Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH...

28

Transcript of Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH...

Page 1: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

An Integrated Geophysical Investigation of Greenland�s

Tectonic History

DISSERTATION

Presented in Partial Ful�llment of the Requirements for

the Degree Doctor of Philosophy in the

Graduate School of The Ohio State University

By

Daniel R� Roman� B�S�� M�S�

� � � � �

The Ohio State University

����

Dissertation Committee�

Dr� Ralph R� B� von Frese� Adviser

Dr� Kenneth C� Jezek

Dr� Hallan C� Noltimier

Dr� Christopher Jekeli

Approved by

Adviser

Department of GeologicalSciences

Page 2: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

c� Copyright by

Daniel R� Roman

����

Page 3: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

ABSTRACT

A new model for the crustal evolution of the Labrador Sea region of southwestern

Greenland ����� ����N and �� ���W was developed from gravityderived Moho

estimates and lithologic and geologic features interpreted from correlative geopoten

tial anomalies� and recent seismic surveys �Chian and Louden� ����� Chian et al��

����a� Chian et al�� ����b� Chalmers and Laursen� ������ Previous kinematic models

�Srivastava� ����� Srivastava and Tapscott� ���� Roest and Srivastava� ����� sug

gested that the opening of the Labrador Sea caused a counterclockwise rotation of

Greenland from ca� �� to � Ma as a part of the opening of the North Atlantic

Ocean� These models were based on interpretation of a ��� km wide zone of crust o�

of the coasts of Greenland and Labrador as oceanic crust with continuous magnetic

isochrons through anomaly ��� The structural implications of the gravityderived

Moho and crustal density models challenge this interpretation� Instead� this region

is interpreted a combination of riftedcontinental and transitional crust� The correla

tion analysis of freeair gravity and magnetic anomalies determined that rocks within

this ��� km zone were more characteristic of riftedcontinental or transitional crust�

and this was further supported by the results of seismic surveys �Chian and Louden�

����� Chian et al�� ����a� ����b� Chalmers and Laursen� ������ The linear magnetic

anomalies interpreted as isochrons �� and �� by Roest and Srivastava ������ were in

terpreted as serpentinization along the crustal rupture and delamination surfaces or

ii

Page 4: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

along the base of regionalscale halfgrabens� The revised model postulates that the

rotational opening of the Canada Basin from ��� to ��� Ma induced a counterclock

wise rotation of Greenland� which extended and thinned the Archean crust between

Greenland and Labrador� This weakened crust was thus well disposed to rifting when

the North Atlantic rift system propagated northward into the region at about �� Ma�

Slow extension rates and an insu�cient supply of magma delayed the initiation of

oceanic spreading until about � Ma when Greenland began to separate from North

America and move with Europe�

iii

Page 5: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

To Lauren for being loving� supportive� and understanding especially when I was

away from home working on all those long� cold� and lonely nights�

iv

Page 6: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

ACKNOWLEDGMENTS

Foremost� I thank Dr� Ralph von Frese for his brilliant tutelage� �nancial support�

and academic advice over the years� I also thank Dr� Ken Jezek who has been another

source of academic wisdom and support� I thank Drs� Hal Noltimier and Chris Jekeli

for their advice and comment during my research and the reviews of this manuscript�

I thank my parents and family for their support and understanding throughout

the long process of obtaining both my graduate degrees� I also thank Lauren for

taking care of the details in life so that I could stay focused on my research�

Thanks also go to Drs� Jeong Woo �Jay� Kim� Francois Baumgartner� Dru Smith�

Bea Csath�o� Li Tan� and countless others who have assisted me greatly in my e�orts�

I also appreciated the camaraderie� advice� and assistance I received from Brent Cur

tiss� Hyung Rae Kim� Jens Munk� Laramie Potts� Tim Leftwich� and others in the

departmental computer labs�

This work has been partially supported by grants from NASA�s Polar Oceans and

Ice Sheets Program and the Geodynamics Branch at NASA�s Goddard Space Flight

Center� as well as by the Byrd Polar Research Center� the Department of Geological

Sciences� the Center for Mapping� and the Ohio State Supercomputer Center� I

also would like to thank the European Space Agency for providing access to radar

altimeter data for this study and the National Imagery and Mapping Agency for

providing access to freeair gravity anomalies in the Greenland area�

v

Page 7: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

VITA

July ��� ��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � Born Barberton� Ohio

December ��� ���� � � � � � � � � � � � � � � � � � � � � � � � � � B�S� Geology� the University of Southern California

Dec� ��� Feb� ���� � � � � � � � � � � � � � � � � � � � � � � O�cer� United States Navy

September �� ���� � � � � � � � � � � � � � � � � � � � � � � � � � �M�S� Geodetic Science and Surveying�the Ohio State University

Sep� ���� Dec� ���� � � � � � � � � � � � � � � � � � � � � � � Graduate Research Associate at theByrd Polar Research Center� the OhioState University

Jan� ��� present � � � � � � � � � � � � � � � � � � � � � � � � � Graduate Research Associate in theDepartment of Geological Sciences� theOhio State University

PUBLICATIONS

Roman� D�R�� B� Csatho� K�C� Jezek� W�B� Krabill and K� Kuivinen� Gravity ValuesMeasured on the Greenland Ice Sheet� BPRC Technical Report ����� Byrd PolarResearch Center� The Ohio State University� Columbus� Ohio� pp� ��� �����

Kim� J� W�� JH Kim� R� R� B� von Frese� and D� R� Roman �Attenuation of TrackLine Noise by Spectral Reconstruction�� Geophys� Res� Lett�� ��� �� ������� �����

Roman� D�R�� B� Csatho� K�C� Jezek� R�H� Thomas� W�B� Krabill� R�R�B� vonFrese� and R� Forsberg� �A Comparison of Geoid Undulation Models for WestCentralGreenland� � J� Geophys� Res�� vol� ���� no� B�� p� ��������� �����

vi

Page 8: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

FIELDS OF STUDY

Major Field� Geological Sciences

Studies in�

Potential Field Geophysics Drs� von Frese � NoltimierGravimetric Geodesy� Spectral Geodesy Drs� Rapp � JekeliGeomathematics� Geostatistics Dr� von FreseRemote Sensing� GPS Drs� Jezek � GoadGeodynamics� Rheology Drs� Noltimier � Whillans

vii

Page 9: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

TABLE OF CONTENTS

Page

Abstract � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ii

Dedication � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � iv

Acknowledgments � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � v

Vita � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � vi

List of Tables � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � xi

List of Figures � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � xiv

Chapters�

�� General Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

�� GravityGeologic Estimation of Bathymetry � � � � � � � � � � � � � � � �

��� Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���� Review of the GravityGeologic Method �GGM � � � � � � � � � � � ���� Application of the GGM in the Barents Sea � � � � � � � � � � � � � ����� Application of the GGM in the Greenland Area � � � � � � � � � � � ����� Conclusions � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

�� Greenland Moho Depths from Spectrally Correlated Gravity and TerrainData � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

��� Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ����� Crustal Modeling Methodology � � � � � � � � � � � � � � � � � � � � ����� Data Description � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

viii

Page 10: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

����� Ice Thickness Model � � � � � � � � � � � � � � � � � � � � � � ������� Bedrock and Ocean Thickness DEMs � � � � � � � � � � � � � ������� FreeAir Gravity Anomalies �FAGA � � � � � � � � � � � � � ��

��� Comparison of Gridded Data Sets � � � � � � � � � � � � � � � � � � � ������ Surface Elevation Comparison � � � � � � � � � � � � � � � � � ������ Ice Thickness Comparison � � � � � � � � � � � � � � � � � � � ������� Gravity Comparison � � � � � � � � � � � � � � � � � � � � � � ��

��� TGE Determination by GLQ Integration � � � � � � � � � � � � � � � ���� Inversion for the Moho Depth Model by GLQ Integration � � � � � ���� Comparisons with Other Available Moho Depth Estimates � � � � � ��

����� Comparison with Seismic Moho Depth Estimates � � � � � � ������� Comparison of Interpolated and BPRC Pro�le Moho Depth

Data � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ����� Discussion � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ����� Summary � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ������� Conclusions and Recommendations � � � � � � � � � � � � � � � � � � ��

�� Crustal Gravity and Magnetic Anomaly Correlations of Greenland � � � � ���

��� Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ������ Greenland Geology and Related Density and Magnetization Variations���

����� Geologic Provinces � � � � � � � � � � � � � � � � � � � � � � � �������� Greenland Rock Densities and Magnetizations � � � � � � � � ��

��� Methodology for Estimating Crustal FAGA and MA and Their Correlations � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

��� TerrainDecorrelated Free Air Gravity Anomalies � � � � � � � � � � �������� Correlation Analysis with EGM� � � � � � � � � � � � � � � ���

��� Terrain Decorrelated Magnetic Anomalies � � � � � � � � � � � � � � �������� Data Description � � � � � � � � � � � � � � � � � � � � � � � � �������� Magnetic E�ects due to Crustal Thickness Variations � � � � ���

�� Gravity and Magnetic Anomaly Correlation Analysis � � � � � � � � ������ Discussion � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ������ Conclusions � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

�� A Regional Model for the Separation of Greenland from North America � ��

��� Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ����� Background � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ����� Regional Modeling � � � � � � � � � � � � � � � � � � � � � � � � � � � ����� Crustal Modeling and Analysis � � � � � � � � � � � � � � � � � � � � ������ New Regional Tectonic Model � � � � � � � � � � � � � � � � � � � � � ����� Conclusions � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

ix

Page 11: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

� Summary and Conclusions � � � � � � � � � � � � � � � � � � � � � � � � � � ���

Appendices�

A� Preprocessing of the FreeAir Gravity Anomalies �FAGA � � � � � � � � ���

A�� Data Description � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���A�� Upward Continuation � � � � � � � � � � � � � � � � � � � � � � � � � ���

B� Preprocessing of the Magnetic Anomalies �MA � � � � � � � � � � � � � � ���

B�� Data Description � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���B�� Upward Continuation � � � � � � � � � � � � � � � � � � � � � � � � � ���B�� Reduction to Pole � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

C� Altimetry Enhanced FreeAir Gravity Anomalies � � � � � � � � � � � � � ���

C�� Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��C�� AltimeterImplied FreeAir Gravity Anomalies � � � � � � � � � � � � ���C�� A case study� The Barents Sea � � � � � � � � � � � � � � � � � � � � ���

C�� Summary � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

D� Tabular control data for Greenland � � � � � � � � � � � � � � � � � � � � � ���

E� Seismic Control Depths for Greenland � � � � � � � � � � � � � � � � � � � ���

E�� Introduction � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���E�� Comparison of Interpolated Moho Model and Seismic Depths � � � ���E�� Signi�cance of the quality of the predictions � � � � � � � � � � � � � ���E�� Summary � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

F� GLQ Geophysical Relationships � � � � � � � � � � � � � � � � � � � � � � � ��

G� Kinematic Model Summary of Events for Greenland � � � � � � � � � � � � ���

H� Acronym De�nitions � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

Bibliography � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

x

Page 12: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

LIST OF TABLES

Table Page

��� Comparison of bathymetric models for the Barents Sea using enhancedgravity GGMderived� standard gravity GGMderived �Andersen andKnudsen� ������ EOTOPO�U DEM� JGP��E DEM� and gridded control data� The control data consist of ��� points� while all models arestatistically compared to ��� check points� The �eld area ranges from������ to ������N and ������ to �����E� Statistics of the comparison include overall correlation �CC � the RMS of the di�erences� andthe mean di�erence� Lastly� an improvement factor with respect to theGGMderived bathymetry is given to demonstrate the magnitude ofthe improvement � RMSi�RMSGGM

RMSi� � � � � � � � � � � � � � � � � � � � ��

��� Comparison of bathymetric models for Greenland� The GGMderivedpredictions from the FAGA of Andersen and Knudsen ������ significantly surpassed all other bathymetric data sets when compared at

the ������� bathymetric check points� Because the FAGA only extendto ���N� the bathymetric predictions likewise only extend to ���N�Bathymetry resulting from the gridding of the ������� control pointsproduced good agreement with the check points but had an RMS difference that was ��� greater than the GGM predictions� The bathymetric predictions of Smith and Sandwell ������ are also given belowbut are compared only for check points south of ���N� Statistics of thecomparison include the coe�cient of correlation �CC � the RMS di�erence� and the mean di�erence� The percent �� improvement of theGGMderived bathymetry relative to the other bathymetric models isalso given as de�ned by � improvement � RMSi�RMSGGM

RMSix ���� � � � ��

xi

Page 13: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

��� Statistical Comparison of Ice Surface Elevation Data� Surface elevation data from the BPRC pro�les and interpolated values from Ekholm����� and NIMA DEMs are compared� All three pro�les are very similar with the interpolated Ekholm and NIMA grids being almost identical� Columns listed compare the pairs of pro�les� giving their correlation� the RMS of their di�erence� the mean of their di�erence� and themaximum value of the di�erence� All heights are orthometric �abovethe geoid � EGM�determined geoid undulations were removed fromthe ellipsoidal heights of the BPRC data� � � � � � � � � � � � � � � � � �

��� Summary of Greenland area crustal rock types and associated densityand volume magnetic susceptibility variations� � � � � � � � � � � � � � ���

��� Chart of generalized correlative crustal lithologies for Greenland� Anomaly magnitudes are relative to source physical property contrasts withgneisses� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

��� Inclination and declination pairs for reductiontopole� Values wereselected at points inset from the four corners and at the middle ofFigure ����� The reductiontopole was performed for each of thesepairs� and the procedure was assumed valid for approximately ��� kmaround each point� Values for points more than ��� km from thesepoints represented a weighted average of the values generated by allthe below pairs� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

B�� Inclination and declination pairs for reductiontopole� Values wereselected at points inset from the four corners and at the middle ofFigure B��� The reductiontopole is performed for each of these pairs�and the procedure is assumed valid for approximately ��� km aroundeach point� Values for points more than ��� km from these pointsrepresent a weighted average of the values generated by all the belowpairs� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

D�� BPRCderived location� gravity and ice thickness data in Greenland�Freeair gravity anomalies were derived by di�erencing the absolutegravity value with a normal gravity value freeair corrected to the observation elevation� A value of ����� in the radar ice thickness columnindicates a null value for that location� � � � � � � � � � � � � � � � � � ���

D�� Height� thickness and FAGA data interpolated from �D data for theBPRC data points on Greenland� � � � � � � � � � � � � � � � � � � � � ���

xii

Page 14: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

E�� Seismic control depths for the Greenland Moho model derived fromseismic survey literature� Listed are the positional coordinates� seismicdepth in kilometers� depth in kilometers interpolated from the Mohomodel� the seismic literature source� and the literature sources� seismicsurvey line and point names associated with the seismic depth values�A dash � in the interpolated Moho depth column indicates that thesource was outside of the region covered by the Moho grid� The ��stations that have depth values are numerically identi�ed below �stat�id� no� and map to the values that are analyzed in Table E��� � � � � ���

E�� Statistical Comparison of the Predicted and Seismic Moho Depths�Columns indicate station identi�cation� predicted Moho depth� seismically determined Moho depth� the numeric di�erence of the two depths�predicted minus seismic � and percentile di�erence �the raw di�erencedivided by the seismic depth � Depths where the numeric di�erencewas greater than ��� �y and depths where the percentile di�erence�num� di�� � seismic depth was greater than ��� �z are noted�Statistical comparisons excluding no depths� y depths� z depths� andboth y and z depths are given at the bottom of the table� � � � � � � � ���

G�� Chronological summary of tectonic events for the Greenland area� Theconventional name of the associated magnetic isochron is given underthe �anomaly� heading� Geologic age and associated time in millionsof years is given under the �age �Ma � heading� Subsequent columnsfor the listed regions summarize the geologic events of their evolution� ���

xiii

Page 15: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

LIST OF FIGURES

Figure Page

��� The GravityGeologicMethod �adapted from Figure � of Nagarajan������ � The top panel shows the gravity �eld components� observedanomalies �solid� gOBS�i � regional anomalies �dashdot� gREG�i �and bedrocktopography anomalies �dashed� gBRT �i � The bottompanel shows the original model �dashed and that predicted by GGM�solid � Asterisks �� mark the �� control measurement locations usedin making predictions at all �� points� A density contrast of ���gm�cm� was assumed between the rock and sediment layer� � � � � � � �

��� Standard FAGA from Andersen and Knudsen ������ �gOBS for theBarents Sea in a Lambert EqualArea Azimuthal Projection centeredon �� E at sea level� The test area is delineated by the red box� � � � ��

��� FAGA from Andersen and Knudsen ������ with enhanced high fre

quency components for the study area �gOBS outlined in Figure ����These data for the Barents Sea are shown in a Lambert EqualAreaAzimuthal Projection centered on ���E at sea level� � � � � � � � � � � ��

��� Barents Sea bathymetric control and check points within the studyarea outlined in Figure ���� Data were digitized by NRL from USSRcharts and Norwegian Polar Institute charts� The blue outline boxdelineates the �eld area� grey circles denote control bathymetry� andred triangles denote check points� � � � � � � � � � � � � � � � � � � � � ��

xiv

Page 16: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

��� Tradeo� diagram for choosing an e�ective density contrast in GGMpredictions of bathymetry for the Barents Sea study area� Solid anddashed curves give the correlation coe�cient and RMS di�erence� respectively� between the predictions and the check points as a functionof varying density contrasts� The tradeo� diagram indicates that little improvement may be expected from predictions based on densitycontrasts greater than ���� gm�cm�� Lower values up to about ����gm�cm� may provide higher resolution predictions with marginal decrease in statistical performance� A geologically more reasonable density contrast of ��� gm�cm� �intersection of the red dotted lines doesnot perform well� matching only ��� of the check points and havingan RMS di�erence of ��� m� � � � � � � � � � � � � � � � � � � � � � � � ��

�� Regional gravity anomalies �gREG determined from control points inthe bathymetry for the Barents Sea study area� The data are shown ina Lambert EqualArea Azimuthal Projection centered on ���E at sealevel� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

��� Residual gravity anomalies �gBRT for the Barents Sea study areashown in a Lambert EqualArea Azimuthal Projection centered on���E at sea level� These anomalies were the result of removing theregional gravity e�ects �gREG shown in Figure �� from the enhancedFAGA �gOBS shown in Figure ���� � � � � � � � � � � � � � � � � � � � ��

��� GGM bathymetric predictions from enhanced FAGA for the BarentsSea� The data are shown in a Lambert EqualArea Azimuthal Projection centered on ���E at sea level� � � � � � � � � � � � � � � � � � � � � ��

��� GGM bathymetric predictions from standard FAGA of Andersen andKnudsen ������ for the Barents Sea study area� The data are shown ina Lambert EqualArea Azimuthal Projection centered on ���E at sealevel� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

���� Bathymetry estimated from gridding the ��� control points for theBarents Sea study area� These data are shown in a Lambert EqualArea Azimuthal Projection centered on ���E at sea level� � � � � � � � ��

���� EOTOPO�U bathymetry for the Barents Sea in a Lambert EqualAreaAzimuthal Projection centered on ���E at sea level� � � � � � � � � � � ��

xv

Page 17: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� JGP��E bathymetry for the Barents Sea in a Lambert EqualAreaAzimuthal Projection centered on ���E at sea level� � � � � � � � � � � ��

���� Bathymetric di�erences obtained by subtracting the standard FAGApredictions �Figure ��� from the enhanced FAGA �Figure ��� for theBarents Sea study area� Data are shown in a Lambert EqualAreaAzimuthal Projection centered on ���E at sea level� � � � � � � � � � � �

���� Greenland bathymetric control and check values shown in a LambertEqualArea Azimuthal Projection centered on ���W� Data coverageincludes ������� points mainly in the eastern areas from the NGDCand ������� points mainly in the western areas from GSC �darker grey �The study area around Greenland is also delineated� � � � � � � � � � � ��

���� Regional bathymetric gravity e�ects �gREG for Greenland from combined NGDC and GSC bathymetric control data� The gravity e�ectsare shown in a Lambert EqualArea Azimuthal Projection centered on��� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

��� Standard FAGA from Andersen and Knudsen ������ �gOBS for Greenland in a Lambert EqualArea Azimuthal Projection centered on ���

W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

���� Residual bathymetric gravity e�ects �gRES for Greenland determinedby removing the regional �gREG from the standard FAGA �gOBS �The gravity e�ects are shown in a Lambert EqualArea AzimuthalProjection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � ��

���� GGMpredicted bathymetry for Greenland in a Lambert EqualAreaAzimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � ��

���� EOTOPO�U bathymetry for Greenland in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � � �

���� JGP��E bathymetry for Greenland in a Lambert EqualArea AzimuthalProjection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � ��

���� Greenland bathymetry fromminimum curvature gridding of the �������control points shown in a Lambert EqualArea Azimuthal Projectioncentered on ��� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

xvi

Page 18: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Greenland bathymetry from Smith and Sandwell ������ in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

��� Greenland surface elevations including rock and ice surfaces above sealevel in a Lambert EqualArea Azimuthal Projection centered on ��� W� ��

��� Greenland subglacial elevations inferred for the bedrock relative tomean sea level in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

��� Ice thickness model �top minus bottom of icesheet in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � �

��� Ocean thickness model for Greenland �reference is MSL in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

��� Bedrock surface model for Greenland �reference is MSL in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

�� Greenland bathymetry from JGP��E data set in a Lambert EqualAreaAzimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � �

��� Original FreeAir Gravity Anomalies �FAGA fromNIMA at the Earth�ssurface in a Lambert EqualArea Azimuthal Projection centered on ���

W� Data are registered to variable surface elevations that are given inFigure ���� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

��� Observation errors provided with the NIMA FAGA in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � �

��� Orthometric heights associated with the NIMA�s FAGA above meansea level in a Lambert EqualArea Azimuthal Projection centered on��� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

���� NIMA FAGA upward continued to Z���km elevation in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � �

���� Greenland gravity and ice thickness measurement locations at about a�� km interval gathered by Byrd Polar Research Center in ����� ���������� and ��� �Roman et al�� ������ Data are shown in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � �

xvii

Page 19: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Ice Surface Elevation Comparisons� a� Orthometric heights given include� BPRC elevations �thickened black � Ekholm elevations �dashedred � and NIMA elevations �dashed green � b� Elevation di�erencesgiven include� BPRCEkholm �thickened black � BPRCNIMA �dashedred � and EkholmNIMA �dashed green � Table ��� gives statistics ofthis comparison� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

���� Comparison of ice thickness pro�les showing data interpolated fromFigure ��� �thickred line and the radar ice thickness measurements�two thinblack line segments obtained by Gogineni and made available by Sohn and Csath�o ������� � � � � � � � � � � � � � � � � � � � � � ��

���� Comparison of FAGA pro�les derived from BPRC �eld observations�thinblack line and interpolated o� the original NIMA FAGA grid�thickred � Note the large disagreement at about the ���� km pointalong the pro�le and that the NIMA FAGA are systematic lower inmagnitude� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

���� Crustal crosssection showing the ice top and bottom with FAGA differences� Top line shows the di�erence between BPRCderived FAGAand NIMAderived FAGA data� Middle line depicts the ice surfaceinterpolated from Figure ���� The bottom line depicts the ice bottominterpolated from Figure ���� Note the large di�erence between theFAGA data sets that originates above a thinning of the ice sheet atthe ���� km point along the pro�le� � � � � � � � � � � � � � � � � � � � ��

��� Initial assumed crustal densities for Greenland used in this study in aLambert EqualArea Azimuthal Projection centered on ��� W� Rangesof density values for prisms of oceanic and continental rock are givenfor the upper crust �u�c� � average for each prism �ave� � and lower crust�l�c� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

���� Gravity e�ect of the ice model at ��km above MSL in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

���� Gravity e�ect of the ocean model at ��km above MSL in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

���� Gravity e�ect of the rock model at ��km above MSL in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

xviii

Page 20: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Terrain gravity e�ect �TGE from the summation of the gravity e�ectsof the ice� ocean and rock models at ��km above MSL in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ��

���� Total of three terrain models �the Earth�s surface in a Lambert EqualArea Azimuthal Projection centered on ��� W �reference is MSL � Elevations on Greenland are from Ekholm ����� �Figure ��� � ocean areasare zero �MSL � and all other terrestrial areas are from JGP��E �Figure �� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

���� Terrain Correlated FAGA �TCFAGA in a Lambert EqualArea Azimuthal Projection centered on ��� W at ��km above MSL� Thesedata represent that component of the reference FAGA that show thehighest positive and negative correlations with the TGE� � � � � � � � ��

���� Terrain Decorrelated FAGA �TDFAGA in a Lambert EqualArea Azimuthal Projection centered on ��� W at ��km above MSL� Thesedata represent that component of the reference FAGA that is left overafter removal of TCFAGA� � � � � � � � � � � � � � � � � � � � � � � � � �

���� Annihilating CTGE �ACTGE in a Lambert EqualArea AzimuthalProjection centered on ��� W at ��km above MSL� These data represent the gravity e�ect of the Moho undulation based upon an assumeddensity contrast� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

���� Moho depth model for the Greenland study area in a Lambert EqualArea Azimuthal Projection centered on ��� W� This model was calculated iteratively using the initial assumed densities for the lower crustshown in Figure ��� and the simpli�ed relationship given in Equation ����� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

��� Adjusted density model for the Greenland �eld area in a LambertEqualArea Azimuthal Projection centered on ��� W� These valueswere determined by �xing the Moho depths to those determined forFigure ���� and modifying the initial assumed densities �Figure ��� to iteratively solve Equation ����� � � � � � � � � � � � � � � � � � � � � ��

���� Root Gravity E�ect for Greenland derived from the Moho depth model�Figure ���� and adjusted lower crustal density contrasts �Figure ��� in a Lambert EqualArea Azimuthal Projection centered on ��� W� � ��

xix

Page 21: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Unmodeled residual ACTGE for Greenland in a Lambert EqualAreaAzimuthal Projection centered on ���W� These data represent the difference between the ACTGE �Figure ���� and the RGE �Figure ���� calculated from the Moho depths and adjusted densities� Remainingunmodeled e�ects are dominantly located along the southern and eastern edges of the �eld area that probably represent edge e�ects� � � � � ��

���� Generalized depiction of Moho depth pro�le generation by Equation����� The ice sheet is shown on top of the subglacial bedrock� whichsits on top of �� km of crust� Beneath this extends the crustal root intothe mantle� The masses in the top two boxes �for ice and subglacialbedrock must equal the mass in the bottom box �in the root to agreewith the assumed Airy method of isostatic compensation� � � � � � � � �

���� Crosssection of interpolated and pro�le Moho depth data� a� Valuesfrom the Moho depth grid �Figure ���� were interpolated to the locations of a Moho depth pro�le �thickred � The pro�le was determinedfrom the ice surface and subglacial models as shown in Figure ���� andgiven in Equation ���� assuming an Airy method of compensation� b�Plot of the di�erence �interpolated pro�le � � � � � � � � � � � � � � � ��

���� Greenland crustal thickness model derived from di�erencing the rockand Moho depth models in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � � � � ���

��� Gravityderived Moho model for the Greenland area in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ���

��� Geographic map detailing Greenland geologic provinces and the locations for the crustal structure in the southwestern Greenland margindetermined in Chapter �� Mapped boundaries in the coastal regions ofGreenland were adapted from Escher and Pulvertaft ������� � � � � � � ���

��� Terrain decorrelated FAGA �TDFAGA in a Lambert EqualArea Azimuthal Projection centered on ���W at �� km above MSL� These datarepresent that component of the reference FAGA that is left over afterremoval of TCFAGA� � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

xx

Page 22: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

��� Percent coherence versus cumulative harmonics� a� The terrain decorrelated portion of NIMA�s FAGA �TDFAGA were correlated withFAGA generated by the cumulative EGM� harmonics from degree�� to � �solid and from degree � to �� �dashed � Note the slopebreak at degree ���� Harmonics lower than ��� generate most of theagreement between the data sets� The e�ects modeled by harmonicsover degree ��� are assumed to be representative of shallower sourcesin the lithosphere� having shorter wavelengths and less power� b� Firsthorizontal derivative of the pro�le for degrees �� to �� The thickenedvertical line highlights the slope break at degree ���� � � � � � � � � � ���

��� EGM�FAGA derived from degrees ��� to �� EGM� coe�cients ina Lambert EqualArea Azimuthal Projection centered on ��� W� � � � ��

�� EGM�FAGA data derived from degrees � to ��� EGM� coe�cientsin a Lambert EqualArea Azimuthal Projection centered on ��� W�Labels refer to geologic features discussed in the text� � � � � � � � � � ���

��� Intracrustal components of the terraindecorrelated gravity anomalies�ICTDFAGA in a Lambert EqualArea Azimuthal Projection centered on ��� W� These data represent that portion of TDFAGA thatmay be more related to lithospheric sources as determined by their coherency with EGM�FAGA data derived from degrees ����� �Figure ��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

��� Residual TDFAGA �TDFAGA ICTDFAGA in a Lambert EqualArea Azimuthal Projection centered on ��� W� These data may bemore related to the lower harmonics �� to ��� of EGM� but alsocontain high frequency signal that may be noiserelated� � � � � � � � ���

��� Residual TDFAGA highpass �ltered at ��� km in a Lambert EqualArea Azimuthal Projection centered on ��� W� These data have littlepower and are oriented along the grid intervals suggesting that theymay be more related to grid processing� � � � � � � � � � � � � � � � � � ���

���� Residual TDFAGA lowpass �ltered at ��� km �MCTDFAGA in aLambert EqualArea Azimuthal Projection centered on ��� W� Thesedata represent the long wavelength component of the NIMAderivedFAGA that are least correlated with surfacial features and most correlated with deeper sources� � � � � � � � � � � � � � � � � � � � � � � � � ���

xxi

Page 23: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Average IGRF declination �solid line � inclination �dashed line � andintensity �grey scale in a Lambert EqualArea Azimuthal Projectioncentered on ��� W� Data were obtained and averaged from the � degree IGRF models for epochs ��������� Blue boxes depict the �veregions that were adjusted based on the inclinationdeclination pairs�Table ��� determined at the red triangle in the center of each box� � ���

���� RTPMA at �� km elevation in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � � � ���

���� Greenland crustal thickness model derived from di�erencing a bedrockDEM �Chapter �� and Moho depth model �Figure ��� in a LambertEqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � ���

���� GLQderived gravity e�ect of the crustal model of Figure ���� in aLambert EqualArea Azimuthal Projection centered on ��� W� � � � � ���

���� First vertical derivative of crustal gravity e�ect FVD�CGE in a Lambert EqualArea Azimuthal Projection centered on ��� W� These datawere generated by applying a standard vertical derivative operator tothe data in Figure ����� � � � � � � � � � � � � � � � � � � � � � � � � � � ���

��� Crustal thickness MA that are the RTPMA components which arecorrelative with FVD�CGE shown in Figure ����� Data are shownin a Lambert EqualArea Azimuthal Projection centered on ��� W�These data were derived from wavenumber components of RTPMAin Figure ���� that correlated higher than ���� with the wavenumbercomponents of FVD CGE � � � � � � � � � � � � � � � � � � � � � � � � ���

���� Intracrustal RTPMA �ICRTPMA in a Lambert EqualArea AzimuthalProjection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � ���

���� First vertical derivative terraindecorrelated �FVD�ICTDFAGA FAGA for Greenland in a Lambert EqualArea Azimuthal Projection centered on ���W� FVD�ICTDFAGA data are derived by taking the �rstradial derivative of the data shown in Figure ���� � � � � � � � � � � � ���

���� Normalized ICRTPMA in a Lambert EqualArea Azimuthal Projection centered on ��� W� Multiplying the normalized amplitudes by NFrecovers the original amplitudes of the magnetic anomalies in Figure �������

xxii

Page 24: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Normalized FVD�ICTDFAGA in a Lambert EqualArea AzimuthalProjection centered on ��� W� Multiplying the normalized amplitudesby NF recovers the original amplitudes of the magnetic anomalies inFigure ���� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

���� Summed local favorability indices �SLFI from adding Figures ���� and���� in a Lambert EqualArea Azimuthal Projection centered on ���W� ���

���� Positively correlated gravity and magnetic features for the Greenlandstudy area in a Lambert EqualArea Azimuthal Projection centeredon ��� W� Noted features are discussed in the text� a The strongerpeaktopeak correlations given by SLFI�ASD���� � b The strongertroughtotrough correlations given by SLFI�ASD� � � � � � � � � � � ���

���� Di�erenced local favorability indices �DLFI from subtracting Figures���� and ���� in a Lambert EqualArea Azimuthal Projection centeredon ��� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

���� Negatively correlated gravity and magnetic features for the Greenlandstudy area in a Lambert EqualArea Azimuthal Projection centered on���W� Noted features are discussed in the text� a The stronger FAGApeaktoMA trough correlations given by DLFI�ASD���� � b Thestronger FAGA troughtoMA peak correlations given by DLFI�ASD� ���

���� Gravityquotient local favorability indices �GQLFI greater than � ina Lambert EqualArea Azimuthal Projection centered on ���W� GQLFI were generated by dividing Figure ���� by the absolute valueof Figure ����� Noted Features are discussed in the text� a Gravitymaxima with null correlated magnetic anomalies� b Gravity minimawith null correlated magnetic anomalies� � � � � � � � � � � � � � � � � ���

��� Magneticquotient local favorability indices �MQLFI greater than �in a Lambert EqualArea Azimuthal Projection centered on ��� W�MQLFI were generated by dividing Figure ���� by the absolute valueof Figure ����� Noted Features are discussed in the text� a Magneticmaxima with null correlated gravity anomalies� b Magnetic minimawith null correlated gravity anomalies� � � � � � � � � � � � � � � � � � ���

xxiii

Page 25: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

���� Composite geology and structural map of Greenland in a LambertEqualArea Azimuthal Projection centered on ��� W� These featureswere compiled from Figures ����� ����� ����� ���� and from the structure implied in Figure ���� � � � � � � � � � � � � � � � � � � � � � � � � ���

��� Green box delineates the study area ����� ����N and �� ���W forsouthwestern Greenland in a Lambert EqualArea Azimuthal Projection centered on ��� W� � � � � � � � � � � � � � � � � � � � � � � � � � ��

��� Moho model for the Greenland area estimated from the spectral correlation modeling of FAGA and the terrain gravity e�ects for integrated mass variations de�ned by the bedrock surface� and water andice thicknesses� The area is shown in a Lambert EqualArea AzimuthalProjection centered on ���W� The green box delineates the study area�Note the deeper roots subparallel to the coast an extending northwardsthrough the Davis Strait region� � � � � � � � � � � � � � � � � � � � � � ���

��� Geologic features for southwestern Greenland in a Lambert EqualAreaAzimuthal Projection centered on �����W and at ���� N� Continental�rifted� transitional� and oceanic zones from Figure ��� are shown alongwith spreading centers and transforms faults� These features weredetermined partially from the Moho predictions �Figure ��� and bythe distribution of correlative freeair gravity and magnetic anomalies�Chapter ��� The numbers and arrows extending away from the extinctspreading ridge generally mark the positions of magnetic isochrons thatare subparallel to the spreading center� Transitional and continentalcrust are both thickened� while oceanic and riftedcontinental crust areboth thin� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

��� Moho model for southwestern Greenland in a Lambert EqualArea Azimuthal Projection centered on ����� W and at ���� N� White triangles indicate the seismic survey points of the R� pro�le from Chianand Louden ������� White lines delineate the boundaries between theoceanic� transitional� rifted continental� and continental crust� � � � � ���

xxiv

Page 26: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

��� Comparison between the R� pro�le of Chian and Louden ������ �diamonds and a coincident Moho pro�le interpolated from Figure ����asterisks across southwestern coast of Greenland� Upper solid lineshows the ocean bottom� Letters refer to seismic stations from Chianand Louden ������� The high velocity zone that Chian and Louden������ suggested may re�ect serpentinized mantle material is locatednear point D� The steep rise just to the right of D re�ects the anomalously shallow Moho root� This feature is also re�ected in the interpolated pro�le but not as sharply de�ned� � � � � � � � � � � � � � � � � � ���

�� Rotational Rift Model for Laurasia � � � � � � � � � � � � � � � � � � � ���

A�� Original NIMA FreeAir Gravity Anomalies �FAGA at the Earth�ssurface shown in a Lambert EqualArea Azimuthal Projection centeredon �� W� Data are registered to variable surface elevations that aregiven in Figure A��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

