journal of Photogrammetry and Remote Sensing.
Wesche, C., Riedel, S., Steinhage, D. Eisen, O. and Oerter, H. (in prep.). An improved DEM and refined ices divides location for Central Dronning Maud Land,
Antarctica, Journal of Glaciology.
List of Tables
2.1
GPS, precise error model, C/A code . . . . . . . . . . . . . . . . . .
21
2.2
Gravity measuring systems
29
4.1
Subunits within the Grunehogna Craton . . . . . . . . . . . . . . . . 100
4.2
Subunits within the Maudheim Province . . . . . . . . . . . . . . . . 108
. . . . . . . . . . . . . . . . . . . . . . .
A.1 DGPS processing parameters . . . . . . . . . . . . . . . . . . . . . . 132
B.1 Gravity base readings, VISA 1. . . . . . . . . . . . . . . . . . . . . . 134
B.2 Tying process, VISA I . . . . . . . . . . . . . . . . . . . . . . . . . . 135
B.3 Sensor drift, VISA I . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
B.4 Gravity base readings, VISA II. . . . . . . . . . . . . . . . . . . . . . 136
B.5 Tying process, VISA II . . . . . . . . . . . . . . . . . . . . . . . . . . 136
B.6 Sensor drift, VISA II . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
B.7 Gravity base readings, VISA III. . . . . . . . . . . . . . . . . . . . . 137
B.8 Tying process, VISA III . . . . . . . . . . . . . . . . . . . . . . . . . 137
B.9 Tie errors, VISA III . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
B.10 Sensor drift, VISA III . . . . . . . . . . . . . . . . . . . . . . . . . . 138
B.11 Gravity base readings, VISA IV. . . . . . . . . . . . . . . . . . . . . 139
B.12 Tying process, VISA IV . . . . . . . . . . . . . . . . . . . . . . . . . 140
B.13 Sensor drift, VISA IV . . . . . . . . . . . . . . . . . . . . . . . . . . 140
125
List of Figures
1.1
The Antarctic continent and the related area of investigation within
the VISA project. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
1.2
The East African Antarctic Orogen and escape tectonics in DML . .
7
1.3
Detailed geological observations in DML . . . . . . . . . . . . . . . .
8
2.1
Polar 2 aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.2
Cross-section of Dornier 228-200 Polar 2 aircraft . . . . . . . . . . .
12
2.3
Polar 2 Radio Echo Sounding instrumentation . . . . . . . . . . . . .
13
2.4
Cross section of RES sounding profile. . . . . . . . . . . . . . . . . .
15
2.5
The Global Positioning System . . . . . . . . . . . . . . . . . . . . .
16
2.6
GPS satellite signals . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
2.7
GPS nominal constellation . . . . . . . . . . . . . . . . . . . . . . . .
18
2.8
Carrier phase tracking . . . . . . . . . . . . . . . . . . . . . . . . . .
19
2.9
Carrier phase positioning
. . . . . . . . . . . . . . . . . . . . . . . .
20
2.10 Principle of scalar gravimeter systems . . . . . . . . . . . . . . . . .
24
2.11 Effects on moving platform . . . . . . . . . . . . . . . . . . . . . . .
26
2.12 Simplified gravimeter and sensor . . . . . . . . . . . . . . . . . . . .
27
2.13 The scalar ZLS Ultrasys S56 Air/Sea gravity meter . . . . . . . . . .
28
2.14 Definition of curvature . . . . . . . . . . . . . . . . . . . . . . . . . .
39
2.15 Sign convention for curvature attributes . . . . . . . . . . . . . . . .
40
2.16 Schematic outline of the Analytic Signal . . . . . . . . . . . . . . . .
42
2.17 Isostatic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48
3.1
Overview of the study area and campaigns. . . . . . . . . . . . . . .
49
3.2
RES Results from the VISA I campaign . . . . . . . . . . . . . . . .
53
3.3
RES Results from the VISA II campaign . . . . . . . . . . . . . . . .
55
3.4
RES Results from the VISA III campaign . . . . . . . . . . . . . . .
57
127
3.5
RES Results from the VISA IV campaign . . . . . . . . . . . . . . .
59
3.6
Free-air anomaly of VISA I campaign. . . . . . . . . . . . . . . . . .
63
3.7
Free-air anomaly of VISA II campaign. . . . . . . . . . . . . . . . . .
65
3.8
Free-air anomaly of the VISA III campaign. . . . . . . . . . . . . . .
67
3.9
Free-air anomaly of the VISA IV campaign. . . . . . . . . . . . . . .
69
3.10 Total Magnetic Intensity map of VISA I campaign. . . . . . . . . . .
71
3.11 Total Magnetic Intensity map of VISA II campaign. . . . . . . . . .
73
3.12 Total Magnetic Intensity map of VISA III campaign. . . . . . . . . .
75
3.13 Total Magnetic Intensity map of VISA IV campaign. . . . . . . . . .
77
4.1
Compilation Ice Topography. . . . . . . . . . . . . . . . . . . . . . .
80
4.2
Compilation Bedrock Topography. . . . . . . . . . . . . . . . . . . .
81
4.3
Compilation Free-air anomaly. . . . . . . . . . . . . . . . . . . . . . .
83
4.4
Complete Bouguer anomaly. . . . . . . . . . . . . . . . . . . . . . . .
85
4.5
Filtering of gravity signals . . . . . . . . . . . . . . . . . . . . . . . .
86
4.6
Isostatic calculations . . . . . . . . . . . . . . . . . . . . . . . . . . .
87
4.7
Compilation Total Magnetic Intensity. . . . . . . . . . . . . . . . . .
90
4.8
Main magnetic units . . . . . . . . . . . . . . . . . . . . . . . . . . .
91
4.9
Analytic Signal analysis. . . . . . . . . . . . . . . . . . . . . . . . . .
93
4.10 Tilt Derivative filter . . . . . . . . . . . . . . . . . . . . . . . . . . .
94
4.11 TMI, Grunehogna Province . . . . . . . . . . . . . . . . . . . . . . .
96
4.12 AS, Grunehogna Province . . . . . . . . . . . . . . . . . . . . . . . .
97
4.13 TDR, Grunehogna Province . . . . . . . . . . . . . . . . . . . . . . .
98
4.14 Bedrock topography, Grunehogna Province
99
. . . . . . . . . . . . . .
4.15 TMI and Bedrock topography, Maudheim Province, central DML . . 102
4.16 AS and TDR, Maudheim Province, central DML . . . . . . . . . . . 103
4.17 TMI and Bedrock topography, Maudheim Province, eastern DML . . 105
4.18 AS and TDR, Maudheim Province, eastern DML . . . . . . . . . . . 107
4.19 Geologic model suggestions . . . . . . . . . . . . . . . . . . . . . . . 110
C.1 Radially averaged power spectra, gravity, VISA I . . . . . . . . . . . 142
C.2 Radially averaged power spectra, magnetic, VISA I . . . . . . . . . . 142
C.3 Euler Deconvolution, magnetic, VISA I . . . . . . . . . . . . . . . . . 143
C.4 Radially averaged power spectra, gravity, VISA II . . . . . . . . . . . 144
C.5 Radially averaged power spectra, magnetic, VISA II . . . . . . . . . 144
C.6 Euler Deconvolution, magnetic, VISA II . . . . . . . . . . . . . . . . 145
C.7 Radially averaged power spectra, gravity, VISA III . . . . . . . . . . 146
C.8 Radially averaged power spectra, magnetic, VISA III . . . . . . . . . 146
C.9 Euler Deconvolution, magnetic, VISA III . . . . . . . . . . . . . . . . 147
C.10 Radially averaged power spectra, gravity, VISA IV . . . . . . . . . . 148
C.11 Radially averaged power spectra, magnetic, VISA IV . . . . . . . . . 148
C.12 Euler Deconvolution, magnetic, VISA IV . . . . . . . . . . . . . . . . 149
D.1 Curvature analysis, regional observation, (A) . . . . . . . . . . . . . 152
D.2 Curvature analysis, regional observation, (B)
. . . . . . . . . . . . . 153
D.3 Curvature analysis, regional observation, (C) . . . . . . . . . . . . . 154
D.4 Curvature analysis, regional observation, (D) . . . . . . . . . . . . . 155
D.5 Curvature analysis, local observation, (A) . . . . . . . . . . . . . . . 156
D.6 Curvature analysis, local observation, (B) . . . . . . . . . . . . . . . 157
D.7 Curvature analysis, local observation, (C) . . . . . . . . . . . . . . . 158
APPENDIX A
DGPS SETTINGS
131
DGPS SETTINGS
A.1
A.1. PARAMETERS
Parameters
Table A.1: DGPS processing parameters, listed for 1 s and 30 s data. Optionally the standard
settings are listed.
settings
primary
static
kinematic
gobal
quality
troposphere
ionosphere
OTF
parameter
elevation mask
ephemerides
solution
min. observation
max. baselength broadcast
max. baselength precise
min. time reference
min. time stat. init
min. init try
min. OTF time
frequency
max. phase jump
max. iterations
max. time calc.
dismiss, if RMS > ...
dismiss, if variance < ...
dismiss, if ref-variance > ...
RMS
modell
interval
solution at
method
1s
15◦
30 s
10◦ /15◦
precise
float/fixed
5 km
3600 s
600 s
600 s
120 s
600 s
10 s
300 s
120 min
90 min
0.2 m
20
Niell
2h
5 km
standard
13◦
broadcast
fixed
120 s
200 km
2000 km
600 s
120 s
3
200 s
L1
600 s
10
30 min
0.03 m
1.5
10
3.5
Hopefield
2h
10 km
optimal
For all calculations, the use of the precise ephemerides was applied. The final observations include observations at an elevation mask of 10◦ . Irregularities are observed
due to this, and so the elevation mask settings are changed to 15◦ . Forcing the
program to calculate a fixed solution was not always possible. Due to this, a float
solution was used. Phase jumps are only corrected if the disturbing effect is longer
then ten epochs (only static). The tropospheric model of Niell was used, including
a parameter interval of two hours. The ionosphere free linear combination of the
measurements will be used at baselength of 5 km.
132
APPENDIX B
GRAVITY READINGS and
TYING-PROCESS
133
GRAVITY READINGS and TYING-PROCESS
B.1
B.1. TYING VISA I
Tying VISA I
Table B.1: Gravity base readings, VISA 1.
date
01.12.2001
01.12.2001
01.12.2001
02.01.2002
02.01.2002
04.01.2002
04.01.2002
04.01.2002
04.01.2002
05.01.2002
07.01.2002
08.01.2002
08.01.2002
10.01.2002
13.01.2002
14.01.2002
16.01.2002
16.01.2002
18.01.2002
21.01.2002
21.01.2002
21.01.2002
22.01.2002
23.01.2002
23.01.2002
24.01.2002
25.01.2002
26.01.2002
27.01.2002
27.01.2002
28.01.2002
28.01.2002
29.01.2002
29.01.2002
09.02.2002
09.02.2002
12.02.2002
13.02.2002
23.02.2002
28.01.2002
station
Poller94
Poller94
Polarstern
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
E-Base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
E-base
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Poller94
land-reading
[mGal]
3159.48
3159.55
3158.67
6249.80
6249.80
6249.91
6224.99
6225.06
6224.99
3160.04
134
S56-reading
[mGal]
12771.9
12771.1
12771.9
12770.9
12771.3
12770.4
12770.4
12771.2
12768.7
12769.1
12768.4
12743.9
12744.1
12743.2
12743.4
12742.0
12741.5
12742.0
12742.1
12741.6
12740.9
12740.5
12740.6
12740.2
12739.8
12739.1
12737.4
12737.4
12739.3
12738.8
12739.5
12763.6
12763.5
12762.9
12763.6
-
GRAVITY READINGS and TYING-PROCESS
B.1. TYING VISA I
Table B.2: Tying process, VISA I, Calculation of absolute gravity readings refers to Poller 94 station
(South Africa) with an absolute g=979657.904 mGal.
station
Boreas Passat
Neumayer Station
E-Base
absolute gravity
value [mGal]
982648.49
982748.67
982723.41
measured relative
gravity value [mGal]
6150.11
6250.28
6225.03
Table B.3: Sensor drift, VISA I. The drift process of the S56 gravity meter was subdivided into 3
stages due to the locations and times of operations. The observed values might be too high, but
take into account that both stations are situated on the ice shelf, which moves continuously by
glacier flow as well as being influenced by tidal movement.
base station
Neumayer (9 days)
E-Base (17 days)
Neumayer (15 days)
135
observed drift
[mGal]
3.5
6.7
0.7
GRAVITY READINGS and TYING-PROCESS
B.2
B.2. TYING VISA II
Tying VISA II
Table B.4: Gravity base readings, VISA II.
