Reconstructing Ancient Landscape S.

Pescarin

Reconstructing Ancient Landscape

S. Pescarin

Preface by A. Addison
Introduction by M. Forte
Reconstructing Ancient Landscape S. Pescarin

Cover by Matteo Pescarin

Appendix by Ivana Cerato, Augusto Palombini, Eva Pietroni and Valentina
Vassallo
Reconstructing Ancient Landscape S. Pescarin

Summary

Preface (A. Addison)

Introduction (M. Forte)
Ch. 1 – Reconstructing Ancient Landscape
Ch. 2 – Interpretation and Reconstruction
Ch. 3 – Archaeological Landscape Reconstruction: mapping the space
Ch. 4 – Potential Landscape Reconstruction
Ch. 5 – Communicating the landscape
Ch. 6 – OpenLand: Open Source for landscape archaeology
Ch. 7 – Webscapes
Ch. 8 – A step into the future
Appendix:
A – Case Studies (I. Cerato, A. Palombini, S. Pescarin,
E. Pietroni, V. Vassallo)
B – Sources and Resources (A. Palombini, V. Vassallo)
C – References (S. Pescarin)
Reconstructing Ancient Landscape S. Pescarin

Sentence
Dedication
This book wouldn't never be published without
Reconstructing Ancient Landscape S. Pescarin

Preface
(Alonzo C. Addison)
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Introduction
(Maurizio Forte)
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Chapter 1
Reconstructing ancient landscape

“Primum, Agricultura non modo est ars,

sed etiam necessaria ac magna;
eaque est scientia, quae sint
in quoque agro serenda ac facienda,
quo terra maximos perpetuo reddat fructus”.

(Varro, De Re Rustica I.4 )

How and where did man use to live in antiquity? What relationship did he have
with his neighbours? How did he use, organize or modify his landscape? These
are some of the questions that landscape reconstruction tries to answer.
But is it possible to reconstruct ancient landscape in a reliable way? Are there
methods, tools, or other indications applicable to this activity?
Although it's impossible to know what real landscape was like in the past,
nevertheless its reconstruction remains a fundamental process, that cannot be
relegated to a “dissemination phase”. Vice-versa it regards the entire
archaeological process, from data acquisition to interpretation, to knowledge
acquisition and dissemination. Communication is a part of it. And archaeology
isn't the only discipline involved.
Recently, interest in environmental aspects has also reached those fields of
archaeology traditionally more focused on “material culture”. We are slowly
moving towards an archaeology of everyday life, where research, adopting a
more ecological perspective, is finally dealing with aspects until now
considered superfluous or unimportant.

1.1 Ancient Landscapes

In recent decades, the word “landscape” has become increasingly used in many
contexts, and on various occasions. As the environment and environmental
issues have assumed a central role in worldwide debates about future
development, attention to landscape has also gained in importance. Even in the
archaeological field, the traditional approach, mainly directed at excavation and
identification of anthropic remains, has been changing, substituted by greater
attention to the surrounding context: the natural environment, traces of
agriculture, land use, etc.
This new approach has evolved into Landscape Archaeology, a relatively new
field ('90s), commonly associated with British studies (D'AGOSTINO 1992: 19),
Reconstructing Ancient Landscape S. Pescarin

although its origins may be traced as far back as the end of the 19th century,
with Metzer’s research (ROBERTS 1987: 78-79).
This discipline has several definitions (BARKER 1992, BERNARDI 1992, LOCK and
STANCIC 1995, ASHMORE and KNAPP 1999, GILLINGS MATTINGLY and VAN DALEN
1999), also because it has been approached in different ways by British and
European, and American scholars. The former have concentrated their studies
more on visual and physical landscapes, while American researchers have
expanded its meaning so as to include a cognitive and perceptive approach.
In this book, landscape archaeology is intended as the discipline that deals with
the relationship between men and their environment, between the people of the
past and their places, as shaped by them consciously or unconsciously. For this
reason I will be adopting a strong (geo)spatial and multidisciplinary approach,
including reference to several different disciplines, technologies and
methodologies (BARKER 1992:265, FORTE 2005: vii, RENFREW 1994). Landscape
cannot take a back seat in archaeological studies, but has to be seen as a
determinant factor for the development of ancient cultures, the result of a
dynamic combination of geophysical, biological, cultural and anthropic
elements in continuous evolution (BERTRAND 1978: 5).
In archaeology, unfortunately, in many cases the fieldwork or excavation is
aimed at reconstructing a single site and not a landscape. In this way it may
become very difficult to acquire sufficient data for a consistent reconstruction
(FORTE PESCARIN 2006).
To study, analyse, interpret and reconstruct this landscape, an ecological and
multidisciplinary perspective is necessary, requiring the application of
disciplines such as remote sensing, GIS analysis, aerial photo interpretation,
field survey, spatial 2d and 3d digital data acquisition, paleo-environmental
researches (geoarchaeology, palaeobotany, etc.), geophysical prospecting, but
also data post-processing and integration in spatial (interactive) systems.
The goal of this process is the reconstruction of the ancient landscape as it
could have been, in its potentiality, and it requires the development of virtual
(conceptual, realistic) models. The result can be useful to get back to
interpretation itself, but it is also a powerful tool to visualise synthetically and
immediately the results of several research projects. It can be integrated into
communicative tools, such as virtual museums, and become part of a three-
dimensional immersive experience, transformed into part of our historical
awareness of the past.
Ancient landscape reconstruction cannot therefore just be considered a marginal
activity, that can eventually be added on to the end of some archaeological
research, just for (relatively unimportant) “communicative” purposes. It is a
part of the research process, a framework for the study, the interpretation and
the communication of many different aspects of ancient man's life, integrated in
Reconstructing Ancient Landscape S. Pescarin

a natural environment and defined by cognitive aspects. It allows us to connect

several layers of information – deriving from history, archaeology, art history,
but also geography, sociology, economy and other disciplines - to places, and to
investigate connections and relationships among and inside ecosystems. We
will therefore be dealing with potential reconstructions and not with
reproductions of past realities, impossible to state scientifically in an extensive
way. More specifically, this book deals with the reconstruction of virtual
ecosystems, which include archaeological, interpreted, potential, perceived and
networked landscapes (table 1.1). Moreover, I will sugggest an open approach:
landscape reconstruction is in fact a never-ending process.

Archaeological Landscape We are in the landscape: actuality, three-dimensionality, spatiality,

dynamism
Interpreted Landscape (Mapscape) We think they were: interpretation
Ancient Potential Landscape (Pastscape) They could have been: potentiality, 3d, spatiality, dynamism
Perceived Landscape (Mindscape) They perceived their territory: perception
Network Landscape (Webscape) We connect over the net

Table 1.1
Over the last ten years I've been working on many projects relating to landscape
research and reconstruction, at CINECA Supercomputing Centre of Bologna
and at the National Council of Research, the Institute of Technologies Applied
to Cultural Heritage (CNR ITABC) in Rome, and in the VHLab, directed by
Maurizio Forte until 2007. Because I had a background in archaeology and
topography, with a specialist qualification in spatial digital systems, I was able
to follow the entire digital pipeline of data acquisition and post-processing and
its further exploitation in communicative networks, from the fieldwork to
virtual reality applications. I've been giving several seminars on these topics,
for post-graduate and post-doctorate students. During these courses I've become
aware of students’ urgent need to acquire practical skills and, most of all, to
experiment with a multidisciplinary approach. At the end of each course they
seemed satisfied, because they had been able to create connections between
specific fields they were already skilled in, “bringing data to life”, changing
them into dynamic information. When I usually start a seminar on this topic, I
can feel the varied expectations in the audience: the defiance of those persuaded
of the usefulness of virtual reconstructions, and also the hope of those interested
in finding reassurance, regarding the worth of the studies they are doing, or
would like to do.
In these years, I've tried to compare the work I have being doing with similar
projects and approaches, to analyse various possible solutions and discuss
problematic aspects. I have found much good work that has reached excellent
Reconstructing Ancient Landscape S. Pescarin

results, but each one in a specific field: GIS, remote sensing, cognitive analysis,
geoarchaeological and palaeobotanical studies, ICT applications, etc.
Dealing with the landscape, on the other hand, requires a broader,
multidisciplinary approach, which forces us to overcome that kind of “solitary
attitude” that often characterizes researchers in the humanities. For this reason,
I have tried to collect and connect some of this specific research work and bring
it into a broad digital methodology, including concepts, hardware and software
tools and formats, and also practical suggestions and technological solutions,
through the analysis of some case studies.
The topic is very complex: it deals with something continuously changing,
uncertain and made up of very many different interconnected aspects. There are
many risks to be taken into account.
Let's start with the title of this book.
What is ancient landscape? Why reconstruct it? How can it be analysed,
interpreted and reconstructed in a coherent way? What are the main problems in
the interpretation process and in the ICT process? How many reconstructions
have been done? How many are widely available? How many are reliable, well-
documented or transparent? How far should we go, in terms of a technological
and epistemological development? What is the state of the art in the field of
ancient landscape reconstruction?
Ancient landscape, in its archaeological and geo-spatial dimension, is one of the
most complex topic both for analysis and reconstruction. Not just from an
archaeological and environmental point of view, but also for Computer
Sciences. This is why it is dealt with from a variety of perspectives, such as
anthropology, archaeology, computer science, history, architecture, geography,
topography, geology, etc. This doesn't mean that anyone interested in landscape
reconstruction necessarily possesses all these skills. But he/she should
understand the entire process and have a basic knowledge of the topics, so as to
address the work correctly and involve the right experts.
Reconstructing ancient landscape includes various perspectives and approaches,
such as cybernetics, cognitive archaeology and ecology (BATESON 1972,
RENFREW 1994, MATURANA AND VARELA 1980 and 1999, FORTE 2008). The process
involves reconstruction at different levels: archaeological landscape, interpreted
landscape, ancient potential landscape and networked landscape.
Archaeological Landscape is the contemporary landscape in its diachronicity;
it includes everything around us, such as archaeological remains and past
historical traces. In this perspective we are in the landscape. The analysis of an
archaeological landscape requires us to deal with actuality, three-
dimensionality, spatiality and relation. Mapping is the primary central activity.
Without it any reconstruction cannot be carried out. A typical bottom-up
approach is commonly followed, since it requires the acquisition of information
Reconstructing Ancient Landscape S. Pescarin

in the field, through surveys, archaeological excavations, photo-interpretation,

geological and botanical explorations, georadar and other geophysical
acquisitions. It is generally studied through landscape archaeology, aerial
archaeology and topography (FORTE PESCARIN PIETRONI 2005: 79-81, CAMBI
TERRENATO 2004; CAMPANA PIRO 2009).
Interpreted Landscape is the Mapscape. It is the geographical spatial
visualization of the interpretation process, the representation of knowledge
acquisition processes and levels through maps. It includes its results:
collections, selections, essential characteristics, but it can also include
excluded elements. Its visual result is made up of 3d and 2,5d maps, layers in
overlay, vector themes, raster grids, and databases. It is also an open working
environment, closely or inextricably connected to archaeological and potential
landscape reconstruction and to spatial and geophysical characteristics. A
mapscape is a visualization of the landscape obtained by a digital classification
process (satellite, aerial images, geophysical, spectral, etc.) (FORTE PESCARIN
PIETRONI 2005: 79-81). It can be a representation indicating spatial relations or a
diagrams in which spatial relationships are configured (RENFREW 2003: 168).
Although it is a symbolic environment, can it be also treated as a real
landscape? We have to be aware that “the map is not the territory, it represents,
but, if correct, it has a similar structure to the territory”, “which accounts for
its usefulness”, as Korzybski stated in 1933 (KORZYBSKI 1933: 58), but we
should also notice that a part of us, probability the right hemisphere of our
brain, normally identifies a representation with the object represented. Even a
flag on a map, with the name of a country, becomes the country itself (BATESON
1979: 47). In this sense a mapscape can be a useful meta-universe of
interpretations, where we can find our direction, our questions, our points of
view.
Ancient Potential Landscape is the Pastscape. It is how the landscape might
have been during a certain historical period, at a defined latitude. It is not a
reproduction of what it was, but a reconstruction of what it might have been,
potentially, based on explicit theories and scientific method, from an analytic,
holistic and ecological perspective (RENFREW BAHN 1991). It requires us to
evaluate three-dimensionality, spatiality and terrain potentiality. Starting from
its main attributes, such as latitude, altitude, climate, morphology, soils and so
on, it is possible in fact to define the natural attitude of a terrain unit (DI FIDIO
1993). Sometimes more useful resources are available, such as Soil Maps or
Land Use Maps, although not for every region (FAO 1976). Potential landscape
can then be compared with other available information and data, regarding the
specific historical period. Archaeological maps or historical maps can be used
to develop an analytic system, based on GIS spatial analysis and consisting of
connected algorithms. During this process we very frequently have to face
Reconstructing Ancient Landscape S. Pescarin

much uncertainty. This is why it is very important to build it on a solid,

scientific, open and transparent interpretation process. This promising approach
was born at the beginning of the New Archaeology, when Binford and other
archaeologists started to support the idea that conclusions should be based on an
explicit approach, developed through logical arguments, and not just on the
authority and prestige of the scholar. In the UK, thanks to archaeologists such
as Clarke and his Analytical Archaeology, the analysis was focused on more
quantitative techniques. (RENFREW BAHN 1995: 29; BINFORD CLARKE 1977). For
the first time, they reconnected archaeology to other sciences, that were used to
building their theories on scientific method.
Sometimes this is not enough.
Logic and quantity have proved to be inadequate tools to describe organisms,
their interactions and their organisations (BATESON 1979: 37). New tools need
now to be tested and adopted. Perceived Landscape is the Mindscape. It is how
the territory was perceived in the past. It requires us to deal with concepts such
as relations and perception (FORTE 2003). Richard Gregory in 1973 stated that
perception has one main characteristic: it is a hypothesis1 (GREGORY 1973: 49-
96) and as such, it can be contradicted. The original idea of considering
archaeology beyond its material culture, including the concept of perception,
comes from Cognitive Archaeology and Anthropology (RENFREW BAHN 1991,
RENFREW 1994). Although archaeology and anthropology are still thought of as
two distinct disciplines, since they use different methods and techniques, they
are also very similar when they deal with the study of long-term temporal
processes based in the landscape. Cognitive archaeology tries to answer
questions such as “what did they think?”. It studies how ancient men thought
and what they believed, starting from the analysis of what they left:
archaeological remains, symbols, ancient representations, rituals, myth and
religious beliefs (RENFREW BAHN 1991: 345-377). It aims at understanding how
past communities organised their lives and their settlements, or how they used
and modified their landscapes. Recent developments in Virtual Reality
applications enable us to adopt new interesting approaches, through the
development of interpretative models that look at how ancient man lived and
perceived his landscape.
Webscape represents specifically relations and connections among
characteristics’ variability, objects, people, characters and cyberspaces.
Networked landscapes (their topology consists of characteristics,
interconnections and feedback) are a result of an iterative application of input
and feedback; the landscape can be altered through the adjustment of network
internal parameters (connection weights). Neural networks algorithms can be
1 For example when we see just a part of a table, because it is hidden behind an object, we
assume it is a table, basing it on the past visual experiences.
Reconstructing Ancient Landscape S. Pescarin

used to develop this kind of landscape.

.B.1 Beyond the view

How can we reconstruct an ancient landscape? Let's start from some basic
considerations. Firstly, we cannot achieve any reconstruction without
interpretation, comprehension, and knowledge. Secondly, we shouldn't work on
the reconstruction of the past, without considering and mapping the present, i.e.
the archaeological landscape. Thirdly, the only way to approach the present is
by changing our perspective and viewpoints. Fourthly, our vision is the main
activity involved in the entire process. Fifthly, knowledge acquisition depends
upon the comparison of different characteristics and on the identification of
differences.
The final purpose, interest in and focus on studies and research regarding our
past landscape and our history, can be found in the structure of our mind. Day
by day we try to acquire information (and knowledge) from the complexity that
is around us, using our brain. We specifically use a part of it, that is connected
to our vision (visual brain), whose main function is exactly "to acquire
knowledge about this world", as stated by neurobiologists (ZEKI 1999: 4). We
need to observe carefully all the details we can see, in order to classify them, to
find essential “ambiguous” characteristics. We do this primary for biological
reasons (Chapter 2).
We use an analytical process consisting of a de-composition phase, where we
distinguish different aspects and we systematize them, and a re-composition
phase. In the first case we organise information into archives, databases, or GIS
systems. In the second we use more advanced visualisation systems, such as
virtual reality environments or 3d models. Unfortunately there is a risk implicit
in the process: each time we re-compose the “picture”, we lose information.
The problem is that if we want to use the reconstruction process as an active
part of the interpretation, we need to work out how to find a way to preserve the
details. Secondly, we should also be aware that visualisation of 3d realistic
models, through movies or other broadcasting products, is often incredibly
effective, in terms of knowledge transfer, since it can transform a “lesson” into
an “experience.” On the other hand, it also creates “dependency” on this opaque
visualisation (TURKLE 1996). There aren't any good or bad visualisation systems
or media, we should just take into account which is the right communication
system for a specific purpose (Chapter 7).
Although there is a lot of resistance and suspicion in cultural heritage, due
mainly to a lack of credibility, of authority reference, and peer-review control,
there are at least three good reasons why we should persist, intensifying the
research in this field. First of all, we should be aware that a wider public is
increasingly asking for a better understanding of past history and archaeological
Reconstructing Ancient Landscape S. Pescarin

landscape, through the employment of visual interactive tools. Secondly,

research teams are increasingly using VR tools as working tools, as we will be
seeing in the next chapters, and not just for presentation purposes. Finally,
representation is often the basis of protection. Although it is not the only way
to propose areas for protection, it does contribute to draw public attention to
archaeological finds, providing a consistent way to identify them through their
knowledge.
Moreover it is quite evident that the sharp resistance to a certain theory, to a
new methodology or approach, is part also of its cultural impact. Academic
culture has confined philosophy inside seminar classes, but new computer
systems are bringing it into everyday life (TURKLE 1991: 167-174).
Although we need to follow a clear and schematic interpretation process,
reducing complexity, abstracting essential characteristics of the objects (ZEKI
1999: 5), we can also develop and implement in our visual systems complex
models, to better explain living and dynamic phenomena, getting closer and
closer to complexity. A cybernetic approach will be more adopted in the future
even to landscape reconstruction, through the application of artificial
intelligence algorithms, neural network programs, chaos theories, fuzzy logic,
etc.
At the end of each interpretation process, we can deal with ideas and
hypotheses regarding past landscapes inside visual spatial systems, where, in
interactive sessions, we could select and gather together all available data,
interconnecting it in the same cyberspace, and define constants and stable
properties. Thanks to the selection of these constants it is possible to classify
and order objects in categories, eliminating those not pertinent, in a continuous
battle between complexity and classification.
Probably in the near future a new generation of VR applications, as virtual
shared dynamic environments, will contribute to the development of this field,
offering new solutions for different data integration, and multidisciplinary
collaborative work for more reliable reconstructions.

.B.2 Digital Pipeline

With the continuous evolution of technologies, applications in the field of
Virtual Archaeology are increasing. Some of them seem to be driven simply by
an interest in a specific technology (in a top-down approach); others by the
consideration of a single typology of sources (in a bottom-up approach).
Nevertheless, we should consider those that provide outstanding results, trying
to bring them as much as possible into a single pipeline.
Ancient landscape, in its archaeological and geo-spatial dimension, is a
complex topic to deal with, to analyze and reconstruct, not just for the
archaeological interpretative process, but also for computer science. That's why
Reconstructing Ancient Landscape S. Pescarin

it is viewed from a variety of perspectives, involving anthropology,

archaeology, computer science, history, architecture, geography, topography,
and geology. Moreover it requires the reconstruction of a virtual ecosystem,
with its environment and its multiple relations and interconnections. Virtual
Reality applications can be successfully used for its exploration and interaction,
and also for its creation itself. To obtain a coherent result, this process needs to
be carefully defined by a digital pipeline, since it requires the interconnection
of several methodologies, formats, software, inputs and outputs.
A digital pipeline is an explicit definition of all digital processes and sub-
processes involved. It can be a flow-chart representation, a guidelines document
or a simple drawing. The main issue is that it should be shared by all the teams
involved in the project. It is commonly divided into principal activities, such as
“acquisition”, “post-processing” and “reconstruction”. It includes a list of
software and hardware used in all the activities. It should also include
import/export formats. Connection lines can be drawn between sequences of
activities in order to clarify which formats are needed, together with
specifications and requirements connected to the final use of the reconstruction
(table 1.2). Spending some time in the initial phase of a project on pipeline
definition, helps save a lot of time in case of mistakes or difficulties, cropping
up at the end. Many common problems can be easily avoided in this way, such
as: differences in geographic coordinate systems or projections, the wrong level
of simplification or optimization of 3d models, naming mismatching, not
supported file format, etc.

Table 1.2 - Digital pipeline.tif

In the case of landscape reconstruction, the process is usually focused on the

acquisition of four categories of data:
I Digital Terrain Models (DTM),
II geospatial imagery (Geoimages),
III vector thematic layers, representing interpretative levels,
IV other thematic layers (3d models) representing the natural and anthropic
aspects of the landscape: monuments, forests, crops, etc.
DTM and Geoimages are essential components in the creation of the three-
dimensional terrain model and in the definition of analysis dedicated to
potential landscape reconstruction. To these data, we should eventually add also
metadata and multimedia information.
A Digital Elevation Model (DEM) is “a digital and mathematical representation
of an existing or virtual object ant its environment. It is a generic concept that
Reconstructing Ancient Landscape S. Pescarin

may refer to elevation of ground but also to any layer above” . DTM, Digital
Terrain Model, is generally used as referred specifically and “limited to ground
elevation”, providing information about elevation of any points on ground
(KASSER AND EGELS 2002:160-161). DSM, Digital Surface Model, is an elevation
model that represents the highest available elevation of each points (top of trees
or of houses for instance). These digital elevation data can have three data
structures: (1) regular grids, (2) triangulated irregular networks, and (3)
contours, depending on the source and/or method of analysis2.
The pipeline starts with “mapping” activity, acquisition of information
regarding the archaeological landscape (Chapter 3). The creation of a GIS
project accompanies this activity, constituting the spatial base of the entire
process. It is essential to take into account the spatial dimension of all the data:
geospatial images, historical maps, any vector information on archaeological
sites or remains, plans, and so on. Even 3d models, acquired on the field with
different techniques (such as photogrammetry, 3d scanning and so on) should
be connected to geospatial information (georeferenced plans, the georeferenced
centre of the model).
Although mapping technologies are becoming more and more efficient in terms
of speed and accuracy (a scanner laser can acquire an entire area of a territory
with sub-millimetric resolution, in accordance with the chosen instrument, in
very short time), the longest part of the work is surely that carried out in the lab
(“post-processing” activities: Chapter 3). Here raw data are downloaded and
processed in order to obtain a first digital prototype of models: the terrain
model, the 3d model of a monument, plans, etc. At the end, single elements of
the archaeological landscape are ready to be visualised, interactively or
otherwise, further analysed, or become part of a collaborative environment.
Ancient landscape reconstruction starts at this point (Chapter 4 and 7).
Along the pipeline, there are some characteristics that should be stored, while
carrying on the activities, since they influence the final result:
• acquisition method,
• processing method,
• data resolution,
• accuracy,
• reconstruction reliability level,
2 Square-grid digital elevation models have some disadvantages: the size of the grid mesh
often affect the quality of the results, not handling abrupt changes in elevation. TINs are
based on triangular elements (facets) with vertices at the sample points, usually constructed
with Delauney triangulation. They are quite efficient because the density of the triangles can
be varied to match the roughness of the terrain. The third structure divides landscapes into
small, irregularly shaped polygons (elements) based on contour lines. Good reviews of
digital elevation data sources and data structures are presented by Carter 1988,Weibel and
Heller 1991, and I. D. Moore et al. 1991
Reconstructing Ancient Landscape S. Pescarin

• sources references,
• institution and team member in charge.

.B.3 Scale, level of detail, accuracy, and reliability

The result of a reconstruction is necessarily connected and conditioned by the
scale, the level of detail, the accuracy and reliability of available data (input)
and processed data (outputs). It can be conducted, in fact, even on a single
level, such as the local level (site, intra-site, farm, single field, etc.) project level
(inter-site, settlement), or regional level. Each level requires different kinds of
survey, different acquisitions, different resolutions, and produces different
outputs: from a very detailed reconstruction of the local level, to the semi-
detailed or reconnaissance level (FAO, 1985) of the regional level.
Each measurement and acquisition is in fact limited in accordance with the
instrument and the method we are using.
The scale is commonly related, in cartography, to the ratio or relationship
between a distance or area on a map and the corresponding distance or area on
the ground, commonly expressed as a fraction or ratio, but it is also used to
indicate the degree of detail or content represented, i.e. in a GIS. Scale refers to
the “level of spatio-temporal” representation in archaeology. With the current
use of digital cartography and GIS, which are substituting the limits imposed by
the printed map, this concept (spatio-temporal scaling) is also opening new
methodological problems . Any representation of the landscape, indeed, is
linked to the level of accuracy and resolution in available data: any further
analysis, therefore, should be very careful over stating “absolute” results. When
we move then from GIS to the realm of Virtual Reality, even greater care
should be taken, since in VR the accuracy level may be very difficult to define
(LOCK 2000: xix-xii).
Accuracy and precision are used when a measurement instrument is evaluated.
“Any measurement system involves errors that can be roughly divided in
systematic and random errors. Each type of error originates a corresponding
wrong behavior which is identified as accuracy and uncertainty (or precision),
respectively” (GUIDI REMONDINO ET ALII 2008: 60)
Accuracy is connected to the quality of the obtained results, while precision
relates to the quality of the operation by which the results are obtained.
Precision concerns the reproducibility of an acquisition done several times on
the same sample (ARTIOLA PEPPER BRUSSEAU 2004: 19-20). In spatial systems
accuracy concerns the level of data correctness in accordance with some
agreed-upon level of precision in the coordinate system in use.
Resolution is the amount of spatial detail that can be observed. It is the minimal
mapping unit for vector data and the dimensions of the pixel (or cell) for raster
data. In satellite images, the spatial resolution can vary “by order of magnitude,
Reconstructing Ancient Landscape S. Pescarin

(such as Spot of 10 meters, Ikonos of 1 meter). There is also a temporal

resolution that is the minimum duration of an event that can be observed.
Thematic resolution refers to the precision of the measurements of the
attributes, it can depend upon a measuring device or the number of categories.
Spectral resolution refers to the number, spacing and width of wavelength
bands of electromagnetic energy that a sensor can detect (ARON GLASS 2001:85-
86).

3d photogrammetry (imaged based semi automatic) > = 1 cm

3d Laser scanner < 7 mm
DGPS dual frequency < 1 cm
DGPS with Racal satellite correction < 25 cm

Table 1.3
Consistency is to the absence of contradiction in the data, in terms of space,
time and theme, while completeness regards the absence of missing
information.
Reliability concerns to what extent a reconstruction can be called correct, in
accordance with resources used (historical, archaeological, etc.), input data
accuracy and resolutions, the processing method followed (level of
simplification, etc.), and scientific method applied.

.B.4 Virtual Archaeology: state of the art

Virtual Archaeology, in the definition of Maurizio Forte, regards the
reconstructive digital archaeology, computational epistemology, and the
reconstruction of three-dimensional archaeological ecosystems (FORTE 2000,
FORTE 2007: 8). Although it was originally used for 3d computer models of
ancient buildings and artefacts (REILLY 1990: 133-139), the concept goes
beyond the simple digital treatment of archaeological data, since it is related to
the increased value of the information. The reconstruction of ancient landscape,
although it has a wide perspective, is part of Virtual Archaeology.
Within this specific field, in the last 15 years there have been several examples.
Unfortunately we cannot obtain a complete list, since most of these projects
have been developed inside research groups, seen by just a limited number of
specialists.
Nevertheless, I've started to create an archive of virtual archaeology projects3,
analysing papers presented during well-known conferences in the field and
published in the proceedings, such as CAA (Computer Application in
Archaeology), VSMM (Virtual Systems and MultiMedia), VAST (Virtual
3 This work is still in progress and so has not yet been published
Reconstructing Ancient Landscape S. Pescarin

Reality, Archaeology and Cultural Heritage conferences), Eurographics,

SIGGRAPH and so on. Other projects have been found during Internet searches
inside institutions web-pages. While awaiting a serious, scientific repository, I
will try to summarize some results on projects regarding landscape
reconstruction, emerging from the research. In table 1.4 are described the
parameters taken into account.
The projects examined have various aims and follow a variety of
methodologies. Some of them regard the construction of models and algorithms
to understand the territory better, and reconstruct it better. This is the case, for
instance, of the reconstruction of landscape evolution and cultural
transformation at Durres, Albania (SANTORO MONTI 2003), which followed a
geospatial analytic approach; the reconstruction of the territory and its
catchment limits, during medieval times, at Jure Vetere (Calabria, Italy: fig.
1.1), through the application of a Cost Surface Analysis algorithm (FONSECA
ROUBIS SOGLIANI 2007: 87-132, SOGLIANI et alii 2008); the stratigraphic landscape
reconstruction in Daunia, carried on by the University of Foggia with the
project Itinera and the creation of a VR environment “Time Machine” (DE
FELICE ET ALII 2007: fig. 1.2).

Figure 1.1 [territorio jure-329Gabellone.jpg]

Figure 1.2 [TM-abbandono2 copia.jpg]

Some projects have a stronger ecological approach, for example in the case of
the study and reconstruction of ancient woodlands in Hungary (SZABÓ 2005;
SZABÓ MÜLLEROVÁ 2008), or in the Ca' Tron project (fig. 1.3) where
archaeologists, geologists and botanists have worked together to reconstruct
20,000 years of history (BONDESAN ET ALII 2009). In some cases, interactive
applications have been developed to enable a wider public to access cultural
knowledge, inside physical museums or on line, as with the “Narrative Museum
of Appia archaeological Park”, the “Virtual Museum of the Ancient Via
Flaminia” (fig. 1.4) and “Virtual Rome” projects (Appendix A).

Figure 1.3 [CaTron_Rom_ovile_ok.tif]

Figure 1.4 [flaminia_landscape.jpg]

Although it is a simplification, we can divide projects into two main kinds of

applications: “GIS-oriented” and “VR-oriented”. GIS-oriented projects are
developed mainly by archaeologists, historians, geographers, geologists and
Reconstructing Ancient Landscape S. Pescarin

ecologists. The products are used for the visualisation of scientific results aimed
at the scientific community, for knowledge dissemination, and, in some cases,
for preventive and predictive archaeology. Examples of products are
geospecific GIS, webGIS and repositories, drawings and digital images, paper
and traditional publications and videos. Tools commonly used are GIS
software, database, terrain generators or ecosystem generators (Appendix B).
VR-oriented projects are mainly built by computer scientists, architects or
designers interested in landscape, and the computer-game industry; only rarely
by archaeologists, art historians or artists. The products are mainly used for
scientific visualisation or entertainment, although there are also some examples
in the field of collaborative environments (CVE) (Chapter 7). These projects are
based on: advanced 3d programming, massive use of computer-graphic and 3d
tools for off-line or on-line interaction; in some cases game engines (Torque,
OGRE, Quake engine, Unreal, etc.) authoring platforms for VR and games
creation (VirTools DEV, Unity 3d, etc.), 3d graphics toolkits
(OpenSceneGraph, OpenSG, etc) or social network platforms (Second Life,
Active Worlds). They are more oriented to small-medium scale projects, more
than to a wider scale (large territories). Examples of products are: videos
(interactive or not-interactive), multimedia, VR application on line and off line.

Parameters Sub parameters

GIS based or geospatiality In input; In output; in input and output; none
Three-dimensionality of final results
Interactivity Interpretation process; Resulting application; Research or/
and dissemination purposes.
Web-based Research process/resulting application
Collaborative environment and networking On line/Off line
Final results availability and accessibility On line/off line; publication/multimedia/VR
Sources availability
Acquisition methods availability and accessibility
Processing methods availability and accessibility

Table 1.4

.B.5 Conclusions
As we have seen, the approach to landscape reconstruction is absolutely
multiple. Landscape reconstruction is a complex process that includes tools for
analysis, interpretation, visualisation and communication. It concerns the
reconstruction at different levels: the archaeological landscape, mapscape,
mindscape and webscape.
Reconstructing Ancient Landscape S. Pescarin

There are many interesting projects, but in most cases they are known only at a
local or national level. Even in the case of international publications, final
results remain rather unknown, often confined inside laboratories or research
groups. For this reason, there are projects which develop the same topics.
Transparency therefore is still a long way off in terms of commonly, applied
scientific practice. Although almost every project has underlined the importance
of multidisciplinary approach, there are few examples of data sharing or
collaborative environment. Three-dimensional visualisation is considered, by
researchers themselves, useful mainly: for the presentation of scientific results,
to enhance the general interest in archaeological research, to communicate
contents that facilitate interpretation or the grasping of ancient territory
distribution for visitors. The E-learning perspective is also considered in the
creation of virtual worlds for immersive facilities or for web based learning
environments. The 3d interactive visualisation of reconstructed landscapes is
also felt to be important to attract the attention of cultural authorities. There is
also an increasing awareness over considering 3d and VR applications as useful
for research, if aimed at exploring and studying past living spaces through
virtual and simulation models, or at developing perceptual analyses and
alternative modes of appreciating and constructing places.

I put forward some useful guidelines at this point that will be further discussed
in the book, ten golden rules for ancient landscape reconstruction:
1. Use scientific methods, while defining your digital pipeline
2. Acquire all available archaeological, historical, architectonic but also
environmental information (a multidisciplinary approach is required)
3. Start mapping the contemporary archaeological landscape, then work on
interpretation, and finally propose a hypothesis on potential landscape
reconstruction
4. Consider the spatial and temporal dimension of all data (2d and 3d)
5. Consider relations among data, objects, sites, etc.
6. Organize your data in a structured way (database, GIS etc.): get used to
annotation during the acquisition, interpretation and reconstruction
phases
7. Use an open approach: landscape reconstruction is a never-ending
process. Use open standards and formats as much as you can: landscape
reconstruction is a collective process and you probably need to share
data, information and results
8. Make use of virtual spatial systems
9. Practice observation, comparisons and critical debates to correct or
improve your reconstructions
10. Use Internet and publish your results in a transparent way
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Chapter 2
Interpretation and Reconstruction

.B.1 Vision, interpretation, and knowledge

Interpretation comes from the Latin intĕrprete(m), whose meaning is originally
'intermediary' or 'mediator'. In fact it consists of inter (in the middle) and
pretium (price) (CORTELLAZZO ZOLLI 1999). Interpretation plays an important role
in the cognitive process, because it creates a link, a bridge between vision and
knowledge, among archaeological findings, observations or sources and
knowledge.
Landscape reconstruction final output is a visual result: the reconstruction or
reconstructions. It is therefore a visual activity. In addition, it is a product of our
brain and must follow its laws. Although an archaeological reconstruction is
commonly labelled “dangerous” by scholars since it freezes one personal vision
of the world, shaping it as the real truth, without possessing enough data, we
will see that this is inaccurate, first of all from a neurobiological point of view.
How do we acquire knowledge of the world? This is a central issue in the
process of analysing available information and producing reliable
interpretation, in order to obtain a better knowledge of the past. As we have
seen in the previous chapter, we observe to obtain a knowledge of the world.
The world around us is complex, because of the quantity of information, but
also because of the speed this information changes. We must survive without
going mad, keeping what we need. Our brain’s activity is continuously focused
on the identification of constant characteristics. It is essential, in fact, to classify
what we see into categories, in order to understand. The classification takes
place in three separate processes: selection, exclusion and comparison. We
select constant properties from different variable information, then we eliminate
what is not relevant and finally we compare it with what is already visually
registered in our brain (ZEKI 1999: 4-23). Even a philosopher such as Gregory
Bateson has defined the process of knowledge acquisition as an activity of
comparison among differences (BATESON 1972).
We can build reconstruction of the landscape that can really help the acquisition
of knowledge about our past. The use of vision is extremely useful in the
extraction of essential information from the quantity of available data. The
landscape itself is a perfect example of inconstancy: it is continuously
changing. The attempt to define its essential characteristics, selecting constant
elements, opens up a more durable interpretation of reality (ZEKI 1999: 28).
From a neurological point of view, the discoveries of the last thirty years have
changed what we know about our brain and about the way it works. At the
Reconstructing Ancient Landscape S. Pescarin

beginning of the 20th century scientists like Henschen and Flechsig developed a
theory which distinguished a main area of the brain, devoted to vision -
interpreted as passive and primitive activity - and a second area around it,
known as the "associative cortex", devoted to understanding - an active task -,
obtained from the association of the received image (main area) and other
images stored in this second area. In this first approach, passive observation
was distinguished from active interpretation (FLECHSIG, 1907: 75-89). Recently,
new observations have changed this theory. It is quite accepted today that there
is a main area, known as V1 or "primary visual cortex" and located in the
bottom left hemisphere of the brain, whose function is vision, and several
distinct areas around it (V2, V3, etc.), committed to specific visual aspects,
such as: shape, colour, movement, depth and relation. The entire observation
process is therefore an active task, starting from vision itself, since it is a
selective activity, defining essential characteristics (ZEKI 1999: 38). While we
are watching a scene, we are already making choices, selecting constant
aspects, referring to the shape, the colour and also to the relation among
objects, situations or concepts.
Maps are perfect examples of the representation of essentiality, as they are built
after a process of observation, selection and exclusion. Interpreted landscape
should make use of interactive maps, exploiting their power to create
mapscapes. A useful constant concerns the identification of similar
characteristics, common to a variety of situations, which can help to define a
representative condition, called by neurologists “situational constant” (ZEKI
1999: 41). A mapscape is a depiction, a description of a representative
landscape, a simultaneous representation of many realities, valid for most cases.
Furthermore, it can really help in the interpretation process, proposing various
potential visions. Ambiguity in this sense becomes a positive aspect, since it lies
in the presentation of many different visions that contribute to the
interpretation. A mapscape is also an interactive vision. Through the connection
to sources and data it acts as a bridge between observation and knowledge, as
an interpretation space.
How reliable is a reconstruction?
We should say that every interpretation uses the “comparison” with what is
already stored in our brain. It is therefore intrinsically personal, since it is based
on previous knowledge. Many archaeological interpretative processes are
commonly considered subjective, since they depend on the skill of the
interpreter, their experience and training. But they are subjective firstly in
another sense: they are part of a personal process. Moreover, there is no
uniquely “true” situation or correct answer. Nevertheless, what we have
described may grant that the visualisation of reconstructions isn't restrictive.
Virtual reconstructions should be striven for to obtain constants and permanent
Reconstructing Ancient Landscape S. Pescarin

shapes from multiple situations, an essential representative vision, reaching for

a greater awareness (SCHOPENHAUER 1819: §38). Virtual Reality can represent the
world in an essential shape, with such intensity and evidence that it can be
considered as reality itself.

The use in the reconstructions of ordinary situations, ordinary characters,

ordinary and not specific objects can contribute to maintain this ambiguity and
representativeness. Ambiguity drives a user to put new questions, activating
connections with previously stored information or stimulating to find new
answers from new data.
Another neurological trick that enables a more effective interpretation is the
“unfinished”, an implicit constant. Artists such as Michelangelo left many
pieces of art unfinished, revealing the power of multiplicity; the observer
participates in the work with his imagination, adapting different concepts and
representations of his mind (SCHULZ 1975: 366-373, ZEKI 1999: 48). In this sense
a reconstruction can be considered close to art, when it helps interpretation with
the power of evocation, widening its horizons and its cognitive potential.
Observing that reconstruction, we should be able to acquire information of the
represented categories. Nevertheless, we should also remember that any
represented objects cannot substitute the incredible wealth and complexity of a
brain representation.
Reconstructions should present, if possible, detailed visualisations. The brain is
interested in details, aiming to bring them into a more general scheme. But
images are stored in our brain if they are recognized. Memory is connected to
the temporal cortex and to the several specialized areas of the brain, above
mentioned. Probably the specialization of these areas (the identification of
colours, shapes, etc.) contributes to the memorization and the cognitive process
(LOGOTHETIS et alii, 1995: 552-565).
In this perspective, the use of a cybernetic approach, which takes into account
the complexity, will help to maintain a detailed visualisation.
The word “cybernetics” was introduced by the mathematician Norbert Wiener
in 19484. It comes from the Greek word kybernetes (steersman, governor). It
was introduced as a way to solve a military problem: how to create an efficient
anti-aircraft mechanism for artillery, capable of hitting a moving object (such as
an aeroplane)? The required system had to take the trajectory into
consideration, but also predict any possible irregular changes in its position,
integrating a feedback mechanism for a continuous correction in accordance
with its real-time behaviour. The problem therefore was the management of

4 Norbert Wiener (1948), Cybernetics or Control and Communication in the Animal and the
Machine, (Hermann & Cie Editeurs, Paris, The Technology Press, Cambridge, Mass., John
Wiley & Sons Inc., New York, 1948).
Reconstructing Ancient Landscape S. Pescarin

complex situations in a future perspective.

Immoderate simplification can preclude a real understanding, and it is in this
sense that complexity should be taken into account. The goal in fact isn't, again,
the impossible reproduction of past realities, but a contribution to their
interpretation, through the construction of models of the past. A technique that
in the future will be increasingly adopted for this purpose is procedural
modelling. It uses and develops algorithms for producing environments, using
shape grammar, starting from a set of rules that can be part of the algorithms
and also modifiable through parameters. An example applied to Roman
architecture and the procedural generation of Roman cities has been applied
successfully. In this test case, the Roman city of Pompeii was completely
reconstructed, using a “shape grammar” (called CGA shape), with the
production of rules that evolve, creating more and more details (fig.2.1). An
advantage of this method is that the hierarchical structure and the annotation of
the models can be reused, creating different procedural variations (MULLER et
alii, 2006)
.
Figure 2.1 – cap2_fig1b.jpg

Dynamic, chaotic and not linear systems, based on neural networks algorithms,
can be used to build very complex environments which can evolve,
independently from initial conditions. Neural networks, as explored by the
mathematician and pioneer of chaos theory Edward Lorenz, can be very useful
to deal with complex systems and to build relations. Today there are also new
approaches which can be applied even to landscape reconstruction. While at the
beginning scientists were much more interested in finding a way to create a
replica of the human mind, today studies are more focused on the development
of relations. There are new generations of neural networks, known as smart
adaptive systems, that can better represent dynamic scenarios, continuously
changing. These systems are built to adapt themselves to an evolving (over
time, space, etc.) environment, recognizing changes and reacting accordingly,
but they are also designed to answer problems very familiar to those who work
in the field of landscape archaeology: is it possible to build a system, starting
from very little information, through incremental learning?5.
Some projects on Artificial Life (AL) are trying to answer this question. AL is
the discipline which is in charge of creating organisms and systems, that might
be considered alive if found in nature (TURKLE 1997:174). One of the first
applications, known as “the blind watchmaker”, was developed by Richard

5 See EUNITE - the European Network of Excellence on Intelligent Technologies for Smart
Adaptive Systems which from 2001 is working on the field (http://www.eunite.org)
Reconstructing Ancient Landscape S. Pescarin

Dawkins, while others have been created as computer simulations of nature

development, such as “Tierra”, created by Thomas Ray in the early 1990s, or
“SimLife”, a sort of ecological virtual system and genetic game. These
simulations can be really challenging, to understand better how natural realms
work.
Genetic algorithms represent another example of artificial intelligence, since
they are built to include different variables, each one representing the solution
to a problem. These algorithms create several similar solutions, maintaining
some basic characteristics. They have no auto-organisation, although they
evolve in different ways, maintaining some essential basic characteristics. In the
near future, I believe they could be successfully applied also to the
reconstruction of Ancient Potential Landscape, since they seem to be very
powerful means of enhancing explanation and communication. A good example
of the application of such an approach is the so-called Generative Art.
Generative Art creates artificial digital objects whose common characteristic is
that they have the same “genetic code” (fig, 2.2). We can say that in these kinds
of artworks what is stored are the essential characteristics of objects. As the
artist Celestino Soddu states, a generative project is a concept-software that
works producing three-dimensional non-repeatable events as possible
expressions of the generating idea identified by the designer as a subjective
proposal of a possible world (SODDU 1991).

Fig. 2.2 – CelestinoSoddu_generazioneCittaMedievali.jpg

Some scientists have applied artificial intelligence to sciences more related to
landscape archaeology. Lovelock and later Margulis have developed the Gaia
Hypothesis, defining the earth as a huge single organism that can self-regulate6.
All the components (organic and inorganic elements) of the earth are strictly
connected and can be thought of as one organism, a single system in which
many retro-action rings keep chemical and physical conditions stable, thanks to
an automatic control for example of the global temperature, transforming it into
an inhabitable place to live. All the components normally evolve together: life
maintains conditions suitable for its own survival7. Lately other scientists
suggested that this auto-organisation can modify evolution speed, and this can
be seen by a progressive increase of complexity. Although we live in a chaotic
6 James Lovelock, 2006, The Revenge of Gaia: Why the Earth Is Fighting Back - and How
We Can Still Save Humanity. Santa Barbara (California); Lynn Margulis, 1982, Early Life,
Science Books International, Margulis, Lynn, 1998, Symbiotic Planet: A New Look at
Evolution, Basic Books, Margulis, Lynn and Eduardo Punset, eds., 2007 Mind, Life and
Universe: Conversations with Great Scientists of Our Time, Sciencewriters Books
7 http://www.gaiatheory.org
Reconstructing Ancient Landscape S. Pescarin

and not casual system, where does the organisation we see everywhere around
us and inside us come from? Complexity generates very much of the order of
the natural world. Selection then acts to further refine it. Every system is in the
middle between chaos and order, at the boundary (KAUFMANN 1995). The
interpretation of ancient landscape should take into account organised and
chaotic elements and it should consider schematic and evolutionary approaches
since a successful exploration of our past lies in a sort of transition between
order and disorder. We will often have to consider, for example, how a
population has shaped its territory. Even in this case, we should know that
communities that have not limited themselves to the exploration of a territory,
and arrive at more remote areas, will find better and more stable conditions. We
often face the problem of neighbouring communities (some GIS analysis, such
as Thiessen Polygons, involves the identification of areas of influence), where a
co-evolutionary approach should be considered: “in co-evolving systems, each
partner clambers up its fitness landscape toward fitness peaks, even as that
landscape is constantly deformed by adaptive moves of its co-evolutionary
partners” (KAUFFMAN 1995: 20-27).
The way we work on the interpretation of our past has changed in the last
century. Computers allow us to manage complex systems and a large quantity
of information together, which was impossible to handle before. The continuous
and rapid access to different information opens up new perspectives in the
knowledge of ancient landscape, but it can also be “dangerous”, since it hides
the risk of superficiality. In the simulation era, understanding depends upon
exploration and upon personal paths, not just upon analysis. In the continuous
navigation from one piece of information to another, we might stop at the vision
of the surface, viewing the word as a collection of signifiers with no meaning.
The risk of being manipulated by our simulations should be considered in the
rapid and multidisciplinary access to information, for a greater awareness of
research contributions (TURKLE 1997).
Another aspect highlighted by researchers such as Maturana and Varela is the
concept of autopoiesis (MATURANA VARELA 1980). This new approach can give
useful suggestions also in our case, since it stresses the prominence of
relationship, rather than properties of living elements. The analysis of relations
should necessarily follow that of separate characteristics and properties.

.B.2 Interpretation: a theoretically explicit scientific approach

Reconstructing ancient landscapes is an interpretative and open process, where
science, but also imagination and art, are important. Its value lies in its
reliability, in the process of available data analysis through an explicit
methodology (the scientific approach) and an interpretation process
Reconstructing Ancient Landscape S. Pescarin

(imagination – art). As Schopenhauer stated, the most important things should

always be left to the imagination (SCHOPENHAUER 1819: § 34).
It is, moreover, an open process, since each "interpretative cycle" leads to the
production of new data, observations, and analysis.
Landscape reconstruction doesn't just produce a "communicative” result for an
end-user. More or less realistic and immersive visualisation inside territories
and meta-territories can be considered part of the research activity. And this is
thanks to interaction and transparency potentiality. Communication is
integrated in the research process of interpretation, when it is used, for instance,
to test theories. Chris Langton is considered the founder of Artificial Life
(ALife), director of the Santa Fe institute in New Mexico. He worked on
developing programs that evolve autonomously in the computer. He creates
communities of individuals who initially learn some basic rules and then evolve
autonomous ways. Langton tried to recreate “life as it could be” and not “life as
it is”, opening up to the reconstruction of its potentiality (LANGTON 1989).

Also ancient landscape can be considered an ecosystem, a complex system that

needs interdisciplinarity and integration (of data, tools, formats, etc.) to be
engaged in. It is strongly connected to space, time and relation concepts. It is
something more than a simple “object” that needs to be reconstructed. It is a
dynamic system whose aspects, whose creation and evolution are connected
with each other, and depends upon geomorphology, climate characteristics,
altimetry, anthropic modifications, and also cognitive and perception aspects.
In this perspective, each analysis and interpretation that takes into account the
territory, should consider the spatial dimension of all data as fundamental.
Moreover, it should start with mapping the existing archaeological landscape.
Without the archaeological and observed landscape, in fact, it won't be possible
the get essential information on ancient landscape reconstruction.

We started working on these concepts in the project of the “Virtual Museum of

ancient via Flaminia”. The aim was the realisation of a multi-user VR
application regarding an archaeological landscape, in northern Rome, crossed
during Roman times by the road via Flaminia. Although the final result of the
project was a public installation inside a museum, with evident communication
priorities, the reconstruction process of the landscape was based on a serious
interpretative work, based on a GIS project, on the processing of all available
data, on raster map analysis and on new data definition. At the end of the
process, inside the reconstructed landscape, high resolution 3d models of
archaeological sites have been integrated (fig. 2.4; Appendix A).
The reconstruction of Bologna during Roman Times has been another project
following the same direction. Archaeological urban landscape has been first
Reconstructing Ancient Landscape S. Pescarin

created and used as reference for the reconstruction of Roman geomorphology

(potential landscape), based on all available archaeological excavations. The
simulated geomorphology has then been used to perform further GIS analysis
on the potential hydrography and on the use of urban space (Fig. 2.3; PESCARIN
2002).

[Figure 2.3: Bononia.tif]

In the case of the Via Flaminia project the aim was to reconstruct the actual
archaeological landscape, but also the ancient potential landscape. The work
was carried out in a wider perspective, with the idea of analysing also a possible
and more general method to reconstruct ancient landscapes, through the use of a
formal explicit process and of a series of connected "algorithms”. The reason
for this approach was the idea of defining an open process, where observations
and assumptions could be controlled step by step, and where it was possible to
modify, almost automatically, the final result, as initial input data changed
(modifications, new introductions). At the end, a VR application based on final
outputs of such a process was used to test the hypothesis, working with
interpretations. A variety of tools were used to obtain inter-disciplinarity,
reliability and updatability. Although reliability depends strictly upon available
data (quality/quantity), the creation of an updatable system was connected to
the method used.
The final and accepted result was then used and implemented inside the
MultiUser Virtual Reality application, created appositely inside a new room of
the Roman National Museum at the Diocletian Thermal Baths in Rome.

[fig. 2.4 _fig05.tif]

Ancient landscape reconstruction is achieved through various phases: the

reconstruction of the archaeological landscape, of the ecological attitude, of
interpreted landscapes and the reconstruction of ancient potential landscapes.
In the case of historical landscapes, through the study of the ecological attitude
of a territory it is possible to reconstruct classes and related sub-categories, such
as woodlands and agricultural lands (mainly for cereals, vineyards and olive
trees, but also pasture) (DI FIDIO 1990: 215, VOLPE ARNOLDUS HUYZENDVELD 2005,
DRAMSTAD OLSON FORMAN 1996, CHIUSOLI 1999). In order to obtain these classes,
spatial analyses should be carried out on different raster grids. At the end what
we obtain are vector layers, that can be further modified during following
processing. These layers can be used as input in 3d terrain generation software
and VR application development.
Reconstructing Ancient Landscape S. Pescarin

Although uncertainty is part of archaeologist’s daily life, the importance of

scientific method is quite evident. A scientific approach allows reliable analysis
and reconstruction, with the final goal of getting to know new elements,
propose new explanations, generate criticism and discussion, and correct/emend
existing data, thus providing a positive impetus to research. The definition of an
accurate and consistent representation of ancient landscape is an open and
collective process which allows us to formulate an interpretation model of past
landscape. Ancient landscape cannot be considered real or true, but we can
work on the potential landscape, to be thought of as an open process, each time
closer to the truth, with continuous approximations. The creation of potential
landscapes will be further described in Chapter 4.
Table 2.1 proposes an example of how the scientific approach can be used, as
referred to a Roman landscape. The initial input was how the landscape should
have appeared, during the early Imperial period, in the suburbium (the area
outside the city). This question was then subdivided into sub-problems easier to
solve: such as the identification of Roman vegetation, of climate conditions for
that period and latitude, of anthropic elements (archaeological sites and remains
excavated or surveyed or predicted in archaeological maps), of cultural
elements (funerary habits, and so on). The archaeological landscape was
reconstructed, a GIS project developed, and vector information overlaid.

Scientific Method Scientific Method in Steps in Roman potential landscape

Landscape Archaeology reconstruction

1. Identification of Identification of What did the Roman Landscape look like

significant question significant question during the early Imperial age?

1B. Subdivision into Subdivision into sub- Type of vegetation, of climate, of

sub-questions questions anthropic or cultural elements?

2. Collection of Collection of information Distribution of archaeological sites; paleo-

information and main (published, ancient botanic analysis; geological analysis, etc.
data sources, on the field) and
main data. Georeferencing

3. Definition of a Definition of a hypothesis Ancient landscape can be defined by

hypothesis that could that could explain the potential landscape and anthropic/cultural
explain the question question: reconstruction modification; how can theory formation
of ecological attitude of be simplified with algorithms?
the territory and
integration of anthropic
Reconstructing Ancient Landscape S. Pescarin

and cultural elements.

4. Hypothesis or Testing with VR 3d tools 3d terrain generation, main ecosystems,

explanation testing Tools of Virtual Reality anthropic elements; implementation in 3d
enabling us to carry out real-time engine for testing in 3d.
further verification and
observations

5. Testing of Testing of hypothesis Terrains 3d generation, main ecosystems;

hypothesis even with VR tools for anthropic elements; implementing in a
further observations real time engine

6. Acceptance or Get back to previous step Get back to GIS project and re-
rejection of the for further work and new implementing of new data also in the 3d
hypothesis tests engine

7. Theory definition Definition of possible Reliable reconstruction of ancient

theories on the landscape
reconstruction of ancient
landscape

8. Experimenting, test Theory verification on Use of the same algorithm with other case
of the theory and other landscapes and other studies
acceptance or data. Accept, modification
rejection or rejection

9. Results published Results published through Results published in scientific

and data validation traditional scientific publications and publication of results
through peer review publications, the web and over the web through VR webGIS open
mechanism other open tools source application

Table 2.1

The archaeological landscape can be employed during “interpretative sessions”

involving archaeologists, computer scientists, geologists and botanists.
Interactive 3d and 2d tools might be used for this purpose, such as VR systems,
webGIS and VR webGIS. New thematic layers are then defined so as to
identify main landscape classes (such as vegetation, hydrography, roads, etc.).
These classes represent specific sub-questions, valid in a certain historical
period, latitude, climate, geomorphology, etc. Each time a new set of data is
created (specific DTM, farming terrains, potential land use maps, hydrography,
location of archaeological sites, etc.) it is also tested inside a 3d or shared
environment, where verification and observations are carried out from "inside"
Reconstructing Ancient Landscape S. Pescarin

and are employed to create further analysis (new areas of interest, predictive
maps, etc.). At the end of the process 3d terrains and main ecosystems are
generated, then implemented in a 3d real time application obtaining a final
result. If the entire process is formalized and discreted, through the
implementation of single analytic algorithms, connected to each other, it is
possible to have a different final result each time initial inputs change or are
modified (fig. 2.5). Despite the solid confidence that emerges from the
application of this process, it should be said that every hypothesis can be
discussed and demonstrated as false. Nevertheless, as Popper observed in 1934,
a hypothesis should be counterfeitable, and a theory cannot be called scientific
if it doesn't admit the possibility of being demonstrated as false. Theoretically,
in fact, it should be always possible to make an observation that can
demonstrate that the theory is false, even if this observation hasn't yet been
made (POPPER 1934).

[fig. 2.5 alternative_landscapes.tif

After the phases described, it is possible to obtain a more general theory, a self-
consistent explanation of the initial question, based on a valid hypothesis. The
definition of a theory helps to trace the new predictions for further testing
At the end, it is also essential to publish results, for validation by the scientific
community through peer-review mechanisms, even using digital publications or
more advanced interactive systems. Unfortunately, in the humanities it is almost
impossible to define “general laws” and to get to definitive interpretations,
without any uncertainty. Also for this reason archaeologists tend to postpone
this moment as long as possible, although if we consider the development of
research in a broader and collective sense, it would be better to take greater care
over documentation production and results sharing (PESCARIN 2006: 137-155).

.B.3 Conclusions
Every reconstruction is the result of an objective and subjective process at the
same time. Final results’ transparency of interpretation, therefore, visible
through simulation and reconstruction models, is obtained through interactivity
and an open approach (PESCARIN 2006, FORTE PESCARIN 2007).
Following a scientific approach, maintaining as much as possible the detail, can
help in overpass risks and the limits of virtual reality simulations.
The work done in recent years by the Virtual Heritage community has
demonstrated that the convergence of different disciplines in landscape
archaeology can bring about a more efficient, valid study of the territory,
Reconstructing Ancient Landscape S. Pescarin

providing at the same time new interpretation keys of ancient landscape.

This convergence also allows a more appropriate and aware use of available
technologies, using and modifying them in accordance with requirements.
The projects described can demonstrate how far ancient landscape
reconstruction can go, and how reliable and valid they can be, if based on an
explicit interpretation process (PESCARIN RUZZA 2004, GUIDAZZOLI ET ALII 2004,
PESCARIN 2000, PESCARIN 2001, PESCARIN ET ALII 2007, PESCARIN 2007, FORTE,
PESCARIN, PIETRONI 2005: 79-92, FORTE ET ALII 2006.).
Reconstructing Ancient Landscape S. Pescarin

Chapter 3
Archaeological landscapes reconstruction:
mapping the space

Every project concerning ancient landscape reconstruction should start from the
archaeological landscape, that is the contemporary landscape, considered in its
diachronic dimension, consisting of monuments, sites, scattered remains, signs
of agriculture, natural characteristics and so on. It is therefore necessary to start
“mapping the space”.
As we outlined in chapter one, the archaeological landscape is made up of
several elements (natural, anthropic material and immaterial, etc.) and by the
relations which connect these elements.
The reconstruction process does not involve only “visible” aspects (reality-
based reconstruction), but also “invisible” ones, such as cultural and relational
aspects, or all those elements that might be deducted in the interpretation
process or discovered with the use of specific technologies. In the past, any
visualisation was “restricted” by its scale: objects beyond a certain threshold,
unable to be represented on a map, were drawn at a certain scale. Moreover, the
territorial scale was usually treated separately from the intra-site scale. Sites or
monuments were often reconstructed as single detailed monographic studies,
rarely even placed in relation to their absolute geographical dimension (see
chapter 1). Also in terms of computing capabilities, the two realms, territory
and site, were treated separately in an attempt to solve the various problems
arising, and in developing different technologies to acquire them. Today, there
is increasing interest in a sort of “global” integration, driven also by the
technical capabilities of new hardware and software: terrestrial and aerial 3d
scanners, the developed potential of photogrammetric applications, the accuracy
of new cameras and DGPS, etc. integration of dataset at different resolution is
leading, slowly, to new problems, in terms of precision, data structure, etc., as
has emerged from projects such as Virtual Rome or 3d-ARCHs (PESCARIN ET
ALII 2008, REMONDINO ET ALII 2009 and figure 3.1). How can we visualize models
with different resolutions, and built to different scales, in a single frame? Can
we merge aerial views with the internal reconstructions of monuments? Should
we express the difference explicitly or had we better develop more
homogeneous landscapes? Could these be threshold visualised, e.g. through
transition surfaces (AGUGIARO KOLBE 2009), and how ? Also from a “perceptive”
point of view, this problem has great impact. The process will probably remain
slow, at least until high quality acquisition devices, such as 3d scanners, though
already considered mature, become more affordable for Cultural Heritage and
Reconstructing Ancient Landscape S. Pescarin

Environmental applications (CIGNONI SCOPIGNO 2008).

[figure 3.1- (CASTLE_TN.jpg)]

A common characteristic of all this data is spatial dimension. The geo-location
of all elements, although often neglected, is central. For this reason it is
important to have a basic knowledge of archaeology, cartography, image
processing and 3d modelling issues. These are very different fields that
commonly make use of different languages, different approaches, and produce
different results. Landscape modelling requires us to manage and interconnect
them. In spite of the fact that, in the development of the specific tasks required
by a project, these fields should be approached by specialized experts, results
can be more satisfactorily obtained if the latter have a basic knowledge also of
interconnected disciplines.
The following sections will analyse various techniques and methodologies for
data acquisition and archaeological landscape reconstruction. Communication
and final visualisation will be further described in chapter 5.
Table 3.1 proposes a sequence of activities related to the creation of the
observed landscape.
General activities Methodology/data/sources
GIS project creation GIS software selection, projections and coordinate system, thematic layers
and levels of detail definition
Generation of contemporary DTM Elevation data mosaic
Processing of geoimagery Geoimagery georeferencing, mosaic, equalization (satellite, aerial, etc.)
Import (in the GIS) of acquired and Archaeological map, historical maps, soil maps, etc.
georeferenced data
Overlay with acquired data on the field Data acquisition on site with integrated technologies (GPS, 3D scanner, etc.)
and creation of detail survey of the monuments.
Identification of critics in the datasets Data validation based on the comparison with all dataset.
Data correction and new layers creation Data correction and creation of new vector layers (GIS and modeling
software).
Final thematic layers creation Export of useful thematic layers: hydrography, road system, etc.
3D terrain generation 3D model generation of the territory, based on processed geoimagery and
DTM.
Vegetation generation 2D/3D vegetation integration at different details
3D models integration Further postprocessing of 3D models and integration on correct location (lat/
long/elev).

Tabella 3.1 Steps of archaeological landscape reconstructions.

Reconstructing Ancient Landscape S. Pescarin

.B.1 Visible landscapes

The reconstruction of visible landscapes is obtained mainly with mapping
activities. The first step is to verify the existence of available data, such as
cartographic maps, thematic maps (vegetation, geology, etc.), aerial or satellite
images, elevation models, planimetries, sections, etc. or any other sources
useful in the construction of the basic digital dataset. A basic dataset is made up
of at least one geoimage (or mosaic of geoimages), and one Digital Terrain
Model. Thematic layers, vector or 3d models, are complementary. In the past,
the usual way to obtain this data was to find printed images or a paper map,
scan them, georeference and orthorectify them, and finally manually digitize
contour levels or elevation points in order to produce DEM, or other required
themes.
These datasets can easily be purchased from specialized companies at various
rates, mainly depending on data resolution. In some cases, they can be freely
downloaded from on line repositories, such as Global Land Cover Facility
(GLCF) or USGS8.
GLCF is a centre for land cover science focused on the analysis of cover change
through remote sensing. It is an on line database of many satellite images and
elevation models, accessible through the Earth Science Data Interface (ESDI:
figure 3.2)9. Here, Landsat series imagery can be downloaded: Landsat MSS
(Multi-spectral Scanner), TM (Thematic Mapper), and ETM+ (Enhanced
Thematic Mapper Plus). The highest available resolution is of 30 mt (15 mt for
the Panchromatic10 L7: table 3.3). There are today many types of satellite
images, with a variety of characteristics. Knowing these characteristics helps to
find the right source for a project. Table 3.2 reports available images on the
market, with indications of their main use and resolution. Other available
imagery products are aerial ortho-photos11.
DTMs with grid sizes of 500 up to 1m are increasingly available for different
parts of the globe, such as through the US Geological Survey (USGS), the
Ordinance Survey, the Italian Istituto Geografico Militare (IGM)12 or Parma
CGR13 repositories. Low resolution Digital Terrain Models are also available on
line from the USGS site, e.g. the GTOPO30, available since 1996 thanks to the
work of the USGS Center for Earth Resources Observation and Science

8 http://edc.usgs.gov/products/elevation/
9 http://www.landcover.org; http://glcfapp.umiacs.umd.edu:8080/esdi/index.jsp
10 Panchromatic: Imagery Single-band or monochrome imagery.
11 Examples at: Terraserver (www.terraserver.com); FlashEarth (www.flashearth.com);
DigitalGlobe (www.digitalglobe.com); TerraFly (www.terrafly.com); Italian CGR
(www.cgrit.it), and in many other countries.
12 http://www.igmi.org/
13 http://www.cgrit.it/prodotti/modello_digitale.htm
Reconstructing Ancient Landscape S. Pescarin

(EROS). This global DEM, based on the Shuttle Radar Topographic Mission,
has horizontal grid spacing of 30 arc seconds (approximately 1 kilometer)
deriving from several raster and vector sources of topographic information.
Unfortunately the absolute vertical accuracy of the elevation data is 16 meters
(at 90% confidence). Most of the available DTM are the result of
photogrammetric data capture, performed with stereoscopic interpretation of
aerial or satellite images through manual or automatic systems. DEMs can be
acquired also by digitizing contour lines on old topographic maps and alsao by
conducting ground surveys, with GPS devises14.

[fig. 3.2 srtm_comparison.jpg]

Sensor Description Resolution

ALOS /2006 It has 1) Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) 2.5m to 10m
for DEMs mapping, 2) Advanced Visible and Near Infrared Radiometer
(AVNIR-2) for land coverage observation, 3) Phased Array type L-band
Synthetic Aperture Radar (PALSAR) for land observation.

ASTER /1999 It monitors cloud cover, land temperature, land use, vegetation patterns, etc. The 15m to 90 m
multispectral images are in 14 different colours, used for change detection and
land surface studies including vegetation and ecosystem dynamics.
CARTOSAT- Built by the Indian Space Research Organization as part of the Indian Remote 2.5 m
1 /2005 Sensing series (IRS). It has 2 panchromatic cameras that take black-and-white
stereoscopic pictures in the visible region of the electromagnetic spectrum, used
to create accurate 3d maps.

CBERS-2 /2003 Born of a partnership between Brazil and China. Used for deforestation and fire 20m to 260m
control, soil occupation, etc. It includes cameras for optical observations.

CORONA photographic surveillance satellite used from the late 50's through the early 70's, 2.75 m to 1.8 m
after de-classification 860,000 images are available through Earth Explorer15

EROS /2000 provides 1.8 meter panchromatic imagery (areas: 12,5x12,5 km) 1.8 m

FORMOSAT-2 It is used for land distribution, natural resources research, forestry, environmental 2m to 8m
/2004 protection purposes.
GeoEye-1 /2008 Capable of acquiring image data at 0.41 meters panchromatic (B&W) and 1.65 0.41m to 1.65 m
meters multispectral resolution.

GeoEye-2 Forthcoming 0.25-meter

IKONOS /1999 Used for urban and rural mapping of natural resources, agriculture and forestry 0.82m to 3.2m
analysis. Available Ikonos Stereo Images for the production of Digital Surface
Models (DSM's) or Digital Elevation Models (DEM's) with postings of ≤5m.

14 There are several archives of old aerial or satellite images in many countries, such as
Historical Aerial Photography in UK (www.oldaerialphotos.com) or the Italian Fototeca
(www.iccd.beniculturali.it/Istituto/Organizzazione/aerofototeca).
15 http://edcsns17.cr.usgs.gov/EarthExplorer/
Reconstructing Ancient Landscape S. Pescarin

IRS acquired over the visible green to near infrared 5.8 m

LANDSAT 7 + first earth observation satellite. It has seven spectral bands that range from visible 30m
ETM /1999 to thermal infrared regions.
QuickBird / high resolution satellite: Pan: 61 cm (nadir) to 72 cm (25° off-nadir); MS: 2.44 m 0.61 to 2.88 m
2001 (nadir) to 2.88 m (25° off-nadir)
SPOT-5 /2002 It has VEGETATION 2 instruments which provide continuity of environmental 5m to 10m
monitoring. An image covers 60 x 60 km or 60 km x 120 km, and there is good
balance between high res. and wide-area coverage

WorldView-1 / A high-capacity, panchromatic imaging system features 0.5m resolution 0.5m

2007
WorldView-2 Forthcoming: it will provide a high resolution Panchromatic band and 8 1.8m to 2.4m
Multispectral bands

Table 3.2 List of main satellite imagery. (Source: Satellite Imaging Corporation).
Satellite Sensor Spectral Range Band #s Scene Size Pixel Res
L 1-4 MSS multi-spectral 0.5 - 1.1 µm 1, 2, 3, 4 60 m.
L 4-5 TM multi-spectral 0.45 - 2.35 µm 1, 2, 3, 4, 5, 7 30 m.
L 4-5 TM thermal 10.40 - 12.50 µm 6 120 m.
185 X 185 km
L7 ETM+ multi-spectral 0.450 - 2.35 µm 1, 2, 3, 4, 5, 7 30 m.
L7 ETM+ thermal 10.40 – 12.50 µm 6.1, 6.2 60 m.
L7 Panchromatic 0.52-0.90 µm 8 15 m.

Table 3.3 list of Landsat imagery sensors.

When it is impossible, in fact, to use or find the desired spatial datasets, or
when more details are needed, we can still obtain them by planning fieldwork
and organising a direct acquisition. Traditionally, archaeology has used specific
investigation techniques such as excavations and field surveys (RENFREW BAHN
1981). Technology today offers a wide range of techniques to acquire, model
and visualize archaeological landscapes with many different scales and varied
resolution, limiting excavations to few important cases.
Archaeological and environmental data are acquired by means of many kinds of
survey and analysis, obtained with terrestrial or aerial acquisitions, direct or
indirect. Usually it is convenient to use them in integration, in order to get
better results on large-scale reconstructions for both intra-site and inter-site
aspects. Terrain acquisitions can be used for detailed cartography, planimetric
restitutions and the 3d modelling of sites, monuments, right up to single
remains. On the other hand, aerial acquisitions can provide their context and
location.
The main aim, for all of the techniques described below, is to acquire
information to produce a model of the archaeological landscape. There are
basically two types of system in the creation of 3d models. The first is based on
measurements, acquired either with contact or non-contact techniques. The
second one doesn't use measurements, e.g. in the case of computer graphics
Reconstructing Ancient Landscape S. Pescarin

(through 3d modelling software such as 3D Studio Max or Maya, etc.) or

surveying (REMONDINO EL-HAKIM 2006).

Terrestrial acquisition
Terrestrial acquisitions are used for reality-based modelling. They make use of
several different techniques and devices, and the most widely used are: GPS,
laser total stations, 3d scanners and digital cameras for photogrammetric
purposes. Instead of entering into a detailed description, I will provide general
information with bibliographic references.

The widespread use of Global Positioning Systems (GPS)16 provides many

opportunities for the definition of the position of point on Earth surface and for
the collection of elevation data, producing reference points for integrated
surveys or detailed micro-topographic surveys. The introduction of Differential
GPS (DGPS) has particularly increased survey accuracy. This enhancement of
the traditional GPS, originally developed to solve the problem of Selective
Availability (SA), can obtain better results, thanks to the use of a ground-based
reference stations network, in charge of correcting the satellite signal through
an in-built UHF band radio modem. Estimated accuracy can be under 10 cm.
Differential GPS measurements can also be computed in real-time by some
GPS receivers, when the correction signal is received using a separate radio
receiver, as in the case of Real Time Kinematic (RTK) surveying.
There are many advantages to using GPS. For example, it is possible to
selectively acquire position or elevation information. This operation can be
further simplified by the use of PDAs, or other portable PCs, with GIS software
installed in it, connected to the GPS. In this way it is possible to directly create
a GIS in the field, at the same time controlling accuracy and the results of the
work in progress (figure 3.3). Nevertheless, the use of this system to create
Digital Terrain Models seems to have been made obsolete by 3d scanners, for
speed and precision. Yet despite its well known problems, GPS remains the
fundamental tool to geo-locate objects on the earth. For this reason it has been
integrated into other kinds of acquisition devices, such as Total Station or Laser
Scanner. Recently the market also offers a combination of Laser Scanner and
GPS mounted on a car, for mobile mapping17.

16 Some references to GPS in archaeology: Wheatley, D. Gillings, M. Spatial technology and

archaelogy, London-New York 2002; Campana S., Sordini M., Laser Scanner e GPS in
archaeologia: geograpfia dei servizi e delle risorse in Internet. In Laser Scanner e GPS.
Paesaggi archeologici e tecnologie digitali 1 (Campana S., Francovich R. eds), Firenze,
2006.
17 Such as the LYNX Mobile Mapper system by Optech (www.optech.ca/pdf/PressRelease-
Lynx-Optech.PDF )
Reconstructing Ancient Landscape S. Pescarin

[fig. 3.3 appia_DGPS.jpg]

There are other techniques that don't require direct contact with the object and
use various systems to obtain a measurement, such as light, X Rays,
microwaves (i.e. radar), or ultrasounds.
Light is used by Range-Based Modelling (RBM) and also by Image-Based
Modelling (IBM), through active sensors (3d scanner, laser total station) or
passive sensors (photogrammetry or shape recovery from edges, texture, or
shading) (REMONDINO, EL-HAKIM 2006).
3d scanner technology is relatively widespread in the field of archaeology and
is an example of RBM18. It uses active sensors, which acquire objects in 3d,
through triangulation or time-delay (TOF) methods. How do they operate?
“They shoot some sort of light over the surface of the artefact and reconstruct
its geometry by checking how the light is reflected by the surface”. Time-of-
Flight (TOF) scanners are largely employed in landscape reconstruction. They
“compute the time elapsed between the emission of a pointwise laser beam and
the detection of the return beam reflected by the surface” (CIGNONI SCOPIGNO
2008: 3). They are generally used for large-scale acquisition projects.
Triangulation 3d scanners use a different technique: they project a single spot or
a pattern on to the object. Reflected images are acquired by an imaging device
nearby, allowing you to compute the 3d position of the points by geometric
triangulation. These scanners are commonly used for small-scale objects.
A typical sequence of activities include: scanner acquisition in the field, post-
processing in the lab where Range-Maps19 can be aligned, merged, edited and
simplified), colour and texture mapping. These activities are supported by
commercial software, although academic tools can present more advanced
features20. There are several advantages in the use of RBM for landscape
acquisition and reconstruction, such as the relative speed in the acquisition
process and the accuracy of acquired data and of the final 3d model.

18 An overview on 3D scanning is offered by Curless, B. Seitz, S., 3D photography. In ACM

SIGGRAPH Course Notes, Course 19. There is a review article on 3d scanning in the
Cultural Heritage field: Cignoni, P. Scopigno, R. Sampled 3d Models for CH Applications:
a viable and enabling new medium or just a technological exercise? In the ACM Journal on
Computing and Cultural Heritage, Vol. 1, No.1, Art.2, Jun.2008;
19 Range Maps are single views of the acquired object that encode sampled points' geometry
(CIGNONI, SCOPIGNO 2008:4)
20 For a review of the software available on the market: CAMPANA, SORDINI 2006 (cited).
Examples of commercial software are: RapidForm, Geomagic, PolyWorks. Examples of
academic tools are: MeshLab, TexAlign, Vrippack, Volfill, PointShop, Scanalyze.
Reconstructing Ancient Landscape S. Pescarin

Unfortunately there are also many disadvantages, due to the still very high cost
of devices and of commercial processing software, to the complexity of post-
processing, and to difficulties in managing very big 3d datasets. There is
another important reason why this sector is still not found very widely, although
successfully tested, in landscape archaeology; this is the “monolithic” way data
are captured, not allowing an easy object extraction and selection. These
problems emerge increasingly in aerial acquisitions21.

Photogrammetry and 3D automatic modelling

An example of Image-Based Modelling (IBM) is Photogrammetry22. It is
commonly used to document, reconstruct, and visualise sites. It is a
“measurement technique” that allows “the modelling of a 3d space using 2d
images” (KASSER AND EGELS 2002 :1), through the identification of
corresponding points in the images (automatic, semi-automatic or manual). A
mathematical model can be used for this purpose, or other methods such as
shape from shadows, from texture, from specularity, from contour or from 2d
edge gradients (REMONDINO EL-HAKIM 2008: 271). Given the specific purpose of
this book, I just mention some techniques useful for the reconstruction of 3d
territories. Generally different views are needed to obtain a 3d model, although
some attempts have been made in the direction of using a single image (EL-
HAKIM 2001).

Aerial acquisitions
The reconstruction of the archaeological landscape requires the development of
models representing the territory (DTM and Geoimages), its natural aspect, the
identification of preserved archaeological sites, and also of anthropic/natural
traces. We have seen that sources can be purchased or built in various ways.
This section is dedicated to those techniques which can be used to obtain aerial
bi- and three-dimensional representations of a territory and its characteristics,
through the acquisition of 2d images or 3d datasets.
Acquisitions can be taken with different sensors mounted on satellite, airplane,
kite, balloon (figure 3.4), small helicopter, plane or paramotors. While it's
impossible to have control over satellite programs, we can, on the other hand,
directly plan air-campaigns, organising and participating in flights over a

21 Many of these problems are handled in: Laser-Scanners for Forest and Landscape
Assessment. In Intern. Archives of Photogrammetry, Remote Sensing and Spatial
Information Sciences (Thies, M. Koch, B. Spiecker, H. and Wienacker, H. Eds), Vol.
XXXVI, PART 8/W2, 2004
22 A good general review can be found in: REMONDINO, EL-HAKIM 2008 and specifically directed
at site reconstruction, in GRÜN 2000.
Reconstructing Ancient Landscape S. Pescarin

territory, when this is possible, or using smaller devices, remotely controlled23.

They provide data on the partial or complete extent of a territory, enhancing
specific spatial and spectral characteristics. The detail is usually lower than that
required for terrestrial acquisitions. For this reason it is good practice to
integrate different models at various scales and resolution.
Since 1960, aerial archaeology is considered an important field for the
interpretation of archaeological landscape, and not just for its representation. It
uses a variety of techniques, according to the goal of the project and the kind of
sensors (passive or active) mounted on board. It is characterized by two phases:
the “observation and photography in or from the air” and the “interpretation of
the images” (BOURGEOIS 2005: 15).
Passive sensors, such as cameras, can be used both for vertical and oblique
photography. Vertical acquisitions, usually taken with cameras pointing straight
down at the ground and fitted inside an aircraft, are used to produce
orthographic images that may be useful for many archaeological purposes, such
as topography (WILSON, 2000: 32, PICARRETA: 1987: xi). Unfortunately, even if
the photo is taken with great care, it still retains some deformations. Usually
what is acquired isn't a single picture, but a series of overlapping photographs.
These can be used for any photogrammetric purpose, such as the creation of
three-dimensional models of the landscape. Models like these reproduce what is
visible in the aerial images and at the same resolution, that is to say: the terrain
and all the structures on it, such as trees, buildings, etc. Various measurements
can be achieved: from single-point measurements up to a regular terrain model,
such as DSM, Digital Surface Model (KASSER EGELS 2002: 169 -190).

[figure 3.4]

Analysis Aerial photo Corona Landsat Spot ASTER QuickBird Ikonos SRTM
Visual interpretation x x x x x x x
Contrast enhancement x x x x x x
Georeferencing x x x x x x
NDVI x x x x
Classification x x x x
Thresholding x x x x
Principal Components Analysis x x x x
Land Use x x x x x
DEM creation x x x x
Hyperspectral analysis x

23 An historical overview is available in: BOURGEOIS 2005. A review of several techniques can be
found in: PICARRETA 1987, BEWLEY 2002.
Reconstructing Ancient Landscape S. Pescarin

Filtering x x x x x

Table 3.4 Technique of geoimage analysis as referred to satellite imagery (source : PARCAK,
2009: 82)
Oblique photography is very useful for landscape interpretation and is much
more used directly by archaeologists. It is a well known technique enabling you
to recognize patterns on the ground which are otherwise impossible to see from
the ground. Although in fact it gives a “distorted” representation of the features
(perspective deformation), it is employed by archaeologists in the interpretation
process. Through photo-interpretation, in fact, it has been possible to identify
several archaeological sites, networks of tracks, fields and settlements, all over
Europe, and of many historical periods, from the Neolithic onwards. An
example of the use of this technique is the identification and reconstruction of
an Iron Age fort at Bloodgate Hill, South Creake in north Norfolk (figure 3.5)24.
[Figure 3.5. bloodgate_ok.tif]
In photo-interpretation, most of the traces originally derived from excavated
features, such as boundaries, pits or post holes, can be seen as soil marks
(variations of colour and texture of the soil) or crop marks (differential crop
growth: positive marks over a ditch or negative ones over wall or stone
foundations). More consistent remains on the ground can be recognized as
earthworks. From the air, these features appear as dots or lines or areas of
differently coloured crops and soil. Since these traces cannot be easily dated, it
is necessary to verify them on the ground and integrate as much information as
possible: some crop marks, in fact, do not belong to archaeological features, but
are formed, for example, by geological features or modern agricultural practices
(WILSON 2000:163-209). Another problem is correcting those images to draw
the feature in the GIS, as vector layer. Although there are several softwares that
can perform such transformations, two specific programs have been developed:
Aerial and AirPhoto, included in the Bonn Archaeological Software Package
(BASP)25.
In many countries there are accessible aerial archives, such as the atlas of
oblique photos of the University of Siena in Italy26; the Austrian archive27;
several archives in the UK, Scotland and Wales28, etc.
24 Wade-Martins, P. (ed.), 1997. Norfolk from the Air, plates 12-15. And:
http://www.norfarchtrust.org.uk/bloodgate/index.htm
25 To download the software: http://www.uni-koeln.de/~al001/airdown.html
26 http://shaq.archeo.unisi.it/oblique/
27 http://www.univie.ac.at/Luftbildarchiv/archiv/aa_ent.htm
28 Collection at Cambridge University (http://www.uflm.cam.ac.uk), the National Library of
Aerial Photographs (http://www.english-heritage.org.uk), the RCAHMS in Scotland (http://
www.rcahms.gov.uk), the RCAHMW in Wales (http://www.rcahmw.org.uk), the Aerial
Reconnaissance Archives of the University of Keele with several RAF World War 2
Reconstructing Ancient Landscape S. Pescarin

Active sensors can be mounted on aeroplanes or satellites. An example of these

sensors are those used for airborne laser scanners. Lidar (Light Detection and
Ranging or Laser Imaging Detection and Ranging) is a well-known technology,
and although not so widely used, it has gained general acceptance as an
accurate and rapid method for three-dimensional surveying of the Earth’s
surface29. The Lidar system has several advantages, such as the ability to view
subtle landscape changes and high resolution feature detection (resolution of 3
cm). It can be combined with vertical survey cameras or airborne digital
sensors, to record the entire landscape. Its main disadvantage, on the other
hand, is its high cost, together with the fact that it is quite impossible to fly in
every place or country (PARCAK 2009: 76-77). Another problem of aerial laser
scanning, as mentioned for terrestrial acquisitions, is the way data are acquired,
in a “monolithic” way. The entire visible information is captured and stored
without distinguishing vegetation, buildings, etc. This often means a long post-
processing work to clean up, divide up and sometimes re-model part of the final
3d mesh. New algorithms and approaches to this problem are currently debated
in specialized conferences.
There are other airborne sensors used in archaeology, such as the RadarSat,
AirSar (Sythetic Aperture Radar) or thermal radiometry. It can acquire at a
resolution of up to 3 meters and identify, no matter the weather conditions or
darkness, pathways or sites even under dense vegetation. Unfortunately even
this system is quite expensive30. Hyperspectral acquisitions are quite useful in
various landscape applications, for example in the definition of natural features,
such as vegetation (the classification and mapping of species). They are used, in
combination with fieldwork and ground analysis, for the identification of
culture characteristics (density, humidity, etc.), the analysis of relations
between vegetation and geomorphology, and the monitoring of parks. In
landscape archaeology they have been used to identify palaeo-channels or
structures hidden under the ground or even to characterize superficial materials.

.B.2 Invisible landscapes

Excavations aren't the only way to define what has been preserved under the
ground. There are several techniques that can help us to analyse ancient natural
or anthropic remains. Geophysics offers several methods to obtain information
from underground (CAMPANA PIRO 2008).
Geoarchaeology can also be successfully employed to identify the geological,
lithological, and geomorphological characteristics of a territory and to
photographs (http://www.evidenceincamera.co.uk).
29 A tutorial can be found at: http://home.iitk.ac.in/~blohani/LiDAR_Tutorial/
Airborne_AltimetricLidar_Tutorial.htm
30 http://www.space.gc.ca/asc/eng/satellites/radarsat1
Reconstructing Ancient Landscape S. Pescarin

understand how or when its shape (geomorphology) has changed through time
(CREMASCHI BARONI 2000, GOLDBERG MACPHAIL 2006, PANIZZA 2005).
Another field that has been integrating increasingly into landscape
reconstruction is Paleobotany. Vegetation biologists can analyse, within
specific required conditions (such as the presence of water, or sediments), the
remains of seeds or pollen and thus identify plant species, crop cultivation or
deforestation, in a certain area and over a certain period of time (table 3.5).

Type of remain Type of sediment Obtainable knowledge

Soil All Type and environmental characteristic of the deposit
Pollen Buried surfaces, deposits with Vegetation, land use
traces of water
Phytolith All Vegetation, land use
Bacillariophyta Submersed deposits Level of water salinity and pollution
Vegetable not Dry and drenched Vegetation, diet, natural material used in constructions
burnt remain
Vegetable All Vegetation, diet, natural material used in constructions
burnt remain
Wood Dry and drenched Dendrochronology, climate, materials and building techniques

Table 3.5 - (from: Renfrew Bahn, 1991: 223)

The first invisible landscape is the one that cannot be seen. Fortunately in recent
decades detectors have been developed to see for us, where we ourselves can't
see. This is the subject of the Remote Sensing field31. These detectors can be
mounted on aeroplanes, satellites or other devices, but also used for terrain
applications. The basic principle is well known: the earth reflects the spectrum
of sunlight, providing us with a lot of information about the composition of this
surface, and revealing also traces of past human activities. The surface is made
up, in fact, of elements which emit heat at various levels, visible to sensors such
as multi-spectral scanners. Differences in the soil or in vegetation can be
detected by variation in temperature. The analysis of infra-red radiation at a
variety of wavelengths, for instance, can help in identifying pathways in the
landscape, irrigation ditches, buried stone walls (stone can absorb more heat
with respect to its surroundings). Moreover, analysis through radar sensors
allows us to study the ground even if it is covered by clouds or canopies or

31 Wiseman, J. El-Baz F., Remote Sensing in Archaeology, Springer, 2007; Williamson, A.

Nickens, P.R., Science and technology in historic preservation, Ed. Springer, 2000; Miller
W.F., Sever T.L. & Lee D. (1991). Applications of ecological concepts and remote sensing
technologies in archaeological site reconnaissance. In Behrens C. & Sever T. (Eds)
Applications of space-age technology in archaeological site reconnaissance 121-136. A
good tutorial can be found in http://www.cnr.berkeley.edu/~gong/textbook/
Reconstructing Ancient Landscape S. Pescarin

hidden by darkness. Several commercial softwares have developed features to

manage these analyses, identifying distinctive "signatures" characterizing
specific features32.
The basic work one can do on satellite images is band combination (table 3.2
and 3.3). Each image, in fact, is characterized by several bands, each one
representing a range of the electromagnetic spectrum (Aster: 15, Landsat
ETM+: 8; Spot: 4, etc.). We should in fact consider various parameters when
we plan to buy geoimagery, such as: spatial resolution, spectral bands, program
history, image surface area, price and licensing. If we want to obtain, from a
Landsat image, a true-colour or visible image, we should for instance combine
bands 3, 2 and 1, assigning respectively a Red, Green and Blue value. The
combinations of these bands enables us to distinguish vegetation, water or
geological features better. These distinctions can also give interesting inputs to
identifying archaeological features, often related to changes or discontinuities.
Some useful combinations (Landsat imagery) in archaeology are:
II bands 3, 2, 1 (RGB): visible image, useful to identify details;
III bands 4, 3, 2 (RGB): false-colour image (pixel values not representing
the true colour photo) where vegetation is enhanced, appearing in red;
IV bands 5, 4, 3 (MIR, NIR, Red) to highlight differences between areas
with or without vegetation.
V bands 4, 5, 6 (RGB): useful to identify hills, barrows etc.;
VI bands 4, 7, 5 have been used to identify, for instance, rammed earth.
VII IHS Fuse of Bands 5,4,3: used to incorporate Band 8 of a Landsat TM
image (15 m Panchromatic) to obtain a 'false' colour image of a 15 m
resolution.
Other remote sensing analyses useful for archaeological landscape
reconstruction are (PARCAK 2009: 91-99):
1. Normalized Difference Vegetation Index (NDVI): used to measure, in
different images for different times of the year, the health of vegetation
possibly affected by buried archaeological remains, or the outline of a
settlement, or a ditch. This has been used, for instance in the Jure Vetere
(fig. 1.1).
2. Classifications: to identify land cover or other kinds of clusters. It can
use “Supervised” (human-assisted) and “Unsupervised” techniques. The
former requires the control and identification on the ground of “training
regions”, that are then reported in digital classification in order to define
different cells better. “Unsupervised” classification, on the other hand,
creates clusters of multispectral images, each one representing
similarities in the spectral space (FORTE, 2003: 85).
3. Thresholding image segmentation: classification in which it is possible
32 ErMapper, GRASS, etc. (Appendix B)
Reconstructing Ancient Landscape S. Pescarin

to specify parts of image values that remain visible.

4. Principal Component Analysis (PCA): helps in clarifying classifications
and band combinations. It has been used in archaeology for better
feature recognition.
5. Land Use Land Cover Changes (LULC): land use variation analysis as
observed in different time periods.
6. DTM modelling: obtained from SRTM data, stereo-pair SPOT, ASTER
or stereo CORONA images.
7. Hyperspectral analysis: useful in case the spectral signature of a
specific material (such as stone) is known. Hyperspectral satellites
datasets are needed, such as OMIS (Optical Monitor Imaging
Spectrometer) sensors.
8. Radar analysis: SAR or LIDAR techniques.
In the Aksum project, directed by the Oriental University of Naples and Boston
University, with the cooperation of CNR ITABC, remote sensing analysis was
employed to identify archaeological features (FORTE 2003: 81-93, BARD ET ALII
2007). This project aimed at the reconstruction of the archaeological landscape
of Aksum (Ethiopia), the capital of one of the most powerful Red Sea kingdoms
during the 1st millennium AD, and at its visualisation in a Desktop Virtual
Reality (DVR) application. The aim was achieved thanks to an integration of
fieldwork, ethno-anthropological investigations, paleo-environmental analysis,
aerial photo interpretation and remote sensing. Thanks to the multispectral
classifications of a Spot XS image, various thematic maps (soil use, vegetation,
geomorphology, and archaeology) were obtained, and an interpretative map
was developed. This was used in the reconstruction of the archaeological
landscape which was used to build a VR environment (Figure 3.6).

[Figure 3.6 – Aksum_ok.tif]

.B.3 Post-processing
At this point, acquired spatial data needs to be further processed (table 1.2). The
goal in fact is to manage and visualize them, developing a model of the
archaeological landscape, made up of its anthropic and natural features, but also
of the interpretative analysis carried out. This model will also be the reference
point for any interpretation layer emerging out of the post-processing work. At
the end of the process, in fact, on top of the reconstructed archaeological
landscape, a “mapscape” will be created. It will represent all the interpreted
features, identified after remote sensing analysis or aerial photo interpretation,
and will be the basic source of potential ancient landscape reconstruction.
Since all datasets should be considered in their spatial dimension, every
Reconstructing Ancient Landscape S. Pescarin

reconstruction process starts by building a GIS project, which will be the

reference point for all activities, continuously updated.
Geographical Information Systems (GIS) are tools, or rather sets of instruments,
initially developed by and for the army, for spatial dataset management. They
are now considered essential also in archaeology, since they are used for many
activities: archiving, analysing, querying, or just as reference and middleware.
They have been described as “ the most powerful technological tool to be
applied to archaeology, since the invention of Radiocarbon dating” (MARIS AND
TE BOEKHORST 1996, cited in CONOLLY LAKE 2006: 5). GIS describes space in
terms of location and attribute. Each element inside a GIS needs two
descriptors, which indicate what it represents (attribute) and where it is located
(location), as referred to a space, either absolute or relative (CONOLLY LAKE
2006: 3-4). There is an extensive bibliography specifically regarding the use of
GIS in archaeology (WHEATLEY GILLINGS 2002, LOCK 2003, LOCK STANCIC 1995).
GIS is also basic during the reconstruction of archaeological and past
landscapes. It can manage to overlay various pieces of information and, most
importantly, it enables to compare data and carry out calculations on different
parameters (grid calculations, vector analysis, etc.). In Appendix B there are
specific GIS analyses, described in detail. The entire reconstructive process is
based on GIS analysis, as described in Chapter 4.
WebGIS are commonly considered as Internet “extensions” of traditional GIS,
although they often do not have the same analytic potential, but are used as
digital on line archives dedicated to geo-spatial data, such as archaeological
maps, digital images, grids and vector data. New projects are under way whose
goal is precisely to extend GIS potentiality, mainly boosting editing and
analysis capabilities. The MapServer Open Source platform, for instance, has
several projects of tools specifically thought out for these purposes33. A project
such as Fasti on Line is a webGIS, built on MapServer technology. It derives
from Fasti Archaeologici (1946-1987) edited by AIAC. It allows registered
archaeologists autonomously to add references to, and the location of,
archaeological excavations in Italy, Albany, Bulgaria, Macedonia, Malta,
Morocco, Romania and Serbia34. Other Open Source projects, whose evolution
could offer interesting applications to landscape reconstruction, are: Ka-Map,
an application Server Open Source35; Embrio, a web interface which combines
map interactive visualisation, and PyWPS, enabling GRASS and QGis
integration and real time analysis, such as Viewshed, Slope, and Cost
Distance36; OpenLayers, a javascript tool which enables the use of dynamic

33 http:// www.mapserver.org
34 http:// www.fastionline.org
35 http:// ka-map.maptools.org
36 http:// pywps.ominiverdi.org/demo/embrio/ka-map/htdocs/index_wps_qgis.html
Reconstructing Ancient Landscape S. Pescarin

maps and layered dataset, located in distributed servers and loaded “under”
personal geographical data37. This last example, as in the case of GoogleMaps,
opens up several possibilities in the field of landscape reconstruction, since it
allows the sharing of various data, without actually “moving” them, but
maintaining them in original servers.
GIS or webGIS are used to store a variety of spatial information during its
process (raster, DTM, vector), to perform analysis, and finally to export spatial
dataset to be used in the 3d modelling process. As mentioned, a basic spatial
dataset is made up of at least one geo-referenced raster image and one digital
elevation model of the terrain.
Generally, architecture or object modelling is carried on separately from terrain
modelling. This is mainly due to the differences in the required computing
work, and of elementary ontologies.
We have already described the way a monument, a site or archaeological
remains can be modelled: mainly by measurement techniques, as in the case of
a 3d scanner (Range Based Modelling) or photogrammetry (Image Based
Modelling), and by non-measurement techniques, as in the case of computer
graphics. Which technique is worth adopting is a question of the aim of the
project, the required post-processing time and level of expertise and, last but not
least, the available budget. No matter which technology is used, the crucial
issue is if and how to integrate the models into the landscape.
As mentioned in chapter 2, the best place to test hypotheses and theory is a 3d
virtual environment, made up of virtual ecosystems. Here the third dimension
gets back to interpretation, and inter-relations and dynamics can be seen. A VR
environment can get across communication in a more direct and simple way,
turning a lesson into an experience, a study into a hyper-research. We should
therefore handle not only GIS datasets in landscape reconstruction, but also
integrate 3d models.
The most problematic (and still open) question is how to integrate those models
in the landscape. A first decision to be taken concerns the use of the
reconstructed environment as a whole: for real time purposes, or not? Real time
applications, in fact, require great care in model optimization, polygon
lowering, levels of detail, and paging generation. Various optimization
techniques are continuously under development. Computer graphic rendering
and animation, on the other hand, can handle much more complicated and
realistic scenarios, using (relatively) less effort in optimization. Another
question concerns the development of on line or off line applications. Even in
this case, bandwidth requirements for on line projects play a central role in the
definition of the dimension of objects. In any case, when we decide to work
with 3d models, the entire team should take responsibility for their geographical
37 http://openlayers.org
Reconstructing Ancient Landscape S. Pescarin

spatial dimension and location. This is usually done inside the GIS, mainly in
two ways: creating a vector layer representing their planimetry (polygons) or
their centre (0,0 axis coordinate). In this way it is easy to locate the models on
the ground automatically. With large models, some software may provide
algorithms to level and smooth the terrain, in order to avoid a “flying” effect, an
operation that can be also done in the GIS, although it requires a modification
of the original DTM. A better solution is normally to avoid saving and using
big models, preferring their subdivision into smaller pieces. With non real-time
applications concerning smaller areas, modellers usually prefer to work in
computer-graphic software38 with a single file, integrating models directly in
the terrain, to provide better control in rendering the effects of the entire scene.
Terrain modelling, on the other hand, requires specific software to be
performed, as well as tools that can handle its complexity, maintaining its
spatial characteristics at the same time, and potentially enabling a final
interactive exploration. It usually follows three steps: 3d terrain generation
(from geoimage and dem), culture generation (both natural and anthropic
characteristics) and 3d models integration. The first programs were created for
the army to train soldiers to fly and hit a target (flight simulators), or drive a
tank without losing direction. Reality-based reconstructions were essential, such
as the geo-referencing of the digital scenario. Software for developing these
kinds of applications were quite expensive at the beginning39. A complete
review of available programs is published every year by the U.S. Army
Topographic Engineering Center (Survey of Terrain Visualisation Software)
and made available on line40. The goal of such a survey is clearly identified in
the premise: “to provide the U.S. Army and DoD with more effective methods
of merging, visualizing and analyzing battlefield terrain and environmental
information.”
In the usual procedure41 (figure 3.7), acquired data are processed in a GIS
software to obtain a geoimage, or mosaic of images, a DTM (or more than one),
and vector datasets representing landscape characteristics (rivers, lakes, roads,
sites, plans , vegetation areas or plants, monument locations, etc.). At this point
a Terrain Generator is required, a software that can import spatial dataset and
produce a fully 3d terrain. Terrain complexity (both in terms of geometry and
texture) is treated with the following techniques: Different Levels of Detail
(LOD), paging and tiling. The final result, the Terrain Database (TD), has a
hierarchical structure, with a “master file” that usually opens all the tiles-files.
Paging capability, which means that it is possible to tile together any number of

38 3d Studio Max, Maya, Blender, Cinema 4D, etc. (Appendix B).

39 Principal software have been Multigen Creator Terrain Studio (CTS) and Terrex Terravista.
40 http://www.tec.army.mil/research/software/TD/tvd/survey/index.html
41 In Chapter 7 is reported an example of this procedure.
Reconstructing Ancient Landscape S. Pescarin

terrains to make extremely large worlds, is important especially in the case of

on line publications (LUEBKE, 2003: 187). A tile is a rectangular portion of the
grid used to organise the TD. Each tile has several LODs, which represent
polygon density: less detail for distant visualisation and more detail for near and
steep visualisation.
Known terrain generators, which automatically produce TD with these
characteristics, are: Terravista, Creator Terrain Studio and, on the Open Source
side, OpenScengraph OSGdem42. Open Scene Graph is a graphic toolkit to
develop graphic applications such as a flight simulator, scientific visualisation,
or augmented reality environments. It is based on OpenGL. It has several
features such as: crossplatform, Real Time optimisation, a wide range of input
format support (flt, 3ds, obj, osg), extensibility through-plug-in architecture,
built-in support of paged LOD terrain generation and navigation. OSG already
provides both a tool to generate hierarchical paged terrains (OSGdem) and a
network loader capable of providing the browsing of such hierarchies with
reasonable bandwidth requirements (.NET).
While OSGdem is very essential, most advanced products can obtain direct
DTM modifications in accordance with vector layers (river beds, road
pavements, automatic bridge generation over rivers, etc.).
Other products, such as Visual Nature Studio, allow to process spatial datasets:
adding effects and modifiers, managing complex ecosystem generation, adding
3d models; to create renderings and animation; to export the final landscape in
formats (table 3.6) useful for: direct real-time visualisation (such as
OpenFlight), computer-graphics manipulation (i.e. 3D Studio) or processed GIS
datasets, ready to be further implemented with a Terrain Generator.
[figure 3.7 terrain_generation.tif]

Elevation data DTED (levels 0,1,2); USGS 1:250k; Arc Info BIL; ESRI Grid Ascii; ESRI Grid Float; GeoTIFF
(as 16 bit elevation data); BitMap (as 8 bit elevation data); etc.
Imagery GeoTiff; TIFF + Tfw; GIF; BMP; JPEG; etc.
Vector data DFAD (all levels); Shape (import & export); DXF; etc.
Textures SGI .rgb .rgba; TIFF; GIF; BMP; PNG; etc. (OpenGL usually requires power of 2 dimension:
256, 512, 1024, 2048 pixel)
3D models (OpenFlight) FLT, 3DS, DXF, OBJ, WRL, Collada, X3D, OSG/IVE; etc.
Tiled Terrain DB OpenFlight; Terrapage; OSG/IVE; Quantum3D VT; ViewTec Web Streaming TVW; etc.

Table 3.6: Landscape generation process file formats

The final terrain can be now visualised with a 3d viewer. Most of the terrain
generators offer their viewer, which can normally open proprietary files but
sometimes also other formats. OpenSceneGraph, for example, has the
42 Other tools which support terrain paging are: Demeter
Reconstructing Ancient Landscape S. Pescarin

OSGviewer tool that enables you to explore OSG/IVE files in real-time.

Another interesting open project is Virtual Terrain, that developed several tools
for terrain handling, such as Enviro, for spatial datasets interactive exploration,
with some interesting editing functionalities43.
In some cases the work isn't finished, but requires a further integration of 3d
models, vegetation characteristics, environmental maps, behaviour and
interactivity such as: link to external databases and other media, avatars,
environmental effects, reference maps in overlay, etc.
In the Delta Po River project, Terrex Terravista was used to generate the large
TD representing the archaeological landscape. The 3d reconstruction was also
used as a sort of 3d interface, where further GIS datasets were overlaid,
becoming a simulation space, to study the evolution of the coastline during four
different periods: the Roman age, and the 14th, 17th and 21st Centuries (fig. 3.8).
Two archaeological sites, with excavated Roman villas, were then modelled in
detail and added to the landscape, developing a second level of walk-through
exploration behaviour (GUIDAZZOLI ET ALII 2004).
[figure 3.8 delta.tif]

.B.4 Data integration: problems and solution

As we have seen, we are moving towards a massive integration of many
different ontologies. Let's start to consider some possible solutions to the
commonest problems. Integration can be considered mainly from two
perspectives: the so-called Layer Fragmentation (LF), and Zonal Fragmentation
(ZF). LF concerns elements belonging to various categories (i.e. a DTM and a
3d model), while ZF refers to the same feature class but with different
characteristics (two DTMs not perfectly overlapping or at different resolutions,
but also parts of a monument) (LAURINI 1998, AGUGIARO KOLBE 2009).
How can we embed high resolution objects, such as monuments or
archaeological remains, in a generally low resolution Digital Terrain Model?
How can we merge together DTMs of different resolution? More in general,
how can we integrate acquisitions made at different resolution in a single
visualisation and management platform? Integration can in fact generate errors
or misunderstanding.
LF is still an open field. In real time applications a variety of solutions have
usually been developed to integrate 3d models in a 3d scene. The more complex
the scene the more care should be taken with respect to polygon numbers.
Although model simplification seems to be a good solution, if exaggerated it
can lead to the elimination of important detail. Usually other systems are used
to avoid this problem, such as multiresolution techniques. A final solution is
43 http://www.vterrain.org
Reconstructing Ancient Landscape S. Pescarin

always a balance between target user/hardware/purpose, in order to obtain

scenes as realistic as possible (complex, highly detailed), but also through rapid
(sometimes interactive) rendering. The goal of multiresolution is to extract the
necessary details from complex models and get rid of unnecessary ones44.
Recently CINECA Visit Lab has developed an integrated solution to enable the
modeller to integrate high resolution 3d models into the terrain database. It is
composed of a plug-in for 3D Studio Max (OSGexp) and a PostProcessing tool.
The basic concept is that every modeller who works for realtime applications,
can be limited in terms of file-dimension (geometry and texture). He can still
obtain a complex and high resolution 3d model by subdividing the object into
several separate files, each one with different LODs. In this way, the real time
exploration can fit the actual perceptive capability of the users and also their
interest in a specific area, whose details can be downloaded sequentially
(PESCARIN ET ALII 2008). It might also be possible to connect terrain LODs with
3d models LODs, so as to give a better feeling of continuity during
visualisation, and to avoid dramatic resolution differences (fig. 3.9).
Figure 3.9 multires_model.tif

Various solutions have been proposed concerning ZF, such as the enhancement
of a part of a DTM (KAREL KRAUS 2006), or of the entire DTM, taking
information from a 2D vector dataset of known features and parameters (roads,
rivers), thus increasing the number of known elevation data; in the same way,
others have suggested increasing elevation information by capturing it from
building planimetries. Multiple DTMs are currently supported in many
landscape applications, such as CityGML (GROGER KOLBE ET ALII 2008) and
OSGdem, and successfully applied in projects such as Virtual Rome (CALORI ET
ALII 2009) (fig. 3.10).
Recently Agugiaro and Kolbe have attempted to solve the problem of
integrating laser scanner high resolution and low resolution DTMs in more
detail; the solution presented is addressed to the creation of a correct and clean
“transition surface”, influenced by local parameters of both terrain models.
This surface is generated in order to create a smooth transition in terms of
geometry, precision and density (AGUGIARO KOLBE 2009).

figure 3.10 – multi_dtm.tif

44 The commonest multires techniques are: image pyramids, volumetric methods, vertex
decimation, vertex clustering, edge contraction (mesh optimisation, progressive meshes),
simplification envelopes, wavelet methods, etc. Although not updated, a good reference
point is: http://www.cs.cmu.edu/afs/cs/user/garland/www/multires/bib.html
Reconstructing Ancient Landscape S. Pescarin

Chapter 5
Communicating the landscape

This chapter presents specific techniques used to visualize landscape

reconstructions, from a twofold perspective: research and dissemination.
Although we often consider “communication” as a final, non-scientific and not
even essential step, we need to reconsider this assumption. Most of the time the
word “communication” is used instead of “dissemination”. There can be no real
distinction, in fact, between research and communication, since we need to
communicate while we develop our research and while we disseminate its
results.
Dissemination regards the diffusion of scientific results to both a restricted and
broad-based community, the circulation of acquired knowledge. It uses various
communicative languages and media, such as publications, drawings, movies,
interactive applications, etc. It can be directed toward a heterogeneous
audience, with the goal of making the information on our past landscape
accessible. In order to be as effective as possible, it requires specific
technologies and techniques, adequate to the kind of communicable
information and to public access and understanding.
Technologies for research regard a more limited group of people, the scientific
community, whose interests are closely connected to the research subject and its
aims, within a multidisciplinary framework. Communication, as previously
mentioned, is part of the research process. It makes use mainly of visualisation
and natural languages - often in combination - during data acquisition,
interpretation, analysis, model development, simulation and cross-networking.

.B.1 Dissemination

Drawings
Images are commonly used in science to present data. There is still some debate
about what image type should be used, whether 2d-digital or 3d-digital or hand-
drawing. Archaeological drawings of human artefacts, such as stone tools, are
sometimes still preferred to photographs. The human drawing system, in fact, is
an observational tool. Drawings are synthetic representations of selected
characteristics. When we consider the innumerable scientific images we can see
a systematic connection between (1) image types, (2) purposes, and (3) contexts
of use (ADKINS 1989, PIGGOTT STUART 1978).
Professional drawers have produced wonderful artistic reconstructions of past
landscapes. The best, along with their artistic skills, often work together with
Reconstructing Ancient Landscape S. Pescarin

archaeologists and palaeo-environmental experts. These experts in fact provide

exact references and information which enable correct representation, which is,
nonetheless, a personal vision developed by combining the perspectives of the
archaeologist, the artist and the naturalist.
If we follow the career of these professionals we find very impressive
reconstructions of ancient landscapes. A well known historical reconstruction
artist is Alan Sorrell (1904-1974), who travelled England and Wales, bringing
vividly to life ancient monuments from the prehistoric up to those of relatively
recent times. The Imperial Rome drawn by Gilles Chaillet or the Egyptian
landscapes by Jean-Claude Golvin are other examples of amazing visualisations
(HARRIS, RUGGLER 2007; GOLVIN 1999; GOLVIN 2007).
These drawings, thanks to their pictorial technique, help create emotional
reconstructions. Their main limitations lie in the bi-dimensionality of the media
and, most of all, in the static nature of the representations. The effort put into
redrawing, whenever new data or interpretations appear, create a serious
problem of updatability in some cases.
Today, artists and designers use a mixed technique. The drawing is first done
with a traditional technique, on paper, and afterwards refined in various phases.
3D simple digital reconstructions of principle monuments are sometimes
developed. They are used to gain different perspectives and therefore to choose
striking viewpoints for the final version. This system permits a sort of
“updatable” version of the drawing, whichcan be more easily re-drawn or
repainted from the same viewpoint, changing just a few details. An example of
this technique is used by Ink Link Studio45.

Not interactive media

The film industry still represents the most mature media for cultural heritage
dissemination, with its schools, experience, comparisons, feedbacks and
success. And also a colossal market. Movies can create involvement, thanks to
skilful shots, effects, sounds, plot and with a shrewd direction, creating in the
spectator a sort of personal experience. With a movie, you can get the public to
understand many concepts in a short time and with immediacy. Moreover,
story-telling techniques, as used in film production, are richly communicative,
obtaining good results in terms of public satisfaction and knowledge transfer.
This technique employs a mix of video sequences, images of real places,
ancient maps, drawings, computer animations, characters and digital effects.
Nevertheless, on the other hand films are linear media. This characteristic
makes them immediate, but it is also their limitation, from a certain perspective.

45 http://www.inklink.it
Reconstructing Ancient Landscape S. Pescarin

When we watch a film, i.e. a computer animation of a reconstructed territory,

the realism of the reconstruction is taken as reliability. We trust in the accuracy
of the director or the computer scientists. We cannot see what there is in the
backstage, how the reconstruction is done, which sources are used, which detail
and accuracy level is followed. In this sense, we end up in complete passivity in
front of the screen (TURKLE 1997). This doesn't mean we shouldn't make use of
films, obviously. Videos are created, in fact, in almost every virtual
archaeology project. Moreover virtually all TV programs on archaeology or
history today use computer graphic animations and renderings. We just need to
be aware of the risks as well as the benefits when we plan a communication
project dedicated to a public.
In the Imperial Forum Museum, in Rome, the reconstruction of the territory is
proposed in panels and in a video, composed using computer graphic
renderings, traditional watercolour drawings and real photos (fig. 5.1). Other
examples are the videos created for the Seabed Project, commissioned by
Wessex Archaeology, or the Stonehenge Landscape reconstructed from Lidar
data46. Also in the Ca' Tron project, the entire work of reconstruction of 20,000
years of history was summarized in a short video (Appendix A). Photo-realistic
reconstructions and videos have been developed by CNR IBAM within the Jure
Vetere project, where spatial analysis was used to reconstruct the Medieval
potential rural landscape (fig.1.1, FONSECA ROUBIS SOGLIANI 2007, ROUBIS ET ALII
2008)), or in the reconstruction of the territory of Metaponto, in Italy (fig.5.2).
In the Virtual Museum of Ancient Via Flaminia, accessible in Rome in the
Roman National Museum at the Diocletian Thermal Baths, a mixed technique
has been developed. Real time VR interaction, in fact, is mixed with videos on
specific aspects of ancient landscape, characterized by a “scientific-descriptive”
language and with other videos created with a different and more personal
“story telling” style (Appendix A) (FORTE ET ALII 2008, FORTE ET ALII 2007).
Story-telling, although a well tried research field, has recently been taken into
account in a new interdisciplinary perspective: virtual story telling, the use of
this collection of techniques in Virtual Reality applications47.
Figure 5.1 imperial_forums.tif

Figure 5.2 - metaponto2.jpg

46 http://video.google.com/videoplay?docid=1246762325049132604;
http://www.youtube.com/watch?v=pzYUx4l80m8
47 There is an annual international conference dedicated to interactive story telling
(International Conference on Interactive Digital Storytelling): ICIDS 2008 Erfurt, Germany,
Nov.ember 26-29, 2008, Proceedings, Vol. 5334, Spierling; Szilas, Nicolas (Eds.), 2008
Reconstructing Ancient Landscape S. Pescarin

Landscape Virtual Museums

How can we overcome passivity and opacity? How can we access “the dark
side” of reconstructions? Through interactivity, transparency and openness.
Examples of a possible use of all these characteristics are Landscape Virtual
Museums. They are Virtual Reality (interactive, real time) applications
specifically for landscape exploration and public access. In the article written in
2006 with Maurizio Forte on “Virtual Museum of Landscape” we defined it in
this way: “A virtual museum of landscape regards, first of all, the process of
virtualization of dynamic relations concerning the ecosystem, humans, animals,
plants, soils, earth, water, etc. It is an artificial ecosystem, map and alphabet of
the landscape itself”. Such a museum is “focused on generating a holistic view
of the environment, because without environment we cannot describe a
landscape, and an ecological model” (FORTE PESCARIN 2006: 87).
If we want to communicate what a territory was like in the past, who used to
live in it, or which archaeological monuments could be seen, we can visit the
place (in this case the landscape will be the museum of itself), read a book or
visit a “non-place”, such as a virtual museum. Thanks to specific applications,
the user can interactively interrogate elements, models, materials, becoming
familiar with dynamic and static aspects. Through the creation of maps, it is
possible to provide the user with interpretation instruments, which can help to
decode something that no longer exists and is, therefore, very alien to our
contemporary awareness. Without a code, in fact, we cannot achieve an
interpretation. In a VR museum, the landscape is approached synchronically
and diachronically. It becomes “an open and evolving model, a scenario of
simulation of artificial life, integrating different information ontologies” (FORTE
PESCARIN 2007: 87).
Landscape Virtual Museums consist of a variety of components:
I Maps (ancient maps, cybermaps, etc.);
II GIS (vector thematic layers);
III Geoimagery;
IV Environmental information (soil maps, land use maps, geological maps,
potential ecosystems etc.);
V Anthropic information (archaeological map, cadastre, etc.);
VI Digital archives (databases, repositories, digital libraries),
VII Interactive behaviour (querying, walking, flying, picking, slicing,
editing, etc.);
VIII Viewpoints (the landscape seen from specific perspectives);
IX Places;
X Relations;
XI Dynamic simulations;
Reconstructing Ancient Landscape S. Pescarin

XII Story telling (video, past literature or sources, contemporary stories and
memories);
XIII Artificial life.
How can we build a Landscape Virtual Museum? Unfortunately there are not
many definitive answers or well-defined guidelines, also because available
examples are in most cases the prerogative of restricted groups or scientific
communities.
Nevertheless, many projects of the VHLab team at the National Research
Council (CNR) or of other institutions and companies such as Visual
Dimension, University of Foggia, University of Virginia, CINECA, etc. have
demonstrated the potentiality of these VR museums: Aksum project (FORTE
2003), Certosa Virtual Museum (GUIDAZZOLI ET ALII 2004), Narrative Museum of
Archaeological Appia Park (FORTE PESCARIN PIETRONI 2005), Delta Po River VR
museum (GUIDAZZOLI ET ALII 2004), Virtual Bononia (PESCARIN ET ALII 2007),
Virtual Rome (PESCARIN ET ALII 2008), Flaminia Virtual Museum (FORTE ET ALII
2007), Ename TimeLine exhibition, Itinera Time Machine, Rome Reborn,
In the case of Aksum, the archaeological landscape was reconstructed by
remote sensing analysis and ethnographic studies, collecting the results of ten
years of excavations in Ethiopia carried out by Boston University, Naples
Oriental University and CNR. The entire territory was generated in 3D, using
Terravista simulation software48, while models of the main archaeological
structures were placed in relation to roads, rivers and geomorphology. A
Desktop VR (DVR) application was developed using the VTree (CG2) OpenGL
library and a Vision Station was used to visualise it, simulating a 360° real time
immersive exploration of the territory (figure 3.6: bottom left). A “sound GIS”
was also created in order to give a deeper perception of being immersed in the
landscape: the sound of water near rivers, of wind on leaves and birds close to
woods, the voices of rituals approaching the villages. The DVR application was
shown during an exhibit on the history of cartography in 2002 in Rome (BARD
ET ALII 2007, FORTE 2003). On the same dataset, another VR application was
developed for a Virtual Theatre, an immersive space characterized by a big
semicircular screen and stereo projection, specifically for researchers’
interaction. This application, in fact, developed with Multigen Vega49, was
strongly oriented to database search and real time visualisation of DB results.
For instance, it was possible to identify in the database all archaeological
structures that shared similar characteristics or belonged to the same
chronological period, and visualize them in 3D in the archaeological landscape,

48 http://www.terrex.com
49 Vega Prime by Multigen Paradigm is a tool for the creation and deployment of visual
simulation, urban simulation, and general visualization applications
(www.multigen.com/products/runtime/vega_prime).
Reconstructing Ancient Landscape S. Pescarin

together with multimedia meta-information (figure 3.6: right)50.

Another VR landscape museum was created more recently for the Delta Po
river project. In this case the final application was used to create two interactive
installations (kiosks) in small visitor centres of Rovigo province, in Italy. The
landscape itself became the interface to access information on the archaeology
of two Roman sites, and also the history of the evolution of the coastline over
the centuries (GUIDAZZOLI ET ALII 2004, PESCARIN RUZZA 2004).
In Ename, Belgium, a long lasting project has reconstructed the history of the
village from the 11th century. The entire work has been exhibited in the local
museum thanks to the TimeLine application (fig. 5.3)
Another virtual environment dedicated to a scientific exploration of a territory
is the Itinera Time Machine project. The system, developed in the Digital
Archaeology Lab of the University of Foggia, in Italy, is based on the scientific
and archaeological documentation acquired in the field from the site of
Faragola (Ascoli Satriano, FG). The Time Machine brings the visitor and his
avatar back in time to explore different hypothetical reconstructions of the site
and of the stratigraphic process (figure 5.4). He can also watch the threshold
between archaeological reality and reconstructive imagination, through the
visualisation of the inter-connections of stratigraphy and interpretative traces
(DE FELICE ET ALII 2007).
An interactive space was recently opened at the Roman National Museum of
the Diocletian Thermal Baths in Rome, dedicated to the virtual multi-user
exploration of the ancient Via Flaminia (figure 5.5). In the Virtual Museum,
located inside a real museum, visitors can choose one of four avatars and
explore together the archaeological landscape, discovering its aspect during late
Republican and early Imperial times, and also meeting historical figures, such
as the emperor Augustus himself or his wife, Livia (FORTE ET ALII 2006, FORTE ET
51
ALII 2007) .
An example of a Virtual Museum dedicated to an immersive exploration of the
archaeological site of Sacred Angkor, in Cambodia, or dedicated to Place-
Hampi in India, are those developed by Sarah Kenderdine and Jeffrey Shaw,
with the Museum Victoria, Swinburne University, Royal Melbourne Institute of
Technology (RMIT), Monash University and Adacel Technologies in Australia.
The environment (stereographic panoramas of the temple complex) is visible in
the Museum, thanks to a state-of-the-art display environment, a unique place

50 The Virtual Museum of Aksumite landscape was developed by CNR ITABC, under the
direction of Maurizio Forte, Aracnet (VTree library and Vision Station) and CINECA (Vega
and Virtual Theater). More information on: www.vhlab.itabc.cnr.it/Projects_Aksum.htm.
The exhibition took place in Rome at EUR during the annual conference of MondoGIS (22-
24 May 2002).
51 http://www.vhlab.itabc.cnr.it/flaminia
Reconstructing Ancient Landscape S. Pescarin

for the visualisation of scientific and historical ideas, and associated objects52.
figure 5.3 Ename.tif

figure 5.4 itinera_ok.tif

figure 5.5 Flaminia virtuale_museo_nazionale.jpg

Social network and Virtual Communities

In 2004, I followed the preparation of the Digicult Thematic Issue 5, dedicated
to “Virtual Communities and Collaboration in the Heritage Sector” (GESER ET
ALII 2004). This report divided Virtual Communities into two main typologies:
1. Interest groups or discussion groups (web-based discussions) focussed
mainly on creating discussions or participation on specific topics, using
mainly simple tools such as: newsletters, newsgroups, chat, IRC
(Internet Relay Chat), Web-cam in real-time video conferencing
meeting, groupware, MUD & MOOs
2. Multi User communities, whose main goal is to create a meeting
experience among participants, through advanced 3D tools. 3D
environments are re-created and users are projected into the cyberspaces
by means of an “avatar”, who represent themselves even while
communicating. The first well-known platforms were Blaxxun, Second
Life and ActiveWorlds53.
Belonging to this second group, there is an interesting project is Quest Atlantis.
With its 200.000 students, QA was developed by Indiana University. In this
project artificial three-dimensional worlds, as in the case of the OPEN Virtual
Parks Project, are re-created with learning purposes to immerse children, ages
52 http://www.vroom.org.au, http://place-hampi.museum,
http://www.vroom.org.au/pdf/kenderdine_avatars.pdf
53 http://www.blaxxun.com; http://www.secondlife.com; http://www.activewolrds.com
Reconstructing Ancient Landscape S. Pescarin

9-15, in educational tasks (quests) (figure 5.6). “Quest Atlantis leverages multi-
user (online gaming) technologies to have QA students and their families
virtually explore the rich histories and beautiful natural environments of
National Parks. Through such engaging learning environments we hope more
people will develop a greater appreciation of our shared global heritage”. The
virtual experience inside the landscape enables young people to learn without
really noticing that they are. Each one, in fact, is appointed to act as a specific
character (the scientist, the ranger, the archaeologist), solving, alone or together
with other students, impelling problems that even threaten the end of an
ecosystem (BARAB ET ALII 2007)54.
Another example of the possible use of multi-user virtual communities comes
from the University of California, Merced, where the reconstruction of the
Roman Villa of Livia in Rome is used to teach undergraduate students, inside
Second Life world, or in the case of the “Akragas doors project”, developed by
No Real, where visitors can dress up as a Greek hoplite and take part in an
attack, thus gaining understanding of the military tactics of 6th century BC
Akragas in the Valley of Temples near Agrigentum, Sicily (figure 5.7) or also
in the case of the Digital Humanities island developed by King's College of
London (Appendix B).

[figure 5.6 quest_atlantis_ok.tif]

[figure 5.7 Akragas.tif]

Key words of these communities are: communication, exchange, common
interests, confidence, interaction, intermediation, authenticity, identity, and
sustainability
Apart from very few experiences, significant examples dedicated to ancient
landscapes are lacking. 3D communities are commonly used as a sort of 3D
chat room, as in most Second Life islands, where what people commonly do is
just meet and talk.
Nevertheless, there are institutions which could be really interested in creating
Landscape Virtual Communities. Museums of territories or Archaeological and
Naturalistic Parks may obtain positive results in the creation of communities
centred on e-learning or social communication and participation, as in the cases
above mentioned. They can help to create common interests and new synergies
between public and institutions, also with the exchange of memories and
stories. The creation of new landscape virtual communities, based on narration,
can give importance also to social features, not just cultural, thus helping to
produce new knowledge, public involvement and awareness through shared

54 http://www.questatlantis.org
Reconstructing Ancient Landscape S. Pescarin

participation in cultural policies, within democratic contexts.

Through these cyberspaces, powerful learning and tourist tools, it might be
possible to visit archaeological territories virtually, meeting other visitors or
guides for a greater and more intense understanding of ancient landscapes. But
it would also be possible to shift a traditional perspective, creating an “upside-
down” view of cultural digital portals with participants interacting directly and
adding, not only receiving, content and information. In Virtual Communities
cyberspaces become places, not just tools, and the system is the people
(REINGHOLD 1994).
If the communicative level of VR communities can be very high, there is a
problematic aspect that should be taken into account, i.e. the geospatial
dimension. They are more suitable for the creation of scenarios or small
environments rather than landscape. These cyberspaces are focused, in fact, on
exchange and experience much more than on the scientific value of the
geographical dataset. In most cases, scenarios are really simplified, so as to
enable a diffusion over the Internet, and sometimes they should be created
inside the chosen platform (as in Second Life), since to directly import
geographical data is not allowed.
Another problem of these projects is that their success depends strictly upon
how much they are used. In sociological studies, it was shown how similar
virtual communities are to real communities. To have success they need to have
a goal, be limited to that goal, have fixed rules, build habits, and be based on
the idea of sharing and exchanging. Unfortunately, it is difficult to obtain an
“exchange” when institutions try to exert close control over the communities,
limiting them just to a museum activity for visitors.

Mobile communication
We cannot talk about communication without touching on mobile technology.
It was the first to be developed for environmental and tourist purposes with car
navigators, thanks to the integration of commercial GPS in low cost devices.
Today this field is rapidly evolving in terms of digital content, moving toward
future generation devices, such as smart mobile phones. Archaeological sites
can be explored while visiting the areas through systems that integrate geo-
localisation of visitors and information provided directly on PDA or mobile
phones. Precursors of these kinds of application were ArcheoGuide, an
augmented reality systems developed to enable a virtual visit of archaeological
sites such as Olympia in Greece.
Recently there are examples of interactive exploration of archaeological areas
through integrated systems based on a centralized Multi-channel Service, as in
Reconstructing Ancient Landscape S. Pescarin

the GITA project developed by Infobyte55, which allows presentation of

multimedia content on demand or automatically on mobile devices (mobile
phones, PDA, SmartPhones) or immersive mobile displays (Augmented Reality
devices for outdoor applications).
Another new and very promising field is 3D interactive content, such as
educational games, available for smartPhones, like IPhone. This market,
although at its very beginning, is already very surprisingly active. In the IPhone
and IPod AppStore, at the end of April 2009 there were almost 1 million
downloaded applications!

.B.2 Research, communication and simulation

3d webGIS and VR webGIS

After the success of Geographical Information Systems in archaeology, and
their increasing use over the web through webGIS applications (chapter 3),
today there are new typologies of web-based geospatial applications, such as
3D webGIS and VR webGIS.
While GIS and webGIS are typically considered as 2d (or 2,5d) systems, 3d
webGIS fully manage the third dimension, enabling at the same time geospatial
data on line access. Thematic layers commonly used in GIS are wrapped on the
3D terrain (vectors, raster maps). In principle, 3d webGIS possess spatial
analysis potentiality.
VR webGIS are applications which have attributes typical of Virtual Reality
(real time 3d interaction behaviours), of web-based tools (networking, on line
access) and also of GIS systems (geographical dimension, vector-raster data
management). They can integrate different types of information such as: 3d
terrains, vector layers, 3d models and metadata, vegetation, multimedia
contents, etc. The landscape can be explored and queried through graphical
interfaces. Chapter 7 is dedicated to these specific applications.
Examples of VR webGIS were developed in the Appia and Virtual Rome
projects (Appendix A). In these cases, the reconstructed archaeological and
ancient landscapes were published in Internet, through a web plug-in,
osg4web56 which enhanced browser potentiality of 3d data exploration,
querying and interaction. Users, no matter who and where they are, can access
geospatial information and digital content through an interface, exploring the
territory dynamically in different ways (flying, or walking on it), uploading
other spatial dataset (2d, 3d, vectors, multimedia) on the landscape, to
understand and analyse spatial relations better (fig. 5.8).

55 http://www.infobyte.it
56 http://3d.cineca.it/storage/demo_vrome_ajax/osg4web.html
Reconstructing Ancient Landscape S. Pescarin

Figure 5.8 appia_virtualrome_osg.tif

Research Virtual Communities and Collaborative Environments

Landscape reconstruction is a multidisciplinary field which requires spatiality
and interaction. For this reason, the emergence of interdisciplinary contexts is
creating a need for research into a variety of fields. If interaction plays an
important role in the dissemination of cultural information, this is even more
important in VR applications for scientific communities, as in the case of
Collaborative Virtual Environments (CVE). CVEs are a class of networking
applications which support cooperation among remote users, through a common
spatial environment and using 3D graphics; CVEs are updated so as to reflect
the actions and the movements of the participants. In these communities, the
exchange of information and data through the net, and the development of
shared multidisciplinary interpretation sessions, based on scientific geospatial
datasets, is essential. The future of the net itself seems to be connected with the
future of community, democracy, education and science (REINGHOLD 1994). “A
virtual community is a space where people can bring in their own objects and
with these (digital) surrogates, their own interpretation [..] thus enriching shared
knowledge” (HAZAN 2004: 8).
Unfortunately, we don't have many examples of these specialized communities,
focused on the sharing of specific knowledge and topics. But we shall be seeing
several developments of these platforms in the near future.
The main characteristics of CVE are interaction, spatiality, inclusivity, editing,
sharing, multi-disciplinarity, networking, and repository and library integration
(table 5.1). In these contexts researchers can exchange ideas and analysis,
sharing the same dataset and environment as well, sometimes even the same
perspective. Each participant can cooperate in the creation of the archive, in
digital collection, and the creation of the virtual museum, bringing with him his
own contribution (such as digital objects) and preserving those of others.

Characteristics of VR museum for Characteristics of VR system for the

dissemination scientific community
Experience, story-telling Research
Learning Analysis, selection and comparison
Game Openness and transparency
Interaction (natural) Complex interactions
Perception Cognition
Simplicity Complexity
Involvement Sharing
Reconstructing Ancient Landscape S. Pescarin

Interfaces Flexibility
Design Updatability
Realism Reliability
Information Data
Content Metadata
Communication Communication
Immersivity 3d geospatial dimension

Table 5.1 (FORTE, PESCARIN 2005)

While in the modelling, design or military fields there are several examples of
CVEs, there aren't as many examples in the Cultural Heritage field, and
specifically in landscape archaeology. The causes may be mainly three: the cost
of these solutions, their complexity and lack of flexibility, and also the lack of
knowledge of their existence. Dealing with landscapes means having to manage
a variety of data, such as: 3d models of different dimensions and resolution,
detailed 3d terrains, high resolution geospecific images, vector thematic layers,
vegetation, natural characteristics, etc. Each one has different attributes and
might become a problem, when it is implemented into a VR system. MultiUser
and On Line access amplify this problem.
There is also a fourth reason. Although there are still many technical problems
in the development of such platforms, most complicated aspects aren't
technical. As in the case of Landscape Virtual Communities, a successful CVE
should be used by researchers. Unfortunately not everybody is interested in
sharing information or data. Some might think that information de-
contextualized and considered just as an exchange object is deprived of
relevance, even dangerous (HAZAN 2004: 9). We need to face the development
of new CVEs also from a sociological point of view, to understand what
interfaces and exchange mechanisms should be used.
In the research field, CVEs might be developed as fully editable environments
or just as interactive spaces used to access complex scientific databases.
Recently, VHLab team at CNR ITABC has been involved in two projects,
whose aim is the reconstruction of ancient landscape: a FIRB project on
Robotics and Virtual Environments (FORTE PIETRONI DELL'UNTO 2008) and
Virtual Rome (PESCARIN ET ALII 2008, CALORI ET ALII 2008). In these projects, as
an archaeologist, I have been able to experience personally the complexity of a
real multidisciplinary approach. For this reason, we have been pushed to
experiment a more efficient approach, testing and developing examples of low-
cost on line Cooperative Virtual Environments, based on Open Source and
Commercial platforms. The FIRB project has developed a multiuser
cooperative environment, available on line, which enables researchers to work
Reconstructing Ancient Landscape S. Pescarin

together in the reconstruction of archaeological sites. It is based on VirTools

Dev57 and is available on line, with a common Internet browser (figure 6.10).
Virtual Rome, on the other hand, has developed an Open Source web VR
application, based on geospecific data, 3d models and multimedia contents,
with front-end and back-end on line solutions, for the interpretation,
reconstruction and 3d exploration of Roman landscapes (Appendix A, figure
5.9).
[Figure 5.9 landscape_KR.jpg]

Simulation models
Reconstructions can be used also as simulation spaces. This is quite a
widespread application field in the hard sciences, such as chemistry, physics,
etc. but is still not very diffused in landscape archaeology. Nevertheless, in
architecture it is a well known practice, even when dealing with ancient
monument reconstruction. An example is provided by a project of the
University of Caen, in France, directed by Philippe Fleury, where a 19th century
plaster of about 70 square metres representing Rome and made by the architect
Paul Bigot is being scanned and completely reconstructed in 3D, becoming an
interactive laboratory on Roman architecture (Fig. 5.10)58. A similar project was
carried out by UCLA, Virginia Universiy and the Politecnico of Milan, on the
plaster conserved in the Museum of Roman Civilization in Rome and
representing this city in the 4th century AD, when Constantine was emperor.
The result of the complex 3D scanning was post-processed and used in various
media, also for communication and educational purposes (GUIDI ET ALII 2005)
A simulation of landscape dynamics was developed in the Exploris project, on
the eruption of the Vesuvius volcano in Italy. In this case, a mathematical
model of the eruption temperature surface was calculated by supercomputers
and visualised in a VR application. The territory around the volcano was
reconstructed in 3D and connected to dynamic databases, containing e.g.
population density. The result of the simulation was then visualised in real-time
and connected to a time display, so as to analyse the effects of the eruption on
the territory (figure 5.11)59. The same simulation could be performed also on
the Roman landscape, reconstructing the original volcano’s shape, as it
probably was before 79 AD .
57 http://a2.media.3ds.com/products/3dvia/3dvia-virtools/
58 Fleury, P. Madeleine, S. Réalité virtuelle et restitution de la Rome antique du IVe siècle
après J.-C., in Histoire urbaine. Société française d'histoire urbaine, vol. 18, 2007: 157-
165; http://doc.ocim.fr/LO/LO044/LO.44(6)-pp.20-23.pdf
59 http://exploris.pi.ingv.it/, Macedonio G., Costa A., LongoA. 2005, A computer model for
volcanic ash fallout and assessment of subsequent hazard. In Computer & Geosciences, 31,
N.7, 837-845.
Reconstructing Ancient Landscape S. Pescarin

These simulations can go beyond the field of research: they can be successfully
applied also in environmental services, for cultural policy evaluations and
preventive archaeology.
Figure 5.10 fleury.tif

figure 5.11 exploris project

Preventive and predictive archaeology

The use and application of preventive archaeology, changes significantly from

country to country. In some cases, even within the same country, as in Italy, the
rules change in accordance with individual regional regulations, due to the lack
of clear central directives and effective coordination. From a theoretical point of
view, preventive archaeology should be carried out by archaeologists and
environmental experts in close cooperation with the administrators in charge of
urban planning (PESCARIN 2008b, CARDARELLI ET ALII 2004). The results in fact
should be used, for example, in the definition of the archaeological risk or of
building restrictions for patrimony preservation, while their use for predictive
archaeology is still quite controversial. An archaeological map, in fact, is a
necessary and reliable tool since it is based on real-based knowledge, acquired
during scientific investigations. Unfortunately, it is not enough to define
portions of the territory potentially rich in remains. Predictive archaeology is,
rather, based on theoretical models provided by environmental, cultural and
geographical observation.
In the United Kingdom the general application of preventive archaeology is
defined by governmental rules, since 1990, through the PPG-16 (Planning
Policy Guide)60. The results of archaeological investigations are then
implemented in the Site and Monument Record, an on line archive of known or
potential remains. This information has also a public level of visibility, through
the on line ADS ArchSearch61. It gives the exact location of archaeological
sites, visible also on google maps embedded in the page, as well as basic
information and other connected resources such as historical maps62, road maps
and aerial images63.
Also in France preventive archaeology is a well established practice. Early
examples date back to the 70s. There is a national institute in charge of these
kinds of investigation: INRAP, the Institute National de Recherches

60 http://www.communities.gov.uk/index.asp?id=1144057
61 http://ahds.ac.uk/archaeology
62 http:// old-maps.co.uk
63 http://www.multimap.com
Reconstructing Ancient Landscape S. Pescarin

Archeologiques Preventive64. It contributes to the creation of the National

Archaeological Map prepared by following clear rules, a document used by the
State itself for urban and environmental planning.
Another example in this field is from Italy. CART (Carta Archeologica del
Rischio Territoriale) is the territorial management tool, developed by the
Institute of Artistic Cultural and Natural Heritage (IBC) in cooperation with
Emilia-Romagna Archaelogical Superintendency and several public
administrations. It is used as a tool for the knowledge and safeguard of our
heritage and as a support for the planning of territorial interventions. CART is a
GIS-based system connected with a geographical relational database, accessible
on line for registered users. CART enables to add to and search archive
information on the basis of various factors such as depth or location; to insert
digitized excavations, researches or just traces recognized in historical maps or
aerial images; to georeference data; to add geophysical and paleo-botanical
analysis; to carry on spatial analysis for quantitative evaluation and simulation
of potential reconstruction; virtual reality reconstructions; open access through
on line interface at various levels; archaeological potentiality simulation for
urban planning (depth or density simulation of possible archaeological
findings).
The systematic classification and digitalization of all known archaeological data
of a territory, in a geospatial system with editing, analysis and querying
functionalities, is fundamental for several reasons: research, reorganization,
standardization, availability of information and revision based on spatial
analysis. The goal is to build a knowledge instrument on the state of the art of
archaeological discoveries (archaeological maps), and a predictive instrument
(risk map) useful to reconstruct ancient landscape dynamics in various
historical periods. In this case, the reconstruction process produces preventive
maps which can be used to support preservation, and our collective memory, for
a better, more aware urban and territorial development.

64 http:// www.inrap.fr
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Chapter 6
OpenLand: open source and landscape archaeology

.B.1 Introduction
In 1999, in his famous book “The Cathedral and the Bazaar”, Raymond wrote
that the style of the Linux community looked like a confusing bazaar, full of
projects with different approaches. Although it would seem that only a miracle
could bring about a stable and coherent system, this style works. The
development of an open source project was compared by Raymond to a bazaar
because it was different from a centralized approach like the common software
development style. His objection referred to the need for a centralized
approach, in relation to projects with a high level of complexity. In these cases,
the traditional approach was directed at leaving the entire work of software
development to individual genius or small bands of wizards, without any
intermediate version available (RAYMOND 2000: 19-20). But the history of
Linux and of some open source projects has indicated a valid alternative
approach. The considerations that originally referred to software development
might be of interest also to the archaeological field. Landscape reconstruction
process could be inspired by some developments of Open Source tools.
In Cultural Heritage, projects do not have always a “linear” development, for
several reasons, such as the continuous lack of substantial funding or the
difficulties over sharing, or the way new data are found. This means that the
"natural" development of a project is often subverted by more urgent priorities.
Sometimes the projects themselves have a certain vagueness to allow for a
necessary flexibility. If on the other hand we compare its approach to the one
adopted in industrial and commercial fields, as in all those areas with a more
pronounced economic impact, we can see the latter follows a “vertical
management", from the initial construction phase of a project to the hierarchical
management of human resources. Several business models based on this
approach have been analysed by economists for adoption by the humanities.
There is another reason why we can associate open movement and virtual
heritage: the social aspect. The archaeological landscape, sites, parks, and
historical monuments are heritage properties belonging to the worldwide
community.
Researchers, scholars, and curators also have a social and public mission. This
mission regards the study and conservation of our heritage to enable it to be
transmitted to future memory.
Scientific research itself has a public dimension, in the creation of studies of our
collective heritage and cultural landscapes, and also in the communication of
Reconstructing Ancient Landscape S. Pescarin

such studies.
What is often underestimated is that progress in research is also founded on
information sharing criteria. This will be increasingly clear in Europe, where, in
a sort of “global competition”, will emerge those who can get access to
knowledge. The extension of an open source approach to the field of
archaeological landscape may open up new horizons, in terms of available tools
used and methodologies to be applied to sustainable projects, as EU projects
such as Epoch65 have demonstrated. The question, whether a “conversion” to
open source and a rediscovery of a real public dimension of research in the
actual economic panorama is really possible, is still open (LASER 2005). We
could think about a softer approach, directed at the adoption of alternative
possibilities to already available commercial products.
Although the Open Source movement started from software development, there
is perhaps a lesson to be learnt also by cultural heritage (STALLMAN 2001,
TORVALDS DIAMOND 2001): “visiting a bazaar” might produce good results in
terms of research’s global (not just individual) growth. Open Source therefore
can represent a possible alternative to commercial software and offer different
models of project development and of business (AA.VV. 1999).
Open Source concerns the way a software is distributed, together with its code.
It doesn't just mean access to the source code, as stated by the Open Source
Initiative66, but should satisfy the following requirements:
I free re-distribution;
II source code availability;
III opportunity for modifications and derived works;
IV integrity of the author's source code;
V no discrimination against persons or groups or against fields of
endeavor;
VI continuity in the distribution of licenses;
VII no license specificity for a product;
VIII no restrictions for other software;
IX neutral development of technology.
In archaeology, we can treat Open Source as a movement that offers open (and
often free) software, but we can also use it as a reference methodology for the
development of a project or an application.
As we have seen in chapters 1 and 2, transparency in the reconstruction process
is crucial. As a method, Open Source can offer several solutions to maintain
transparency in the results.
In the next sections, I will try to describe its pros and cons, comparing a typical
Open Source project with one belonging to landscape archaeology.
65 http://www.epoch-net.org
66 http:// www.opensource.org
Reconstructing Ancient Landscape S. Pescarin

At the beginning, programmers freely shared the software they were

developing, together with the source code. The result of this practice is
described as an incredibly rapid development of the field, with many
consequent positive effects on final users (STALLMAN 2001). At the end of the
80s this development started to slow down, perhaps in part due to the
introduction of software patents, thus giving only binary programs to final
users. This commercial strategy led to the creation of probably more stable,
robust and documented software, but it also blocked the creation of new
algorithms and tools. In 1984 those programmers who believed in a different
philosophy decided to join the Free Software Foundation and the Open Source
Initiative (STALLMAN LAWRENCE 2002). Those who subscribe to the initiative and
decide to publish their tools under a public license (such as GPL or LGPL67)
believe that this can contribute to the creation of better software and to its rapid
evolution. For an end-user, it means that it is possible to download and use tools
(binaries or precompiled versions), modifying them (or asking for
modifications) and adding new functionalities, appropriate for a specific
purpose. It is also possible to continuously access new releases or updates and
participate in the international community, as users, with requests and
suggestions.

.B.2 Open Source and Landscape Archaeology

While Information Technology is related to software and hardware, landscape
archaeology concerns cultural and environmental information. Researchers in
the former field are involved mostly in software development, while the
interpretation process absorbs archaeologists, anthropologists and
environmental experts. What I propose here is that this field can probably
benefit from the application of an open approach. To what extent?
As described in the previous section, participation is a key element, typical of
an open project, and connected to any further potential sustainable
development. A project in fact can be maintained by the scientific community
itself, if it is a research project, or by a users community, in the case of
knowledge dissemination and communication. Involving people and taking
them into the projects makes them aware of the scientific cognitive process or
memory, enabling an easier continuous updating, preventing obsolescence. An
open approach institutes a dynamic involvement process between players,
encouraging a wider participation of the entire working group. Some analysts
think that a new way, based on sharing and exchange, is slowly emerging,

67 For a description of most common licenses: http://www.gnu.org/licenses/

Reconstructing Ancient Landscape S. Pescarin

beside the typical private industrial economics, price-driven (BENKLER 2005).

Other studies have underlined that the use of technologies in various fields,
such as in archaeology, involves a continuous state of tension between a
“working-together” and a “working-apart” philosophy (SURMAN REILLY 200368).
This dichotomy emerges even more clearly, when we consider the research that
uses the network or that are web-based. In these cases, four main approaches
can be identified: formal, informal, distributed and centralized (table 6.1). Each
approach tends to use different applications and tools, making a different use of
reconstructions.

FORMAL
Research Networks, Traditional Media
Collaborative (web page, drawing,
Environments, movie, broadcasting
Repositories products), Intranets
DISTRIBUTED CENTRALIZED
Social Networks, Alternative Media
Wikis, Blogs, P2P, (Interactive Kiosks,
MuDs Virtual Museums),
mailing lists, games
INFORMAL

table 6.1 – schematic approach in cultural heritage

Institutions, such as museums, universities, government bodies, etc. tend to
have a more formal centralized approach, controlling information and
communication, in a top-down perspective, through “linear” systems such as
traditional media. In a few cases, there are experiments regarding a more
informal and interactive use, for example with Virtual Museums. They use
mainly commercial products. The scientific community follows a more formal-
distributed approach and employs some commercial tools, but also many open
projects. The communication in this case, one-to-many, when directed at a
public, or many-to-many, in the case of working environments, can in the future
benefit from the diffusion of an open approach. In fact the construction of open
systems which allow the exchange of data and information and the sharing of
ideas inside multidisciplinary groups, can trigger a virtuous cycle, encouraging
“working-together” initiatives, to the disadvantage of “working-apart”.
Interdisciplinary activities, in fact, are essential in the development of
innovative and useful projects. In its initial phases, a project should involve a
variety of people and experts, to indicate knowledge bases, necessities, interests

68 On line version of the report is available at:

http://programs.ssrc.org/itic/publications/knowledge_report/final_entire_surman_reilly.pdf
Reconstructing Ancient Landscape S. Pescarin

and requirements. The definition of a common language is another requirement,

the premise for and the consequence of this approach.
An open project of landscape reconstruction should therefore be characterized
by an exchange and sharing practice, from its beginning distinguishing what the
general development plan requires from the software development/use plan.
The use of exchange tools (repositories, content management systems, etc.) and
media (Internet) makes it really feasible (LUKE et alii 2004).

.B.3 Open Networks

The use of a network is a fundamental requirement of an open and
multidisciplinary approach.
The Internet is increasingly becoming a working environment and not only a
communication media, useful to publish, often marginally, the results of
research.
The network, thanks to the widespread development of physical infrastructures
and the widespread availability of broadband connections in many countries, is
today one of the most powerful media to access and diffuse information and
knowledge.
The possibility of integrating different data, thanks to the network, increases
validation and criticism of the research itself, preventing wasted effort and
resources duplication. Moreover, it enables rapid development through sharing.
The Internet is therefore an instrument for knowledge, communication and
dissemination, but also for study and research (SURMAN REILLY 2003).
Available tools are: on line database, Content Management Systems (CMS), on
line multimedia repositories, webGIS, VRwebGIS and, in the future, also
Collaborative Virtual Environments (chapter 7).

.B.4 Open tools for ancient landscape reconstruction

There are several open source tools already in use or potentially useful in
archaeology, either as alternatives to commercial software or as starting points
in the development of new applications specifically orientated to the field.
How can open tools be found? One of the principal sources is SourceForge69,
with its 100.000 projects and one million registered users. It is a space
dedicated to open source projects, where programmers can exchange source
codes and users can access binary software. It is considered the largest
repository of available codes and applications. It is a public and open resource
where it is possible to publish projects and share their codes or information on
them, working together through the net. The use of this kind of shared space
strengthens the quality of delivered applications, because they are also open
69 This repository is available at: http://www.sourceforge.net
Reconstructing Ancient Landscape S. Pescarin

spaces, where other skillful developers belonging to a broad worldwide

community, can freely express criticisms. The idea itself of putting forward a
project in such a “glass case”, with tremendous visibility, represents a further
motivation in producing products that are stable, robust, and error free.
It was said above that open source software is developed by several
programmers contemporaneously, working on the same program (code), or who
add sub-programs to the main one. Within this structure, the only way to
produce reasonable results is to align and keep track of various versions,
merging them, but preserving, at the same time, a history of the project. For this
purpose, programmers use Software Configuration Management tools (SCM),
such as CVS (Current Versions System70), Subversion (SVN) or Tortoise71.
These softwares are exactly designed to control different versions and align
them. Also in the landscape reconstruction process there are several researchers
working on the same territory or on the same site or even model. Sometimes it
can be very difficult to cooperate, avoiding wasted effort or duplication. A
simple example of this approach is the shared writing of a document through an
on-line management system, such as GoogleDocs. In 3d modeling this can
already be seen. When different modelers have to work, for instance, on the
same monument, but concentrating on different parts, it is quite helpful to use
tools which can help to align them to the resulting merged model, keeping track
of the different versions. An experiment in this approach was made during the
Virtual Rome project, with the OSG Post Processing tool (fig. 6.1) (PESCARIN et
alii 2008). Unfortunately there are no relevant examples in the “terrain” field,
although this could be really challenging. Recently, there are some interesting
initiatives that can provide stimulating suggestions for the scientific
community. One of these is the Open Archaeology initiative, which hosts a
number of sub-projects dedicated to tools specifically designed for archaeology,
such as StereoPhotogrammetry, Digisite, and so on72. Another one is the Virtual
Terrain Project, probably the most complete review web-site regarding terrain
generation and visualization.

Figure 6.1 osg_postproc_tool.jpg

The complexity of landscape reconstruction makes acquisition and post-

processing activities quite problematic. The risk is to create huge monographic
studies, far from the spatial or temporal context, or which cannot be integrated

70 http://www.nongnu.org/cvs/
71 http://subversion.tigris.org, http://tortoisesvn.tigris.org/
72 https://launchpad.net/openarchaeology
Reconstructing Ancient Landscape S. Pescarin

into other research on the same topic. Integration is another central issue in data
processing; it can be obtained by working inside a single digital “framework”,
collecting and connecting data, procedures, technologies, formats, and
methodologies, from different disciplines. In this way greater care is taken over
exchange formats more than on standards, or on the methodological approach
rather than on software lists. The more we use open exchange formats, the more
researchers can be free to choose tools better suited to their uses. Projects and
data can therefore be more easily updated and moved from one platform to
another73. Users dependency on the software company in the case of
commercial products is very great and makes for insecurity. If it fails, or
decides not to produce a specific software we are using any more, we will no
longer be able to read our data.
In table 6.2 there is a schematic list of the most common open source tools,
useful for landscape reconstruction, together with their commercial references.
In the Appendix there is a more detailed description of most of these tools.

Category Commercial tools Open Source tools

Operative System Windows, Mac Linux, (CygWin)
Server Microsoft IIS Apache
Suite Office Microsoft Office OpenOffice, KOffice
Web browser MSExplorer, Safari, Opera Mozilla Firefox, Konqueror
Desktop Publishing Macromedia Freehand, Xpress, Inkscape, Scribus
FrameMaker
Image Processing Photoshop Gimp
CMS (Content Management System) Oracle Mambo, Joomla, Drupal,
PHP-Nuke, MediaWiki
CAD AutoCad Qcad, BRLcad
Panorama Tools QuickTime, PhotoVista PanoramaTool, PanoTools,
PTStitcher OpenQuickTime
(Linux), Hugin
GIS ESRI ARCGIS, ARCpad GRASS, QGis
webGIS MapServer
VR webGIS and 3d webGIS Skyline OSG4WEB
Virtual Reality CG libraries, VirTools, Vtree, Unity3D OpenSceneGraph, OpenSG,
development toolkits Delta 3D
Game Engine Torque OGRE
Terrain generator and Ecosystem Terravista, Multigen Creator Terrain OSGdem, Demeter, VTP
generator Studio, Visual Nature Studio, Bryce
GPS software SkyPro OpenSourceGPS,
GPStoolkit,
3d modeling 3D Studio Max, Maya Blender, Art of Illusion,

73 On open formats: Open-document Standard Initiative (http://www.oasis-open.org/specs)

Reconstructing Ancient Landscape S. Pescarin

Wings 3d
Laser Scanner (acquisition, Cyclone, RapidForm Grass, MeshLab, Scanalyze,
processing) MeshAlign, PointShop
3d viewer and web 3d viewer
Software Configuration Management CVS, Turtoise, SVN
tools

Table 6.2
Landscape reconstruction requires the contribution of various disciplines, as we
have explained, in a dynamic process. An open approach contributes therefore
to creating a research network, through which various and distant working
groups could cooperate, using Internet. Moreover, the entire digital pipeline
described in chapter 2 can be successfully designed on Open Source software,
as can be seen in the proposed scheme (fig. 6.2).

Figure 6.2: fig12_1_digital_pipeline_opensource.tif

Data acquired in the field, with GPS, Laser Total Station or Laser Scanner, are
initially treated with specific software provided by the instruments. Almost all
hardware companies, in fact, do not let their users freely access the source code
of the accompanying software. Nevertheless, there are some projects which are
developing tools to process raw data captured by instruments.
Raw data may be initially uploaded on a server, in order to store the original
data’s accuracy and resolution.
After the acquisition, raw data are processed in the lab with various software
categories. For 3d modeling processing, tools might be used such as
Scanalyze74, MeshLab75 or, although not open but free, the ARC3D webservice
(Automatic Reconstruction Conduit)76, and in the following phase a computer
graphic software more oriented towards realistic rendering and computer
animation, such as Blender77. For terrain generation, the first step requires the
use of GIS and Remote Sensing tools. The most famous and mature GIS
package is GRASS78, that can be used both for vector and raster data. It also has
powerful algorithms for remote sensing analysis. To generate 3d terrains and to
create entire ecosystems consisting of forests, monuments and other objects,
there is for example the Virtual Terrain Project or OSGdem, a tool that belongs

74 http://graphics.stanford.edu/software/scanalyze/
75 http://meshlab.sourceforge.net/
76 http://www.arc3d.be/
77 http://www.blender.org
78 http://grass.itc.it/
Reconstructing Ancient Landscape S. Pescarin

to the OpenSceneGraph library79. Metadata and other textual information can be

stored in a database such as MySQL or, in the case of spatial data, PostGRE
SQL.
Processed data can be uploaded in a structured repository, built on a
geographical database, and accessible to non-programmers through a web-
based interface. A Content Management System, based on various connected on
line databases, can be successfully used in order to let users access repositories
and add their contents (3d, 2d, multimedia, text), or modify those already
uploaded. Although the majority of CMS are dedicated to text or image
content, some experiments are actually in progress to enable a user to access
and edit also 3d or 2d-geospecific data, embedding or integrating plug-ins for
3d visualization, such as OSG4WEB-backend,80 or webGIS editing, such as
ARK81. Through these systems data can be visualized, edited, dynamically
modified, and connected to other repositories or database. Other useful tools for
shared work are Wikis82. The entire set of processed information will be
integrated in shared repositories, which can be successfully used in the
following reconstructive phases (archaeological, interpreted and potential
landscapes). The reconstruction work can be configured as a collective process,
developed on shared resources inside 2d or 3d geospatial environments.
The digital pipeline ends with the landscape reconstruction and with its
visualization on line or off-line, through an interactive real-time system (VR –
AR) or through a non-interactive visualization (video, digital image). The
creation of a Virtual Reality application, where users can dynamically explore
the reconstructed landscapes activating various behaviors, requires significant
programming activity, especially in the Open Source domain. 3D platforms
dedicated to VR, such as OpenSceneGraph83 or OpenSG84, or game and
simulation engines, such as OGRE, Crystal Space85 or Delta3D86, although very
powerful, often do not offer development and deployment utilities. VR
(web)GIS and 3d (web)GIS represent a good compromise for those who need to
visualize geographical dataset and 3d models simultaneously and interactively,
although open source solutions are still under development. One example is the
OSG4WEB plug-in developed by Italian CNR and CINECA Visit Lab (CALORI

79 http://www.openscenegraph.org/projects/osg/wiki/Support/UserGuides/osgdem
80 See Virtual Rome project: http://www.virtualrome.net
81 See http://ark.lparchaeology.com/
82 There are several open source Wikis (http://en.wikipedia.org/wiki/Wiki), such as
MediaWiki (http://www.mediawiki.org)
83 http://www.openscengraph.org
84 http://opensg.vrsource.org/trac
85 OGRE game engine: http://www.ogre.org; Crystal Space: http://www.crystalspace3d.org
86 Delta3D is a simulation engine that manages geospatial information:
http://www.delta3d.org/
Reconstructing Ancient Landscape S. Pescarin

et alii 2008). OSG4WEB provides a framework for in-browser openGL-based

application wrapping. The framework allows the development of OpenGL and
OSG based applications embedded in the browsers, allowing JavaScript
bidirectional interaction with surrounding page elements. It is particularly
suited to landscape visualization over the web, since it supports paged LOD
terrains (chapter 7).
Realistic rendering, animations and video compositing are still quite
problematic, in both non-open and open communities. Landscape
characteristics, in fact, make the management of large territories inside
traditional Computer Graphics software such as 3D Studio Max or Blender,
very difficult. One solution is to use a game or simulation engine that supports
terrain rendering, or import terrain generation outputs.
A full and updated review of the most common open source tools is available
also in the Wikipedia, divided into categories87.

.B.5 Other open case studies

During the annual CAA (Computer Application in Archaeology) international
conference held in Prato in 2004, some colleagues and I organized a workshop
for the first time on open source and free ware in archaeology (“Open Source
and free ware applications to archaeological research”). During the workshop
the increasing tendency of the worldwide community to try out this relatively
new approach emerged quite clearly. Some of the projects presented on that
occasion are still maintained and developed, demonstrating that in the near
future more and more users will start using and developing open source
solutions for archaeology.

The following case studies represent interesting examples.

ArcheOS, developed by the ArcTeam, offers archaeologists the chance to have
a Linux distribution which includes fundamental specialized tools for CAD
drawing, GIS, database creation, GPS, Scanner Laser, statistics, stereo-
photogrammetry, image processing and computer-graphic88.

I have already mentioned GRASS (Geographic Resources Analysis Support

System), a powerful GIS and remote sensing tool at one and the same time,
continuously maintained and updated by its main developer community based
in Trento, Italy. This software was initially developed for the US Army in 1982
by CERL (Construction Engineering Research Laboratory). In 1997 it was
released to the worldwide community open and free. Since that moment, this
87 Open Source software:
http://en.wikipedia.org/wiki/List_of_open_source_software_packages
88 http://www.arc-team.com/archeos/wiki/doku.php?id=home
Reconstructing Ancient Landscape S. Pescarin

project, now available also under Windows, has created new functionalities for
terrain creation and visualization, for GPS and Laser Scanner data
management89. Now it is part of a wider framework, OSGeo.

In the field of VR application dedicated to terrain interactive exploration we

should also mention the VISMAN project, a software framework for Managing
Virtual Scenarios. It is currently developed at CINECA Visit laboratory and is
mainly devoted to cultural heritage. It is written in C++ and uses the wxWidgets
and OpenSceneGraph libraries, so it relies on open-source software only. It is
available on Windows and Linux systems, being scalable from an average PC
to a virtual theater with stereo capabilities. It has been used to develop
advanced systems within virtual cultural environments, as in the case of the
simulation of the results of a possible new eruption of the Vesuvius volcano90
(fig. 5.10). The same simulation might be used to recreate the 79 AD eruption
in Pompeii. Visman's goal is directed to the integrated fruition of complex 3D
and multimedia contents on multi-platform technologies and in various
contexts. The 3D virtual worlds become starting points for a straightforward
interaction with complex environments that can be navigated, modified and
queried through the link to more than one relational Data Base at a time (GIS,
Multimedia, etc.), becoming an interface for accessing multiple information91.

.B.6 Conclusions and future perspectives

Today software and contents are more and more integrating into advanced
systems, available on line. The creation of programs, through an open approach,
dedicated to the specific needs of the discipline, enables the creation of
sustainable projects and the rapid evolution of the research itself.
We are slowly moving toward a shared on-line cyberspace, where
interdisciplinary exchanges can happen and allow new ideas and methodologies
to emerge. New cooperation agreements in the field of archaeology, computer
science, psychology and the environment should be created.
The archaeologist’s role itself is changing, and it will change all the faster

89 Useful references can be found in the FOSS4G conferences or in the GRASS users
conferences, such as
http://www.ing.unitn.it/~grass/conferences/GRASS2002/proceedings/proceedings/pdfs/Brov
elli_Maria_Antonia.pdf
90 http://exploris.pi.ingv.it
91 http://www.cineca.it/resources/files/visman_en.pdf; to download the software:
http://www.cineca.it/resources/files/visman.zip
Reconstructing Ancient Landscape S. Pescarin

when they are involved inside (open) collaborative environments, where a more
efficient and pragmatic perspective will assure more reliable results.
Archaeologists will still occupy a very important position in these projects even
if they are not computer experts. Their perspective can help in addressing and
testing applications and software development.
From this viewpoint, open source should not be intended only as a collection of
softwares, but as a project that is also a network of people, methods, and
technologies, adaptable to the various applicative contexts. Landscape
reconstruction will be a pre-eminent field in the development of new
applications web-based and GIS oriented, if it follows a few basic rules in the
development of the actual project:

1. get your data out, perfection will come later;

2. avoid locking your data in proprietary formats;
3. prefer free and open source software for data editing and storage.
Reconstructing Ancient Landscape S. Pescarin

Chapter 7
Webscapes

.B.1 Internet connections

Internet is the most widely used planetary inter-connection system. If we
examine the evolution of the Internet landscape (the webscape) over the last 10
years on a map, we will be very surprised92.
Although it started as an intra-net to connect leading research institutions93, its
wide diffusion is mainly due to the use that has been made of it by ordinary
people at home. In the 80's, the world began to be transferred from the outside
to the inside. In this sort of “Escher” game, the perception of time and space
changed. Near or now are today much more expanded concepts than they were
in the previous era.
The internet is now giving new meaning to concepts of equity, of impact (also
in terms of politics), of trust through collaboration processes, and sustainability
(economic impact) (SURMAN REILLY 2003).
Figure 7.1 – webscape.tif

Internet has also exasperated the way information is sought. Interaction is on

the increase through hyperlinks and windows interfaces. Understanding doesn't
just depend on analysis, but also on navigation and on the creation of new
personal paths.
In this new era these are also new risks: the risk of limiting our approach to
surface explorations, to rapid navigation, losing our capacity to concentrate and
analyse; the risk of dealing with the world as a collection of signifiers without a
signified; the risk of being manipulated by simulations (TURKLE 1997: 121-140).
This Internet era is in fact the era of simulations and of new freedoms, which
become merely apparent and superficial, if we don't achieve greater awareness
through the creation of critical instruments, tools to increase our potential for
memorization and understanding. On the one hand opacity does enable us to
deal with a higher level of complexity, and to create connections, which would
otherwise be impossible. Internet is a parallel world, a new type of organism;
new objects live in its connections, “lives on the screen” (TURKLE 1997: 42).
While we work studying and simulating the landscape we are part of a complex

92 Map of connections: http://www.telegeography.com/products/map_internet

93 Further information and links at: http://en.wikipedia.org/wiki/Internet
Reconstructing Ancient Landscape S. Pescarin

system of relations and connections (fig. 7.1).

Internet is everywhere in most of the world, it is part of us, and it will be a
persistent aspect of everyday life more and more in the future. It is not a matter
of if we want to use it in or for our research. The question is whether we want to
be connected, or isolated and left apart, whether we want to dialogue, to
disseminate, to measure the merits and results of our studies. Internet is today a
medium of science, beside traditional scientific publications.
In terms of social change and global scientific development, it is necessary for
new concepts to be diffused as far as possible. In some cases a capillary
diffusion isn’t necessarily connected to content validity, but to the way these
concepts are diffused and the medium used. The way parallel programming
(PDP) started to be so widely diffused was due to the free distribution of
software to be tried out on floppy disks, in conformity with the successful
model “try it, play it, modify it” (TURKLE 1997: 163-164).
The use and further development of web applications in this new and complex
field of ancient landscape reconstruction depends, therefore, on social
dynamics. From a certain perspective, the resistance to reconstruction activities
we still experience in much of the Cultural Heritage field is a positive factor,
part of the cultural impact of a new approach. This resistance tries to separate
emotions and aesthetics from cognition and logic, returning our attention to
essential characteristics, creating a distance between virtuality and the reality of
ancient landscape.

What can be done for landscape reconstruction through or with Internet? There
is no simple answer to this question. In this book we have analysed the various
aspects of the field: people (scientists and the worldwide community, teachers
and learners, etc.), data (geographical, spatial, bi- or three-dimensional,
multimedia, ecosystems etc.), processes (visualisation, interpretation,
transparency, interaction, exchange, cooperation, land evaluation, data
acquisition, post-processing, etc.) and systems (simulation and communication
systems, interaction, virtual reality, complex systems, etc.). I'll try to answer the
question above by taking as reference this distinction, indicating a few general
categories, analysing problems and state of the art, and finally proposing some
solutions. Table 7.1 proposes a summary.

.B.2 webscapes

People
Social networks represent an interesting possible evolution for scientific
communities to exchange ideas and results. As described in chapter 5, these
Reconstructing Ancient Landscape S. Pescarin

networks aren't necessarily visual environments. In most cases they are places
where people can exchange information, through text-base tools, and
multimedia data, such as pictures or videos. One of the most famous social
networks is facebook. With millions of members, its owes its success to its
structure. First, it was designed for college students, by college students.
Second, it allows information to spread widely but also protects privacy: you
can trust it. Third, it emphasises clusters (socio-economic) and groups, thanks
also to low-involvement communication; and last but not least, it offers
simplicity and speed. It can be used by people interested in specific topics, such
as ancient landscape. In fact the success of a social network depends to a great
extent on having many users. It works when it creates social aggregations with
people joining public discussions and forming personal relationships in the
cyberspace (REINGOLD 1994).
Virtual Communities are more focused on the concept of experience (chapter 5).
They include 3d visual environments at different levels. The user participates in
the VR community through a digital alter-ego or an avatar, employed also to
create a relation and an involvement with other users. A virtual community is
based on the same concept as the social network, from a certain perspective: the
creation of relations among people. In a Virtual Community the cyberspace
becomes place.
In these cases, the language used, the interfaces, and the levels of interaction are
crucial. Virtual reconstructions of ancient landscapes will be interesting for a
wider community of humanists when developed applications start to speak the
language of the humanities, and for a community of scientists when it starts to
speak the language of logic and mathematics. As Sherry Turkle observed in her
“Life on the Screen”, sometimes it is necessary to create a Trojan horse to
enable some ideas or approaches to be introduced into “enemy territories”
(TURKLE 1997: 169).
E-Learning applications are commonly used for online education at various
levels. They involve teachers/students in learning activities. Although in most
cases they are used as the simple re-production of academic lectures, they could
go much beyond the simple lesson, when they become an experience. They
often are based on VR online platforms, such as Second Life or ActiveWorlds,
but they are carefully put together by teachers. These 3D online e-learning
platforms can be particularly successful, as in the Quest Atlantis described in
the following section, since they personally involve students in the solution of
problems, helping them to get to know and understand the basic or specific
information they require.

Data
On Line repositories are particularly useful for landscape reconstruction, since
Reconstructing Ancient Landscape S. Pescarin

they enable us to access basic shared resources, such as geoimages (like the
Global Land Cover Facility), digital elevation models, 3d models, vector files,
etc.. They can be simply FTP spaces, accessible widely or restricted to
registered users, or they can be accessed through more advanced and structured
interfaces, and included in digital libraries and WCMS.
Digital Libraries are structured typological archives of digital content, as in the
case of Roman vegetation library or 3d terrains or 3d architectonic parts library.
They are based on online database structures (like MySql or Oracle), geo-based
in the case of geographical datasets (such as PostGreSQL). They are used in
content management systems.
A web content management system (WCMS or Web CMS) is a content
management system (CMS) software, usually implemented as a Web
application, for creating and managing HTML content and connecting html
pages dynamically to digital libraries and online repositories. It is used to
manage and control a large, dynamic collection of Web material. It facilitates
content creation, editing and maintenance. There are many available CMS
platforms (like Joomla, etc.). Usually this software provides authoring tools
designed to allow users to easily create and manage contents. The most
advanced WCMS can include 3d visualisation and DB visual editing
functionalities, embedding 3d plug-ins.
Other well known and fundamental tools for landscape online visualisation are
webGIS. They started as web extensions of GIS software, used to manage
cartography online, but they are more widely used now and have broader
communicative attitude. A webGIS is used to disseminate geo-spatial
information, but also to archive and share archaeological and environmental
information with other users. In some advanced cases, they have a higher
interaction level, including editing functionalities which more closely involve a
variety of professionals, e.g. in public administration, higher education, tourist
operators, etc.
An evolution of webGIS is 3d webGIS. They enable on line 3d visualisation of
geographical information with basic functionalities such as interactive
exploration and in some cases questioning or problem raising. The first 3d
webGIS platform was Skyline, an Israeli company which first believed in this
promising field as early as 1997. Nevertheless, Google Earth is surely the most
well-known application. Its developments are leading to a further development
of this project directed at VR functionalities creation (VR webGIS).

Processes
There aren't many applications or examples which handle geo-spatial processes.
This specific field, although at its very beginning, is very promising.
A recent tool, used also by specialists to exchange information or share specific
Reconstructing Ancient Landscape S. Pescarin

knowledge (such as software development or program use), is Wiki. The best

known example is the wikipedia, but Wikis can be successfully integrated in the
process of landscape reconstruction since they enable the scientific community
to share, edit, modify, or correct textual information, integrating images or
videos and referring to other tools, such as digital libraries or online
repositories. They are also commonly used in the cooperative work of project
development.
Less familiar applications are cooperative environments, also known as CVEs
(Collaborative Virtual Environments). They are still at a very experimental
stage. CVEs are a class of networking applications that support collaboration
between remote users, through a common spatial environment and using 3D
graphics; CVEs are updated so as to reflect the actions and the movements of
each of the participants. They are based on the concept of data and knowledge
sharing, and on editing and validating functionalities. They include 3d and real
time characteristics, typical of virtual reality applications, and communication
paradigms of VR communities. In the near future they could be very promising
applications, useful to reconstruct and represent potential Roman landscape,
through the integration of different sources, the definition of models to be
loaded and the identification of content connected to 3d models. The entire
process can be completely updatable and editable. Some attempts have been
made in this direction, enabling 3d editing, apart from the 3d terrain that needs
to be pre-processed to be handled online as paged. There are also inspired
examples that have explored the possibilities of exchanging digital information
in a sort of virtual reality networking matrix, as in the case of City Cluster.
Another interesting field for landscape reconstruction is 3d web service. This
includes a variety of services, from the distribution of stitched pictures, and
geographical data on demand, including also post-processed remote sensing
data, to the digital publication of scientific works. Recently there are some
interesting new developments in the creation of 3d digital models online, from a
series of pictures provided by users. This service, described below, is based on
computer vision technology. Although not yet available, in the future there will
be more and more of this kind of service, dedicated to 3d digital modelling or
also to 3d terrain modelling, on demand. The success and diffusion of such
complex projects will probably depend upon the service itself and the source
code being (initially) freely and openly available .

Systems
In chapter 5, I dedicated some space to the description of Landscape Virtual
Museums. One of the most promising communication systems, in terms of
potential diffusion, of knowledge dissemination and cognitive impact are online
Virtual Museums. If virtual museums, despite being known about and used
Reconstructing Ancient Landscape S. Pescarin

since the early 90s, are still considered at an embryonic stage, this is even more
true for online VR museums. In the inclusive Internet dimension, where
anything can be accessed with a click, it is essential to avoid hyper-information,
which in the end may turn out to be no information. These applications should
therefore provide the careful design of behaviour, the differentiation of levels of
access to information, a well planned storyboard and well constructed
storytelling.
Online computer games are surely more mature applications. The powerful role
that they are playing in culture and society cannot be denied, as well as the
issues that may be raised about the boundaries between the ‘real’ and the
‘virtual’. It doesn't simply involve a “migration” of previous board games into
digital online format, but has a wider social impact. Games have also gained a
place at the academic level today, with thousands of courses held on the subject
throughout the world. One specific type of game is the so-called MMORPG, an
acronym for "Massively Multiplayer Online Role-Playing Game". In a
MMORPG, thousands of players exist in the same game world at the same
time. They play a role game together in a 3d environment. World of Warcraft is
one of the best known MMORPGs. It enables thousands of players to come
together online and battle against the world and each other. Players from across
the globe can leave the real world behind and undertake quests and heroic
exploits in a land of fantastic adventure. It allows players to play the game at
their own pace; it offers interesting quests with story elements, and dynamic
events. There are several game engines which can be used to create interactive
3d experiences, even on line94. Most of them are costly and proprietary, but rich
in available behaviours, such as physics management, multi-user functionalities
or artificial life implementation. Although a few of them are dedicated to
geographical simulation, the majority aren't focused on the treatment of “real-
based” spatial datasets. In order to maintain a high level of speed (frame rate)
and interactivity, especially on line, they use specific techniques to visualise
and create texture and geometry (procedural modelling, procedural texture
creation, normal maps, shaders and similar effects).
While computer-games aren't usually based on geo-spatial scientific datasets,
there is a type of application which includes some aspects typical of Virtual
Reality systems and others of GIS: VR webGIS. These applications can be used
to visualise large territories online, interactively, to explore sites in an intra-site
and inter-site dynamic, to add to and query elements of the environment, to
create personal interpretation paths, and to test simulation models, including
artificial life evolution. They are spatial geographical applications and for this
reason they can be successfully used by the scientific community as working
tools. They can be developed through a cooperative virtual environment or a
94 A complete list is available at http://en.wikipedia.org/wiki/List_of_game_engines
Reconstructing Ancient Landscape S. Pescarin

web content management system (WCMS). Unfortunately, there aren't many

examples of them, and a lot still needs to be done. In the panorama of 3d
applications rendering platforms “browser embedded” there is no clear leader of
the market, or even generally accepted standard. This panorama is currently led
by game applications, which have also influenced its development (hardware
and software), driven mainly by the high budget that characterizes these
“productions”. This has implied the development of various hardware and
software platforms and the creation of privately owned and expensive
development systems, often customized to specific devices so as to obtain very
high performance. On the other hand, Internet tends to define a different mode
to access content, an open standardized one (e.g. html, Javascript, Xml, etc.).
Open Source and cross platform projects are therefore fostered (see the Firefox
project). For this reason, joining these two worlds (of 3d interactive and of web
access) isn't easy, and there are still no consolidated approaches or standard
technologies to use as reference (fig. 7.2). One of the projects that will be
described in the following section, Virtual Rome, has proposed a solution to
this problem.
We can subdivide available browser embedded systems into: VRML-X3D
based systems, global earth geo-viewers, flash based systems and general
purpose Scene Graph based systems (CALORI ET ALII 2009).

Figure 7.2 – pipeline_web.tif

As we have seen there are several available applications, some of them based
on well-established technology, other on experimental high-end technologies. If
we are interested in the online publication of a landscape, it is important to start
assessing what is available. In order to evaluate a web project or application we
should take into account communication potential, interaction, transparency,
functionalities and also intuitiveness. The following parameters can be
considered in the evaluation process:
• Web browser integration (applications embedded in the web browser or
stand alone);
• technology used to develop applications (original software; VRML-
X3D based systems; flash based systems and general purpose Scene
Graph based systems; global earth geo-viewers; Java based applications
or game engines based applications);
• Operative Systems and Browsers (Windows/ Explorer; Windows/
Explorer and Firefox; Windows and Linux/Explorer and Firefox;
Windows, Linux and Mac/Explorer, Firefox and Safari; other);
Reconstructing Ancient Landscape S. Pescarin

• graphic card use (OpenGL; DirectX);

• interaction devices (mouse; graphic interface; keyboard; natural
interaction; other);
• content that can be loaded in final application (GIS 2d or 3d data (DTM,
geoimages, etc.); vector data; labels; 3d models; multimedia; metadata;
avatars; other);
• type of loaded terrain (paged or not paged);
• type of available behaviours (fly or walk functionalities; hyperlinks;
reference maps; overlay of various layers; switching of layers, terrains,
etc.; dynamic evolutions, simulations; animations; virtual communities;
artificial life; etc.);
• type of licence (Open Source, commercial, application built on open
libraries and released as freeware; other);
• type of downloading (streaming or complete download).

.B.3 State of the art

This section is dedicated to examples, case studies, projects and web
applications dedicated to the use of Internet for ancient landscape online
publication and interaction. The most room is given to interactive 2d and 3d
projects, whose utility has been extensively discussed in previous chapters.

category type Examples

People Social Networks Facebook
Virtual Communities FIRB
E-learning Quest Atlantis, Akragas doors in Second Life
Data Online repositories Global Land Cover Facility, Open Street Map, etc.
Online Content Management Systems Virtual Rome
Digital Libraries Roman vegetation library; 3d model libraries
webGIS ECAI TimeMap; ADS; Fasti On Line; CART
3D webGIS SkyLine, Nasa WoldWind, Google Earth (Rome
Reborn)
Processes Cooperative Virtual Environments FIRB, Virtual Rome, City Cluster
3d web services Epoch 3d web service
Wiki Media wiki,
Systems Online virtual museums Donald Sanders; Clarke, Flaminia on line virtual
museum
VR webGIS Appia VR webGIS, Salerno archaeological district,
Virtual Rome
Online computer games Rome

Table 7.1
Reconstructing Ancient Landscape S. Pescarin

Examples of online repositories dedicated to geographical or spatial

information, such as the Global Land Cover Facility or the USGS Earth
Explorer, are listed in Appendix B.

WebGIS is quite a stable and well-known technology. Specific applications,

based on webGIS platforms, have been developed for ancient landscape
visualisation or archaeological site location, at different levels of interaction.
All these projects are browser-embedded and their portability depends on the
software or library used. The three cases given below are based on three
different tools.
ECAI, the Electronic Cultural Atlas Initiative, developed the TimeMap™ Java
application. It is a set of tools created by Ian Johnson and Artem Osmakov 95. It
is now an Open Source project that enables the generation of complete
interactive maps with a few lines of html. It's easy for beginners, yet provides
completely customisable power and distributed backend database connectivity
for the expert. It's free for personal use. The peculiarity of TimeMap is the time-
based maps delivery system, that allows you to combine space and time
dimensions, to visualise urban growth, environmental changes, etc. Several
projects have been developed with TimeMap, such as the PHALMS or Ankor
projects, demonstrating the potentiality of multi-temporal study with multiple
datasets.
Another already mentioned example is the Fasti On Line project96. Fasti was
created by LP archaeology with the aim of continuing the “Fasti Archaeologici”
publication, published by the International Association for Classical
Archaeology (AIAC) between 1946 and 1987. It provides a list with the
location of archaeological excavations since 2000. It is a database (MySQL)
driven website, based on a MapServer Open Source library. PHP and
Mapserver, on the server, connect to these data to produce the maps and the
results pages97 (fig. 7.3).
Fig. 7.3 fasti.tif
CART is an example of webGIS and database developed and used by public
administrations in order to archive, manage and also produce “risk-maps” for
urban planning purposes. CART is a geodatabase built in 2000 and based on a
(privately owned//commercial application (Highways) and on open standards
(Xml, Html). It enables a dynamic bi-directional connection with a desktop GIS
(ESRI ArcGIS) that is used by public employees at the local level. Data are
95 ECAI: http://ecai.org; TimeMap: http://www.timemap.net; the software is available at:
http://sourceforge.net/projects/timemap.
96 http://www.fastionline.org
97 For technical references: http://www.fastionline.org/about.php?view=techsummary and on
MapServer: http://mapserver.org/
Reconstructing Ancient Landscape S. Pescarin

archived online, through an Internet Explorer browser interface, as coordinates

x,y,z (points, lines, polygons are described as sequences of coordinates). All
this information can be exported as ascii open file98.
A similar and more advanced webGIS system was created by the National
Monument Record of English Heritage: the Archaeology Data Service. After
the opening of the OASIS project (Online AccesS to the Index of
archaeological investigationS), all archaeological data have been archived at
national level in a unique geodatabase. The access to this database is possible at
various levels. Institutions and professionals have user passwords, and they can
query the entire archive or edit it to add new information on excavations,
prospectings, etc. There is also a public and open access, through the
ArchSearch interface99, where the database can be queried in different ways and
basic information is provided, together with the location of the archaeological
sites visible on Google Maps. This archive is also connected to other online
repositories, such as the old-maps archive100.
There are some interesting developments of the webGIS application, aiming to
make it a more interactive and editable environment.
Connected to MapServer library there are two interesting projects: KaMap and
Open Layer. KaMap is an AJAX application that offers an intuitive web
interface. It aims at providing a javascript API for developing highly interactive
web-mapping interfaces using features such as interaction, continuous panning
without reloading the page, keyboard navigation, optional layer control from
the client’s side, etc. These functionalities make this project really interesting
for archaeological and environmental dataset visualisation and interaction101.
Open Layer is dedicated to web dynamic mapping. It can display map tiles and
markers loaded from any source. MetaCarta developed the initial version of
OpenLayers and gave it to the public to further the use of geographical
information of all kinds. It is free, Open Source JavaScript, released under a
BSD-style License, and is part of the Open Source Geospatial Foundation
(OSGeo)102. It implements methods for geographical data access, such as the
Web Mapping Service (WMS) and Web Feature Service (WFS) protocols103.
Also the Italian Military Geographical Institute has developed its online

98 GUERMANDI, M.P. 2000. Rischio archeologico: se lo conosci lo eviti (IBC, Documenti/31). In

Atti del Convegno di studi su cartografia archeologica e tutela del territorio, Ferrara 2000;
D'ANDREA, A. GUERMANDI, M.P. 2008. Strumenti per L’Archeologia Preventiva. Esperienze,
Normative, Tecnologie, Archeolingua, Budapest 2008.
99 http://ads.ahds.ac.uk/catalogue/
100http://www.old-maps.co.uk
101Examples are available at: http://www.ominiverdi.org/index.php/kamap
102http://www.openlayers.org/; http://www.osgeo.org/
103WMS: http://www.opengeospatial.org/standards/wms;WFS:
http://www.opengeospatial.org/standards/wfs
Reconstructing Ancient Landscape S. Pescarin

repository based on Open Layer104.

A specific multidisciplinary project on landscape archaeology has developed
the ArcheoServer PO-BASyN, which intends to define a base for the
experimentation of innovative solutions to exploit new conceptual tools,
models, techniques and technologies to assist in the management of cultural
resources. The choice of the context to experiment with this research approach
is the study of the settlement dynamics of the Bronze Age in the Po river valley
(northern Italy). It has experimented two online spatial analysis tools:
settlements’ spatial density tool (visualisation of the density in different Bronze
Age phases) and sites’ territorial tessellation (it calculates and represents a
Thiessen polygon analysis on the sites: fig. 7.4) 105.

Figure 7.4 - Po_Basyn2.tif

Just as GIS software 3d features began to be more widely used, also with
webGIS new applications and platforms were developed so as to implement the
3d visualisation of geographical data. There are several categories of 3d
applications, divided into various categories. There are two important
classifications: the first relates to geo-based or non geo-based spatial
applications, the second refers to available behaviour and to the level of
interactivity, also in terms of editing functionalities. Although these are not
always easy to distinguish, we can recognize: 3d webGIS, VR webGIS and VR
applications, the latter including online games, virtual e-learning communities
and collaborative environments (CVE). 3d webGIS enables the interactive
exploration of territories and of geospatial datasets, with limited behaviour and
editing functionalities, while VR webGIS includes a higher level of interaction,
typical of Virtual Reality systems, such as fly through and walk through
navigation functionalities or interactive querying. They are both strongly based
on geospatial datasets: geoimages, digital elevation models and vector layers.
VR webGIS may include virtual ecosystems and 3d models. Global earth geo-
viewers or Scene Graph based systems belong to these categories.
Although Skyline Software Systems Inc. developed one of the first earth geo-
viewers, complete with authoring tool and server support106, Google Earth, in
its stand-alone version, is still the reference point for this sector107. It has been
created and distributed since 2005. It is a stand-alone application which runs on
Windows, Linux and Mac, based both on OpenGL and DirectX. It provides
users with an immediate interface with various functionalities such as search

104http://www.igmi.org/ware/
105http://www.archeoserver.it/pobasyn/webgis/index.php
106http://www.skylinesoft.com
107http://earth.google.it/
Reconstructing Ancient Landscape S. Pescarin

panel, overview map, hide/show sidebar, placemarks, polygon or path inclusion

on the scene, images in overlay, environments, navigation tools, etc.
An example of an archaeological project published using Google Earth is the
reconstruction of 4th century Rome based on the Constantine plaster preserved
at the Roman Civilization Museum in Rome (Fig. 7.5). It is the continuation of
the Rome Reborn project, directed by Bernard Frischer (GUIDI ET ALII 2005)108.
figure 7.5 romereborn_google01.tif
Google Earth has also published a plug-in that has a JavaScript API which
embeds the 3d Google Earth into the web pages. Using the API it is possible to
draw markers and lines, drape images over the terrain, add 3D models, or load
KML files109. An example is Florence on Earth project, where it is possible to
access one century of archaeological excavation in Florence regarding four
periods: Roman, Late Roman, Early Medieval and Medieval. It is possible to
explore also 3D reconstructions loaded on the 3d map110.
There is another recent GoogleEarth project dedicated to an open source
version of the plug-in, O3D, for creating rich, interactive 3D applications in the
browser. This API is shared by a broader developer community interested in
establishing an open web standard for 3D graphics. Google Earth enables urban
building and architectural coverage, also supporting Collada111.

Another global earth geo-viewer is Nasa Worldwind. It was developed in 2004

as an Open Source project by NASA, and is especially interesting as it permits
the layered visualisation of scientific datasets. It is a stand-alone application,
available as a full install under the terms of the NASA Open Source Agreement.
It runs on Windows on PCs with a 3D Graphics Card with a fast Internet
connection. It requires the .NET runtime environment and DirectX installed. It
comes with some basic datasets (from NASA and TerraServer-USA of
Microsoft Research), which are integrated and further downloaded after a
certain point, during the navigation. Through Nasa Worldwind it is possible to
explore datasets such as: Blue Marble, Landsat imagery, SRTM, NASA SVS
dedicated to hurricane dynamics, MODIS (Moderate Resolution Imaging
Spectro-radiometer) catalogue of fires, floods, dust, smoke, storms and volcanic
activity, GLOBE with earth temperature dataset, country boundaries and place
names, and a visual tool for latitude and longitude (fig. 7.6). The latest version
of WorldWind is Java based with better cross platform support112.
108http://www.romereborn.virginia.edu/; http://earth.google.it/rome/
109 http://code.google.com/apis/earth/
110http://florenceonearth.com/
111 http://code.google.com/intl/it-IT/apis/o3d/
112http://worldwind.arc.nasa.gov/features.html
Reconstructing Ancient Landscape S. Pescarin

In 2007 Microsoft also created its own 3d earth-viewer: Virtual Earth 3D. It is a
stand-alone application that runs only on Windows and mainly on Internet
Explorer, and is based on DirectX113. The Virtual Earth platform consists of two
interfaces: Virtual Earth Map Control that lets users make requests via
JavaScript to an AJAX map object and a MapPoint Web Service API using
SOAP XML to communicate with other applications. Recently there is a new
version of VE embeddable in browsers with sound Javascript SDK. Both
NASA WorldWind and Microsoft VE unfortunately do not allow user 3d model
insertion.

Figure 7.6 Nasaworldwind.tif

Virtual Reality applications are mainly not based on geographic datasets,

although in some cases they start from geospecific data. Their characteristic is
the complexity of available behaviour that extends the simple exploration of a
3d geographical viewer to a real cognitive experience, to the creation of
collaborative spaces, and socially dynamic environments. They can be based on
various technologies, although for those embedded in web browsers they are
commonly VRML-X3D or Flash based, or even game-engine based systems. In
some cases there are some experimental projects, scene graph based, that try to
combine 3d webGIS (geospatial dataset in input, large 3d territories processing,
etc.), with VR characteristics (interactivity, inclusivity, simulation, exchange,
sharing, dynamics). I've called these applications VR webGIS.
Several projects, especially in the past, were developed on VRML, whose
current improved version is X3D. NuME, the New Electonic Museum of the
city of Bologna, was initially created with VRML and Javascript, although it
moved to Performer and then to OpenSceneGraph libraries 114 (GUIDAZZOLI
BONFIGLI 1999). Other interesting projects using the same technology, relative to
archaeological sites’ virtual exploration and interaction, have been created by
Learning Sites Inc., as in the case of the reconstruction of the prehistoric
settlements at ancient Nemea (fig. 7.7). Unfortunately X3D doesn't appear to
be adopted sufficiently widely nor considered as a standard, lacking
performance in world database paging (CALORI ET ALII 2009).

Figure 7.7 - java_based_sanders.jpg

The use of a Flash-Shockwave based application has the undoubted advantage

113http://www.microsoft.com/virtualearth/
114http://3d.cineca.it/3d/Nume/nume_3d.php
Reconstructing Ancient Landscape S. Pescarin

of widespread availability and good browser integration. Examples of

geodatabas are the Italian Pagine Gialle Visual and the GeoMind products
(3DIMap technology)115.
A different case, based on Skyline technology, is Bologna 3d, a three-
dimensional interactive version of the city developed by Bologna City Council
itself. To explore the city it is necessary to download the plug-in, Terra-
Explorer, a stand-alone application that runs on Windows, or can also be used
as an integrated ActiveX component. It features automatic download and
installation for web users, efficient streaming for terrain and overlays over the
net an auto-pilot feature allowing pre-defined routes to be replayed
automatically, easy navigation and "fly-to-location" using the information
panel, a 2d map orientation synchronized with the 3d terrain, control speed,
altitude and viewing angle via the mouse, keyboard, joystick, control panel,
view of objects, symbols or geo-referenced information layers created within
TerraExplorer Pro or TerraBuilder, web content integration, tools to measure
distances and areas in the 3D world, and a saver of screenshots116.
The Sardinia region has created a 3d stand-alone version of the entire island
fully explorable in 3d, thanks to CRS4 research Institute in cooperation with
CORE. This project is part of the Regional GIS117. The software is based on the
BDAM technology, efficient for large terrain visualisation and developed by
CRS4, in cooperation with CNR ISTI. Sardegna 3d runs on Windows, Linux
and Mac, on 3d graphic cards, and it enables mouse or interface command
panel interaction. It allows to calculate distances and paths, to locate a site
through a layer-menu and to place a marker to remember a specific location, as
well as to access the archive of aerial photographs taken over the years.
Recently CRS4 has released as open source RATMAN, a framework for terrain
streaming and rendering, a real-time terrain rendering framework able to
asynchronously access terrain information from remote servers. It contains a
simplified version of our BDAM technology118.
Another example of VR viewer has been developed by the Scuola Sant'Anna of
Pisa: the XVR player. Based on this technology is the project dedicated to
Piazza dei Miracoli in Pisa (fig. 7.8). It is a browser embedded application
where users can walk in time and space in the city. It is an ActiveX plug-in
(only on Windows and Explorer), based on OpenGL. Models are completely
downloaded on the client PC before starting the VR session. XVR imports 3d
models from 3DSMax, including animations, and it has script language, vertex
and pixel shaders support, real-time physics using the Tokamak(TM) physics

115http://www.visual.paginegialle.it/3d; http://www.geomind.it
116http://www.skylinesoft.com
117http://www.sardegna3d.it/
118The source code is available as open source from http://ratman.sourceforge.net
Reconstructing Ancient Landscape S. Pescarin

engine, HTML pages interaction using Javascript or VBScript, video textures

and audio (also 3d audio) support 119.

Figure 7.8 - xvr.tif

Some projects have been built on privately commercial or open source game
engines120 or on development and deployment platforms, mainly proprietary,
dedicated to interactive 3D content and game creation121.
Exhibits Player, based on VirTools and Direct X, has been developed to support
3d online exploration of sites and monuments and to access multimedia
contents, improving the communication value. It is a stand-alone software that
runs on Windows, downloading all data at the beginning of the session, an
operation taking some time, but avoiding any long wait after having data
downloaded. Many examples can be seen in the web pages of the Italian
Ministry of Cultural Heritage (MIBAC), Cultural Internet (Internet Culturale),
such as Bologna 3d122. Another example has been built for the Virtual Museum
of Ancient Via Flaminia (Appendix A), where the individual user version of the
VR museum, dedicated specifically to the Villa of Livia, is available online123
(fig.7.9).

Figure 7.9 – giardino_rep2.tif

Applications built on game engines are the TiE project, an EU funded project
whose goal is to create an environment where you can have challenging and
compelling game experiences by interacting with virtual representations of the
European heritage. Specifically in the reconstruction of the scenario of the
Mação archaeological site, that is part of the project, the Instituto Politécnico de
Tomar in collaboration with the Museo Municipal de Mação used an integrated
approach, including GIS data in a Game Engine, with the aim of developing a
credible and educationally valid game adventure contextualized in the settings
of an archaeological site124.
A recent low-cost engine, with great potential for landscape online
visualisation, is Unity 3d. It is an open-ended 3D game/interactive software
engine for web, Windows, Mac OS X, the iPhone, and Nintendo Wii. It
includes a multi platform game development tool, designed from the start to

119http://piazza.opapisa.it/
120Most known game engines are: Doom, Kaneva, Torque, Unreal: proprietary; Delta 3D,
Ogre, Panda 3D: open source.
121Such as VirTools, Unity 3d, etc.
122http://www.internetculturale.it/genera.jsp?s=12&l=en
123http://www.vhlab.itabc.cnr.it/flaminia/VR_DEMO/Release_Ver_1_en.rar
124http://www.tieproject.eu
Reconstructing Ancient Landscape S. Pescarin

ease creation. A pilot project was started by CNR ITABC and SoftLogic with
the aim of building Cultural Heritage VR applications, specifically designed for
IPhone. An example of this approach is the Villa of Livia interactive application
for mobile phone, available on line125.
There are also other case studies not specifically developed for or by the
archaeological or environmental fields, but that are currently exploring new
possibilities of using the network to exchange experience and information. This
is the case of the City Cluster project, an artistic virtual installation created by
Franz Fishnaller
. City Cluster is a virtual reality networking matrix, in which multiple
environments or cities (Chicago and Florence have been the test cases), can be
hosted, coexist and be connected through a common virtual territory, an high-
speed network and a graphical interface, enabling distant users to collaborate in
a shared environments (fig. 7.10).

Figure 7.10 -citycluster.tif

Projects specifically built for communication, dissemination, and e-learning
purposes are based on platforms dedicated to virtual communities or online
games, such as Second Life or Active Worlds. The reconstruction in Second
Life of the Akragas doors in Sicily has become a way to acquire knowledge of
Greek military tactics (fig. 5.7), while the “island” where the villa of Livia
stands is used for teaching about Roman architecture and cybernetics at the
University of California – Merced. The natural and archaeological landscape of
Mesa Verde, built on Active Worlds platform, is one of the learning activities
schools all over the world can do within the Quest Atlantis project (fig. 7.11).
figure 7.11 questatlantis.tif
Advanced applications which integrate online 3d exploration, multi-user
functionalities like those required by virtual communities, VR behaviour and 2d
and 3d editing functions, are Collaborative Virtual Environments (CVEs). They
are, as already mentioned, new and promising tools, particularly for landscape
reconstruction purposes. An experimental project was presented in 2009 by an
Italian team (University of Pisa, Scuola Sant'Anna and CNR ITABC):
“Integrated Technologies of robotics and virtual environment in archaeology”.
The project, described in Appendix A, concerns the development of a multiuser
domain on the web for a multidisciplinary scientific community of
archaeologists, historians, experts in the human and social sciences,
topographers and communication experts, and it was founded by the Italian
ministry of University and Research. The 3d environment, built on VirTools
125http://unity3d.com/unity/
Reconstructing Ancient Landscape S. Pescarin

platform, can be accessed by scientists in charge of specific tasks regarding the

reconstruction of an archaeological site (figg. A4: 1-4), or by teachers
interested in preparing a lesson in the “Virtuoteca”. They are projected in
cyberspace where they can be discussed using chat and start editing sessions, to
add, modify, or eliminate digital content from the scene.

Other solutions are based on the use of a scene graph, a middleware library
with a level of abstraction between a base graphics library, like OpenGL or
Direct3D, and a complete application as a game engine. It can be a good
solution when multithread capability is needed to enable efficient data paging
and when there is no particular need for advanced interactions, such as physics.
VR webGIS such as the Appia project, Salerno Archaeological District, Cannes
3d or Virtual Rome are all based on a scene graph solution. There are several
problems arising from the publication over the Internet of large 3d geospatial
terrains, interactively explorable through complex behaviour: the dimension of
the dataset (often GB of geoimages or DEMs); the coordinated systems
management; the integration of high resolution 3d models, vegetation and
vector thematic layers; the inclusion of multimedia content; the development of
efficient and scalable navigation tools, useful fly or walk through
functionalities; the creation of the 3d scene through editing tools, and the co-
existence and harmonization of a research space with a communication space.
The development of a specific plug-in, OSG4WEB, by an Italian team
(CINECA and CNR ITABC), tried to offer solutions to these problems. It is
based on the Open Source library OpenSceneGraph, and on previous work
carried out for the Appia project, the Esaro and Salerno Districts (CALORI FORTE
PESCARIN 2005, PESCARIN CALORI 2005, PESCARIN ET ALII 2005). Features of this
library (paging quality, cross platform, Internet publishing tool, large input
formats, active development community, etc.) oriented the choice. It has also
been extended with the osgEarth126 plugin, which is able to page in directly
from OGC repositories as well as Google maps and Nasa servers (CALORI ET ALII
2009).
The Appia archaeological park project, directed by CNR ITABC, besides the
Virtual Narrative Museum described in Appendix A, also created a VR webGIS
for the team and the funding institution, the Superintendency, to offer a view of
the state of the art of the survey work (years 2003-2006) (FORTE PESCARIN
PIETRONI 2005, DELL'UNTO ET ALII 2006). The original idea was to publish in real
time and over the web all the digital dataset, vector layers, DTM and 3d models
acquired in the fieldwork, month after month, enabling the team to visualize
126 OsgEarth builds whole-earth or localized OSG terrain models at run-time: http://wush.net/
trac/osgearth
Reconstructing Ancient Landscape S. Pescarin

them interactively (with free and open access), and to download the raw data
(restricted access). The goal of the project in fact was to acquire, with different
techniques and technologies (DGPS, Total Station, Scanner Laser,
Photogrammetry), several sites and monuments of the archaeological park. The
acquired data were processed with open source software, QGis and GRASS, to
produce Digital Elevation Models, geoimagery and thematic layers. They were
then processed with OSG, obtaining the 3d terrain model, and paged with
various levels of detail, published on line. The landscape, although quite
extensive, could be easily accessed online, maintaining the geographical
information, and explored in 3d, thanks to the Net plug-in of the OSG library.
3d models and vector layers were then added to the landscape using VTP
Enviro127, which was appositely modified so as to use the OSG paged terrain
already created, and to export in the OSG/IVE format other 3d needed content
(fig. 7.12). Thanks to Enviro, which already had a ready-to-use interface and
several functionalities, archaeologists and architects have found a simple way to
dynamically interact with the landscape, adding 3d models (3ds, obj, osg, flt,
etc. file formats) or vegetation taken from a digital library (shp, xml, png),
working at a continuously open and updated reconstruction. All the added or
modified objects could then be exported in .osg or .ive formates and added to
the 3d terrain for online exploration. The first plug-in built by CINECA, an
ActiveX component for Internet Explorer, needed a further integration of PHP
and JavaScript to add some more behaviour (switch, upload, view points). This
plug-in, OSG4WEB, was further developed during two other projects, Esaro
Cultural District and Salerno Archaeological District128, which coincided with
the development of the Mozilla Firefox version of the plug-in (PESCARIN ET ALII
2005, PESCARIN ET ALII 2005b).
Figure 7.12 fig_C_web01.jpg

Based on the same technology and on the same OpenSceneGraph library, there
is another project, Cannes 3D, on the exploration of the city of Cannes, in
France. It is a browser embedded application where users can interact with the
3d scene with the keyboard, the mouse, or simply by using the graphic
interface. It runs only on Windows and on Explorer and Firefox, with an
OpenGL graphic card and a good Internet connection (ADSL)129.

127Virtual Terrain Project is an open source project available at: http://www.vterrain.org

128http://www.vhlab.itabc.cnr.it/esaro/; http://www.vhlab.itabc.cnr.it/salerno/
129http://3d.cannes.fr/
Reconstructing Ancient Landscape S. Pescarin

The most updated version of the OSG4WEB software was released for the
Virtual Rome project. This project, described in Appendix A, aims to study,
interpret and reconstruct the archaeological landscape of ancient Rome, and
enable distributed and interactive visualisation through a web-based VR
application, based on Open Source libraries, on Remote Sensing and GIS data,
and on 3d models (VR webGIS: fig. 7.13 left). It includes a front-end
interactive interface, browser embedded, and a back-end 3d CMS online lab
which enable professionals of various disciplines to update and modify a
variety of parts that include archaeological and ancient landscapes (VR
webLAB: fig. 7.13 right). Virtual Rome enables visitors to explore the
landscape at different scales and resolution, thanks to the OSG4WEB plug-in
and its functionalities, such as paged geospatial dataset support, coordination
and projection handling (both in input and output), large 3d terrain dataset
management, 3d models integration (modelled with software such as 3D Studio
Max, etc); natural elements, such as vegetation integration, vector layers
integration; on-line 3d data publication and interaction, possibly embedded into
a web browser; fly and walk navigation tools; behaviour integration: terrain,
model switching, vector information and 3d model loading, selecting and
loading external pages or multimedia content; environment integration (CALORI
ET ALII 2008; CALORI ET ALII 2009). The procedure followed, described in figure
7.16, starts with data processing in a GIS (GRASS). Here vector layers
regarding natural aspects of the landscape are prepared, as described in chapters
3 and 4, and used as input in the ecosystem and terrain generators (Visual
Nature Studio and OSGdem). New geoimages, representing the archaeological
or potential landscape, and digital elevation models are built (river beds, roads
or other modifications are calculated) and exported in GIS formats (geoTiff,
Ascii grid). The vegetation is also exported as a GIS vector file (shape) and as a
XML/database file containing coordinates of each plant, the name of the
species, and the texture associated with it. 3D models are prepared apart, as
multimedia information, but a shape file is also created with the indication of
the position of contents in the territory. All these data are finally joined in the
back-end, where the graphic appearance of the site (the front-end) is
established.
Figure 7.13 - virtualrome_cms.jpg

.B.4 From Mapscape to Webscape

This section provides a practical example of territorial dataset processing and
3d landscape online publication, using open source tools.
The reconstruction of ancient landscape is the last step in a complex process
which involves the acquisition, analysis, and interpretation of the environment,
Reconstructing Ancient Landscape S. Pescarin

of its natural, anthropic and perception features, and of the dynamics and
relations which intersect it. This process is developed through the
reconstruction of the archaeological landscape, its ecological significance,
anthropic modifications, and the ancient potential landscape.
The communication of this final result and the creation of collaborative
environments where this process can be approached in a real multi and inter-
disciplinary mode will require, more and more in the future, the use of the Net.
In this section I will try to give a practical indication of the steps, data, and
techniques required to build a simple example of VRwebGIS. The use of open
source, free ware or low cost software will be taken as given, to enable every
interested reader to start “having their finger in the pie”. This section is written
as a tutorial for the creation of a simple three-dimensional terrain ready to be
published on line through the osg4web plug-in. Let's start from the beginning.
What do we need? We need a variety of software and the basic knowledge of
various disciplines:
• A GIS, to create vector layers, georeference geoimages, or re-project
them from one coordinated system to another, to build digital elevation
models of the terrain;
• a 3d modelling software: to include also 3d models on the landscape;
• an Image Processing software to eventually modify raster data;
• the OpenSceneGraph library compiled;
• an html editor to modify web pages and Mozilla Firefox as Internet
browser;
• the osg4web plug-in installed.
It is clear that the basic knowledge required is quite extensive: cartography and
topography to understand scale, projections, detail or accuracy issues; image
processing to edit and modify digital images; GIS to handle geospatial datasets;
photogrammetry, 3d modelling and computer-graphics to acquire, build or
modify 3d models, handling geometric and texture optimization and terrain
modelling.
In order to simplify the work, you can download a complete ready-to-use
packet (OSG_PACK130), created by CINECA, containing OpenSceneGraph
already compiled, and other useful tools such as OpenEV, OSGdem, Gdal_info,
Gdal_translate, Gdal_warp, and osg4web. To use this packet, it is necessary to
unzip the file in a directory and then to double click on the file _osgshell.bat.
This operation will open a DOS shell that will be used to lunch all the
commands of the programs, such as Gdalinfo, Gdal_translate, Gdal_warp,
osgdem, and osgviewer
We should then define the goal of the project, and analyse available datasets, in
130The packet, created by CINECA Visit Lab and L. Calori, is available at:
http://www.vhlab.itabc.cnr.it/openVRwebGIS/software/osg_pack.zip
Reconstructing Ancient Landscape S. Pescarin

order to eventually identify any further data required.

To build a VRwebGIS we would need, at least:
• one Digital Terrain Model (DTM);
• one georeferenced image (geoimage): aerial or satellite images,
historical maps or other digital georeferenced images.
Complementary and accessory data are:
• Thematic vector layers (shape format) describing natural or anthropic
characteristics of the landscape, such as rivers, road, etc.;
• 3d models (*.3ds, *.obj, *.osg, *.ive), of which we need to know the
geographical location;
• multimedia data (html documents, video, images, etc.).
As fully described in the book, it is quite helpful to define a digital protocol in
which all steps are indicated, together with required input and output. Table 7.2
summarizes the activities required by the creation of a basic VRwebGIS.

Activities Software

1. Spatial dataset creation: DTM, geoimages, vector layers GIS Software (GRASS, ESRI ARCGIS, etc.)

2. Pre-processing of spatial dataset GIS Software; Gdal library (Gdalinfo, Gdal_translate)

3. 3D models creation and export in OSG/IVE format Plug-in OSG for 3D Studio Max (OSGexp)

4. 3D terrain generation OSGdem, OSGviewer

5. Web Data integration (terrain, thematic layers, 3d PHP, JavaScript, osg4web, web-design tools
models)

6. Exploration of the landscape online Internet Explorer/Mozilla Firefox, osg4web plug-in

Table 7.2

Spatial dataset creation

To build a 3d terrain we need to describe its shape, in terms of clivometry, and
its characteristics. We therefore need a DTM, raster georeferenced images, and
eventually also thematic layers which describe the landscape. All this data can
be processed with a GIS software. In case of operations such as geo-
referencing, band combination or colour calibration, a remote sensing software
is preferable131. We should take great care over the projection of the original
dataset, eventually re-projecting in the desired output coordinate system. To
start working with a sample dataset, it is possible to download geoimages from
free online spatial repositories, such as the Global Land Cover Facility, through

131Some GIS products, such as GRASS, cover various spatial operations, including remote
sensing analysis and georeferencing.
Reconstructing Ancient Landscape S. Pescarin

the Earth Science Data Interface (ESDI)132, where a wide selection of Landsat
imagery is available. In these repositories files containing different bands
(geotiff format) can be downloaded. A remote sensing software, such as
GRASS or ErMapper or even OpenEV133, can be used to perform data fusion,
combine bands 3, 2 and 1, assigning respectively a Red, Green and Blue value,
as explained in chapter 3 (fig. 7.14).

Figure 7.14 – openev.tif

We should now obtain the DTM. We might already have elevation data (points,
contour lines, grid, tin) or need to interpolate them in a GIS. They can also be
downloaded from a repository, although at a low resolution (90 mt), such as
from the above mentioned ESDI or directly from SRTM134. In this example a
raster DTM is required (ascii grid or geotiff format). The DTM can be used to
carry on further spatial analysis such as: slope, aspect, buffer, viewshed,
Thiessen polygons, cost analysis to identify resource accessibility, site
catchment or site exploitation territory analysis, and cluster analysis (Appendix
B). Vector thematic layers might be acquired from repositories or libraries, or
on the field, or they can be digitized from raster maps in various ways. What we
would need are shape135 files both in input and in output. Vectors can be used in
an ecosystem generator, such as Visual Nature Studio, where they can
contribute to creating new geoimages representing the visual aspect of the
landscape (potential, archaeological, or interpreted).

Pre-processing of spatial dataset

We can now verify whether the files are correct and then export them in a
format compatible to the Gdal library136. Gdalinfo and gdal_translate tools can
be used for this purpose. Gdalinfo gives information on the raster files, and
fundamental for the use of OSGdem are: Coordinate System, Origin, Corner
Coordinates: upper right and lower left (fig. 7.15). The syntax is:
gdalinfo [options] [file path]137
Gdal_translate performs raster files transformations. The syntax is:
132http://glcf.umiacs.umd.edu; http://glcfapp.umiacs.umd.edu:8080/esdi/index.jsp
133http://openev.sourceforge.net
134See also Appendix B for a list of geospatial repositories; http://srtm.csi.cgiar.org
135Shape is currently considered a standard in GIS.
136I recommend the use of the geotiff, *.tiff (with *.tfw georeferencing file) or *.ecw compress
file for geoimages and Ascii grid or Geotiff for DTM.
137The commands must be launched from the shell. A complete list of Gdalinfo options is
available at: http://www.gdal.org/gdalinfo.html
Reconstructing Ancient Landscape S. Pescarin

gdal_translate [options] [input file] [output file]138

figure 7.15 gdal.tif

3D models creation and export

3d models, natural elements (vegetation), visual quality effects (lighting,
shadows etc) are usually part of a reconstruction. The integration of these
elements depends also upon the flexibility of the system: if there are ways to
import models generated with off the shelf modellers, if the paging or streaming
system can handle 3d models of interior scenes; if vegetation rendering is
supported, if there are state of the art rendering techniques such as shadows and
other visual effects supported. A rough qualitative measure is obtained by
comparing the visual quality of the system on modelled input with that attained
by commercial game rendering engines (CALORI ET ALII 2009). 300-500 Kb per
file (model or part of a model) could be considered an average limit for a
acceptable download. For this reason it is recommended, for instance, to
subdivide the model into different sub-objects.
3d models that will be placed on the landscape and published online should
have certain characteristics to be interactively managed. They should be small
files characterized by many different levels of detail of textures and geometry;
they should have the centre of the model in geographic coordinate x,y,z
(according to the projection chosen by the project) or, alternatively, have the
centre corresponding to 0,0,0 coordinates.
Models are generally built with software such as 3d Studio Max, Blender,
Maya. etc. The final output should be in the format required by the VRwebGIS
platform. In our case *.osg or *.ive that can be obtained using a plug-in for the
3d modelling software139.

Terrain generation
We now have all the necessary data to proceed in the creation of the 3d terrain
(Fig.3.7 b). In order to produce a 3d territorial model useful for online
interaction, the model should have a hierarchical structure, different Levels of
Detail (LOD), various tiles, and be paged. Paging is the system which enables
you to access a multi gigabyte hierarchical dataset of geospatial data (images,
terrain, or features) in real time, adapting the image quality produced to the
available bandwidth. It involves image and terrain compression multithreading,
caching etc. Ideally it provides a player component, as well as a server able to

138A list of Gdal_translation options is available at: http://www.gdal.org/gdal_translate.html

139OSGexp, the plug-in for Max are available at: http://osgmaxexp.wiki.sourceforge.net/; for
Maya: http://www.cs.ucla.edu/~sunghee/osg2maya/user_guide.htm
Reconstructing Ancient Landscape S. Pescarin

build a hierarchical database out of standard GIS georeferenced data. The

performance of such a system can be qualitative, compared by looking at how
fast the perceptual quality increases when a completely new viewpoint is
selected, given a fixed amount of bandwidth and hardware resources. Another
point is the amount of visual artefacts such as popping, aliasing, or seams that
the multi resolution visualization generates (CALORI ET ALII: 2009).
The terrain is created with OSGdem, launching the program from the shell, as
we have seen before with Gdal tools. OSGdem has several options, but the
basic syntax is:
OSGdem -t [ texture with path] -d [dem file with its path] -o
[output]ad
Example:
OSGdem -t c:\...\mosaico.tif -d c:\...\dem.tif -o terreno.osg
Other important options concern the coordinate system (--cs: indicated also
using the EPSG codes140, the geodetic parameter dataset known by all GIS-
based software internationally); the extent of the area to be generated (-e)141.
A complete example:
OSGdem --cs EPSG:23033 --tile-terrain-size 128 --tile-image-
size 128 -v 2 -e 444482 4167928 375200 299600 -d dem.grd -l
10 --levels 0 10 -t image1.tif --levels 5 10 -t image2.tif -o
terrain.ive
To visualise the result, we can launch the program osgviewer from the shell and
explore it:
osgviewer terrain.ive

Web data integration

This includes all the aspects related to web usability and ease of integration
with a typical web portal and applications. The comparison would again be
qualitative and in some ways subjective. To name some criteria: browsers,
operating systems and graphics platform supported, browser integrated or
stand-alone application, the requirements of a special streaming server.
140Frequently used EPSG codes in Italy are: WGS 84 / UTM zone 32N: <32632> ; WGS 84 /
UTM zone 33N: <32633>; WGS 84 / UTM zone 32S: <32732>; WGS 84 / UTM zone 33S;
ED50 / UTM zone 32N: <23032> ; ED50 / UTM zone 33N: <23033> ; Monte Mario:
<4265>; Monte Mario (Rome): <4806>; Monte Mario/Italy zone 1: <3003>; Monte
Mario/Italy zone 2: <3004>; Monte Mario (RM)/Italy zone 1: <26591>; Monte Mario
(RM)/Italy zone 2: <26592>
141The extent might be written using this formula: -e = X [lower left x coord], Y [lower left y
coord], X [upper right x coord – lower left x coord], Y [upper right y coord – lower lefty
coord]
Reconstructing Ancient Landscape S. Pescarin

If we copy the generated terrain in a directory on a web-server, we can simply

visualize it with osgviewer typing the URL:
osgviewer http://www.landscape.org/terrain.ive
To create an Internet application, with several possible behaviours, we should
use a plug-in for the browser, in the case of a browser embedded project, or a
stand- alone software. In this case we can use the osg4web plug-in for Mozilla
Firefox142. There are several possible interactive levels with the 3d scene: view
points, objects selecting and loading html, or other multimedia content, guided
tours, model switching. The interaction can be programmed in JavaScript-Php,
as in the case of Virtual Rome (Appendix A)143.

142Actually the version for Firefox is fully supported (http://3d.cineca.it/storage/osg4web/

htdocs/download_plugin/OSG4Web_1.0.2.xpi), while the ActiveX for Internet Explorer
isn't currently supported. The source code, licensed under the GPL and LGPL, is available
at: http://3d.cineca.it/storage/prove_repo/osg4web_bazar Tests have also been carried out
by CINECA for Linux: http://3d.cineca.it/storage/bazaar_repo/BrowserEmbed/test
143An example of programming interaction javascript-osg4web is available at:
http://www.vhlab.itabc.cnr.it/openVRwebGIS/tutorial/osg4web.html
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Chapter 8
A step into the future

The ministry of culture has recently decided to start an extensive campaign of

investigations, acquisitions and valorization of the region. The project is
directed at the identification of the areas subject to a higher level of protection,
and therefore to a narrower series of limitations in urban development and to a
different kind and degree of exploitation (parks, etc.). It is a significant
initiative, financed by the government and by leading private businesses and
organisations of the territory itself, interested in its future prospects for tourism.
The ministry has involved universities, public institutions and private
companies in this work, but also the most important museums and a few
schools. Moreover, for the first time, it has created a commission that has
singled out a list of valid scientific work and activities, already produced by
researchers, scholars, and various organisations in both public and private
sectors. 12 months from its beginnings, the commission has published its report
and produced an online available version, where the territory is explorable
interactively in 2d and 3d and the whole project is geo-linked to the territory,
subdivided for typology, chronology, resolution and results. This browser-
embedded geo-report has been left open, in accordance with ministerial
directions, to further updating by the developers themselves. Everyone may
access the geo-report and indicate his own work by sending a mail, or locate it
directly in the interactive map, clicking on the interested point or drawing an
area. A new section has recently been updated, dedicated to the second phase of
the project. As researcher of the European Council of Researches, I could ask
for user password and be part of the team involved in the study of the north-east
area. Although departments and institutes are involved, the ministry has singled
out individual scholars who will be in charge of the work (as reference points
and with responsibility for it). In addition to my lab, committed to 3d
reconstructions and virtual heritage applications development, I'm actually
working with a researcher of the Earth Science department and another of the
Botanic, an officer of the archaeological superintendency and a couple of
people in charge of 3d programming and of the web geo-database. The system
is still being tested, but it seems quite promising. We are now working on the
development of shared libraries (vegetation, archaeological remains, etc.) and in
the mapping of the archaeological landscape. The botanists have finished
identifying the main species in the area, from various periods, and they are
working with geomorphologists and geographers to define the main
Reconstructing Ancient Landscape S. Pescarin

ecosystems. Those who are probably finding the most difficulties are architects
and in general modellers. The latter are in fact required to develop their final
models at various levels of detail, at least in the last stage adopting common
tools requiring an Internet connection, to check model validity (naming
conventions, group definitions, geometry, texture, metadata, etc.). The final
result, produced in several formats (standard and/or open), should certainly
achieve remarkable results, butat the present stage what is visible in the
webscape is just a series of simplified objects that are increasingly being
substituted by more detailed ones. The way the webscape is changing day by
day is impressive. It has never been “empty”: the initial reported projects could
(and can) be visible in this “restricted” section and may be consulted. But after
this first starting point, the cyberspace has started to grow. I can arrive at the
office and see that the site where I was working the day before has changed.
New thematic layers and objects have been added and new signs indicate that
there are studies or fieldwork going on at that moment. In this way, I can verify
a problem I'm facing in a very short time, contacting the colleague directly or
checking the archive. My lab will soon be able to start working also on ancient
landscape (mainly the Iron, Bronze and Roman ages). We will be working on
two different interactive projects. The museum's director has signed an
agreement to include an interactive installation, in the new section he is
creating, and this will ensure that the work will be widely communicated as
well. There will also be a second project more oriented to educational purposes,
we hope in a couple of years.

Something like this has never happened before.

In the first part of this final chapter I tried to give a substantial idea of possible
evolutions of the applications and processes we have been discussing in the
book. Most of the technologies are already available, although not integrated or
not off the shelf.
The use of Internet can really be addressed to various objectives: to diffuse
results or just new approaches, to co-operate with people at the same level
(scientists, tourists, students, etc.) or at different levels, to disseminate cultural
information and the memory of our territories. Is it possible to strengthen this
use, in connection with past landscapes? To diffuse the ecological approach in
archaeology? And most of all to support the collaborative model within
landscape reconstruction? A model adopted in a distant field has worked. It was
proposed by the first programmers on Parallel Programming, adopting the
slogan “try it, play it, modify it”. This could be successfully used in our case,
trying hard to create open systems, tested and used by different professionals,
as happened in the Virtual Rome or the FIRB projects I have described in the
Reconstructing Ancient Landscape S. Pescarin

previous chapter144. In the future most of the services will become Internet
provided, as in the case of arc3D145, but also most of our resources will be
moved from our external hard drives to the web and, finally, even programs
will be used through the Net (GRANELLI SARNO 2007).
Virtual ecosystems might be seen as interactive online environments where it is
possible to experience multiple voices, elements, relations, diversifications, and
where new and different cognitive paths, directed to “decentralization”, might
be proposed (TURKLE 1997: 167).
Models emerging from virtual ecosystems, like those discussed in this book,
might support a more integrated vision of ancient landscape itself. But we also
have to be aware of the risk represented by “opaque” systems, where virtual
reality incorporates emotion and aesthetic. Nevertheless the introduction of
complex systems in VR might also contribute to the harmonization and
integration of emotion with logic, and game with analysis, for a better
understanding and for a more efficient interpretation of our landscape, our
memory.

144A full description of these case studies is available in Appendix A.

145http://www.arc3d.be
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Appendixes
(I. Cerato, A. Palombini, S. Pescarin, E. Pietroni, V. Vassallo)
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Appendix A
Case Studies

.B.1 - 20,000 years of history at Ca' Tron

Location: Ca' Tron (TV), Italy
Time: Upper Palaeolithic, Bronze Age, Roman Times, Middle Ages
Period: Summer

Sources, methodology and techniques

The research, conducted since 2000 by the University of Padua in this area of
the Veneto plain, was addressed to the reconstruction of the "total history”,
from antiquity to the present, of a stretch of land (so far ignored by historical,
archaeological and environmental research), and its valorisation by conveying
the potential, historical landscape of this portion of the lowlands which was at
last recovered from the marshes in the years 1930-40.
The area, of about 1137 hectares, is situated between the Sile river and the
northern lagoon of Venice. It includes the Roncade municipality (TV) and a
small part of Meolo (VE), now part of Ca 'Tron estate farm, so named by a
noble Venetian family in the early seventeenth century. It was once formerly
included in the east Altinum Roman Ager, and the ancient Annia Street crossed
through the southern sector of the estate along the stretch uniting Altinum and
the site of Iulia Concordia.
Interest in this area of the Venetian lowlands involves both its geographical and
historical features: the former is linked to its pivotal position between the large
Brenta and Piave alluvial conoid and its location on the internal lagoon edge;
the second is related to the neighbouring topography of the ancient city of
Altinum (which is only two miles away) and to the road network realized by the
Romans in the territory, partly exploiting the existing connections (in particular
Annia Street).
In recent years the methodology used to study an area has evolved, absorbing
interdisciplinary trends: information from various fields and research areas
allows us to have highly scientific data available for a virtual reconstruction of
the ancient landscape in its totality.
The study of Ca' Tron lends itself to this multidisciplinary logic due to the
complexity of the morphological and human history of the area. Certainly one
of the most advanced features of research into the past is the integration of
historical and archaeological studies with studies of the the environmental, and
the relationships between environment and culture (Environmental
Archaeology), reconstructing human history through the recovery of the
Reconstructing Ancient Landscape S. Pescarin

temporal dimension of the environmental dynamic.

Archaeological research is carried out in close relation with palaeo-
environmental research, making use of the most advanced technologies of
investigation, detection and data processing, with the primary aim of grasping
the relationship between man and his environment.
An interdisciplinary project of this kind, in a landscape context, among the most
unspoiled of Venetian valleys, has stimulated reflection on the issue of the
valorisation of acquired knowledge, both in scientific and communicative
terms. One of the most innovative solutions in this area requires the use of
multimedia: "virtual reality" techniques enable us to increase the amount of
information and the quality of our perception. Precisely because of the wealth
of data from the research at Ca' Tron, it was possible to verify the potential
offered by a virtual reconstruction system.
The aim of the project was the reconstruction (starting from the research
results) of the Ca' Tron archaeological landscape using visual simulations as a
tool for hypotheses testing, and the geo-spatial contextualization in a diachronic
way of the geological, botanical and archaeological information collected
during these years of research. Unable to perform a continuum transformation
of the landscape, four periods of significant changes in geomorphology,
vegetation and human presence were identified:
4. the Last Ice Age Maximum (LGM), about 18,000 years ago,
5. the Bronze Age (second millennium BC) (fig. A1.1),
6. the Roman Age (first century BC.) (fig. A1.2),
7. the Middle Ages (XIII-XVI century BC).
Figure A1.1. Catron_bronze.tif

Figure A1.2 Catron_roman.tif

The work consisted in:
XIV Data Collection: geophysical exploration, probings, sediment
analysis, interpretation of aerial and satellite photos, survey of the land
by airborne laser scanner, palinological and other plant analysis,
excavation samples, surface reconnaissance, detection campaigns.
XVData processing: creation of digital terrain model and its preparation for
the four chronologically reconstructed phases, study of plant species
documented in similar geographical and climatic contexts, study of the
structures, surface and finds, 3D reconstructions.
XVI Creation of virtual system (fig. A1.3): creation of a GIS system of
the "observed landscape", identification of the most significant phases
(LGM, Bronze Age, Roman Age, and Middle Ages); 3D virtual
modelling of the main archaeological evidence, drafting in GIS of the "
interpreted landscape " for the different phases; virtual landscape
Reconstructing Ancient Landscape S. Pescarin

generation and its ecosystems (software VNS), data export into a virtual
interactive GIS (VTP open source software, CNR-CINECA).
Figure A1.3 Catron_VR_archeovirtual08.tif

Acknowledgements
The project is directed by the University of Padova, Italy under the scientific coordination of: F.
Ghedini; Archaeological coordinator: P. Basso, J. Bonetto, M.S. Busana; Geomophology and
Remote Sensing: A. Bondesan, P. Mozzi; Gephysics: E. Finzi, R. Francese; