AMT and CSAMT methods for

hydrocarbon exploration at Nile Delta,

Egypt

Abdellatif Younis, Gad El-Qady,

Mohamed Abd Alla, Mohamed Abdel
Zaher, Amin Khalil, Mohamed Al Ibiary
& Alexander Saraev
Arabian Journal of Geosciences

ISSN 1866-7511

Arab J Geosci
DOI 10.1007/s12517-014-1354-6

1 23
Your article is protected by copyright and
all rights are held exclusively by Saudi
Society for Geosciences. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

1 23
 Author's personal copy
Arab J Geosci
DOI 10.1007/s12517-014-1354-6

ORIGINAL PAPER

AMTand CSAMT methods for hydrocarbon exploration at Nile

Delta, Egypt
Abdellatif Younis & Gad El-Qady & Mohamed Abd Alla &
Mohamed Abdel Zaher & Amin Khalil &
Mohamed Al Ibiary & Alexander Saraev

Received: 26 August 2013 / Accepted: 5 March 2014

# Saudi Society for Geosciences 2014

Abstract The offshore Nile Delta Basin, Egypt, has been estimated and imaged with a maximum thickness of 4,000 m
known for a long time as a significant source of gas and oil. at profile 1 in the northwestern part of the study area.
The study area is located on the Middle Eastern part of the
Nile Delta near Mansoura City. The magnetotelluric method Keywords Hydrocarbon exploration . TEM . MT . Nile
(MT) has emerged as a promising tool for oil exploration than Delta . Egypt
any other geophysical technique. Hydrocarbon reservoirs typ-
ically exhibit higher electromagnetic resistivity than their
surroundings. In this paper, the MT method was used to Introduction
investigate the resistivity subsurface structure that related to
hydrocarbon exploration, and the results were integrated with The Nile Delta is one of the most well-known deltas of the
seismic data collected in the same area. The MT survey was world and has attracted the attention of many geologists due to
performed using two frequency ranges: high (10 Hz to its potential hydrocarbon reserves. This basin contains a thick
100 kHz) and low (0.1 Hz to 1 kHz). Low-frequency natural sequence of potential hydrocarbon source rocks that generate
MT waves were recorded using AMT only, and the measure- essentially gas and condensate. The area of study is located in
ment was then repeated using an artificial signal source for the Middle Eastern part of the Nile Delta of Egypt around
CSAMT. These surveys were carried out along three profiles Mansoura City. It lies between latitudes 30° 57′ 30″ and 31°
having 22 sites. The recorded time-series data were trans- 10′ N and longitudes 31° 20′ and 31° 36′ E with an area of
formed to the frequency domain and processed to determine 566,480 km2 (Fig. 1).
apparent resistivities and phases at each site in which the Magnetotellurics (MT) is an electromagnetic geophysical
determinant average of the impedance tensor was then used method of imaging the earth’s subsurface by measuring natu-
for 1D and 2D inversion. A good coherence was found be- ral variations of electrical and magnetic fields at the earth’s
tween the electromagnetic and seismic profiles and shows that surface. The depth of investigation ranges from 300 m below-
the hydrocarbon is exiting in the Abu Madi and Qawasim ground by recording higher frequencies down to 10,000 m or
channel which represent the main gas-containing layers in the deeper with long-period soundings. Recently, electromagnetic
Nile Delta. The depth and extension of these layers were methods have a wide range of deep investigation especially in
hydrocarbons to treat the defect in traditional seismic
A. Younis (*) : G. El-Qady : M. Abd Alla : M. Abdel Zaher methods, and the MT method can provide information com-
National Research Institute of Astronomy and Geophysics, Helwan, plementary to that derived from seismic exploration.
Egypt Additionally, the MT method is low cost, lightweight, accu-
e-mail: [email protected]
rate, and wide ranging, and involves a minimum of work
A. Khalil : M. Al Ibiary persons compared with the seismic method. As a conse-
Geology Department, Faculty of Science, Helwan University, quence, the MT method represents an important nonseismic
Helwan, Egypt exploration tool, particularly for reconnaissance surveys and
in areas where the seismic reflection method performs poorly.
A. Saraev
Department of Geophysics, Faculty of Geology, St. Petersburg State The purpose of this work is to investigate the possibility of
University, St. Petersburg, Russia exploration of the hydrocarbon reservoir at Nile Delta of
 Author's personal copy
Arab J Geosci

Fig. 1 Location map of the study

area showing the distribution of
the MT sites as well as the drilled
wells. The distance between
stations ranges from 700 m to
1000 m depending on the suitable
space for spreading the electrodes
of electric field

Egypt using the MT method. The essence of the MT method is simulated the crust of Nile Delta by 2D gravity models and
the computation of electromagnetic earth impedance from concluded that the thickness of the Nile Delta exhibits a visible
measurements of orthogonal horizontal magnetic and electric thinning northward. Omran (2000) obtained the same results
fields at the surface. Estimates of impedance magnitude and and showed that the crustal thickness of the Nile Delta varies
phase at various frequencies allow the investigation of elec- between 20 and 32 km with marked thinning and stretching at
trical conductivity as a function of depth. Impedance mea- the Mediterranean coast, extending northerly offshore. Mekkawi
sured at several locations allows the investigation of conduc- et al. 2005 used the MT method to detect geoelectrical conduc-
tivity as a function of horizontal position. tivity anomalies in the earth’s crust and linked them to local
The reliability of the MT method has improved due to seismic activity in the active Kalabsha Fault, Aswan, Egypt.
advances in data acquisition (e.g., Constable et al. 1987;
Nichols et al. 1988), data processing, and improved 1D, 2D,
and even 3D forward and inverse modeling codes (e.g., Smith Geological setting
and Booker 1991; Wynn et al. 2003; Lin et al. 2011; Wiik et al.
2013). The Nile Delta has attracted the attention of many The Nile Delta is not only the oldest known ancient delta, but
researchers due to its potential gas reserves. Sitto (1991) also is the largest and most important depositional complex in
 Author's personal copy
Arab J Geosci

the Mediterranean basin. Furthermore, it is a unique site in the fringes of the Delta. This thickness reaches about
Egypt that is favored for accumulation and preservation of the 500 m in the middle part of the Nile Delta (Said 1981).
Quaternary sediments (Said 1981). The Miocene rocks in the Nile Delta region were
The Quaternary Nile sediments lie unconformably subdivided by E.G.P.C. (1994) into three formations,
over the Pliocene or older sediments in the Nile Delta. from base to top: the Sidi Salim, Qawasim, and
The Nile trough possesses the most complete record of Rosetta, while the Pliocene rocks in the Nile Delta
the Quaternary in Egypt where the sediments assume region are subdivided into three formations, from base
great thickness and are divided into several units, which to top: Abu Madi, Kafr El Sheikh, and El Wastani. The
are unconformably stratified (Said 1981). According to Abu Madi formation is the gas-producing horizon of the
Rizzini et al. (1978), the maximum thickness of the Nile Delta. Pleistocene–Holocene rocks are subdivided
Quaternary succession attained in the Nile Delta is into two formations: the Mit Ghamr formation at the
about 1,000 m, with thinning southwards and towards base and the Bilqas formation at the top (Fig. 2).

Fig. 2 Simplified stratigraphy

column of the Nile delta
(modified after Schumberger
1984)
 Author's personal copy
Arab J Geosci

Fig. 3 Polar diagrams of the elements of the impedance tensor Zxy and Zyx for some MT sites on Profiles 1, 2 and 3
 Author's personal copy
Arab J Geosci

Fig. 4 One-dimensional inversion of MT data at some sites on all Profiles (lined block refers to Bostick inversion and solid red line refers to Occam
smooth inversion)
 Author's personal copy
Arab J Geosci

Fig. 5 One-dimentional model of

sounding 17 correlated with the
nearest stratigraphic well no. 8

The structural setting of the Nile Delta region oc- Field survey and data acquisition
cupies a key position within the plate tectonic develop-
ment of the eastern Mediterranean and the Levant. It Twenty-two audio-magnetotelluric (AMT) stations were mea-
lies on the northern margin of the African plate, which sured in the study area along three profiles; two of them were
extends from the subduction zone adjacent to the Cretan performed above the two major seismic lines that were previ-
and Cyprus arcs to the Red Sea where it drifted apart ously surveyed. A MT survey was performed using a
from the Arabian plate. Stratagem instrument with two frequencies, high (10 Hz to

Fig. 6 Cross sections of apparent

resistivity versus depth using
Bostick and Occam 1D models on
Profile 1. Station locations are
denoted by the triangles
 Author's personal copy
Arab J Geosci

Fig. 7 Cross sections of apparent

resistivity versus depth using
Bostick and Occam 1D models on
Profile2. Station locations are
denoted by the triangles

100 kHz) and low (0.1 Hz to 1 kHz). Higher-frequency data phase, and coherency curves, which helps to process the
are influenced by shallow or nearby features, whereas lower- acquired data in the field. It also permits users to change signal
frequency data are influenced by structures at greater depth. gains to obtain the optimum signal for both electric and
We chose the best locations for the AMT station avoiding the magnetic fields. Nevertheless, some of the noisiest sites were
noises in the area of study to increase the accuracy of the excluded from the database. The Stratagem system consists of
acquired signals. The distance between MT stations ranged two basic components: a receiver and a transmitter. In the
from 700 to 1,000 m depending on the suitable space for high-frequency range, natural signals are typically weak, so
spreading the electrodes of the electric field. The Stratagem the use of a transmitter can improve data quality.
represents a unique AMT system that uses both natural and
artificial electromagnetic signals to obtain a continuous elec-
tric sounding of the earth beneath the measurement site. MT data processing and analysis
Electrode stakes are used to measure electric fields, and highly
sensitive magnetic coils are used to measure magnetic fields. Data processing and interpretation were conducted on the
The instrument is portable, with a rapid setup and teardown. It AMT data by using WinGLink software (Geosystem Srl).
has the ability to display and print out signal amplitudes, For the preliminary data processing, MT data at different

Fig. 8 Cross sections of apparent

resistivity versus depth using
Bostick and Occam 1D models on
Profile3. Station locations are
denoted by the triangles
 Author's personal copy
Arab J Geosci

Fig. 9 Two-dimensional
inversion resistivity models along
Profile 1and 2 using REBOCC 2-
D inversion code

frequencies for the same MT site were merged. Then, these which considers the main trend in the study area. Some
combined data were used to estimate the elements of the MT stations show that the strike direction varies with frequencies,
impedance tensor, which is finally used to determine the reflecting the impedance changes with depth. For example, at
apparent resistivities and phases for both transverse electric station 09, the strike direction is almost N–S for the higher
(TE) and transverse magnetic (TM) modes. The analysis of frequencies which reflect shallower depths and rotating clock-
different periods allows us to obtain information at different wise with lower frequencies, deeper depths, to a NE–SW
depths and to determine the dimensionality of the geological direction. Also, at station 10, the impedance strike rotates
structures whether it is 2D or 3D, strike direction, and their from N–S counterclockwise to NW–SE. In profile 3, the
depth dependence. regional strike is oriented NE–SW and represents the main
trend in this profile as appears in sites 16 and 17. Additionally,
the impedance tensor was rotated to the strike direction, which
Dimensionality of MT data is almost the same as the xy polarization trend. Thus, the xy
polarization is assigned for TE polarization, and yx polariza-
An important task for any MT study is the analysis of the tion defines the TM polarization.
dimensionality of the data, in particular of the magnetotelluric
tensor, which gives rise to the modeling of the geoelectric
structures. Thus, before the conversion of MT data into a
resistivity model of the subsurface, it is essential to understand 1D inversion of AMT data
the dimensionality of the MT data. The impedance polar plots
provide a measure of the MT data dimensionality (Reddy et al. 1D inversion of AMT data was conducted using two types of
1977). MT tensors have significantly different forms depending algorithms: Bostick and Occam. The Bostick and Occam
on whether the subsurface structure is approximately 1D, 2D, (Bostick 1977) are based on information from stratigraphic
or 3D. In 2D, the MT tensor is decoupled into two independent well data. Examples are shown in Fig. 4, where the lined
modes: the TE mode, where the electric axis is parallel to the blocks refer to Bostick and the bold blocks to the Occam.
strike, and the TM mode, where the electric axis is perpendic- Bostick transformation represents one of the simplest ways to
ular to the strike (Simpson and Bahr 2005). Polar diagrams are invert the MT data. This heuristic inversion scheme generates
given for the magnitude of one particular tensor element a near-continuous resistivity distribution versus depth
plotted versus the angle of rotation of the axes (Bostick 1977). On the other hand, Occam inversion automat-
(Tammemagi and Lilley 1973). These polar diagrams plot ically produces models with a maximally smooth resistivity
a graphical representation of the impedance elements when
the coordinates are rotated for some of the recorded fre-
quencies All MT sites in the study area have polar plots
that indicated 3D character in the lower frequencies
(Fig. 3). These diagrams consist of two curves: the outer
curves are the traces of Zxy, while the inside curves repre-
sent the traces of Zyx. The bold cross beside these curves
reflects the regional strike direction, which guides us to
differentiate between the two polarizations, TE and TM.
Most of the AMT stations, for example sites 01 and 02 in
profile 1, 09 and 10 in profile 2, and sites 16 and 17 in profile
3, show moderate coherence in the polarization directions for Fig. 10 Two-dimensional inversion resistivity model along Profile 3
all frequencies, revealing a regional strike of almost NE–SW using REBOCC 2-D inversion code
 Author's personal copy
Arab J Geosci