date
station
28.12.2002
28.12.2002
29.12.2002
29.12.2002
31.12.2002
31.12.2002
13.01.2003
13.01.2003
13.01.2003
13.01.2003
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
Neumayer
land-reading
[mGal]
6240.42
-
S56-reading
[mGal]
12760.02
12770.00
12760.24
12770.00
12759.88
12759.44
12758.32
12756.90
12757.64
12756.30
Table B.5: Tying process, VISA II. Calculation of absolute gravity readings refers to Poller 94
station (Cape Town, South Africa) g=979657.904 mGal using readings taken the year before due
to the absence of direct measurements.
station
Neumayer Station
absolute gravity
value [mGal]
982738.83
measured relative
gravity value [mGal]
6240.42
Table B.6: Sensor drift, VISA II. The observed drift of the S56 gravity meter, of 3.9 mGal, seems
to be good in view of Neumayer Station’s movement.
base station
Neumayer (16 days)
136
observed drift
[mGal]
3.9
GRAVITY READINGS and TYING-PROCESS
B.3
B.3. TYING VISA III
Tying VISA III
Table B.7: Gravity base readings, VISA III.
date
22.12.2003
23.12.2003
23.12.2003
24.12.2003
25.12.2003
27.12.2003
29.12.2003
30.12.2003
06.01.2004
07.01.2004
07.01.2004
09.01.2004
09.01.2004
09.01.2004
10.02.2004
13.02.2004
station
Novo
Novo
Novo
Novo
Novo
Novo
Novo
Novo
Sanae
Sanae
Sanae
Sanae
Sanae
Sanae(seismo)
Novo(abs.)
UCT
land-reading
[mGal]
6104.93
6105.01
6095.17
6095.06
6095.17
6099.63
6215.32
3254.19
S56-reading
[mGal]
12496.20
12495.98
12492.58
12493.78
12491.92
12492.36
12480.94
12480.72
12480.74
(12752.84)
12480.72
-
Table B.8: Tying process, VISA III. Calculation of absolute gravity readings refer to UCT station
(South Africa).
station
Cape Town (UCT)
Novo (airfield)
Sanae (airfield)
absolute gravity
value [mGal]
979616.80
982467.58
982457.72
137
measured relative
gravity value [mGal]
3254.19
6104.98
6095.12
GRAVITY READINGS and TYING-PROCESS
B.3. TYING VISA III
Table B.9: Tie errors, VISA III. During the same summer season, Mäkinen (2003) made some
absolute g measurements. His results compare to the tying process as listed.
station
Novo (reference)
Novo (airfield)
Sanae (Seismo)
measured absolute
gravity [mGal]
982579.43
982468.35
982463.47
difference gravity
value [mGal]
1.50
0.77
1.23
Table B.10: Sensor drift, VISA III.
base station
Novo (9 days)
Sanae (4 days)
138
observed drift
[mGal]
4.3
0.2
GRAVITY READINGS and TYING-PROCESS
B.4
B.4. TYING VISA IV
Tying VISA IV
Table B.11: Gravity base readings, VISA IV.
date
station
26.11.2004
26.11.2004
17.12.2004
20.12.2004
20.12.2004
25.12.2004
25.12.2004
26.12.2004
27.12.2004
27.12.2004
28.12.2004
28.12.2004
29.12.2004
31.12.2004
02.01.2005
02.01.2005
02.01.2005
03.01.2005
03.01.2005
04.01.2005
04.01.2005
04.01.2005
05.01.2005
05.01.2005
05.01.2005
05.01.2005
UCT
BM3
NM-U
NM
NM
NM
NM
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
Kottas
land-reading
[mGal]
3255.60
3277.68
6372.28
6368.42
-
139
S56-reading
[mGal]
12760.5
12760.8
12760.8
12339.6
12339.4
12399.4
12339.1
12339.3
12338.5
12337.1
12336.5
12337.1
12337.4
12340.5
12337.6
12337.0
12337.0
12336.7
12337.1
12337.6
12337.4
12336.4
GRAVITY READINGS and TYING-PROCESS
B.4. TYING VISA IV
Table B.12: Tying process, VISA IV. Calculation of absolute gravity readings refer to UCT station
(South Africa). Measurements and calculations are done by TU-Dresden.
