www.geosoft.com
The software described in this manual is furnished under license and may
only be used or copied in accordance with the terms of the license.
Geosoft Incorporated
8th Floor
85 Richmond St. W.
Toronto, Ontario
M5H 2C9
Canada
Tel: (416) 3690111
Fax: (416) 3699599
Create a Workspace 11
Open a Database 12
References 27
Glossary 28
1
Grant of License. This License Agreement allows you to use one copy of the Program(s) for which
you purchased a license, which may include user documentation in printed or online (electronic) form,
on any single computer at any one time. If you have multiple licenses for the Program(s), you may have
as many copies in use as you have licenses. The Program(s) is in use when loaded into Temporary
memory (i.e. random access memory) or installed into the permanent memory (e.g. hard disk or other
storage device). If the anticipated number of users will exceed the number of applicable licenses, you
must ensure that the number of people using the Program(s) does not exceed the number of licenses.
Upgrades. If the version of the Oasis montaj Program(s) you purchased is an upgrade from an Oasis
for DOS Program(s), you may use the Oasis montaj Program(s) only in conjunction with the Oasis for
DOS Program(s) for which you are licensed.
Copyright. The Program(s) is owned by Geosoft Inc. or its suppliers and is protected by Canadian
copyright laws and international treaty provisions. Therefore, you must treat the Program(s) like any
other copyrighted material (e.g. a book) except that you may make one copy of the Program(s) solely
for backup or archival purposes.
You may not rent or lease the Program(s), sublicense or otherwise transfer or encumber the rights and
obligations hereunder without the prior written approval of Geosoft.
You understand and acknowledge that this license is strictly limited to its terms and, in particular,
grants no rights to any other person and is restricted in its application solely for use by you. You may
not transfer, transmit, disclose, modify, translate, adapt, reproduce, decompile, reverse engineer or
disassemble the Program(s) or any documentation without the prior written approval of Geosoft.
You shall indemnify and save harmless Geosoft from and against any and all claims, suits, actions,
costs (including reasonable legal fees and disbursements) or other liabilities to any person whatsoever
arising in any way out of or in respect of any breach by you of your obligations or the restrictions set
out in this License Agreement.
To use this package effectively, you should be familiar with Fourier domain filtering
concepts and methods including Fast Fourier Transforms. You should also have an
appreciation of how to apply filters for numerical analysis and interpretation of line
data.
By transforming data to the Fourier domain (through the use of a Fast Fourier
Transform, or FFT), we can deal with the data as a function of wavenumber, or
wavelength. In this form, there are a number of manipulations that can be applied to
enhance information that is of interest, remove information that is not of interest, or
transform the data (see Bhattacharya, 1966, and Spector and Grant, 1970).
For example, in the Fourier domain we might apply a Butterworth lowpass filter to
data to remove short wavelength features that are not of interest. We might also
upward continue or downward continue data to a new level that is more sensitive to
the information of interest, or we might transform magnetic data from an observed
geomagnetic latitude to the magnetic North pole. All of these procedures can be
carried out using Fourier techniques.
The Oasis montaj 1D FFT Processing System is a suite of Geosoft eXecutables
(GX applications) that you can use to apply Fourier domain filters to one dimensional
(line) data stored in an Oasis montaj database. Designed primarily for potential field
data, Oasis montaj  1D FFT provides you with the tools required to:
Perform profilebased processing for interpretation and modeling purposes
Interactively select filtering parameters, display power spectra and identify
optimal filters for processing and interpretation
Rapidly apply filters to one or more lines of data
4 Chapter 1: System Capabilities and Concepts
1. Sharpening filters that enhance the shorter wavelength features in the data. These
include highpass filters, downward continuations and vertical and horizontal
derivatives. Such filters are normally used to enhance information from shallow
geologic features.
2. Smoothing filters that enhance longer wavelength features in the data, normally by
removing or attenuating the shorter wavelengths features. These include lowpass
filters, upward continuations and integrations. Smoothing filters are normally
used to remove shortwavelength noise in the data or to remove the effects of
shallow geologic features.
3. Geophysical transformations that convert data from one physical form to another.
These include reduction to the pole for magnetic data and the calculation of
apparent magnetic susceptibility or apparent density.
