GRAV-MAG SUITE: An open source MATLAB-based program for processing potential

field data
Castro, F.R., UFPR; Oliveira, S.P.*, UFPR; de Souza, J., Secretary of Education/PR; Ferreira, F.J.F., UFPR

Copyright 2018, SBGf - Sociedade Brasileira de Geofísica

Program interface
Este texto foi preparado para a apresentação no VIII Simpósio Brasileiro de Geofísica,
Salinópolis, 18 a 20 de setembro de 2018. Seu conteúdo foi revisado pelo Comitê
Técnico do VIII SimBGf, mas não necessariamente representa a opinião da SBGf ou de GRAV-MAG SUITE is composed by a set of MATLAB
seus associados. É proibida a reprodução total ou parcial deste material para propósitos functions. The program functions are categorized into two
comerciais sem prévia autorização da SBGf.
____________________________________________________________________ types: GUI functions, and processing functions. The first
type are functions responsible to generate the program
Abstract graphical user interfaces, and the second type are
GRAV-MAG SUITE is a MATLAB-based program that has functions which perform some processing task.
been developed for processing potential field data. The Among the GUI functions there is a main function which
program is composed by several tools, which involve generates the core interface. This graphical interface is
miscellaneous field transformations and enhancing filters, composed by two fields. The first field is the menu field,
semiquantitative methods, and simple shape forward which links the core interface with all other program GUIs.
modeling algorithms. GRAV-MAG SUITE provides a The second field presents the current data the user is
graphical user interface (GUI) that facilitates the user’s processing and has some GUI components to customize
interaction with the program. In order to illustrate the the data visualization (Fig. 1).
program functionalities, a set of tools will be applied on a
synthetic data composed by six prisms representing the
most typical geological structures.

Introduction
Airborne potential field data has been a cornerstone of the
mineral industry through the last decades. The use of
gravity and magnetic data plays an important role in
identify subtle features related to sources with economic
value. However, potential field data has a frequency
content that may evidence details regarding both shallow
and deep causative sources, and the extraction of
information related to specific frequency bands depends on
the enhancement technique applied to data. In this context
a set of qualitative and semiquantitative techniques has
been developed (Nabighian 1972; Roest et al. 1992; Figure 1 – Main GRAV-MAG SUITE graphical user
Cordell and Grauch 1985; Miller and Singh 1994; Cooper interface with the synthetic data loaded.
and Cowan 2006; Ferreira et al. 2013; Thompson 1982;
Salem et al. 2007, Cooper 2015). In the File menu, the user can find options to open a
dataset disposed on xyz column format or in a matrix form,
Currently, the processing of potential field data is carried view the data statistics, and export the data in a common
out mostly by commercial software packages. However, figure formats. The Profile menu provides a set of options
another strand, encompassing open source software has to process a profile data, for instance, to perform numerical
been taken strength since new technologies have made its differentiation and apply some enhancement filters, and to
development possible (Mendonça and Meguid 2008; extract profiles from gridded datasets. The Grid menu
Arisoy and Dikmen 2011). In this scenario we present a provides some grid operations like interpolation,
MATLAB-based open source software for processing both windowing, regrid, and trend removal.
gravity and magnetic data.
In the Field Transformations menu there is a set of linear
Several program functionalities will be tested on synthetic Fourier filters, for example, derivative filter on 3D
data composed by prismatic sources whose dimensions orthogonal directions, upward/downward continuation,
simulate common geologic features. In order to validate the vertical integration, reduction to the magnetic pole and
implemented methods, the source parameters estimated equator, directional cosine filter, anisotropic diffusion filter,
by the qualitative and semiquantitative methods will be among others.
compared to the exact parameters used to generate the
synthetic model. The Enhancement Filters menu is divided into two
categories. The first category groups qualitative filters
commonly used to enhance or suppress some data
frequency bands of gridded datasets. The second one
gathers the semiquantitative methods to estimate some

VIII Simpósio Brasileiro de Geofísica

 GRAV MAG SUITE: AN OPEN SOURCE MATLAB-BASED PROGRAM FOR PROCESSING POTENTIAL FIELD DATA
____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ 2

source parameters like horizontal location, depth to the top horizontal thin slab, and a culvert, and prism F represents
and bottom, physical property contrast, and dip, etc. a contact model.

Program functionalities
In order to illustrate the program functionalities, a synthetic
model with six prisms was generated to simulate interfering
dike-like bodies, contacts, and prismatic bodies with
different strike directions, and depths to the top (Fig. 2).

Figure 2 – 3D representation of synthetic model.

The model parameters are shown in Table 1. The strength,
declination, and inclination of the induced field vector are
57000 nT, 0, and 90 degrees, respectively, and no
remanent magnetization was considered. The magnetic
anomaly was computed in grid with mesh ranging from 0
to 20 km in both x and y directions, with a squared grid cell Figure 3 – Total magnetic intensity of synthetic data. a)
of dimension 100x100 meters. Linear color distribution (MATLAB’s default). b) Histogram
Table 1 - Spatial and physical parameters of the synthetic equalized color distribution.
model. X, and Y are the coordinates of the prisms, and Z The magnetic anomaly shown in Fig. 3 was generated
are the prisms depth to top. using GRAV MAG PRISM, a MATLAB-based forward
modeling program for both magnetic and gravity fields
Strike direction
Thickness (km)

generated by prismatic bodies (Bongiolo et al. 2013).

Length (km)

Width (km)

GRAV MAG SUITE offers a GUI function that calls the

(degrees)
X (km)

Y (km)

Z (km)
Prism

original GRAV MAG PRISM scripts. (Fig. 4). GRAV MAG

PRISM GUI has entry components for both field and
source parameters, needed for the computation of the
gravity and magnetic anomaly fields.

A 10 10 0.1 50 0.1 20 -45

B 4 15 0.3 50 0.05 20 45
C 7.5 10 0.4 10 0.2 0.5 20
D 15 15 0.5 7 3 0.2 120
E 15 4 0.8 6 1.5 5 50
F -1.5 -17 0.7 50 10 20 0

Figure 3 shows the total magnetic intensity (TMI) computed

using the model of Figure 2. Prisms representing the
bodies A and B have dimensions which characterize dike-
like bodies, while prisms C, D, and E characterize a duct, Figure 4 – Graphical user interface of Grav Mag Prism.

VIII Simpósio Brasileiro de Geofísica

 Castro, F.R.; Oliveira, S.P.; de Souza, J.; Ferreira, F.J.F.
____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
3

GRAV MAG SUITE performs forward modeling of two-

dimensional bodies with simple shapes, for example, dike-
like, spherical, and cylindrical bodies using closed formula
(Telford et al. 1990), and irregular cross-section bodies
using the line integral approach (Hubbert 1948; Talwani et
al. 1959; Won and Bevis 1987) (Fig. 5).

Figure 7 – Profile extraction GUI. a) Main profile extraction

interface. b) Extracted profile.
Figure 7 illustrates the profile extraction tool present in the
Profile menu of GRAV-MAG SUITE. This tool allows the
user to extract profiles in any direction of the anomaly map,
and the profile can be a single line or a polyline, which
comprises a multiple line segments profile.
To select a region rather than a profile, one can use the
Window Grid GUI present in the Grid menu. This tool
provides some options that allow the user to apply a
window to a gridded dataset. This window can be
rectangular, polygonal, or irregularly shaped (Fig. 8).
Figure 5 – Other forward modeling GUIs. a) Dike-like body
forward modeling GUI with magnetic anomaly caused by a
45 degrees dipping dike. b) Irregular shaped body forward
modeling GUI with both gravity and magnetic anomalies
caused by three irregular cross-section bodies.
To grid scattered data the program offers traditional
interpolation methods like bilinear, bicubic, nearest
neighbor, and natural neighbor (Verbyla 2003). When an
anisotropic sampling rate is verified, as in airborne surveys,
these interpolate methods generate spatial aliasing of
short-wavelength features, like boudinage effects. This
issue can be overcome by using an anisotropic diffusion
filter (Smith and O’Connell, 2005) (Fig. 6).

Figure 6 – Aeromagnetic data of southeast São Paulo

State. a) Survey gridded by bilinear interpolation method.
b) Same survey after applying anisotropic diffusion filter.
Once the anomaly maps are generated, a set of processing
tools can be applied, for instance, a profile extraction (Fig. Figure 8 – Window grid GUI. a) Main window grid graphical
7), windowing some anomaly map (Fig. 8), derivative filters user interface. b) Anomaly map windowed by a rectangular
(Fig. 9), and field continuation filters (Fig. 10). window (white line rectangle in a). c) Anomaly map
windowed by an irregular window.

VIII Simpósio Brasileiro de Geofísica

 GRAV MAG SUITE: AN OPEN SOURCE MATLAB-BASED PROGRAM FOR PROCESSING POTENTIAL FIELD DATA
____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ 4

GRAV MAG SUITE has support for different derivative Several enhancement filters or qualitative methods are
algorithms. It is possible the take the derivative in the three implemented in GRAV-MAG SUITE. Figure 12 illustrates
orthogonal directions, products necessary for computing some of these filters: ASA (Nabighian 1972; Roest et al.
several enhancement techniques (Fig. 9). All filters 1992), THDR (Cordell and Grauch 1985), TDR (Miller and
implemented on GRAV-MAG SUITE prepare the input grid Singh 1994), TDX (Cooper and Cowan 2006), and TAHG
by expanding, filling NaN/dummy values, and tapering, to (Ferreira et al. 2013).
prevent edge noise introduced by Fourier operations.
Furthermore, a set of semiquantitative methods are also
implemented in GRAV-MAG SUITE, such as, Euler
deconvolution (Thompson 1982), tilt-depth (Salem et al.
2007), source distance (Cooper 2015), and Signum
transform (de Souza and Ferreira 2015; Oliveira et al.
2017).

Figure 9 – Derivative filter. a) Graphical user interface of

derivative filter. b) First order derivative in x direction. c)
First order derivative in y direction. d) First order derivative
in z direction.

Figure 11 – Depth solutions estimated by semiquantitative

methods. a) Tilt-depth solutions. b) Euler deconvolution
depth solutions for a structural index of 1 and window size
of 10x10 grid points (1000x1000 meters).
Figure 11 shows the depth estimates of two
semiquantitative methods. Although Tilt-depth method
(Salem et al. 2007), represented in Figure 10a is
recommended only to approximate the depth of contact
models, a good depth estimation of the other sources was
verified. The same was verified in Euler deconvolution
depth solutions except for the contact model whose depth
was overestimated.

Conclusions
A set of processing tools, usually restricted to commercial
software, has been implemented for use in MATLAB in the
Figure 10 – Field continuation filter. a) Field continuation form of an integrated system with a graphical user-friendly
GUI. b) Anomaly of Figure 3 upward continued by a height interface.
of 500 meters.

VIII Simpósio Brasileiro de Geofísica

 Castro, F.R.; Oliveira, S.P.; de Souza, J.; Ferreira, F.J.F.
____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
5

By enabling the access of commercial software algorithms Hubbert, M. K., 1948. A line-integral method of computing
to the community, we facilitate the reproduction of the gravimetric effects of two-dimensional
academic works for validation purposes, as well as its masses. Geophysics, 13(2): 215-225.
application and improvements in new researches.
Mendonça, C. A., Meguid, A. M. A., 2008. Programs to
Semiquantitative methods implemented on GRAV-MAG compute magnetization to density ratio and the
SUITE have estimated the sources parameters with magnetization inclination from 3-D gravity and magnetic
acceptable error, even when the methods have intrinsic anomalies. Computers and Geosciences, 34(6): 603-610.
limitations on situations whose source’s shape leads to
poor estimates. Miller, H. G. and Singh, V., 1994. Potential field tilt - A new
concept for location of potential field sources: Journal of
The program should be available by April 2019, upon Applied Geophysics, 32: 213-217.
completion of the related master’s degree research. Until
the final release, more functionalities are planned to be Nabighian, M. N., 1972. The analytic signal of two-
included, such as more interpolation methods like dimensional magnetic bodies with polygonal cross section:
minimum curvature (Briggs, 1974), improvements of Its properties and use for automated anomaly
source distance method (Cooper, 2015), among others. interpretation: Geophysics, 37: 507-517.
Oliveira, S. P., Ferreira, F. J. F., de Souza, J., 2017.
Acknowledgments EdgeDetectPFI: An algorithm for automatic edge detection
in potential field anomaly images–application to dike-like
The authors thank the support of Laboratório de Pesquisas magnetic structures. Computers and Geosciences, 103:
em Geofísica Aplicada (LPGA) for providing the 80-91.
infrastructure required for the implementation of GRAV-
MAG SUITE. SP Oliveira and FJF Ferreira are supported Roest, W. R, Verhoef, J. and Pilkington, M., 1992.
by CNPq (grants 313100/2017-9 and 306978/2015-6, Magnetic interpretation using the 3-D analytic signal:
respectively). Geophysics, 57(1): 116-125.
Salem, A., Williams, S., Fairhead, J. D., Ravat, D. and
References Smith, R., 2007. Tilt-depth method: A simple depth
Arisoy, M. Ö. and Dikmen, Ü., 2011. Potensoft: MATLAB- estimation method using first-order magnetic derivatives:
based software for potential field data processing, The Leading Edge, 26(12): 1502-1505.
modeling and mapping. Computers and geosciences, Smith, R. S., O'Connell, M. D., 2005. Interpolation and
37(7): 935-942. gridding of aliased geophysical data using constrained
Bongiolo, A. B. S., de Souza, J., Ferreira, F. J. F. and anisotropic diffusion to enhance trends. Geophysics,
Castro, L. G., 2013. GRAV MAG PRISM: a Matlab/Octave 70(5): V121-V127.
program to generate gravity and magnetic anomalies due Talwani, M., Worzel, J. L., Landisman, M., 1959. Rapid
to rectangular prismatic bodies: Revista Brasileira de gravity computations for two‐dimensional bodies with
Geofísica, 31: 347-363.
application to the Mendocino submarine fracture
Briggs, I. C., 1974. Machine contouring using minimum zone. Journal of geophysical research, 64(1): 49-59.
curvature. Geophysics, 39(1): 39-48.
Telford, W., Geldart, L., Sheriff, R., 1990. Applied
Cooper, G. R. J. and Cowan, D. R., 2006. Enhancing Geophysics. Cambridge University Press.
potential field data using filters based on the local phase:
Thompson, D. T., 1982. EULDPH: A new technique for
Computers and Geosciences, 32: 1585-1591. making computer-assisted depth estimates from magnetic
Cooper, G. R. J., 2015. Using the analytic signal amplitude data: Geophysics, 47(1): 31-37.
to determine the location and depth of thin dikes from
Verbyla, D. L., 2003. Practical GIS analysis. CRC press.
magnetic data: Geophysics, 80(1): J1-J6.
Won, I. J., Bevis, M., 1987. Computing the gravitational
Cordell, L. and Graunch, V. J. S., 1985. Mapping basement and magnetic anomalies due to a polygon: Algorithms and
magnetization zones from aeromagnetic data in the San
Fortran subroutines. Geophysics, 52(2): 232-238.
Juan Basin, New Mexico. In: Hinze WJ (Ed). The Utility of
Regional Gravity and Magnetic Anomalies Maps. Society
of Exploration Geophysicists: 181-197.
de Souza, J., Ferreira, F. J. F, 2015. The application of the
Signum transform to the interpretation of magnetic
anomalies. ASEG-PESA 2015, Geophysics and Geology
together for Discovery, 24th International Geophysical
Conference and Exhibition, Expanded Abstracts.
Ferreira, F. J. F., de Souza, J., Bongiolo, A. B. S. and de
Castro, L. G., 2013. Enhancement of the total horizontal
gradient of magnetic anomalies using the tilt angle:
Geophysics, 78: J33-J41.

VIII Simpósio Brasileiro de Geofísica

 GRAV MAG SUITE: AN OPEN SOURCE MATLAB-BASED PROGRAM FOR PROCESSING POTENTIAL FIELD DATA
____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ 6

Figure 12 – Qualitative enhancement filters. a) ASA; b) THDR; c) TDR; d) TDX; e) TAHG.

VIII Simpósio Brasileiro de Geofísica