structure. The Occam algorithm leads to a simple model

containing the essential properties of all possible models
fitting the MT data (Constable et al. 1987). The D+ smoothing
technique was used to smooth resistivity and phase curves,
instead of some other numerical smoothing techniques.
Namely, this technique relates the apparent resistivity and
phase of the same component through a D+ function
(Beamish and Travassos 1992). In essence, this finds the 1D
earth which best fits both parameters. We found that smooth-
ing the curves by using the D+ technique produces more
accurate results for 1D inversion than using a numerical
smoothing method.
Generally, the better the fit between measured and predict-
ed data, the better the quality of the inversion based on prior
information about the geology and stratigraphy from litholog-
ical wells. Figure 5 shows a 1D model of site 17 constructed
based on stratigraphic well 8 which is located near the site. All
soundings show the increase of resistivity with depth. This can
be interpreted as the presence of highly resistive rock type. For
easy interpretation, cross sections were created using Bostick
Fig. 11 misfits of modeled to observed aperant resistivities of all MT
sites on profile 1, 2 and 3. Station locations are denoted by the triangles and Occam 1D models along each profile (Figs. 6, 7, and 8).
These sections elucidate the subsurface geological situation
below the study area by correlating MT soundings with each

Fig. 12 The interpreted seismic lines cross profile 1 and 2, made by El Mansoura Petroleum Company and Melron Company
 Author's personal copy
Arab J Geosci

other along each profile and hence illuminating the lateral Conclusion
change of resistivity along the profiles.
The Nile Delta of Egypt, both onshore and offshore, shows
promise and a huge potential as a hydrocarbon source, and it is
considered as the most important gas-producing province in
2D inversion of MT data Egypt. The processing and analysis of the MT data acquired in
the study area permitted us to characterize the subsurface
The determinant average of the impedance tensor from electric structure. The determinant average of the impedance
the best MT sites was inverted by the reduced basis tensor was inverted for 1D and 2D and reveals the occurrence
Occam’s inversion (REBOCC) 2D inversion code. The of a resistive region where the resistivity is above 500 Ω m.
code is based on a variant of the Occam algorithm (de This resistive zone is related to the presence of a gas-bearing
Groot-Hedlin and Constable 1990), in which the solution layer of the Abu Madi and Quasim formations. An agreement
of the nonlinear inversion is obtained in the model pa- was found with interpreted seismic lines which were made by
rameter space of the model by the Gauss–Newton meth- the El Mansoura Petroleum Company and Merlon Company,
od. Techniques of space reduction are used to solve the also with oil well logs that are located in the study area.
resulting system of linear equations (REBOCC,
Siripunvaraporn and Egbert 2000). The determinant data Acknowledgments We acknowledge the staff of the National Research
provides a useful average of the impedance tensor for all Institute of Astronomy and Geophysics, Egypt, for offering continuous
current directions (Pedersen and Engels 2005). The goal help in this work. Sincere thanks go to the El Mansoura Petroleum
of the inversion is to find the minimum structure model Company for giving data and permission for publication.
subject to a desired misfit level. The obtained model
fitted well the experimental data. Figures 9 and 10 show
the 2D inversion model along profiles 1, 2, and 3 (RMS References
equals 3.7, 3.5, and 1.95, respectively). The 2D inversion
gives reasonable results for the subsurface structure of Beamish D, Travassos JM (1992) The use of the D+ solution in
the study area. The 2D models can help in delineating magnetotelluric interpretation. J Appl Geophys 29:1–19
Bostick FX (1977) A simple almost exact method of MT analysis.
the subsurface reservoir which is corresponding to the Workshop on Electromagnetic Methods in Geothermal Exploration,
shallow conductive region (20–60 Ω m) above the highly Snowbird, Utah, U.S, Geol, Surv., Contract no, 14080001-8-359.
resistive region. Constable SC, Parker RL, Constable CG (1987) Occam’s inversion: a
The 2D inversion shows that the subsurface sequences practical algorithm for generating smooth models from electromag-
netic sounding data. Geophysics 52:289–300
along the three profiles can be divided into three main layers de Groot-Hedlin C, Constable SC (1990) Occam's inversion to generate
according to the resistivity and geological information from smooth, two-dimensional models from magnetotelluric data.
oil well logs. The upper layer concerns the sand layer of Mit Geophysics 55: 1613–1624
Ghamr and El Wastani formations, which represent the main E.G.P.C. (Egyptian General Petroleum Cooperation) (1994) Nile Delta and
North Sinai: A Field, Discoveries and Hydrocarbon Potentials (A
aquifer of the Nile delta. Below this layer lies the shale layer of Comprehensive Overview), Egyptian General Petroleum, Cairo,
the Kafer El Sheikh formation which extended to 3,000 m and Egypt.
overlay on the Abu Madi and Qawasim formations. The Abu Lin C, Tan H, Tong T (2011) Three-dimensional inversion of CSAMT
Madi and Qawasim have depths ranging between 1,000 m and data using conjugate gradient method. International Workshop on
Gravity, Electrical & Magnetic Methods and Their Applications,
more than 4,000 m in profiles 1 and 2. However, this layer Beijing, China
ranges between 500 and 4,000 m in profile 3 from south Battra Nichols EA, Morrison HF, Clarke J (1988) Signals and noise in measure-
to west Dikirnis. The gas bearing of the subsurface is founded ments of low-frequency geomagnetic fields. J Geophys Res 93:743–
in the third of the section related to the shaley sand of the Abu 754
Mekkawi M, Schnegg P, Arafa T, Elathy E (2005) Electrical structure of
Madi and Qawasim formations which have highly resistive the tectonically active Kalabsha Fault, Aswan, Egypt. Earth Planet
values that related to the hydrocarbon which is characterized Sci Lett 240:764–773
by high resistivity. Omran MA (2000) Crustal modeling of the continental margin of the Nile
Figure 11 shows the data misfits of all inverted sites at Delta region. Ann Geol Surv Egypt v. xxiii:381–387
Pedersen LB, Engels M (2005) Routine 2D inversion of magnetotelluric
different periods that define the difference between the ob- data using the determinant of the impedance tensor. Geophys 70:33–
served and modeled data. The inverted data of all sites show 41
acceptable misfit at short periods while some sites, for exam- Reddy IK, Rankin D, Phillips RJ (1977) Three-dimensional modeling in
ple site 10, show poor misfit at long periods. The 2D inversion magnetotelluric and magnetic variational sounding. Geophys J R
Astron Sot 51:313–325
of the MT data shows matching with seismic lines measured Rizzini A, Vezzani F, Cococcetta V, Milad G (1978) Stratigraphy and
before by the El Mansoura Petroleum Company and Merlon sedimentation of Neogene-Quaternary section in the Nile Delta area
Company (Fig. 12). (A.R.E.). Mar Geol 27:327–348
 Author's personal copy
Arab J Geosci

Said R (1981) The geological evolution of the river Nile. Springer, Smith JT, Booker JR (1991) Rapid inversion of two- and three-
Netherlands, p 151 dimensional magnetotelluric data. J Geophys Res 96:3905–
Schlumberger (1984) Well Evaluation Conference, Egypt. Geol Egypt 1: 3922
1–64 Siripunvaraporn W and Egbert G (2000) An Efficient data- Tammemagi HY, Lilley FEM (1973) A magnetotelluric traverse in
subspace inversion method for 2D magnetotelluric data. Geophysics Southern Australia. Geophys JR Astr Sot 31:433–445
65,791–803. Wiik T, Hokstad K, Ursin B, Mütschard L (2013) Joint contrast source
Siripunvaraporn W, Egbert G (2000) An Efficient data-subspace inversion of marine magnetotelluric and controlled-source electro-
inversion method for 2D magnetotelluric data. Geophysics 65: magnetic data. Geophysics 78(6):E315–E327
791–803 Wynn JC, Gray F, Nordstrom TE, Dexin Liu, Reed EF, Villaseñor FA,
Sitto AA (1991) A crustal model for Nile Delta, Egypt. Egypt J Geol Gerry Connard (2003) Using analytic signal analysis on aeromag-
34(1–2):279–292 netic data to constrain AMT inversions, Sonora San Pedro Basin,
Simpson F, Bahr K (2005) Practical magnetotelluric. Cambridge Mexico. Symposium on the Application of Geophysics to
University Press, Cambridge, 270p Engineering and Environmental Problems, 1–11