A�� Observation errors provided with the NIMA FAGA shown in a LambertEqualArea Azimuthal Projection centered on �� W� � � � � � � � � � ���

A�� Orthometric heights associated with the NIMA FAGA above mean sealevel shown in a Lambert EqualArea Azimuthal Projection centeredon �� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

A�� NIMA FAGA upward continued to Z��� km elevation shown in aLambert EqualArea Azimuthal Projection centered on �� W� � � � � ���

B�� Magnetic anomaly data provided by GSC from Open File ����b forGreenland shown in a Lambert EqualArea Azimuthal Projection centered on �� W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

B�� Corrected magnetic anomaly data for Greenland shown in a LambertEqualArea Azimuthal Projection centered on �� W� Data are reducedtopole to clarify the location of magnetic sources and are upward continued to �� km elevation for comparison with gravity data� � � � � � ���

B�� Average IGRF derived Declination �solid line � Inclination �dashedline � and Intensity �grey scale shown in a Lambert EqualArea Azimuthal Projection centered on ��W� Data were obtained and averagedfrom the � degree IGRF models for epochs ��������� The graticuleis left out for clarity� � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

xxv

Page 27: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

B�� MA reducedtopole and upwardcontinued to �� km elevation shownin a Lambert EqualArea Azimuthal Projection centered on �� W� � � ���

C�� Altimetry analysis �owchart� Various processing steps are necessary togenerate enhanced FAGA� Ultimately� the long wavelength componentof the reference FAGA data is combined with the short wavelengthdata from the altimetry� � � � � � � � � � � � � � � � � � � � � � � � � � ���

C�� ��day ERS� altimetry for the Barents Sea� Ascending �top anddescending �bottom altimeter coverage can be seen to be su�cientfor the test area �red box and is shown in a Lambert EqualAreaAzimuthal Projection centered on �� E at sea level� � � � � � � � � � � ���

C�� Andersen and Knudsen FAGA for the Barents Sea shown in a LambertEqualArea Azimuthal Projection centered on �� E at sea level� Thetest area is shown with the red box� � � � � � � � � � � � � � � � � � � � ���

C�� Andersen and Knudsen FAGAderived geoid undulations for the Barents Sea shown in a Lambert EqualArea Azimuthal Projection centered on �� E at sea level� The test area is shown with the red box� � ���

C�� Comparison of altimeter �solid and reference geoid �dashed derivedpro�les �top and the di�erence between them �bottom � Di�erencesare primarily due to long wavelength ���� km errors� spikes� andhigher frequency features that represent both noise and crustal signals�Pro�le shown is ERS� ��day mission track ���� and is sampledroughly every � meters� � � � � � � � � � � � � � � � � � � � � � � � � ���

C� Power spectrum and tradeo� diagram showing cuto� correlation �CCk selection� Increase in overall track pair correlations and improvementin signal to noise ratio are nearly linear �upperright and upperleft diagrams and the drop o� in power is nearly linear except at the highestvalues �bottomleft diagram � The slope of the power drop o� �bottomright diagram more clearly shows that the maximal in�ection point isat about a CC of ��� to ���� For this reason� ��� was chosen for theCCk � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

C�� Residual geoid undulation for Barents Sea test area shown in a LambertEqualArea Azimuthal Projection centered on �� E at sea level� � � � ���

xxvi

Page 28: Daniel R Roman - U.S. National Geodetic Survey · 2017-05-16 · Roman DR B Csatho KC Jezek RH Thomas WB Krabill RRB v on F rese and R F orsb erg A Comparison of Geoid Undulation

C�� Residual FAGA for Barents Sea test area shown in a Lambert EqualArea Azimuthal Projection centered on �� E at sea level� � � � � � � � ���

C�� Total enchanced FAGA for the Barents Sea shown in a Lambert EqualArea Azimuthal Projection centered on �� E at sea level� Comparisonof these features to those in the red box in Figure C�� show that themajor features are still present� � � � � � � � � � � � � � � � � � � � � � ���

E�� Comparison of � seismic pro�les around Greenland� Values interpolated from the �nal Moho model �Figure E�� are shown as solid whitepro�les with the seismic estimates given in the red dashed pro�les�The correlation coe�cient �CC � standard deviation �St� Dev� � andmean di�erence �mean are given at the top of each pro�le� a� Chianand Louden ������ b� Chian and Louden ������ c� DahlJensen et al������� d� Fechner and Jokat ����� e� Gregersen et al� ������ f� Jackson and Reid ������ g� Reid and Jackson ������� Asterisks �� markpoints �agged for removal� � � � � � � � � � � � � � � � � � � � � � � � � ��

E�� Moho model for the Greenland �eld area shown in a Lambert EqualArea Azimuthal Projection centered on �� W� � � � � � � � � � � � � � ���

E�� Location of Greenland control Moho depths listed in Table E�� andshown in a Lambert EqualArea Azimuthal Projection centered on ��W� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ���

F�� Depiction of GLQ Geophysical Relationships� a� The initial location ofthe prism� its nodes �black dots � and the observation surface� b� The�nal location of the prism and its nodes� This simpli�ed case depictsthe e�ects on the change of nodal location only for a point immediatelyabove the node in the observation grid� If h� � h� is su�ciently small�then l� � l� and the relationship between the prism and the observationpoints may be linearly modeled� � � � � � � � � � � � � � � � � � � � � � ���

xxvii