Smoothing and sharpening filters are often combined to meet the needs of a specific
problem. For example, a vertical derivative (sharpening filter that enhances near
surface geology) might be combined with an upward continuation (smoothing filter
that reduces the effect of noise in the data).
Filters can further be described as geophysical filters  those whose outcome has a
geophysical basis or mathematical  those that have a mathematical definition only.
Since geophysical filters are based on the physics of potential fields, they are ideally
suited to gravity and magnetic data whereas mathematical filters can be applied to any
kind of data.
In addition to filters, power spectra are often required. The Oasis montaj  1DFFT
system enables you to create a power spectrum of the data and to place the real and
imaginary components of the Fourier transform into channels of the database. You
may then apply your own mathematical algorithms to the data using the math
functions of Oasis montaj.
Chapter 1: System Capabilities and Concepts 5
f (w ) = f ( x ) e iwx dx

The Fourier transform of the potential field due to a prismatic body has a broad
spectrum whose peak location is a function of the depth to the top and bottom
surfaces and whose amplitude is determined by its density or magnetization. You can
6 Chapter 1: System Capabilities and Concepts
relate the peak wavenumber (w') to the geometry of the body according to the
following expression:
ln( hb ht )
w = where
hb  ht
For a bottomless prism, the spectrum peaks at the zero wavenumber according to the
expression:
f (w ) = e  hw
where w is the angular wavenumber in radians / ground_unit and h is the depth to the
top of the prism. For more information, refer to Bhatacharia, 1966.
For a prism with top and bottom surfaces, the spectrum is:
f ( w ) = e  htw  e  hbw
where ht and hb are the depths to the top and bottom surfaces respectively. As the
prism bottom moves closer to the observation point at surface, the peak moves to
higher wavenumbers as illustrated in the following figure:
1
no bottom
top = 4
bottom depth
36
20
12
0
0 1
wavenumber
Considering the spectrum of a fixed size prism, as the prism depth increases, the peak
of the spectrum shifts to lower wavenumbers (the space domain anomaly becomes
broader) and the amplitude of the spectrum decreases:
Chapter 1: System Capabilities and Concepts 7
1
thickness = 4
top = 4
16
0
0 1
wavenumber
When looking at the spectrum above, it is important to note that the amplitude of a
deep prism does not exceed the amplitude of the same prism at shallow depth at any
wavenumber. The effect of increasing the depth is to shift the peak to lower
wavenumbers. Because of this characteristic, there is no way to separate the effect of
deep sources from shallow sources of the same type by using wavenumber filters.
You can only distinguish sources if the deep sources have greater amplitude or if the
shallow sources have less depth extent.
When considering a line that is long enough to include many sources, you can use the
log spectrum of this data to determine the depth to the tops of a statistical ensemble of
sources using the relationship (see Spector and Grant, 1970):
log E ( k ) = 4phk
You can determine the depth of an 'ensemble' of sources by measuring the slope of the
energy (power) spectrum and dividing by 4p. A typical energy spectrum for magnetic
data may exhibit three parts  a deep source component, a shallow source component
and a noise component.
8 Chapter 1: System Capabilities and Concepts
The following figure illustrates the interpretation of an energy spectrum into these
three components:
ENERGY SPECTRUM
deep sources
Nyquist
shallow sources
noise
The Fourier domain filtering process (as outlined above) is built in to every IDFFT
filter.
Note: You can use the FFT1DAdvanced Usage menu item to go through the step
bystep Fourier domain filtering process. This stepbystep method enables
you to apply a user defined filter formula or math expression.
10 Chapter 1: System Capabilities and Concepts
The list below provides contact information for Geosoft Technical Support around the
world.
North America Europe and North Africa
Geosoft Inc., Geosoft Europe Ltd.
85 Richmond St. W., 8th Floor 20/21 Market Place, First Floor
Toronto, Ont., Wallingford, Oxfordshire
Canada United Kingdom
M5H 2C9 OX10 OAD
Tel. (416) 3690111 Tel: 44 1491 835 231
Fax (416) 3699599 Fax: 44 1491 835 281
Email: tech@geosoft.com Email: tech.eu@geosoft.com
South America Australia and Southeast Asia
Geosoft Latinoamerica Ltda. Geosoft Australia Pty. Ltd.
Av. Rio Branco, 156  Gr. 2615 350 Hay Street
CEP 20043900, Centro Subiaco, WA 6008
Rio de Janeiro, RJ, Brasil Australia
Tel: (5521) 25320140 Tel. 61 (8) 93821900
Fax: (5521) 25327197 Fax 61 (8) 93821911
Email: tech.sa@geosoft.com Email: tech.au@geosoft.com
South and Central Africa
Geosoft Africa Ltd.
Buren Building, Second Floor
Kasteelpark Office Park
c/o Nossob & Jochemus Streets
Erasmuskloof X3, Pretoria
Tel: 27 12 347 4519
Fax: 27 12 347 6936
Email: tech.za@geosoft.com
Chapter 2: Quick Start Tutorial 11
Create a Workspace
In order to access the 1DFFT menu in Oasis montaj, you must have a workspace
open. The procedures below describe how to do this.
T O C REATE A W ORKSPACE :
T O LOAD A M ENU :
1. On the GX menu, click the Load Menumenu option or the menu button ( ).
The system will display the Load Menu dialog box.
2. Select the 1D_FFT.omn file and click [OK].
12 Chapter 2: Quick Start Tutorial
3. The system will display the FFT1D menu on the main toolbar.
Butterworth Filter FFTBUTTW.GX
Gaussian Filter FFTGAUSS.GX
Cosine Rolloff Filter FFTCOSN.GX
Lowpass Filter FFTLOWP.GX
Highpass Filter FFTHIGHP.GX
Bandpass Filter FFTBANDP.GX
Vertical Derivative FFTVDRV.GX
Vertical Integration FFTVINT.GX
Horizontal Derivative FFTHZDRV.GX
Horizontal Integration FFTHZINT GX
Downward/Upward Continuation FFTCONT.GX
Magnetic Pole Reduction FFTRPOLE.GX
Magnetic Susceptibility FFTSUSC.GX
Density for Gravity Data FFTDENS.GX
Analytic Signal FFTAS.GX
Power Spectrum FFTPSPEC.GX
Open a Database
You must have a database open in your working directory to apply the 1DFFT filters
to a data channel.
T O O PEN A D ATABASE :
1. On the Data menu, click Open Database. The system will display the Open
Database dialog box.
2. Select the fft_mag.gdb file and click [Open].
3. The system will display the fft_mag.gdb database.
Chapter 2: Quick Start Tutorial 13
Note that, the entire Fourier domain filtering processes (see Understanding the 1D
FFT Filtering Process page 9) is built in to every 1DFFT filter.
Note: All 1DFFT GXs (utilities and filters) have their own dialog boxes with
unique parameters specific for each GX. To find out detailed information on
each GX. Click the dialog [Help] button.
T O APPLY F ILTER TO D ATA C HANNEL :
1. On the FFT1D menu, click Butterworth. The system will display the FFT
Butterworth Filter dialog box.
2. Select the channel to filter (mag) from the dropdown list. Specify the name of the
output channel (butter_filt). You can specify values for central wavelength
14 Chapter 2: Quick Start Tutorial
cutoff, and degree of filter and select either a low or high pass filter. For more
information on these parameters click the [Help] button.
3. Click the [SetSampling] button. The Set FFT Sampling Parameters dialog box is
displayed.
4. Here you can specify the required distance increment and the interpolation method
to use. For more information on these parameters click the [Help] button.
5. Click [Back] to return the FFT Butterworth Filter menu. Click [OK] to apply the
Butterworth filter and display the output butter_filt channel in the current
database.
Chapter 3: Preparing Data and Applying Filters 15
It is normally more convenient to work with your data in a real distance unit such as
metres (or your ground_units). To do this, you must make sure the fiducial
numbering of the channel you want to filter is based on distance. If your fiducial
numbering is not based on a real distance unit, we recommend that you resample
your data to a real distance base before applying filters.
The DISTFID GX also creates a channel with the same name as the original channel,
but with extension _fid appended. This channel contains the original fiducial
values for the new data channel. If you want to return the filtered channels back to the
original fiducial base for other processing, you will need the original fiducial values.
This chapter describes the application and mathematics of each of the Oasis montaj
1DFFT filters. The mathematical expressions use the following terms:
This filter calculates the analytic signal of a channel. The analytic signal can be
useful for locating the edges of remanently magnetized bodies and in areas of low
magnetic latitude (Macleod et al., 1994).
as = dz dz + dx dx
where
The vertical derivative is calculated using the FFT process described later in this
chapter. The horizontal derivative is calculated by applying a space domain
convolution filter. The analytic signal is then evaluated from these two sets of data.
This filter calculates the apparent density of the ground that would give rise to the
observed gravity profile. The density assumes that the gravity profile is due to a set of
rectangular prisms with a top at the level of observation of the gravity profile, a
bottom at depth t, and infinite strike length.
You may wish to downward continue the profile to be close to the tops of the
assumed geologic model of interest before calculating the apparent density.
Chapter 3: Preparing Data and Applying Filters 17
You must supply the thickness of the earth model and the background density.
w
L( w ) =
2p G(1 e tw )
Parameters:
1
L( w , q ) =
2pF H ( w ) G( q ) K ( w )
H ( w ) = e  hw
G( q ) = (sin I a + i cos I cos( D  q ) ) 2
sin( aw )
K(w ) =
aw
H(w) Downward continuation to h
Parameters:
I Geomagnetic inclination
D Geomagnetic declination
You can use the Bandpass filter to pass or reject a range of wavenumbers from the
data. However, applying such a simple cutoff filter to an energy spectrum almost
invariably introduces a significant amount of ringing (otherwise known as Gibb's
Phenomena). We recommend that you use a smoother filter such as the Butterworth
filter (FFTBUTTW).
1.0
0.0
k0 k1
Wavenumber (cycles/ground_unit)
L( k ) = 0, for k < k 0
L( k ) = 1, for k 0 k k1
L( k ) = 0, for k > k1
Parameters:
0/1 If 1, pass the defined band. Otherwise, reject the defined band. The default is
to pass the band.
The Butterworth filter is excellent for applying straight forward highpass and low
pass filters to data because you can easily control the degree of filter rolloff while
leaving the central wavenumber fixed. If ringing is observed, you can reduce the
Chapter 3: Preparing Data and Applying Filters 19
degree until you are satisfied with the result. A common, but more complicated
alternative is the Cosine filter (FFTCOSN).
1.0
16
4 8
n= 2
L(k) 0.5
0.0
0 kc N
Wavenumber (cycles/ground_unit)
1
L( k ) =
k n
1 + kc
Parameters:
0/1 Residual/Regional flag to specify if a residual high pass or a regional low pass
is required. By default, the system applies a regional filter.
Downward continuation enhances the responses from sources at a shallow depth (by
effectively bringing the plane of measurement closer to the sources). Note, however,
that it is not possible to continue through a potential field source. If the data contains
short wavelength noise, this noise can appear as very shallow sources in the
continuation. Such noise should be removed before attempting to downward continue
the data. A Butterworth low pass filter set to between 1 and 1.5 times the depth can be
very effective to remove noise before continuation.
You should make a plot of the energy spectrum to determine the wavenumber at
which sources (noise) appears to be more shallow than the depth of continuation. The
energy spectrum is also a good guide for determining the depth to which you can
continue data downward.
20 Chapter 3: Preparing Data and Applying Filters
20.0
h = 16
h=8
L(k)
h=4
h=2
1.0
0 N
Wavenumber (cycles/ground_unit)
L( w ) = e hw
Parameter:
Because this filter has a smooth shape, and it does not alter the energy spectrum
below the start of rolloff (or after the end of rolloff in highpass mode), it is
commonly used for simple highpass or lowpass operations. To reduce ringing, you
can increase the separation between k1 and k0.
1.0
L(k) 0.5
n=2 1 0.5
0.0
0 k0 k1 N
Wavenumber (cycles/ground_unit)
L( k ) = 1, for k < k0
p k  k0
L( k ) = cos n , for k 0 k k1
2 k1  k0
L( k ) = 0, for k > k1
Parameters:
k0 Low wavenumber starting point of the filter. (Cutoff wavenumber for high
pass or start of roll off for lowpass.)
k1 High wavenumber end point of the filter. (Start of roll off for highpass or
cutoff wavenumber for lowpass.)
Chapter 3: Preparing Data and Applying Filters 21
0/1 0 for residual (highpass) filter; 1 for regional (lowpass) filter. The default is
a lowpass filter.
The Gaussian filter is another smooth filter often used for lowpass or highpass
applications.
1.0
L(k)
0.5
0.0
0 ko 2ko 3ko N
Wavenumber (cycles/ground_unit)
( k )
2 2
2 k0
L( k ) = 1  e
Parameters:
As with the bandpass filter, you should use this filter selectively because it can suffer
from Gibb's Phenomena (ringing).
1.0
reject pass
L(k)
0.5
0.0
ko
Wavenumber (cycles/ground_unit)
22 Chapter 3: Preparing Data and Applying Filters
L( k ) = 0 , for k < k0
L( k ) = 1 , for k 0
Parameter:
You can use the horizontal derivative for identifying geologic boundaries in profile
data.
L( w ) = ( w i )
n
Parameter:
n Order of differentiation.
L(w) = (i w) 1
As with the bandpass filter, you should use this filter selectively because it can suffer
from Gibb's Phenomena (ringing).
1.0
pass reject
L(k) 0.5
0.0
k0
Wavenumber (cycles/ground_unit)
L(k) = 1, for k k0
L(k) = 0, for k > k0
Parameter:
Chapter 3: Preparing Data and Applying Filters 23
Reduction to the pole is used in low magnetic latitudes to change an anomaly to its
equivalent as would be observed at the north magnetic pole. This transformation
simplifies the interpretation and visualization of anomalies from low magnetic
latitudes.
1
L( q ) =
(sin I a + i cos I cos( D  q ) ) 2
where
I Geomagnetic inclination
D Geomagnetic declination
Parameter
Ia Inclination to use for the amplitude correction. Default is 20. (Ia = 20, if I
>0; Ia = (20), if I <0 ). If Ia  is specified to be less then I , it is set to I.
Reduction to the pole has an amplitude component (the sin(I) term) and a phase
component (the icos(I)cos(Dq) term). When reducing to the pole from equatorial
latitudes, NorthSouth features can blowup due to the strong amplitude correction
(the sin(I) term) that is applied when Dq is /2 (i.e. a magnetic eastwest
wavenumber). By specifying a higher latitude for the amplitude correction alone, this
problem can be reduced or eliminated at the expense of undercorrecting the
amplitudes of North South features.
Also, you can interpret upward continued data numerically and with modeling
programs. This is not the case for many other filter processes.
1.0
h=2
L(w)
h=4
h=8
h = 16
0.0
0 N
Wavenumber (cycles/ground_unit)
L( w ) = e  hw
Parameter:
The vertical derivative is commonly applied to total magnetic field data to enhance
the most shallow geologic sources in the data. As with other filters that enhance the
highwavenumber components of the spectrum, you must often also apply lowpass
filters to remove highwavenumber noise.
L( w ) = w n
Parameter:
n Order of differentiation.
This filter calculates the vertical integral of the input transform. This is the inverse of
the vertical derivative. The zero wavenumber is set to 0.
L(w) = w 1
Chapter 4: Applying Custom Filters 25
The Advanced Usage menu provides users with stepbystep 1DFFT data processing
submenus. For information on the 1DFFT filtering process see the Understanding
the 1DFFT Filtering Process page 9. These menu items are designed for advanced
users who want to apply custom filters or math expressions to their data.
For detailed information on these or any of the 1DFFT GXs, click the dialog [Help]
buttons.
The channels created by FFTIN have the same name as the original input data channel
with the suffixes _r, _i and _w added. These are the real, imaginary and
wavenumber channels respectively. The wavenumber channel is in units of
radians/fiducial (2p cycles/fiducial). The fiducial numbering of the Fourier domain
channels is in cycles/fiducial.
Note that the FFTIN process does not remove a trend from the data. If your data
contains an undesired trend, you must remove it before running FFTIN. It is also your
responsibility to replace a trend in the data after processing.
The process of applying your own filter is best illustrated through an example. Say
we have an existing channel named Mag which is already sampled on a distance
base (the fiducials are ground_units). We would like to calculate the first vertical
derivative using the expression:
L( w ) = w n
26 Chapter 4: Applying Custom Filters
where n is the order of differentiation (in this case, n=1). To apply this expression we
must multiply the Fourier transform by the wavenumber:
1. On the FFT1D menu, select Advanced UsageFT space >Fourier and select
Mag as the channel to transform. This creates channels Mag_r, Mag_i and
Mag_w.
2. Select the entire Mag_r channel in the database (click three times on the
Mag_r channel header).
3. Press the = (equal sign) to enter an expression, and enter:
Mag_r*Mag_w
The Mag_r channel is replaced by the filtered real data.
4. Select the entire database for the Mag_i channel. Press = to enter an
expression, and enter: Mag_i*Mag_w. The Mag_i channel is replaced by the
filtered imaginary data.
5. Select FFT Fourier > space... and choose the Mag_r and Mag_i as the
input channels, Mag as the original reference channel, and dz as the output
filtered channel. The dz channel will contain the vertical derivative in
nT/ground_units.
References 27
References
Bhattacharya, B. K., 1966, Continuous spectrum of the total magnetic field anomaly
due to a rectangular prismatic body. Geophysics, Vol. 31, p.97121.
Gupta, V. K., and Grant, F. S., 1985, Mineral exploration aspects of gravity and
aeromagnetic survey in SudburyCobalt area, Ontario. SEG; The Utility of Regional
Gravity and Magnetic Anomaly Maps, W. J. Hinze (Editor) p.392411.
MacLeod, I. N., Vierra, S. and Chaves, A. C., 1993, Analytic signal and reduction 
tothepole in the interpretation of total magnetic field at low magnetic latitudes.
Proceedings of the Third International Congress of the Brazilian Geophysical Society,
November, 1993.
Spector, A. and Grant, F. S., 1970, Statistical models for interpreting aeromagnetic
data. Geophysics, Vol. 35, No. 2, p.293302.
Glossary
Array channels Special channels in columns of the spreadsheet that contain multiple
channels of data. Array channels are represented graphically by a curve in
the spreadsheet. See also channels, subarray channels.
column In the Oasis montaj spreadsheet, a vertical line of cells that contain data.
desktop Background area in the Oasis montaj workspace. You can open and
display a virtually unlimited number of Spreadsheet, Profile and Map
windows in this area.
fiducials Points accepted as fixed bases of reference. Marks indicating the order in
which each reading or sample reading was taken.
Graphical User Interface Interactive software environment where functions are performed by
selecting graphic objects.
grid Collection of points along rows and columns that define a twodimensional
rectangular area on some plan, usually a ground plan.
Grids or Grid file Files containing location (X and Y) and data (Z) values. Values are
typically interpolated to create a regular and smoothly sampled
representation of the locations and data.
groups A set of graphics elements that make up a graphic component of the map.
For example, a line path plot, a contour plot or a profile plot would all be
separate graphics groups within the Data View.
GX or Geosoft eXecutable Programmed process (identified by the *.GX file extension) used to process
data in Oasis montaj.
Images or Image file Files containing location (X and Y) and color values. The values are not
interpolated. Standard PC file types created using imaging or electronic
photoediting techniques.
Main window Primary tool used to create and maintain databases, display data and
process data. Oasis montaj is a Graphical User Interface (GUI) system that
provides all functionality required to process and display virtually any type
of Earth Science data.
Map (*.MAP) Geosoftdeveloped file that integrates all graphics elements (lines, polygons
and text) and layers (base maps, data, grids, plots and images) constructed
in Oasis montaj.
References 29
map description file (*.MDF) File used when making a standard map that has a sheet size and margins
inside which there is a map area with a specified origin and scale. Usually,
MDF files are created with the MAPSET GX. MDF files may also be
created manually by editing an ASCII file with extension .MDF.
menu (*.OMN) Graphical list of commands or functions that a user may choose from.
Oasis database Objectoriented database that stores Earth Science data in a form that
delivers fast access to data and efficient storage required for applications
that deal with very large volumes of data.
Oasis montaj Oasis montaj is the core software platform that provides all functionality
required to manage, manipulate, visualize and map spatially located Earth
Science data.
random XYZ data Located XYZ data that cannot be grouped naturally into separate lines. For
example, regional gravity surveys or geochemical surveys are commonly
considered random data because the locations appear somewhat random in
nature.
row In the Oasis montaj spreadsheet, a horizontal line of cells that contain data.
View Stored snapshot of the screen settings, current line, displayed channels and
displayed profiles in Oasis montaj.