station
Neumayer (construction)
Neumayer (airfield)
Kottas (campsite)
Kottas (airfield)
absolute gravity
value [mGal]
982733.47
982729.61
982311.87
982310.57
measured relative
gravity value [mGal]
6372.29
6368.43
-
Table B.13: Sensor drift, VISA IV.
base station
Kottas-airfield (11 days)
140
observed drift
[mGal]
2.9
APPENDIX C
DEPTH ESTIMATION
SOLUTIONS
141
DEPTH ESTIMATION SOLUTIONS
C.1
C.1. SOURCE DEPTHS, VISA I
Source depths, VISA I
Figure C.1: Radially averaged power spectra of gravity, VISA I. Different ensembles of causative
bodies or layers at depths of 27 km, 15 km and 9 km can be recognized. These will correlate with
the results of receiver function analysis of seismological data, processed and interpreted by Bayer,
2007.
Figure C.2: Radially averaged power spectra of magnetics, VISA I. Suggestive magnetic sources at
depths of 10-12 km, 8 km and 5 km.
142
DEPTH ESTIMATION SOLUTIONS
C.1. SOURCE DEPTHS, VISA I
Figure C.3: Euler Deconvolution, magnetic, VISA I, for sill and dyke structures (SI=1).
143
DEPTH ESTIMATION SOLUTIONS
C.2
C.2. SOURCE DEPTHS, VISA II
Source depths, VISA II
Figure C.4: Radially averaged power spectra of gravity, VISA II. Ensembles of causative bodies
vary with depth and wavelength. Sources may be recognized at depths of 30 km, 16 km and 8 km.
Figure C.5: Radially averaged power spectra of magnetics, VISA II., The approximate depth solution for magnetic sources suggests bodies at 15 km, 10 km, 7 km and 5 km depth.
144
DEPTH ESTIMATION SOLUTIONS
C.2. SOURCE DEPTHS, VISA II
Figure C.6: Euler Deconvolution, magnetic, VISA II, for sill and dyke structures (SI=1).
145
DEPTH ESTIMATION SOLUTIONS
C.3
C.3. SOURCE DEPTHS, VISA III
Source depths, VISA III
Figure C.7: Radially averaged power spectra of gravity, VISA III., Suggestive source horizons exist
at 42 km, 16 km and 10 km.
Figure C.8: Radially averaged power spectra of magnetics, VISA III. Magnetic sources are located
at depths of nearly 12 km and 7 km.
146
DEPTH ESTIMATION SOLUTIONS
C.3. SOURCE DEPTHS, VISA III
Figure C.9: Euler Deconvolution, magnetic, VISA III, for sill and dyke structures (SI=1). As
mentioned, at 74◦ S, the results became less clear and must be interpreted with caution.
147
DEPTH ESTIMATION SOLUTIONS
C.4
C.4. SOURCE DEPTHS, VISA IV
Source depths, VISA IV
Figure C.10: Radially averaged power spectra of gravity, VISA IV, suggests horizons at 40 km,
20 km, 12 km and 5.5 km depth.
Figure C.11: Radially averaged power spectra of magnetics, VISA IV, suggests ensembles of
causative magnetic sources at 12 km, 7 km and 5 km depth.
148
DEPTH ESTIMATION SOLUTIONS
C.4. SOURCE DEPTHS, VISA IV
Figure C.12: Euler Deconvolution, magnetic, VISA IV, for sill and dyke structures (SI=1). Additionally, the strike directions are confirmed within this solution, especially so for the Heimefrontfjella
shear zone [Jacobs and Golynsky, 2001].
149
DEPTH ESTIMATION SOLUTIONS
C.4. SOURCE DEPTHS, VISA IV
.
150
APPENDIX D
CURVATURE DISCUSSION
D.1
Regional and local observations
151
Figure D.1: Curvature attributes, regional observation (50 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
152
Figure D.2: Curvature attributes, regional observation (50 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
153
Figure D.3: Curvature attributes, regional observation (50 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
154
Figure D.4: Curvature attributes, regional observation (50 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
155
Figure D.5: Curvature attributes, local observation (10 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
156
Figure D.6: Curvature attributes, local observation (10 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
157
Figure D.7: Curvature attributes, local observation (10 km window), left: topography, middle: gravity, right: magnetic.
CURVATURE DISCUSSION
D.1. REGIONAL AND LOCAL OBSERVATIONS
158
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement