User’s manual (January 2012)

by María José Ramón and Emilio L. Pueyo

 The VPD software manual: Index

1) Software code and system requirements

- Goals and scope
- Installation, software upgrades, bugs report and acknowledgments

2) Input formats

3) Main menu
- Visualization menu
- Sample information

4) Sample analysis: principal component analysis (PCA)

5) Site processing tools: stacking routine (SR)

6) Site processing tools: linearity spectrum analysis (LSA)

7) Site processing tools: virtual directions (VD)

- Virtual directions window
- Cartesian window
- Filtering window

Cited references

Related softwares

Appendix 1: Input formats

 1) Software code and system requirements

Goals and Scope

The program called Virtual Paleomagnetic Directions (VPD) has been developed in Java to
easily visualize the demagnetization data and to compare the results obtained with virtual
directions (VD) and other classic methods: eye-ball inspection and fitting with principal
components analysis (PCA), linearity spectrum analysis (LSA) and the stacking routine (SR).
Java allows running the program in any operative system (Linux, Mac, Windows). Principal
component analysis (PCA) [Kirschvink, 1980] and part of the visualization menu have been
inspired on the Paldir code developed by the University of Utrecht. The stacking algorithm
[Scheepers and Zijderveld, 1992] is based on software Gamsstack [Pueyo et al., 2003] from
Gams paleomagnetic laboratory (Montanuniversität Leoben, Austria) whereas LSA is detailed
by Schmidt [1982]. The virtual directions method is fully described in Pueyo [2000], Ramón
and Pueyo [2008] and Ramón et al. [2012].

Installation

VPD is an executable file that runs within a Java Virtual Machine. Thus, the software needs
no installation, only the Java that can be easily download and installed from the official site
http://www.java.com/en/download. The Java SE 6 version is required. Software upgrades are
available from http://www.igme.es/internet/zaragoza/aplicaInfor.htm. Bugs should be reported
to [email protected] and [email protected]

Acknowledgments: Research funding during the development of this software was provided by
the Pmag3Drest projects (CGL-2006-2289-BTE, CGL2009-14214) from the Spanish Ministry
of Science and the 3DR3 & GeoPyrDatabases projects (PI165/09 & CTPP01/07) from the
Aragonian Government. MJR held a contract (PTA2007-0282).

2) Input format

Data are opened or appended from the main menu in the program (upper right corner, Fig. 1).
The VPD software was designed to process large paleomagnetic datasets starting from the
demagnetization information at the sample scale. Several formats have been considered and
implemented including the Barcelona, Burgos, Roma, Tübingen, Utrech laboratories as well
as the raw demagnetization format from 2G magnetometers (see appendix 1). Besides, raw
data from most common u-channels were also taken into account.

Others formats could be implemented upon request ([email protected];

[email protected])
 3) Main menu

This main menu contains two areas; the upper part (visualization menu) displays the graphic
demagnetization information at once. The lower part has, in turn, three areas; the core info
box, the PCA box, and the site info box with buttons for other approaches (Fig. 1).

Figure 1. Main menu (sample level). A) Stereographic projection. B) Orthogonal projection. C) Intensity decay
graphic. D) Visualization mode options. Sample) Core/specimen/sample information. PCA) PCA menu. Others)
Other methods: LSA, SR, VD. E) Initial Site Info-Box (Sample selector). F) Final Site Info-Box.

Figure 2. Visualization menu (sample level). A) Stereographic projection, where the pointer offers the numerical
information. B) Present day field orientation, drilling direction and the bedding plane (S0). C) Orthogonal and
decay diagram (intensity in 10-6 A/m). D) Reference system choices.
Visualization menu

In the same window, the stereographic projection (equal area) together with the orthogonal
diagram and the intensity decay along the demagnetization level can be looked at once (Fig.
2). The reference system, geographic (NO TC) or restored (TC, bedding corrected), as well as
the orientation of the orthogonal diagram (up/W to up/N) can be also chosen. Moreover, the
present geomagnetic field and the drilling direction (core axis) can be visualized in both
geographic and paleographic coordinate systems to help evaluating the effect of viscous and
spurious laboratory components. Finally, the bedding plane (S0) and additional plane (S1) can
be also visualized.

Figure 3. Sample box. A) Core/specimen/sample information: name, drilling orientation and bedding plane. B)
Demagnetization steps. C) Paleomagnetic vector information for each demagnetization step: declination,
inclination, intensity and error. D) Step selector.

Sample information

The sample box (bottom left Fig. 1) displays the demagnetization information. The specimen
info is firstly display (Fig. 3A). The drilling direction is measured as the azimuth and (real)
plunge of the cylinder axis. The bedding plane is displayed as strike (following the right hand
rule) and the dip (Fig. 4).

Selecting a given demagnetization step from the box (Fig. 3B) will provide the detailed
demagnetization information (Fig. 3C): declination, inclination, intensity and laboratory error
(if available).

Delete button (and “d” key) will remove the selected demagnetization step for further analysis
at specimen or site scale. Insert button (and “i” key) will select the step for principal
component analysis (Fig. 3D).
To change the sample, go to the sample selector (Fig. 1E) or use the “c” key.

Figure 4. Orientation system used in the VPD software

4) Sample analysis: principal component analysis (PCA)

The manual selection of demagnetization steps (insert button and key) allow estimating by
principal component analysis (PCA) the directions and circles at the sample level, either
including or not the origin (resultant or difference). Erroneous steps can be deleted from
selected specimen. Numerical and graphical information of the PCA estimation is displayed
and can be exported.

“Export Graphics” button exports the sample in the EPS format that includes all directions
and circles calculated in that sample (Fig. 5). Numerical information from the entire site can
be exported at once using the “Export Directions” button. The resultant file is a text file (*.txt)
with VD format and can be readily transformed to many post-processing software (i.e.
Stereonet, SuperIAPD, etc…). “Clear” button will remove the directions of the selected
sample information and “Clear All” will removed all direction calculated in that site.

Exporting Directions - VD format

NO TC
Sample R/D D/C stepMIN stepMAX #steps DEC INC INT MAD
FA4-2a R D 360.0 475.0 7 167.0 59.0 177.0 10.0
FA4-2a D D 360.0 475.0 7 248.0 6.0 31.7 38.4
FA4-4a R D 320.0 500.0 9 156.0 48.0 137.1 8.4
FA4-4a D D 320.0 500.0 9 167.0 45.0 85.7 30.9
FA4-4a R C 320.0 500.0 9 274.0 23.0 0.0 35.5
NO TC/TC: Directions calculated in situ or after bedding correction. For all virtual directions the information
is: Sample name, Resultant/Difference (R/D), Direction/Circle (D/C), Minimum step, Maximum step, Number of
steps to calculate the VD, Declination, Inclination, Intensity, Error (MAD).
 Figure 5. Exported sample demagnetization information

5) Site processing tools: stacking routine (SR)

The stacking routine (SR) is an automatic site-based calculation method [Scheepers and
Zijderveld, 1992]. In the SR approach, an individual mean is calculated from specimen
vectors for each demagnetization step for a given site and builds a stacked (“mean”)
demagnetization diagram for the site. The alfa95 confidence angle is attached to every mean.
Some improvements of the original method have been proposed by Pueyo et al. (2003).

Selected specimens are moved from the initial site (Fig. 1E) to the final site box (Fig. 1F) and
the SR calculation will be done with those cores selected in the final site information box.
Anomalous demagnetization steps (specimen level) can be deleted manually and they will not
be transfer to the final site. In the same way, anomalous samples (site level) can be excluded
in the final site. The final site can be saved with the Utrecht format (TH/AF, appendix 1).
Stacking key will generate the stacked sample in the visualization menu (Fig. 6). An error
graphic that shows the error of every calculated stacked demagnetization step is also
displayed. PCA menu allows fitting directions and circles in the stacked sample.

Figure 6. Stacking menu. A) Stacking button. B) Stacking mode. C) Step information: fisherian mean data and
number of samples considered to run that stacking step. D) Stacking error graphic vs demagnetization level.

6) Site processing tools: linearity spectrum analysis (LSA)

Linearity spectrum analysis (LSA) is also an automatic site-based calculation method

[Schmidt, 1982]. The only prerequisite is that the specimens are derived from a homogeneous
source. The method establishes the demagnetization interval seeking for the quality of the
directions, that is, those with maximum linearity (related to the maximum angular deviation
[MAD] from PCA). After selecting the mean demagnetization step for the whole site,
directions for each specimen are calculated around this step to find the best quality one.

“LSA” button of the main menu opens the LSA window (Fig. 7) and the calculus will be done
with those cores selected in the final site information box. In this window, linearity of each
individual sample is displayed as well as the mean linearity for the site (averaged of all
individuals). The mean linearity diagram helps identifying (maximun linearity values) the
middle point of the demagnetization spectra of the paleomagnetic components. Directions are
calculated using the mean step selected (Fig. 7G) and displayed in stereographic projection.
The default step value is that with maximum mean linearity. The mean direction is also
calculated and displayed with its statistical information.
Figure 7. Linearity spectrum analysis menu. A) Specimen linearity vs. demagnetization step. B) Mean linearity of
the site. C) Stereographic projection of directions obtained with LSA, directions are calculated around the step
selected in the “Mean Step” list (G). The initial value is the step with maximum mean linearity. D) Selection of
directions to visualize: before or after bedding correction (NO TC/TC), resultant or difference (RVD/DVD).
Exporting directions button. E) Statistical information of the mean direction. F) Specimen legend of colors. G)
Mean demagnetization step selector. H) k and alfa95 of all mean directions calculated for each demagnetization
step.

The possibility to change the middle point of the demagnetization interval that will be used by
to compute the direction with the maximum linearity can be especially useful for multi-
component samples. Besides, a graphic with the evolution of the statistical information (k and
alfa95) derived from the automatic calculation perform by the LSA method along the entire
demagnetization routine (Fig. 7H) also permits estimating the main paleomagnetic
components independently. Directions for every sample derived from the LSA method can be
exported in the VD format described before.

7) Site processing tools: virtual directions (VD)

Virtual directions window

Virtual directions and circles are calculated from all specimens included in the final site
information box. Virtual directions key will generate a new window (Fig. 8). Individual
stereographic projections of virtual directions and circles, resultant and difference as well as
the addition of them (R, D and R+D) are accompanied of detailed statistical information. On
one side, the directions are characterized by Fisher [1953] means. They are calculated
separately for vectors with positive inclination (lower hemisphere, plotted as filled ovals; Pos)
and for vectors with negative one (upper hemisphere plotted as empty ovals, Neg), besides, a
combined mean using the positive vectors plus the reversed negative ones (Pos+ Neg-) is also
given. On the other side, directions and circles are characterized by the orientation matrix
[Scheidegger, 1965], orientation and magnitude of the eigenvectors are calculated and
displayed, [Bingham, 1974]. It is worth noticing that we display the polarity of circle poles
although the Bingham [1974] statistics only works on the lower hemisphere.

From “File” directions and statistics can be saved. Virtual directions and circles (before or
after filtering) can be saved in a text file with VD format, equivalent to the one used to export
PCA results. Also the statistical information can be saved in a text file with the statistics
format.
Exporting statistics of mean directions – statistics format
VD STATISTICS, NO TC
Resultant Virtual Directions
Fisher mean: dec, inc, k, a95, n
Pos: 172, 53, 12.0, 0.9 , 2043
Neg: 144, -23, 4.9, 6.6 , 120
Pos+ Neg-: 174, 56, 7.1, 1.2 , 2163
Bingham: e, dec, inc
(1) 0.82, 171, 52
(2) 0.111, 288, 19
(3) 0.069, 30, 31
Difference Virtual Directions
Fisher mean: dec, inc, k, a95, n
Pos: 199, 76, 3.9, 2.0 , 1699
Neg: 125, -57, 1.9, 10.3 , 218
Pos+ Neg-: 209, 77, 3.3, 2.1 , 1917
Bingham: e, dec, inc
(1) 0.556, 191, 76
(2) 0.251, 343, 13
(3) 0.193, 74, 7
Resultant Virtual Circles
Bingham: e, dec, inc
(1) 0.528, 267, 4
(2) 0.424, 360, 40
(3) 0.048, 172, 50
Difference Virtual Circles
Bingham: e, dec, inc
(1) 0.477, 79, 8
(2) 0.35, 343, 35
(3) 0.173, 180, 54
Fisher: Type (Pos, Neg, Pos+ Neg-), Dec, Inc, k, α 95, Number of vectors used to calculate the fisher mean.
Bingham: Axis number, Eigenvalue, Declination eigenvector, Inclination eigenvector.
Figure 8. Virtual directions menu. Upper row) Resultant virtual directions (RVD), difference virtual directions
(DVD) and total virtual directions (VD). Lower row) Resultant virtual planes or circles (RVC), difference virtual
circles (DVC) and total virtual circles (VC). A) Stereographic projection showing declination and inclination
values of the clicked direction. B) Visualization options: geographic (NO TC) or restored (TC) directions,
drilling direction and present field visualization. C) Statistical information: Fisher and Bingham means. D)
Cartesian diagram button. F) Filter options.

