See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/220164292

Potensoft: MATLAB-based software for

potential field data processing, modeling and
mapping

Article in Computers & Geosciences July 2011

DOI: 10.1016/j.cageo.2011.02.008 Source: DBLP

CITATIONS READS

15 1,822

2 authors:

Muzaffer zg Arsoy nal Dikmen

General Directorate of Mineral Research and Ankara University
18 PUBLICATIONS 55 CITATIONS 38 PUBLICATIONS 137 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by nal Dikmen on 19 February 2014.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 Computers & Geosciences 37 (2011) 935942

Contents lists available at ScienceDirect

Computers & Geosciences

journal homepage: www.elsevier.com/locate/cageo

Potensoft: MATLAB-based software for potential eld data processing,

modeling and mapping$
zgu Arsoy n, Unal
M. O Dikmen
Ankara University, Faculty of Engineering, Department of Geophysical Engineering, 06100 Ankara, Turkey

a r t i c l e i n f o abstract

Article history: An open-source software including an easy-to-use graphical user interface (GUI) has been developed for
Received 30 October 2009 processing, modeling and mapping of gravity and magnetic data. The program, called Potensoft, is a set
Received in revised form of functions written in MATLAB. The most common application of Potensoft is spatial and frequency
8 February 2011
domain ltering of gravity and magnetic data. The GUI helps the user easily change all the required
Accepted 9 February 2011
Available online 4 March 2011
parameters. One of the major advantages of the program is to display the input and processed maps in a
preview window, thereby allowing the user to track the results during the ongoing process. Source codes
Keywords: can be modied depending on the users goals. This paper discusses the main features of the program
Gravity and magnetic data and its capabilities are demonstrated by means of illustrative examples. The main objective is to
GUI (Graphical User Interface)
introduce and ensure usage of the developed package for academic, teaching and professional purposes.
Modeling
& 2011 Elsevier Ltd. All rights reserved.
Mapping
Spatial domain ltering
Frequency domain ltering

1. Introduction within the dataset. Filtering procedure can be undertaken in the

frequency domain by means of Fourier Transform (FT) or in the
Gravity and magnetic methods are potential eld methods and spatial domain by convolution. Frequency domain ltering involves
used for a wide range of applications and scales in geosciences. converting the dataset into the frequency domain, performing
Traditionally, they have been used for large scale investigation of an operation on the data by applying the appropriate lter and
geologic structures. Smaller-scale applications of the gravity and then transforming the data back to the spatial domain. The most
magnetic methods are for mining exploration, environmental and commonly used frequency domain lters include reduction to pole,
engineering studies, etc. (Hinze, 1990; Reynolds, 1987; Sharma, pseudogravity transformation, analytical continuations and derivative
1997; Telford et al., 1996). In order to arrive at geologically mean- lters. Convolution methods involve convolving a lter impulse
ingful anomaly values, the survey gravity or magnetic data are response (lter coefcients) with the dataset (Byerly, 1965). In
rstly processed to make appropriate corrections. After these steps, addition to ltering, the main intention of gravity and magnetic data
the survey data is ready for interpretation or displaying as maps. interpretation is calculating the potential eld from the constructed
Interpretation of potential eld data is performed on either prole model (forward-modeling). Various methods have been developed for
or map data. These data are most often interpreted by the use of different aims: simple geometries, sequence of isolated 2-D or 3-D
inversion or data processing techniques. bodies or layers (Blakely, 1996; Jessel, 2001; Parker, 1972; Rasmussen
Spatial and frequency domain ltering, image processing and and Pedersen, 1979; Telford et al., 1996). On the other hand, potential
managing grids are essential tools in gravity and magnetic data eld data visualization has a key role for the interpretation process.
interpretation. A potential eld or image processing lter highlights Technological developments have resulted in faster and more sensi-
different aspects of potential eld data (Bhattacharyya, 1972; tive visualization of potential eld data. Using different types of maps
Clement, 1973; Gunn, 1975; Ku et al., 1971; Jacobsen, 1987; Telford (i.e. pseudocolor, shaded, and contour maps) retains the maximum
et al., 1996; Vaclac et al., 1992). Filters can emphasize boundaries amount of information from the gravity and magnetic data.
between geological contacts, highlight deeper or shallower sources, A number of commercial software packages are commonly
highlight features from different angles or reduce undesirable effects used for ltering, modeling and mapping of gravity and magnetic
anomalies (e.g. Geosoft Oasis Montaj and its extensions, Encom
$
Prole Analyst and ModelVision, Intrepid Software, and Golden
Code available from server at: http://www.iamg.org/CGEditor/index.htm
n Software Surfer). The literature contains a large number of
Corresponding author. Tel.: 90 312 2033355; fax: 90 312 2120071.
zgu Arsoy),
E-mail addresses: [email protected] (M. O papers, including source code or software, for potential eld data
[email protected] (U. Dikmen). interpretation (Bezvoda et al., 1990; Cooper, 1997, 2005, 2006;

0098-3004/$ - see front matter & 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.cageo.2011.02.008
936 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U.

Cooper and Cowan, 2004, 2006; Durrheim and Cooper, 1998; Fedi The project controls the working directory. If the user creates a
et al., 2005; Gibert and Galdeano, 1985; Mendonc- a and Meguid, new project, Potensoft automatically saves all project les in the
2008). These papers commonly address a specic problem in the working directory. Similarly, if the user opens a previously
interpretation of potential eld data. However, there is no open- created project, Potensoft assumes that all project les are located
source software with a GUI, which embodies all of the procedures in the same directory. The user launches Potensoft by running the
described above. The main objective of the current paper is to start function; the program then opens a GUI window, where
eliminate this gap for gravity and magnetic data interpretation. the user can open an existing project or create a new project.
Potensoft is written in MATLAB, using Version 7.6 Release 2008b Selecting either of these two menus opens the main Potensoft GUI
under the Windows operating system. It has not been tested in (Fig. 1). The Potensoft GUI window includes standard bars such
other operating systems (e.g. UNIX, and LINUX). The Potensoft GUI as; title bar, short-cut bar, project path bar, mapping bar and
is designed at 1280  1024 screen resolution. Despite the fact that it information bar. The remaining part of the Potensoft GUI window
is designed using normalized screen resolution and can therefore be is subdivided into two windows, the project explorer window and
used at any screen resolution, the interpreter gets the best GUI view working window.
at 1280  1024 screen resolution. MATLAB was chosen because of The project path bar displays the working project path
its powerful data analysis capabilities, visualization features, exten- (see Fig. 1). Project name, the map type and map coordinates
sive library of mathematical functions and popularity in academic from the tracked mouse positions are displayed on the informa-
applications. During the last few years, MATLAB has become an tion bar (See Fig. 1).
increasingly popular tool in earth sciences (Trauth, 2010). Maps are a key method for interpretation of both gravity and
The main objective of Potensoft is to help interpretation of the magnetic data and many earth science applications. Potensoft
gravity and magnetic data, many types of enhancements (trans- provides a wide range of capabilities for performing both basic
formations, lters, edge detection techniques, etc.) can be applied and advanced mapping. The mapping bar enables users to create a
to the maps to emphasize features of interest. Potensoft includes wide variety of maps. The Potensoft working window is divided
basic processing modules for gridding, modeling, ltering of into two windows, a mapping window and map-editing window.
gravity and magnetic data. It provides some functions such as When the user opens a map using the mapping bar, it is displayed
spatial and frequency domain ltering tools and gridding tools. It in the Potensoft mapping window. When the user double-clicks on
also provides a convenient tool for application in other disci- the created map, the map-editing window is opened (Fig. 2). Many
plines, where the data are presented as map or image. The most features of the maps can be changed using the map-editing
important benets of Potensoft are its ease of use and that it is window. When the user right-clicks on the map, a popup menu
open-source. Specic and major components of the program are with several items is opened (Fig. 2). Two of the highlighted popup
described in the Section 2. menu items are the distance and azimuth tool and the colormap
tool. The distance and azimuth tool displays the distance and
azimuth between two selected points in a map. The colormap tool
2. Program overview enables the user to alter the color zones interactively, by setting the
color limits of the map to specied minimum and maximum values.
Working in Potensoft requires opening an existing project or The project explorer enables the user to browse as well as
creating a new project. A Potensoft project le (n.prj) includes open any project les. Project explorer has 3 windows: a grid les
three items in the working project: grid, model and prole les. window that includes (n.grd) les, a model les window that

Fig. 1. GUI of Potensoft main window. Potensoft main menu, short-cut bar and project bar are displayed under the title bar. Mapping bar is displayed on the right panel,
project explorer window is displayed on the left panel and information bar is displayed on the lower panel of the Potensoft GUI. Created maps are displayed on the central
part of the gure where is called as working window.
 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U. 937

Fig. 2. GUI of Potensoft main window showing voxel view of a model, popup menu and map-editing window.

includes (n.mod) les, and a prole les window that includes in the model les window. The user can open a model le for viewing
(n.xy) les. To access the pseudocolor map of the grid les, the by clicking on the model lename. Three different forms of visualiza-
user should click one of the listed grid les. When the user clicks tion are available: plan view, depth projection view and voxel view.
one of the model les, a model bar will be displayed at the bottom An image of the gravity modeling study using 3-Dimensional model-
of the GUI window. This model bar includes three icons. The rst ing GUI is presented in Fig. 3.
icon displays the plan view, the second icon displays both the The grid menu includes data gridding and basic grid operation
plan view and the depth projections and the third icon displays tools. There are many mathematical methods for creating a grid.
the voxel view of the model. Potensoft offers a set of gridding algorithm choices such as linear,
The Potensoft main menu is displayed under the title bar when cubic and nearest interpolation methods. Basic grid operation
the user opens or creates a new project. Dropdown menus provide tools are divided into seven sections. The math expression tool
direct access to all of the Potensoft capabilities. Using the le allows the user to create a new grid le from a single or two grid
menu, a new project can be created, an existing project can be les. The output grid le is based on a mathematical function. The
opened, a working project can be saved, a map can be exported in function command can use any command within the MATLAB
many image formats, or a map can be printed. The edit menu function library. The smooth tool can use one of the nearest
allows a map to be copied to the clipboard or the working window neighbor, linear, cubic spline or cubic interpolation methods to
to be cleared. Files (n.grd; n.mod; n.xy) can be added to an existing compute new grid nodes. The number of grid nodes added
Potensoft project using the add les to project menu; added les between the existing grid nodes can be selected using the nodes
will be placed in the project explorer window. slider. The transform tool includes three options to modify the
The program structure includes four basic components. Poten- horizontal grid coordinates. This tool can be used to rotate and
tial eld data mapping component was described above. The rst ip the grid nodes. The extract grid tool creates a sub-grid of an
component is 3-D modeling of both gravity and magnetic data. existing grid le interactively. The grid2data tool converts the
Other component is gridding datasets and the ability to perform grid le to an ASCII data le. The grid info tool provides the user
grid operations interactively. The most basic component is spatial with some properties (e.g. path, min, max values and size of the
and frequency domain ltering of potential eld data. These three grid) of a grid le. The noise generator tool adds different types of
basic components will be described in the following paragraphs. noise to a grid. Noise density can be adjusted interactively.
The 3-D modeling menu can be used for both gravity and magnetic The prole menu includes Grid prole and Multi Grid prole
modeling. Due to the dynamic and comprehensible design of the 3-D tools. Grid prole takes up to ve cross-sections from an image of
modeling GUI (both gravity and magnetic), model parameters can be a grid le. The Multi Grid prole tool is capable of taking a cross-
adjusted very quickly and effectively. Subroutines one_prism and section from up to six grids at once. Cross-sections are formed
multi_prism (converted to MATLAB) from Mendonc- a and Meguid when the user enters a polyline. 2-D datasets are automatically
(2008) were used to compute 3-D gravity and magnetic anomalies. interpolated from the grids for generating prole plots and saved
The one_prism.m and multi_prism.m functions calculate gravity as prole (n.xy) les. The generated prole le is immediately
and magnetic anomalies which result from a number of prismatic displayed on the prole le window.
bodies. In addition, magnetic eld components and gravity derivatives The Spatial Domain lter menu includes common digital
are also computed and all synthetically produced maps are displayed image processing lters, including a broad suite of high-pass,
on the bottom-left panel of the 3-Dimensional modeling GUI window. low-pass, band-pass, derivative-based, directional and embossing
All of these maps can be saved separately as grid (n.grd) les. One of lters. The Spatial Domain lter tool also includes a user-dened
the major benets of the 3-D modeling tool is that it provides model lter tool that allows the user to set the width and height of the
les (n.mod). Whenever the user creates a model le, it will be placed lter and the lter coefcients. The user can change the number
938 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U.

Fig. 3. Example of 3-D gravity modeling GUI showing user selection options and pseudocolor maps of synthetically produced gravity eld and its derivatives.

of lter coefcients by clicking the row and column sliders. The the spatial domain map of the input dataset, designed lter and
input grid and ltered grid are displayed on the lower panel of the output dataset. Interactive slider and other lter options allow
window (Fig. 4). users to adjust all lter parameters and preview the results
The Frequency Domain ltering tool provides a broad library of instantly. The tangent hyperbolic sub-menu includes 2-D low-pass,
2-dimensional Fast Fourier Transform (FFT) geophysical and digital high-pass, band-pass and band-stop tangent hyperbolic lters
image processing lters, to allow the application of frequency- (tanh lters). The properties of both frequency and time domain
domain lters to gridded gravity and magnetic datasets. This menu low-pass and band-pass tanh lter were presented by Bas- okur
includes circularly symmetrical, tangent hyperbolic, gaussian, (1998). The main advantage of the tanh lter is controlling the
butterworth, laplacian of gaussian, cosine roll-off, cosine direc- slopes at the cut-off regions using a smoothness parameter. If the
tional, directional, butterworth elliptic, strike and wiener lters. smoothness parameter takes relatively high values, then the slope
Applications of these lters to the interpretation and enhancement of the lter function in the frequency domain decreases. This will
of potential eld data and their advantages and disadvantages signicantly reduce the oscillations of the lter response in the
were discussed in various papers (Cook et al., 2003; Naidu and spatial domain. A special care is needed in the selection of the
Mathew, 1998; Spector, 1968). These lters are also widely used smoothness parameter because a high value may lead unsuccessful
for processing digital images and other scientic datasets extraction of unwanted frequency components from the input
(Gonzales and Woods, 2002). If the user converts any type of data. The decision about the value of the smoothness parameter
dataset format to Golden Software ASCII grid format (n.grd) or can be made interactively depending on the input data. Imple-
Geosoft 2-byte binary grid format (n.grd), the converted dataset can mentation of 2-D tanh lters for the purpose of gravity and
be easily ltered using this tool. The Frequency Domain ltering magnetic data ltering is rstly presented in this paper.
tool has important and specic properties. This tool enables the The gravity and magnetic tools menu includes most of the
interpreters to display radially-averaged power spectrum, 1-D frequently-used spatial and frequency domain processes. RTP
lter spectrum and 1-D ltered radially-averaged power spectrum (Reduction to magnetic pole) and RTE (Reduction to magnetic
on one axes in the upper panel of the frequency domain ltering equator) tools reduce the magnetic anomalies to magnetic pole
GUI window (Fig. 5). In addition, a lter option window can be and equator, respectively. Pseudogravity and Pseudomagnetic tools
seen on the middle panel of the GUI window (Fig. 5). Some of the transform magnetic anomalies to gravity anomalies and transform
lter parameters can be modied interactively by using the lter gravity anomalies to magnetic anomalies, respectively. Upward and
option window. Functions calcrad2.m, fcoef1.m and per- downward continuation tools compute analytical continuations of
gram2.m from Ridsdill-Smith (2000) were used to compute the the gravity and magnetic anomalies. The vertical integration menu
radially-averaged power spectrum. Selection of the high and low calculates the vertical integral (inverse of the vertical derivative) of
cut-off frequencies or transition band zone degrees can be done by the input dataset. The details of the mentioned transformations can
moving the green colored slider that can be seen in the upper panel be found in Blakely (1996). The TMI transformations tool converts
axes. The major benet of the tool is to display the preview total or vertical magnetic eld components to horizontal, vertical or
window in the lower panel of the GUI window. The preview total components. The sunshading tool can be used for enhance-
window includes frequency and spatial domain tabs; The fre- ment of the linear features in potential eld images. The
quency domain tab displays the spectral map of the input dataset, cirlesun.m function from Cooper (2003) was used for sunshading
designed lter and output dataset; The spatial domain tab displays computation. The trend remove tool removes the rst, second and
 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U. 939

Fig. 4. Example of user-dened spatial domain ltering GUI showing lter coefcients panel and pseudocolor maps of input and output datasets.

third-order polynomial trend from the dataset. Frequency domain

processing requires expanding the gravity and magnetic grid to be
square so that grid dimensions will be acceptable for the FFT. The
subsequent process is a grid-lling procedure. The grid-lling
procedure involves padding the edges of the grid with dummy
values and then replacing all dummy values with interpolated
values. The taper2spline function from Cooper and Cowan (2008)
was used for this procedure.
The derivatives menu includes horizontal (x and y) and vertical
(z) derivative tools. Gravity and magnetic data derivatives help to
estimate the physical properties of the sources (Verduzco et al.,
2004). Horizontal derivatives of the potential eld data are
computed in the spatial domain by means of central difference
approximation of the nite difference method. Vertical deriva-
tives are computed in the frequency domain and therefore involve
the use of a FFT. Horizontal derivatives enhance edges and vertical
derivatives narrow the width of anomalies and so enhance the
details in the map. Horizontal and vertical derivatives can be
computed up to 4th order. The most commonly used gravity and
magnetic data derivatives are rst and second order, higher
orders being used less, due to noise problems (Gunn et al., 1997).
Edge detection tools can be used to enhance the edges of
anomalies and other features. The horizontal location of the
boundaries or edges of gravity and magnetic anomaly sources is
requested in potential eld interpretation. Edge analysis of the
gravity and magnetic anomalies is accomplished using a number
of methods involving derivatives. Edge detection tools include all
these procedures, so it is easy to see which lter achieves the best
result for the interpreters specic geological problem. The edge
detection tool includes analytical signal, the basic concepts for the
2D case were discussed by Nabighian (1972) and for the 3D case
by Roest et al. (1992), horizontal derivative (Cordell and Grauch,
Fig. 5. Example of frequency domain ltering GUI showing spectral plots, lter
1985), tilt angle (Miller and Singh, 1994), total horizontal deriva-
options panel and spectral maps of input dataset, designed lter and output tive of tilt angle (Verduzco et al., 2004) and hyperbolic tilt angle
dataset. (Cooper and Cowan, 2006) methods.
940 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U.

3. Application to synthetic and real magnetic data the cosine function was 2, the low frequency starting point of the
cosine roll-off lter was 0.04725 Hz and the high frequency end-
Potensoft was applied to articially-generated magnetic data point of the cosine roll-off lter was 0.3225 Hz. Fig. 6j shows the
and real aeromagnetic data from Turkey. Fig. 6 illustrates the sun-shaded data in Fig. 6d when the elevation of the sun was 101
results of a magnetic modeling study that used the synthetically from the horizontal. Fig. 6k shows the spatial domain lter applied
produced magnetic data to check some of Potensofts spatial and to the data in Fig. 6d. A simple 3  3 horizontal differentiation
frequency domain lters. Fig. 6a shows a magnetic model com- operator was used for computation. Fig. 6l shows the strike lter
posed of four prismatic bodies. Horizontal locations of the prisms applied to the data in Fig. 6h. Features located at 1801 clockwise
can be followed from Fig. 6a and b shows the voxel view of the from North are enhanced.
created model shown in Fig. 6a and c shows the total magnetic Fig. 7a shows aeromagnetic data from Turkey. The aeromagnetic
eld anomalies calculated from the prism model. The direction of data is 340  340 km2 in size and has a grid resolution of 2 km in
the ambient eld is D 01 and I601. The inclination and declina- both horizontal directions. The data mostly covers the Eskis- ehir fault
tion angles of the magnetization are 01 and 601, respectively. The zone, which comprises of successive fault segments (Koc- yigit, 2000).
intensity of magnetization for all prisms is 1 A/m. Fig. 6d shows The Eskis-ehir fault and its segments extend in a Northwest to
the data in Fig. 6c after magnetic pole reduction has been Southeast direction (see Fig. 7a). The area mainly consists of base-
applied. Fig. 6e shows the pseudogravity anomaly transformed ment rocks, marbles, Inon u blueschists and peridotites. The con-
from the data in Fig. 6c and f shows the analytic signal amplitude tinental Middle-Late Miocene sequences that overlie these rocks
of the data in Fig. 6d. The tilt angle of the data in Fig. 6d is shown with angular unconformity are made up of the Porsuk formation
in Fig. 6g and h shows the total magnetic eld anomaly in Fig. 6d, (conglomerate, sandstone, claystone, marl, and lacustrine lime-
continued upward for 10 km. The result of the cosine roll-off high stones). Quaternary alluvium covers the older units unconformably
pass ltering of the data in Fig. 6d is shown in Fig. 6i. The degree of
(Gozler et al., 1984). Fig. 7b shows data after being reduced to

Fig. 6. Application to synthetic magnetic model data. (a) Synthetic magnetic model composed of four prismatic bodies. (b) Voxel view of the model showing relative sizes
and positions of four buried prismatic bodies. (c) The total magnetic eld produced by prism model shown in (a and b). (d) Reduced to the magnetic pole data in (c).
(e) Pseudogravity of the data in (c). (f) Analytic signal of the data in (d). (g) Tilt angle of the data in (d). (h) 10 km upward continued data in (d). (i) Cosine roll-off high pass
ltered (cosine function degree: 2; low frequency starting point of the lter: 0.04725 Hz; high frequency ending point of the lter: 0.3225 Hz) of the data in (d).
(j) Sunshade of the data in (d). The sun elevation was 101 from the horizontal. The lter center location is marked as * in (d). (k) Horizontal spatial domain ltered of the
data in (d). (l) Strike ltered of the data in (h). Features lying at the 1801 clockwise from North are enhanced.
 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U. 941

m_les and synthetic datasets, is available via the internet at:

http://eng.ankara.edu.tr/arisoy/potensoft.htm.

Acknowledgments

The authors thank the General Directorate of Mineral Research

& Exploration (MTA) of Turkey for providing aeromagnetic data.
Some of the MATLAB functions which are freely available on the
internet were reconsidered and modied in designing Potensoft
GUI: gdist.m, getgrd2.m, gridt.m, motionblur.m, uitabpanel.m and
voxel.m. We thank the authors of these free MATLAB functions for
their work and for sharing their code. This work constitutes part of
the Ph. D. thesis undertaken by M.O . Arsoy (Ankara University).
Constructive comments by the three anonymous reviewers were
helpful in improving the software and the initial version of the
manuscript. We thank them for their critiques and comments.

Appendix A. Supporting information

Supplementary data associated with this article can be found

in the online version at doi:10.1016/j.cageo.2011.02.008.

References

Bas- okur, A.T., 1998. Digital lter design using the hyperbolic tangent functions.
Journal of the Balkan Geophysical Society 1 (1), 1418.
Fig. 7. Application to data from Turkey. (a) Aeromagnetic data covering a
Bezvoda, V., Jezek, J., Segeth, K., 1990. Fredpack-a program package for linear
340  340 km2 area from Turkey. Magnetic eld values range from -638 nT to
ltering in the frequency domain. Computers & Geosciences 16 (8),
513 nT. (b) Reduced to pole of the magnetic data shown in (a). (c) Pseudogravity of 11231154.
the magnetic data in (a). (d) First-order derivative with respect to variable x of the Bhattacharyya, B.K., 1972. Design of spatial lters and their application to high-
magnetic data in (b). (e) First-order derivative with respect to variable y of the resolution aeromagnetic data. Geophysics 37 (1), 6891.
magnetic data in (b). (f) Theta map of the magnetic data in (b). (g) Total horizontal Blakely, R.J., 1996. Potential Theory in Gravity and Magnetic Applications.
derivative of the pseudogravity data in (c). (h) Tilt angle of the pseudogravity data in Cambridge University Press, Cambridge, 441pp.
(c). (i) Tanh band pass ltered (low cut-off: 6.125e 5 Hz; high cut-off: 2.60e-4 Hz; Byerly, P.E., 1965. Convolution ltering of gravity and magnetic maps. Geophysics
smoothness parameter: 5.5e 5) of the pseudogravity data in (c). (j) 5 km upward 30 (2), 281283.
continued of the magnetic data in (b). (k) 750 m downward continued of the Clement, W.G., 1973. Basic principles of two-dimensional digital ltering.
magnetic data in (b). (l) Cosine directional ltered of the magnetic data in (b). Geophysical Prospecting 21, 125145.
Features lying at the 451 anticlockwise from North are passed while ltering. Cook, F.A., Lynn, C.E., Hall, K.W., 2003. Cross-strike potential-eld anomalies in the
Canadian Cordillera. Canadian Journal of Earth Sciences 40 (1), 111.
Cooper, G.R.J., 1997. GravMap and PFproc: software for ltering geophysical map
magnetic pole. Fig. 7c shows data after pseudogravity transforma- data. Computers & Geosciences 23 (1), 91101.
tion. Fig. 7d and e shows x and y derivatives, respectively, of the Cooper, G.R.J., 2003. Feature detection using sun shading. Computers & Geo-
sciences 29 (8), 941948.
reduced-to-pole data. A Theta map of the reduced-to-pole data is Cooper, G.R.J., 2005. Analysing potential eld data using visibility. Computers &
shown in Fig. 7fh shows total horizontal derivative and tilt angle of Geosciences 31 (7), 984992.
the pseudogravity data. Fig. 7i shows a tanh band-pass lter of the Cooper, G.R.J., 2006. Interpreting potential eld data using continuous wavelet
transforms of their horizontal derivatives. Computers & Geosciences 32 (7),
pseudogravity data. The low cut-off frequency of the lter was 984992.
6.125e5 Hz, the high cut-off frequency was 2.60e 4 Hz and Cooper, G.R.J., Cowan, D.R., 2004. Filtering using variable order vertical derivatives.
the smoothness parameter was 5.5e5. Fig. 7j and k shows Computers & Geosciences 30 (5), 15851591.
Cooper, G.R.J., Cowan, D.R., 2006. Enhancing potential eld data using lters based
reduced-to-pole data continued 5 km upward and 750 m down-
on the local phase. Computers & Geosciences 32 (10), 15851591.
ward, respectively. Fig. 7l shows cosine directional ltered output of Cooper, G.R.J., Cowan, D.R., 2008. Edge enhancement of potential eld data using
the reduced-to-pole magnetic data. Magnetic anomalies of the normalized statistics. Geophysics 73 (3), H1H4.
Eskisehir fault zone and its segments lying at almost 451 antic- Cordell, L.E., Grauch, V.J.S., 1985. Mapping basement magnetization zones from
aeromagnetic data in the San Juan basin, New Mexico. In: Hinze, W.J. (Ed.),
lockwise from North are enhanced and the effects of other features The Utility of Regional Gravity and Magnetic Anomaly Maps, Society of
are removed from the data using the cosine directional lter. Exploration Geophysicists, pp. 181197.
Durrheim, R.J., Cooper, G.R.J., 1998. EULDEP: a program for the Euler deconvolu-
tion of magnetic and gravity data. Computers & Geosciences 24 (6), 545550.
4. Conclusions Fedi, M., Paoletti, V., Rapolla, A., 2005. The role of multilevel data in potential eld
interpretation. Computers & Geosciences 31 (6), 681688.
Gibert, D., Galdeano, A., 1985. A computer program to perform transformations of
The purpose of this paper is to introduce the Potensoft program, gravimetric and aeromagnetic surveys. Computers & Geosciences 11 (5),
which includes the basic steps of gravity and magnetic data 553588.
processing, mapping and modeling. The program includes a user- Gonzales, R.C., Woods, R.E., 2002. Digital Image Processing, 2nd edn. Prentice Hall,
Upper Saddle River, New Jersey, 793pp.
friendly interface, providing a useful tool for earth scientists to

Gozler, - ukayman,
M.Z., Cevher, F., Kuc A., 1984. The geology and hot water springs
implement basic and advanced data processing. The design of the of Eskis-ehir and surroundings. Bulletin of Mineral Research and Exploration
GUI allows users to easily modify or extend the developed package, 103, 4054 (in Turkish).
depending on their research requirements. We believe that Poten- Gunn, P.J., 1975. Linear transformations of gravity and magnetic elds.
Geophysical Prospecting 23 (2), 300312.
soft will be modied and extended by users to become a popular Gunn, P.J., Fitzgerald, D., Yassi, N., Dart, P., 1997. New algorithms for visually
tool for processing gravity and magnetic data. Potensoft, including enhancing airborne geophysical data. Exploration Geophysics 28, 220224.
942 zg
M. O Dikmen / Computers & Geosciences 37 (2011) 935942
u Arsoy, U.

Jessel, M., 2001. Three-dimensional geological modeling of potential-eld data. Parker, R.L., 1972. Rapid calculation of potential anomalies. Geophysical Journal of
Computers & Geosciences 27 (4), 455465. the Royal Astronomical Society 42, 315334.
Hinze, W.J., 1990. The role of gravity and magnetic methods in engineering and Rasmussen, R., Pedersen, L.B., 1979. End corrections in potential eld modelling.
environmental studies. In: Geotechnical and Environmental Geophysics, vol. 1, Geophysical Prospecting 27 (4), 749760.
Society of Exploration Geophysicists, Tulsa, pp. 75126. Reynolds, J.M., 1987. An introduction to applied and environmental geophysics.
Jacobsen, B.H., 1987. A case for upward continuation as a standard separation lter John Wiley and Sons, NY, 806pp.
for potential-eld maps. Geophysics 52 (8), 11381148. Ridsdill-Smith, T., 2000. The application of the wavelet transform to the processing
Koc- yigit, A., 2000. Seismicity of the southwestern part of Turkey. In: BADSEM of aeromagnetic data. Ph. D. Dissertation, The University of Western Australia,
2000-Symposium on Seismicity of the Southwestern part of Turkey, Izmir, Australia, 197pp.
Roest, W.R., Verhoef, J., Pilkington, M., 1992. Magnetic interpretation using the 3-D
Turkey, pp. 3039 (in Turkish).
analytic signal. Geophysics 55 (1), 116125.
Ku, C.C., Telford, W.M., Lim, S.H., 1971. The use of linear ltering in gravity
Sharma, P.M., 1997. Environmental and Engineering Geophysics. Cambridge
problems. Geophysics 36 (6), 11741203.
University Press, Cambridge, UK, 475pp.
Mendonc- a, C.A., Meguid, A.M.A., 2008. Programs to compute magnetization to
Spector, A., 1968. Spectral analysis of aeromagnetic data. Ph. D. Dissertation,
density ratio and magnetization inclination from 3-D gravity and magnetic The University of Toronto, Canada, 167pp.
anomalies. Computers & Geosciences 34 (6), 603610. Telford, W.M., Geldart, L.P., Sherrif, R.E., 1996. Applied Geophysics. Cambridge
Miller, H.G., Singh, V., 1994. Potential eld tilta new concept for location of University Press, Cambridge, 770pp.
potential eld sources. Journal of Applied Geophysics 32, 213217. Trauth, M.H., 2010. Matlabs Recipes for Earth Sciences, 3rd edn. Springer,
Nabighian, M.N., 1972. The analytic signal of two-dimensional magnetic bodies Heidelberg, 336pp.
with polygonal cross-section: its properties and use for automated anomaly Verduzco, B., Fairhead, J.D., Green, C.M., 2004. New insights into magnetic
interpretation. Geophysics 37 (3), 507517. derivatives for structural mapping. The Leading Edge 23 (2), 116119.
Naidu, P.S., Mathew, M.P., 1998. Analysis of geophysical potential elds. Vaclac, B., Hrabe, J., Segeth, K., 1992. Linear lters for solving the direct problem of
Elsevier, Amsterdam, 298pp. potential elds. Geophysics 57 (10), 13481351.