Content uploaded by Carmen Cuenca-García
Author content
All content in this area was uploaded by Carmen Cuenca-García on Dec 08, 2015
Content may be subject to copyright.
Best Practices of
GeoInformatic
Technologies for
the Mapping of
Archaeolandscapes
Edited by
Apostolos Sarris
Archaeopress Archaeology
Archaeopress Publishing Ltd
Gordon House
276 Banbury Road
Oxford OX2 7ED
www.archaeopress.com
ISBN 978 1 78491 162 1
ISBN 978 1 78491 163 8 (e-Pdf)
© Archaeopress and the authors 2015
All rights reserved. No part of this book may be reproduced, stored in retrieval system,
or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise,
without the prior written permission of the copyright owners.
Printed in England by
This book is available direct from Archaeopress or from our website www.archaeopress.com
i
Table of Contents
Preface ���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������iii
Images of the Past: Magnetic Prospection in Archaeology ���������������������������������������������������������������������������������������������� 1
Kayt L. Armstrong and Tuna Kalayci
GPR: Theory and Practice in Archaeological Prospection ������������������������������������������������������������������������������������������������13
Meropi Manataki, Apostolos Sarris, Jamieson C. Donati, Carmen Cuenca Garcia and Tuna Kalayci
Identification of Shapes and Uses of Past Landscapes through EMI Survey �������������������������������������������������������������������25
François-Xavier Simon and Ian Moffat
Seismic Geophysical Methods in Archaeological Prospection �����������������������������������������������������������������������������������������35
Pantelis Soupios
Locating Graves with Geophysics ���������������������������������������������������������������������������������������������������������������������������������������45
Ian Moffat
Exploring the Interior of Tumuli: Examples from Investigations in Macedonia and Thrace�������������������������������������������55
Gregory N. Tsokas, Panagiotis I. Tsourlos and Georgios Vargemezis
Off-Shore Archaeological Prospection Using Electrical Resistivity Tomography�������������������������������������������������������������63
Kleanthis Simyrdanis, Nikos Papadopoulos and Theotokis Theodoulou
Data Integration in Archaeological Prospection����������������������������������������������������������������������������������������������������������������71
Tuna Kalayci
Overview of Underwater Archaeological Research with Advanced Technologies in Greece �����������������������������������������85
Theotokis Theodoulou
Aerial Reconnaissance in Archaeology – from Archives to Digital Photogrammetry ���������������������������������������������������103
Gianluca Cantoro
On the Use of Satellite Remote Sensing in Archeology ��������������������������������������������������������������������������������������������������115
Athos Agapiou, Dimitrios D. Alexakis, Apostolos Sarris and Diofantos G. Hadjimitsis
Cities and Satellites: Discovering Ancient Urban Landscapes through Remote Sensing Applications ������������������������127
Jamieson C. Donati
More Than Line of Sight and Least Cost Path. An Application of GIS to the Study of the Circular Tombs of
South-Central Crete ����������������������������������������������������������������������������������������������������������������������������������������������������137
Sylviane Déderix
Mixed Reality Applications, Innovative 3d Reconstruction Techniques & Gis Data Integration for
Cultural Heritage ���������������������������������������������������������������������������������������������������������������������������������������������������������149
Lemonia Argyriou and Nikos S. Papadopoulos
Interpreting the Past through Agent-Based Modeling and GIS ��������������������������������������������������������������������������������������159
Angelos Chliaoutakis and Georgios Chalkiadakis
Geomorphometry, Multi-Criteria Decision Analysis and Archaeological Risk Assessment �����������������������������������������171
Athanasios Argyriou, Apostolos Sarris and Richard Teeuw
Adding a Geographical Component in Cultural Heritage Databases �����������������������������������������������������������������������������181
Poulicos Prastacos and Eleni Gkadolou
Historical Maps on the Semantic Web ���������������������������������������������������������������������������������������������������������������������������� 189
Eleni Gkadolou and Emmanuel Stefanakis
Archaeomagnetic Method as a Dating Tool: Application to Greek Archaeological Sites from Prehistoric to
Byzantine Periods �������������������������������������������������������������������������������������������������������������������������������������������������������199
Despina Kondopoulou and Elina Aidona
Geoarchaeology: a Review in Techniques ����������������������������������������������������������������������������������������������������������������������209
Eleni Kokinou
Inorganic Geochemical Methods in Archaeological Prospection ����������������������������������������������������������������������������������� 219
Carmen Cuenca-García
Mineralogical and Petrographic Techniques in Archaelogy ��������������������������������������������������������������������������������������������231
Georgia Karampatsouand and Theodoros Markopoulos
Provenance of Ceramics: Methods and Practices ����������������������������������������������������������������������������������������������������������239
Nikolaos A. Kazakis and Nestor C. Tsirliganis
ii
Sustainable Data Management in the Study of Ancient Materials – Using the Example of Archaeological
Ceramics ����������������������������������������������������������������������������������������������������������������������������������������������������������������������251
Anno Hein and Vassilis Kilikoglou
Laser Tools in Archaeology and Conservation How Far Can We Get? ����������������������������������������������������������������������������261
A. Philippidis, P. Siozos, Z.E. Papliaka, K. Melessanaki, K. Hatzigiannakis, M. Vakondiou, G. Manganas,
K. Diamanti, A. Giakoumaki and D. Anglos
iii
Preface
New geoinformatic technologies have recently had a transformative effect on landscape archaeology, particularly by
facilitating the high resolution acquisition and analysis of data over large areas� These techniques have fundamentally
changed the nature and scope of questions that can be addressed regarding the archaeological record� Despite this
stimulating potential, many practising archaeologists were not trained in these methods and so are not fully aware of
their capabilities or the most appropriate ways to apply them� This volume collates state of the art research in the
fields of geophysics, geochemistry, aerial imaging, dating, digital archaeology, GIS and marine archaeology to present
a comprehensive overview of the specialised techniques which can contribute to landscape scale archaeological
investigations� It is hoped that it will serve as a ‘best practice’ guide for their use and encourage their widespread adoption
by the archaeological community�
Geophysical survey has been revolutionised by the wide scale availability of high quality positioning systems such as
differential GPS and the introduction of fast sampling multi-sensor arrays� These have freed geophysicists from traditional
small and regularly spaced grids and allowed them to collect high density data over large areas extremely quickly�
This significant increase in areal coverage, does not simply increase the amount of data available, but fundamentally
revises the way in which these techniques can examine the spatial distribution of archaeological material� It moves
geophysics from just being a technique of ‘archaeological prospection’ to being one that can answer nuanced questions
about the interaction between the landscape and humans in ways that are impractical through conventional excavation
and pedestrian survey techniques� Not to mention that the playground of the application of these techniques becomes
even more challenging when the archaeological context is not the usual one (e�g� off-shore prospection)� Another major
advance in geophysical survey is the prospects offered by data fusion for quantitatively combining different geophysical
data in a rigorous fashion, which is not coloured by preconceptions about the data or study area�
On a much larger scale, aerial and satellite imaging also provide important new insights into archaeological landscapes�
A particularly popular approach in recent years is the use of structure from motion photogrammetry to create composite
orthophotos and digital elevation models from aerial platforms such as kites, balloons and drones� This has been mainly
driven by the widespread availability of cheap unmanned aerial platforms and high resolution, lightweight digital
cameras� This advance has been augmented by the widespread availability of high resolution satellite imagery, allowing
the subtle surface expression of archaeological features to be mapped remotely� The aerial and satellite perspective of the
archaeological sites is particularly suited to the study of the archaeolandscapes as it positions archaeological material in
the context of the broader area, including the topography, geomorphology, vegetation and hydrology�
Other spatially based digital techniques such as GIS, visualisation and agent based modelling also provide new tools for the
analysis and dissemination of archaeological information� GIS modelling of geomorphic, topographical and geological
parameters coupled with the spatial distribution of the cultural residues provide alternative ways to model the interactive
responses between humans and environment through time� This is especially obvious when agent based modelling (ABM)
is used to simulate the interactions between individual and collective entities in the landscape� Additionally, digital
techniques allow the archiving and analysis of historical information in ways that facilitate new interpretations and mixed
reality approaches provide stimulating means to reconstruct monuments and interpret and represent sites�
A number of other scientific methods have made an important contribution to the documentation of archaeological
landscapes� For example, maritime archaeology has benefited from the wide scale availability of geophysical equipment
as well as new technologies that allow the inspection of archaeological features below the depth of conventional SCUBA
investigation� This includes diving technology such as the Exosuit and unmanned underwater vehicles� The use of
geochemical techniques in archaeology has been facilitated by the use of low cost, high throughput methods such ICP-
AES and essentially non-destructive methods such as laser ablation sampling� Also of great interest are new portable
field techniques such as pXRF which allow immediate feedback on composition during archaeological investigations
and facilitate the informed guidance of sampling for more elaborate analysis� The above techniques coupled with other
approaches such as dating methods and provenance studies can create a more integrated framework for the study of the
archaeolandscapes�
In conclusion, new geoinformatic technologies provide exciting new tools for the investigation and documentation of
archaeological landscapes� This volume summaries the current best practices to enable archaeologists and practitioners
to gain a further understanding of the current ‘state of the art’ across a broad disciplinary range� These methods are rapidly
evolving and many new developments are expected soon however it is hoped that this book will serve as an impetus for
archaeologists and cultural heritage professionals to integrate these techniques within their own research in a useful and
productive fashion�
iv
The compilation of the articles and the publication of the volume was performed in the framework of the POLITEIA
research project, Action KRIPIS, project No MIS-448300 (2013SE01380035) that was funded by the General Secretariat
for Research and Technology, Ministry of Education, Greece and the European Regional Development Fund (Sectoral
Operational Programme : Competitiveness and Entrepreneurship, NSRF 2007-2013)/ European Commission�
Dr Apostolos Sarris
Director of Research
Laboratory for Geophysical-Satellite Remote Sensing & Archaeo-environment (GeoSat ReSeArch LAB)
Institute for Mediterranean Studies
Foundation for Research and Technology, Hellas
Rethymno, Crete, Greece
13
Introduction
Ground Penetrating Radar (GPR) is a non-destructive
electromagnetic (EM) geophysical technique that uses
radio waves, in the frequency range of 10MHz to 2GHz,
to map the subsurface� The first reported attempt of using
radio wave signals to measure subsurface features was by
El-said (1956) who tried to image the depth of a water table�
The development of the method accelerated considerably
after 1970 as a result of the tremendous progress that took
place in electronics and computer technology but it was
after 1985 when the real explosion of the advancement
of GPR occurred (Annan 2002)� During this period GPR
technology became better known worldwide and more
affordable and the scope of its applications expanded�
As a consequence, the strengths and weaknesses of GPR
were better understood and this opened new ways into
hardware development for further improvements� Recent
advancement includes the multichannel system that
greatly improved the speed, the area coverage and the
spatial resolution (Goodman and Piro 2013)�
GPR can be used in a series of applications like the
mapping of the bedrock depth (Davis and Annan 1989),
the determination of the stratum thickness and the aquifer
depth (Doolittle et al. 2006), the location of physical and
artificial cavities in the subsurface (Benson 1995) and
detecting fracture zones (Grasmueck 1996, Theune et al.
2006)� It is widely used in archaeological prospection
for the detection of numerous archaeological features
(Conyers, 2012)� Among them are the mapping of graves
in cemeteries where location of both wooden and metal
coffins can be identified (Conyers, 2012)� GPR has
been successfully used by Goodman and Piro (2013) to
identify the moat’s shape of a burial mount located in
Saitobaru, Miyazaki, Japan� This information derived by
GPR, contributed to the determination of the monument’s
construction date� GPR has also been successfully
applied in mapping with high detail Roman buildings and
structures like cisterns (Spanoudakis et al. 2011), a 1st DC
amphitheatre, living quarters and villas (Goodman et al.,
2004)� It also has proven to be very useful when applied in
urban and industrial areas for archaeological explorations�
Examples can be found in the works of Conyers (2012)
and Papadopoulos et al. (2009)�
In this article the operational principle of GPR is going to be
discussed along with all the important parameters that one
should take into account when it comes to archaeological
investigation in order get reliable results� Additionally,
the basic steps of data processing are briefly described
providing examples derived from real world data� Finally,
three case studies are presented and the capabilities of the
method are discussed�
Theoretical background
GPR is an electromagnetic technique (EM) and its
operation principle has similarities with the seismic
reflection method� A typical GPR system consists of the
GPR:
Theory and Practice in Archaeological Prospection
Meropi Manataki,1,2 Apostolos Sarris,1 Jamieson C. Donati,1
Carmen Cuenca Garcia1 and Tuna Kalayci1
1Laboratory of Geophysical-Satellite Remote Sensing & Archaeo-enviroment (GeoSat ReSeArch Lab),
Institute for Mediterranean Studies,
Foundation for Research and Technology, Hellas (F.O.R.T.H), Greece,
mmanataki@gmail.com, asarris@ims.forth.gr, jcd297@nyu.edu, carmen@ims.forth.gr, tuna@ims.forth.gr
2 Laboratory of Applied Geophysics,
School of Mineral Resources Engineering, Technical University of Crete, Greece
Abstract: Ground Penetrating Radar (GPR) is a near surface geophysical method that has proven to be an appropriate tool in archaeo-
logical prospection. There is a large number of studies in literature where authors manage to map buried antiquities like roads, paths,
public and residential buildings, graves and many other features. The operating principle of GPR lies in the interaction of electro-
magnetic (EM) energy with the subsurface targets. When the conditions are ideal (low conductivity environments, flat surface, lack of
vegetation) GPR can provide highly detailed and accurate results, otherwise it can still perform well as a complementary method to the
rest of the geophysical techniques applied. For all the above reasons, GPR has evolved as one of the main prospection methods used
in archaeological prospection. A number of examples are presented in the particular article to emphasize the importance of the usage
of the GPR in archaeological landscape studies. These examples address the potential of GPR techniques in mapping the underground
features of ancient cities (Demetrias and Mantinea) and Neolithic mounds (magoules Almyriotiki and Perdika 2).
Keywords: GPR, Archaeology, near surface, Mantinea, Demetrias, Thessaly, Neolithic tell, Magoulas.
14
Best Practices of GeoInformatic Technologies
antenna, the timing unit and a portable computer (Fig� 1a)�
The antenna is responsible for emitting and detecting EM
energy (10~2000MHZ) through a transmitter (Tx) and a
receiver (Rx)� The timing unit is the most important part
since it controls the generation of the radar signal and
converts the received signals as a function of time� The
portable computer is used for storing the data and displays
them in real time�
The operation principle of GPR is simple� The transmitter
emits high frequency pulses of short duration into the
ground that ‘travel’ through the subsurface until they meet
a boundary of different material� At this point, part of the
energy is reflected back to the surface and recorded by
the receiver antenna (black arrows in Fig� 1b), while the
remaining energy is diffused deeper (red arrow in Fig� 1b)
until it hits another boundary, where it will be reflected
and diffused again� This procedure reaches an end when
all of the energy is absorbed by the ground� This boundary
or reflector is defined by differences in subsoil’s materials
electrical properties, such as the conductivity and the
permittivity� Both of them affect the EM waves’ propagation
and are of significant importance� Conductivity affects
the energy absorption, thus the signal penetration and the
depth of the investigation, while permittivity affects the
velocity of the signal� In general, GPR is most useful in
low-electrical-loss materials (i�e� very low conductivity
values)� Clay-rich environments or areas of saline water
will affect negatively the method’s effectiveness (Cassidy
2009a)�
Unlike magnetic or electrical methods, GPR doesn’t
directly measure the properties of the ground� What is
recorded by the receiver is the amplitude of the reflected
signals associated with the EM energy of the reflected
pulses in respect to their travel time� This time series is
called trace� The travel time, also known as double travel
time or two-way time, is the time that a signal needs to
cover the route transmitter-reflector-receiver and depends
on the propagation velocity of the EM pulse� When an EM
pulse leaves the transmitter, its energy is spread at different
paths that are illustrated in Fig� 2� In this example, the
subsoil consists of two homogeneous layers of different
electrical properties� The first records are the direct
airwaves and ground waves since in air the EM waves
have their maximum velocity (0�3x109 m/s)� Additionally
they exhibit the highest amplitude values since the energy
loss during this path is minimum� Critical refracted and
reflected waves exhibit slower velocities that depend on
the electrical properties of the medium and their recordings
appeared in greater return times than the direct waves� If
the velocity is known then the depth of the reflector can be
determined�
To better understand the GPR records or traces, we can
consider that the subsurface is a single homogeneous layer
(i�e� without contrast in electrical properties)� If we place
the antenna at some point on the surface and trigger a
pulse, what it would be recorded is the direct waves only
since, due to homogeneity, the pulse is not allowed to be
reflected� Now, if we add a layer of different electrical
properties below the previous one mentioned above
and trigger a pulse again, the resulted trace will include
amplitudes that derive from the direct waves as well as
amplitudes from waves that are reflected on the boundary
within the two layers� The latter will be recorded at a
time that corresponds to the depth of the reflector, i�e� the
boundary (Fig� 2)� In other words, GPR method depends
on the record of the waves reflected on surfaces that divide
regions with different electrical properties�
Figure 1. The operation principle of GPR where a) describes the system components and b) the behavior of the EM waves
when they meet a boundary with different electrical properties from the soil (ε1>ε2). Part of the energy is reflected to the
surface and another one (red arrow) is diffused at deeper levels (Daniels 2000).