Cartesian diagram window

“Dec/Inc Window” button generates a new window to visualize the virtual directions in a
cartesian diagram (inclination vs. declination) as suggested by Juan Cruz Larrasoaña (Fig. 9).
Together with the cartesian diagram, the classical stereographic projection is displayed. In this
diagram directions (resultant or difference) are displayed with a color scale as a function of
several parameters: specimen, number of steps that define a direction, minimum
demagnetization step, maximum demagnetization step, intensity and MAD. This diagram can
help to find clusters of directions and the reason that causes them; which is useful to proper
filter the VD dataset.
Figure 9. Cartesian diagram. A) Parameter selection to determine the color scale: specimen, number of steps,
minimum step, maximum step, intensity and MAD. B) Color scale legend. C) Declination axis from 0º to 360º or
from -180º to 180º. D) Visualization of DVD or RVD.

Filtering window

“Filter” button of virtual directions window generates a new window to proper filter VD
dataset (Fig. 10). This window is designed to filter data with several criteria: a) minimum
number of steps, b) minimum and maximum intensities, c) minimum and maximum error
(MAD), d) interval of demagnetization.

Some useful graphics are also displayed in this menu to help deciding the correct filtering
parameters: a) the specific demagnetization routine (real laboratory data) is shown as a
number of demagnetization steps vs. the demagnetization level or step (Fig. 10B). b) The
frequency diagram of VD intensities (Fig. 10C) that helps to properly select the intensity
filtering parameters (especially useful to detect anomalous samples; v.g. lightning or road
blasting, etc…). c) Frequency diagram of virtual directions and circles MADs (Fig. 10D). d)
Woodcock [1977] diagram (Fig. 10E). Besides a virtual direction and circle counter is shown
in this menu (Fig. 10F).

It is possible to see this window together with the virtual directions window to evaluate the
effect of the filtering using the “Refresh graphics” button every time parameters are changed.
Figure 10. Filtering menu. A) Filtering parameters: Initial and final steps, number of steps, minimum and
maximum intensity, minimum and maximum MAD. B) Number of demagnetization steps. C) Frequency diagram
of VD intensities. D) Frequency diagram of VD MAD. E) Orientation matrix control: Woodcock [1977] diagram.
F) VD counter. G) Graphic exporting.

There is also the option to save the

graphics of this window in EPS format
(Fig. 11).

Figure 11. Exported filtering diagrams

 Cited references

Bingham, C. (1974), An antipodally symmetric distribution on the sphere, Ann. Statist., 2, 1201-1225.
Fisher, R. A. (1953), Dispersion on a sphere, Proc. R. Soc., A217, 295-305.
Kirschvink, J. L. (1980), The least-squares line and plane and the analysis of the paleomagnetic data, Geophys. J.
R. Astron. Soc., 62, 699-718.
Pueyo, E. L. (2000), Rotaciones paleomagnéticas en sistemas de pliegues y cabalgamientos. Tipos, causas,
significado y aplicaciones (ejemplos del Pirineo Aragonés). PhD thesis, Dep. Earth Sci., Univ.Zaragoza.
Spain.
Pueyo, E. L.; Feischl, T.; Scholger, R. y Mauritsch, H. J. (2003). Stacking of paleomagnetic data: an objective
approach to the ChRM. Proceedings of the MAGIBER II Paleomagnetismo en España y Portugal, 32-34.
http://www.ub.edu/paleomag/pdf/Comunicacoes.pdf
Ramón, M. J. and E. L. Pueyo, (2008), Cálculo de direcciones y planos virtuales paleomagnéticas: Ejemplos y
comparación con otros métodos, Geotemas, 10, 1203-1206.
Ramón, M. J., Pueyo E. L., Oliva B., Larrasoaña, J. C. (2012), Virtual directions in paleomagnetism: A global
and rapid approach to evaluate the NRM components, Geochem. Geophys. Geosyst. (in review)
Scheepers, P. J. J. and J. D. A. Zijderveld (1992), Stacking in Paleomagnetism: Application to marine sedi ments
with weak NRM, Geophys. Res. Lett., 1914, 1519-1522.
Scheidegger, A. M. (1965), On the statistics of the orientation of bedding planes, grain axes, and similar
sedimentological data, US Geological Survey Professional Papers, 525C, 164–167.
Schmidt, P. W. (1982), Linearity spectrum analysis of multicomponent magnetizations and its application to
some igneous rocks from south-eastern Australia. Geophys. J. R. Astron. Soc., 70, 647-665.
Woodcock N. H. (1977), Specification of fabric shapes using an eigenvalue method. Geol. Soc. Am. Bull., 99,
1231-1236.
Zijderveld, J. D. A. (1967), A.C. demagnetization of rocks: Analysis of results, in Methods in Paleomagnetism,
Eds D.W. Collinson, K.M. Creen and S.K. Runcorn, Elsevier, Amsterdam.

Related software

Allmendiger, R. (2005). Stereonet.

http://www.geo.cornell.edu/geology/faculty/RWA/OldPrograms.htm
Torsvik-Briden-Smethurst (1996-2000) Super IAPD,
http://www.geodynamics.no/software.htm
Enkin, R. Software of paleomagnetic analysis PMAGSC
http://gsc.nrcan.gc.ca/sw/paleo_e.php
Cogne, J. P. (2011). PaleoMac: A MacintoshTM application for treating paleomagnetic
data and making plate reconstructions
http://www.ipgp.fr/~cogne/pub/paleomac/refmessage.html
Paldir from Utrecht Paleomagnetic laboratory
http://www.geo.uu.nl/~forth/Software/soft.html
Tauxe, PmagPy package
http://magician.ucsd.edu/Software/PmagPy/
See more software at the Mark W Hounslow’s web page
http://geography.lancs.ac.uk/cemp/resources/software.htm
 Appendix 1: Input Formats

TH files. Utrecht input data – “- Paldir” format

The first line of the file contains the information of the specimen. Different samples of the
specimen are separated by the string 9999 and the last sample is followed by “END”. The first
line of each sample has its general information: Name, Information, Trend core (drilling),
Plunge core, Volume of the core, Strike S0 (tectonic), Dip S0. Next lines of the sample are the
information relative to each vector measured in each demagnetization step: Step, X (down), Y
(north), Z (east), Error (of measurement).

"heading","info-file"
"FA1-1a","info-sample",116,63,10,208,87
0,-90.0200,41.9300,-19.6700,0.00," "," "
…
…
500,101.2000,-135.1000,-131.8000,0.00," "," "
9999
…
…
"FA1-na"," ",188,39,10,34,266
…
…

9999
"END"

G2 Magnetometer input data

A file contains the information for a single sample, not for the whole site. General sample
information: Heading / Sample Name / Date / (Size) / Trend core / Plunge core / Dip
Direction S0 / Dip S0 / Trend Fold / Dip Fold / Declination present correction. Step
information for each step: Step number / Demagnetization level / Dec (core coordinate
system) / Inc (core coordinate system) / Dec (in situ) / Inc (in situ) / Dec (after bedding
correction) / Inc (after bedding correction) / M / J / X intensity / X standard deviation / X
number of measurements / X error of standard deviation / Y intensity / Y standard deviation /
Y number of measurements / Y error of standard deviation / Z intensity / Z standard
deviation / Z number of measurements / Z error of standard deviation / Signal/Noise ratio /
Signal/Drift ratio / Signal/Holder ratio / Date. The orientation of axis is different from
previous format, in this case is: –z (down), -x (north), y (east).

G2 Roma input data

Sample ID, AF Z, X intensity, Y intensity, Z intensity, Trend core, Plunge core, Dip Direction
S0, Dip S0, Declination, Inclination, Intensity, X error, Y error, Z error, X drift, Y drift, Z
drift, Temperature.
Michigan input data

Heading, specimen information: Core name, Trend core, Plunge core, Strike S0, Dip S0,
Volume of the core. For each step: Temperature, Intensity, X, Y, Z, Declination, Inclination.

U-Channels input data

Depth, AF Z, X intensity, Y intensity, Z intensity, Declination, Inclination, Intensity. As

well, axis orientation is: –z, -x, y.