15
Manataki et al.: GPR: Theory and Practice in Archaeological Prospection
Survey methodology and important parameters
The GPR method that is most often used in archaeological
investigations is the common-offset reflection� The
transmitter and the receiver have a fixed spacing and
orientation at each measurement (i�e� trace) location�
The data are collected by moving this fixed offset along
the surface (scan axis or Y direction) at regular station
intervals (Δy) as depicted in Fig. 3a. Additionally, in the
common offset reflection systems, antennas come with
a fixed central frequency� Frequency is a very important
parameter which affects both the investigation depth and
the data resolution� The higher the central frequency the
lower the pulse penetration depth, but the resolution on
both vertical and horizontal directions is better� Thus,
prior to survey with GPR, one must know the target(s)
expected depth and select a frequency that will allow to
reach that depth� The antenna frequency selection is based
on previous experience� As an example, a range of 200-
300MHz can penetrate up to 2-3 m if the ground conditions
(conductivity and permittivity) are appropriate�
The data are usually collected by employing grids that
cover the area of interest� An important parameter is the
line spacing, Δx, which is the distance between the GPR
parallel profiles (Fig� 3b)� Line spacing should be a fixed
value during the survey and is set accordingly to the
targets’ expected size and geometry� For detecting buried
structures such as walls or foundations, a line spacing of
0�5m is suitable, since they can be fully resolved without
Figure 2. The paths that the emitted pulse can follow between the transmitter and the receiver are indicated (Annan 2009).
The direct airwaves and ground waves are the earliest records and are located at the top of the trace, while the refracted
and reflected waves are caused by the reflector which in this case is the boundary between the two layers.
Figure 3. Survey parameters of common offset reflection GPR systems. (a) As the antenna is moving along the surface,
traces are recorded with a step that is defined by Δy along the scan axis. Additionally, each trace includes a finite number of
records (black dots) that are obtained with a step of Δt. (b) The survey grids consist of parallel lines that are separated by a
constant distance defined by Δx.
16
Best Practices of GeoInformatic Technologies
losing information about their continuation in space and
unnecessary overlapping is avoided�
GPR transmits continuously but only records discrete
signals� This means that sampling intervals should also be
set� The spatial sampling interval, Δy, defines how often
the traces are recorded along the scan axis or the survey
line (Fig� 3a)� Thus it affects the total number of traces
that the line will include� A small value will result in a
high detail coverage but caution is needed in order to avoid
oversampling that can lead in spurious results� Usually,
this value is selected with respect to the central frequency�
The time sampling interval is the time lapse between two
records in the same trace (spacing among the black dots
on Fig� 3a)� It affects the resolution on the vertical axis�
Similar with Δy, the time sampling interval value is set
accordingly to Nyquist criteria and the central frequency
(Annan 2009)� As an example, for a 250MHz antenna the
recommended values are Δx=0.025m and Δt=0.4ns.
As the antenna moves along the scan axis, traces are
collected forming the image of Fig� 4a� If a colormap is
applied on this image, the outcome is called radargram or
section, and it is actually what the user sees on the display
while surveying (Fig� 4b)� Radargrams contain information
about the subsurface along the survey line� The reflections
from targets can appear as hyperbolas or linear reflections
depending on the orientation of the antenna with respect
to the target geometry� For better understanding, consider
an example of a buried wall (Fig� 5)� If the survey line
is oriented perpendicular to the wall (‘point’ target), its
signature on the section will be a hyperbola� If the scan
axis is along the wall, then the signature will be a linear
anomaly (Fig� 5b)� Also, if the velocity is known, the time
axis is converted to distance indicating the depth of each
feature/reflector� In case the velocity is unknown, it can be
estimated from the hyperbolas appearing on the sections�
This operation is included on every GPR processing
software packages and is usually carried out by fitting a
curve on the hyperbola�
Data processing
Processing is a very important and time consuming
procedure that aims to highlight reflections related to the
target(s) and to remove unwanted information i�e� noise�
Various types of noise are present in GPR data� The most
common are white or random noise and coherent noise�
The former appears usually at deeper levels and hides
reflections from targets� Coherent noise can be caused
by external sources (cell phones, TV antennas, etc�) or
by the remaining energy that escapes the ground to the
air and reflected by objects on the surface (trees, modern
buildings, cars, rocks, electrical cables etc�) back to the
transmitter� Coherent noise appears as echoes similar to
the ones caused by targets, and caution should be exercised
during interpretation�
As described above, radargrams are created by moving
a transmitter-receiver along a profile of the surface and
are 2D images (Distance (m)-Time (ns) or Depth (m))
of the subsurface� When working in grids, 3D images of
the subsurface can be constructed from the radargrams
out of which depth slices (or time slices) are extracted�
Depth slices are also 2D (Distance (m) – Distance (m))
images that provide information about the reflections of
the subsurface at a certain depth�
Data processing of GPR data can be divided into two
major stages: (1) the processing of the radargrams where
signal processing techniques are used and slices are
extracted, and (2) the processing of slices where image
processing corrections are applied� More emphasis will be
given into the first stage that aims to filter the noise from
the data along a profile and enhance the reflections of the
raw data� Some standard processes that are usually applied
regardless the field of application are (Cassidy 2009b):
•Traces reposition that corrects the position of GPR
traces included in a survey line� This correction is
useful to eliminate systematically offsets in survey
Figure 4. Radargram or section, where (a) is the outcome when a survey line is completed, while (b) is the resulting image
after the application of a colormap.
17
Manataki et al.: GPR: Theory and Practice in Archaeological Prospection
lines’ starting and ending positions which usually
occur in rough terrains�
•Timezero correction which allows to estimate
the correct vertical position of the first pulse that
left the antenna and entered the subsurface (Tzanis
2006)� The effect of time zero correction is shown
on Fig� 6b, where (a) is the raw section�
•Dewow filter which removes low frequency noise
derived by low frequency energy near transmitter
and is associated with electrostatic and inductive
fields� The output of dewow filter is presented on
Fig� 6c, and it is applied after time zero correction�
•SEC (Spreading & Exponential Compensation)
gain that enhance signals located at greater depths
and have much smaller amplitude compared with
that of the shallower signals� This correction
emphasizes the reflections but also highlights noise
as it appears in the example of Fig� 6d�
•Background subtraction filter which reduces
random noise from the data and also removes the
direct waves and ringing noise� On Fig� 6e, the
output of background removal filter is shown�
The noise is removed while the hyperbolas are
highlighted� When gain correction is applied,
background subtraction filter is recommended to
be applied next, in order to remove the noise that
the gain emphasized�
•Frequency domain filters consisting of low- or
high-pass are 1D filters that remove high or low
frequency noise correspondingly� These filters can
be combined to retain frequencies at a certain range
and are called bandpass filters (Cassidy 2009b)�
The effect of a bandpass filter is presented on the
slices of Fig� 7� Fig� 7a presents the slice where
all the above corrections have been applied but
noise couldn’t be removed sufficiently, shadowing
reflections from buried structures� By selecting an
appropriate frequency range, the stripping noise
caused by plowing lines is significantly removed
on Fig� 7b highlighting structures that were barely
visible before�
•Migration that removes distortion due to
diffraction, by trying to reconstruct the signal to
fall at its correct position� In such a case, ideally,
a hyperbola signal will be reduced to an isolated
point target�
The above correction, besides timezero and dewow, are
consider to be basic but are not standard, meaning that they
may not always enhance the original data�
By the time the processing on the radargramms is
complete, Hilbert Transform is applied to calculate the
instantaneous amplitude (Spanoudakis and Vafidis 2010)
and to extract depth slices� The slices indicate the changes
on the instantaneous amplitude at a certain depth� High
values designate strong changes in the electrical properties
or high reflectivity�
Case studies examples
Ancient Demetrias, Volos, Greece
The ancient city of Demetrias is located south of the modern
city of Volos� The city was established by the Macedonian
military leader and eventual king, Demetrius Poliorcetes
(337-283 BCE) in 294 BCE (Batziou-Efstathiou, 2001)�
The city became the royal residence of the Antigonid
dynasty of Macedonian kings and flourished as an
international and political center� The city permanently fell
to the Romans following the battle of Pydna in 168 BCE�
The Antigonid dynasty was immediately dissolved and the
Figure 5. Reflections from targets can be hyperbolas or linear anomalies. (a) Hyperbolas are formed from small targets or
walls that the antenna passes perpendicular to their longest dimension. (b) Linear anomalies characterize linear reflectors
and are formed when the antenna is moving along their longest dimension.
18
Best Practices of GeoInformatic Technologies
Roman province of Macedonia, in which Demetrias was a
part, was officially established a few decades later in 146
BCE� During the Roman Imperial period, many central
areas of the city, including the area around the Hellenistic
palace, were used for burials� Demetrias experienced
a brief recovery beginning in the 4th century CE, when
the Roman emperor Constantine the Great made the city
an episcopal sit (Batziou-Efstathiou 2001)� The city was
finally abandoned during the 6th century CE and it was
never reoccupied again�
The geophysical survey at the ancient Greek city
of Demetrias was conducted by the Laboratory of
Geophysical, Satellite Remote Sensing and Archaeo-
environment of the Institute for Mediterranean Studies
(FORTH) during March 2014� Two GPR systems were
used� The first was a single channel Sensors and Software
NOGGIN Plus-Smart Cart system equipped with a
250 MHz shielded antenna frequency (Fig� 8a) and the
second was a multi-channel MALÅ Imaging Radar Array
(MIRA) with 400 MHz antennas (Fig� 8b)� Both radars
were employed for surveying the area of the soccer field
that is located east of the city agora and southeast of the
Hellenistic palace� The area was ideal for using GPR due
to the flat surface and the lack of vegetation� The total
area covered is a 120x60m with a 0�5m line spacing� The
results for both radar investigations are very detailed and
they are presented in Fig� 9�
The overall arrangement of the architectural features
beneath the soccer field recalls Hellenistic and Roman
urban houses with courtyards or gardens in the back and
shared partition walls between houses (Rumscheid and
Koenigs 1998, Zanker 1998)� Two main roads are clearly
distinguished (Features 1 and 2) with a dense collection of
buildings in the rectilinear city blocks in between� Features
Figure 6. Processing example of GPR data where (a) is the raw section, (b) the time zero correction, (c) dewow filter, (d) gain
correction and (e) background removal noise correction, applied to the raw data respectively.
19
Manataki et al.: GPR: Theory and Practice in Archaeological Prospection
3 and 4 seem to be a single structure with at least seven
rooms located at the west and a large free zone at the east
with few walls that functioned as perhaps a backyard
courtyard or garden� A similar arrangement is noted with
Features 5 and 6� Feature 7 reveals another collection
of rooms that take up the whole width of the city block,
but there is no clear evidence for an open courtyard�
The survey found another cluster of rooms described by
Feature 8, while Feature 9 appears to be a large open area
that is not clear if it is connected with Feature 8� More than
a dozen rooms were mapped from Feature 10, and at least
four from Feature 11�
Comparing the two GPRs, the multichannel Mala seems
to exhibit better resolution than the single channel Noggin
and the data are less noisy� Differences can be seen in
Feature 9 where the former managed to map parts of walls
and a semicircular structure at the southeast corner of the
division wall that are not visible on the single channel
GPR� This manifests the obvious advantage of using multi-
antenna GPR systems, as they are collecting information
with a much more dense spacing than the single antenna
GPR units�
Figure 7. The effect of bandpass filtering where (a) is the slice after the basic processing and (b) is the output of bandpass
filtering. Structures that were not visible before are highlighted.
Figure 8. GPR survey at Demetrias modern soccer field. (a) Single channel Noggin Plus GPR system equipped with a 250MHz
antenna and (b) the multichannel Mala Mira GPR equipped with 400MHz antennas.
20
Best Practices of GeoInformatic Technologies
Ancient Mantinea, Peloponnese, Greece
Mantinea was established within a level flood basin of
northeastern Arcadia in the Peloponnese before the middle
of the 5th century BCE� At 385 BCE the city was destroyed
by a Spartan invasion and its citizens were forced to
depopulate� For 15 years Mantinea was abandoned until it
was reestablished in 370 BCE after Sparta’s defeat in the
Battle of Leuctra� The city played a prominent role in the
activities of the newly established Arcadian League during
the 4th century BCE, and along with Megalopolis and
Tegea it continued to have an influential regional presence
in Arcadia and the Peloponnese for several centuries�
The known archaeological features at Mantinea include the
well-preserved elliptical fortification walls, approximately
4 km in circumference, and the agora and theater at the
center (Hodkinson and Hodkinson 1981, Winter 1987,
1989) but very little of the remaining urban area inside
the fortification walls (~120 hectares) has been explored�
A geophysical survey through the use of soil resistivity
and magnetic methods was conducted by the University
of Patras (Greece) from 1988-91 northwest of the theater
(Sarris 1992)� The target area was limited to 1 hectare
but the survey revealed evidence for subsurface streets
arranged at right angles together with various buildings,
possibly domestic in nature�
A geophysical survey was conducted to explore the
structure and urban development of the classical Greek
city of Mantinea in the Peloponnese through an intensive
geophysical fieldwork campaign carried out by the
Laboratory of Geophysical, Satellite Remote Sensing and
Archaeo-environment of the Institute for Mediterranean
Studies (FORTH)� For this task the GPR system Noggin
smart cart of Sensors & Software was also employed using
a 250MHz antenna� An area of 3�41ha was covered in total
with GPR using 0�5m spacing between each transect�
The data acquired with Noggin GPR were noisy but they
exhibit anomalies related to buried structures� In order
to enhance the anomalies related with buried antiquities,
data were processed using the following corrections in
order: trace reposition, time zero correction, dewow filter,
SEC gain, background removal, bandpass filtering and
migration� Here, the results obtained from the eastern
side of the agora are presented (Fig� 10a) where public
buildings including a long ‘L’ shape stoa with columns
appear as high amplitude anomalies with high detail�
Many of these features appear to be related to the public
Figure 9. GPR survey in the soccer field at demetrias. (a) The slice at 0.4-0.5m depth derived from Noggin GPR and (b) the slice
at 0.47m derived from MALA Mira are indicated.
21
Manataki et al.: GPR: Theory and Practice in Archaeological Prospection
buildings excavated by the French in the 19th century but
they were reburied (Fougères 1898)� The similarities and
differences with the French plan are shown in Fig� 10b,
where with black lines are the findings of the excavation,
while with red lines are the interpretation of the GPR data as
occurred from all the depth slices� There is a slight shift in
the orientation of the whole settlement� The ‘L’ shape stoa
has a double row of internal colonnades on the north-east
direction and a single row along the north-south direction�
Two adjoining structures appear as strong linear anomalies
behind the west section of the stoa� These structures are
also present in the French plan� The southern one is almost
a perfect square and is subdivided into smaller rectilinear
rooms on either side of larger central rooms� The French
plan of the agora did not show the internal subdivision of
space� Only part of the other structure could be surveyed
because of a large tree, but it is clear that this building is
smaller than its neighbor and likely had no internal rooms�
The building is also oriented at a diagonal angle (unlike
the southern building), which is a distinct characteristic
not present on the French plan� Another notable anomaly
further to the northeast is a small structure and is also set
at a diagonal angle�
Neolithic Thessaly, Greece
A large-scale geophysical exploration was conducted by the
Laboratory of Geophysical, Satellite Remote Sensing and
Archaeo-environment of the Institute for Mediterranean
Studies (FORTH) during 2013-2015 at a large number of
Neolithic tell sites (magoules) in Thessaly� The purpose
of this study was the identification of intra- and inter-
spatial patterns of Neolithic settlements through the
comparative study of both archaeological and geophysical
data� Non-destructive geophysical methods like electrical
resistivity, magnetics, EM and GPR were used on selected
archaeological sites that have been partially excavated
or have been identified by survey expeditions� The most
interesting results derived from magnetics included
features like ditches, enclosures, paleochannels, burnt
structures, daub and stone structures, etc�
Survey with GPR on such landscapes is not an easy task,
due to the rough terrain and the modern cultivation� The
geomorphology of the area and the soil condition (clay-
rich environments) resulted in noisy data with very limited
penetration (to a maximum depth of 1�0m)� Thus, GPR
Figure 10. Results of the Noggin Plus GPR survey at the eastern side of the agora at Mantinea. (a) GPR slice illustrating the
anomalies at 1.0-1.1m depth. (b) Comparison of the French plans (black color) and interpretation (red color) as occurred
from all slice depths.
22
Best Practices of GeoInformatic Technologies
Figure 12. Comparison between GPR and magnetic results from magoula Perdika 2. (a) GPR slice at 0.7-0.8m
and (b) magnetic results from the same area. GPR revealed a structure, probably stone-made, that is barely
visible on the magnetics data.
Figure 11.Comparison between GPR and magnetic results from Magoula Almyriotiki. (a) GPR slice at 0.7-0.8m and (b)
magnetic slice. GPR exhibits better resolution revealing that the large structure indicated by the magnetic results is a
cluster of individual buildings.
23
Manataki et al.: GPR: Theory and Practice in Archaeological Prospection
was used mostly as a supplementary method to enrich the
information obtained from other geophysical methods�
Such an example is presented in Fig� 11, where the left
image is the depth slice derived from GPR with 250MHz
antenna at Magoula Almyriotiki, while the right image is
the magnetic results from the same region� Even though
GPR could not map all the structures as the magnetics did,
it managed to provide better resolution, revealing that the
large structure appearing in the magnetic data consists of
individual dwellings�
In another survey at Perdika 2, GPR managed to map
a structure with high detail that is barely visible on
the magnetic data due to the geological background
noise� This contrast between the two methods indicates
complementary information, as it seems that the local
building material of the strauctures was not able to produce
significant anomalies in the magnetic measurements� In
contrast to other magoules, where burned clay structures
were appearing as vivid anomalies, the stone structures
show a very weak magnetic signal, but are easily resolved
through the GPR techniques� This is a clear indication
of the fact that GPR relies mostly on the contrast of the
electrical rather than magnetic properties of the targets�
Conclusions
Overall, GPR proves to be an extremely useful tool for
archaeological investigations� When soil conditions
and geomorphology are appropriate, it can provide very
detailed results of the subsurface and map successfully
buried structures, roads, city blocks and other features�
Even if the survey conditions are not ideal, the data can
be significantly improved with proper processing� In such
cases, GPR is better to be used as a complementary method
that will provide additional information with respect to both
the horizontal and vertical extent of the subsurface targets�
This has been manifested by the case studies presented
here� GPR was able to retrieve information from various
targets, at different depths and within diverse geological
contexts� Compared to other methods, the GPR technique
has the advantage to provide information not only about
the lateral extent of the targets but also of the stratigraphy
of the site and the vertical extent of the monuments,
contributing in this way to the 3D reconstruction of the
cultural landscapes�
Acknowledgements
This work was performed in the framework of
two research projects: IGEAN (‘INNOVATIVE
GEOPHYSICAL APPROACHES FOR THE STUDY
OF EARLY AGRICULTURAL VILLAGES OF
NEOLITHIC THESSALY’) project which is implemented
under the ‘ARISTEIA’ Action of the ‘OPERATIONAL
PROGRAMME EDUCATION AND LIFELONG
LEARNING’ and is co-funded by the European Social
Fund (ESF) and National Resources and POLITEIA
research project, Action KRIPIS, project No MIS-
448300 (2013SE01380035) that was funded by the
General Secretariat for Research and Technology,
Ministry of Education, Greece and the European Regional
Development Fund (Sectoral Operational Programme:
Competitiveness and Entrepreneurship, NSRF 2007-
2013)/ European Commission�
References
Annan, A�P�, 2002� GPR—History, Trends, and Future
Developments� Subsurf. Sens. Technol. Appl. 3, 253–
270�
Annan, A�P�, 2009� Chapter 1 - Electromagnetic Principles
of Ground Penetrating Radar, in: Harry M� Jol (Ed�),
Ground Penetrating Radar Theory and Applications.
Elsevier, Amsterdam, pp� 1–40�
Batziou-Efstathiou, A�, 2001� Δημητριάς� Athens Archaeol�
Receipts Fund�
Benson, A�K�, 1995� Applications of ground penetrating
radar in assessing some geological hazards: examples
of groundwater contamination, faults, cavities� J. Appl.
Geophys� 33, 177–193�
Cassidy, N�J�, 2009a� Chapter 2 - Electrical and Magnetic
Properties of Rocks, Soils and Fluids, in: Harry M�
Jol (Ed�), Ground Penetrating Radar Theory and
Applications. Elsevier, Amsterdam, pp� 41–72�
Cassidy, N�J�, 2009b� Chapter 5 - Ground Penetrating
Radar Data Processing, Modelling and Analysis, in:
Harry M� Jol (Ed�), Ground Penetrating Radar Theory
and Applications. Elsevier, Amsterdam, pp� 141–176�
Conyers, L�B�, 2012� Interpreting ground-penetrating
radar for archaeology. Left Coast Press�
Daniels, J�J�, 2000� Ground Penetrating Radar
Fundamentals (Appendix to U�S�EPA, Region V)�
Ohio�
Davis, J�L�, Annan, A�P�, 1989� Ground-Penetrating
Radar for High-Resolution Mapping of Soil and Rock
Stratigraphy 1� Geophys. Prospect� 37, 531–551�
Doolittle, J�A�, Jenkinson, B�, Hopkins, D�, Ulmer, M�,
Tuttle, W�, 2006� Hydropedological investigations
with ground-penetrating radar (GPR): Estimating
water-table depths and local ground-water flow
pattern in areas of coarse-textured soils� Geoderma,
Hydropedology: Bridging disciplines, scales and data
131, 317–329�
El-said, M�A�H�, 1956� Geophysical Prospection of
Underground Water in the Desert by Means of
Electromagnetic Interference Fringes� Proc. IRE 44,
24–30�
Fougères, G�, 1898� Mantinée et l’Arcadie� A� Fontemoing�
Goodman, D�, Piro, S�, 2013� Multi-channel GPR, in: GPR
Remote Sensing in Archaeology, Geotechnologies and
the Environment� Springer Berlin Heidelberg, pp� 175–
185�
Goodman, D�, Piro, S�, Nishimura, Y�, Patterson, H�,
Gaffney, V�, 2004� Discovery of a 1st Century AD
Roman Amphitheater and Other Structures at the
Forum Novum by GPR� J� Environ. Eng. Geophys. 9,
35–41�
24
Best Practices of GeoInformatic Technologies
Grasmueck, M�, 1996� 3-D ground‐penetrating radar
applied to fracture imaging in gneiss� GEOPHYSICS
61, 1050–1064�
Hodkinson, S�, Hodkinson, H�, 1981� Mantineia and the
Mantinike: Settlement and Society in a Greek Polis�
Annual of the British School at Athens 76, pp� 239-296�
Papadopoulos, N�, Sarris, A�, Yi, M�, Kim, J�, 2009�
Urban archaeological investigations using surface 3D
Ground Penetrating Radar and Electrical Resistivity
Tomography methods� Explor. Geophys� 40, 56–68�
Rumscheid, F�, Koenigs, W�, 1998� Priene: a guide to the
‘Pompeii of Asia Minor.’ Ege Yayinlar�
Sarris, A�, 1992� Shallow Depth Geophysical Investigation
Through the Application of Magnetic and Electric
Resistance Techniques: AN Evaluation Study of
the Responses of Magnetic and Electric Resistance
Techniques to Archaeogeophysical Prospection
Surveys in Greece and Cyprus� PhD thesis, Univ� Neb�
- Linc� 1992 Source Diss� Abstr� Int� Vol� 53-08 Sect�
B Page 4002�
Spanoudakis, S�N�, Vafidis, A�, 2010� GPR-PRO: A
MATLAB module for GPR data processing, in: 2010
13th International Conference on Ground Penetrating
Radar (GPR)� Presented at the 2010 13th International
Conference on Ground Penetrating Radar (GPR), pp�
1–5�
Theune, U�, Rokosh, D�, Sacchi, M�, Schmitt, D�, 2006�
Mapping fractures with GPR: A case study from Turtle
Mountain� GEOPHYSICS 71, B139–B150�
Tzanis, A�, 2006� Matgpr: A freeware matlab package
for the analysis of common-offset GPR data, in:
Geophysical Research Abstracts�
Vaughan, C�, 1986� Ground‐penetrating radar surveys used
in archaeological investigations� GEOPHYSICS 51,
595–604�
Winter, F�E�, 1987� Arkadian Notes I: Identification of the
Agora Buildings at Orchomenos and Mantinea� EchCl
31, 235–46�
Winter, F�E�, 1989� Arkadian Notes II: the Walls of
Mantinea, Orchomenos and Kleitor� EchCl Ns 8, 192–
96�
Zanker, P�, 1998� Pompeii: public and private life� Harvard
University Press�
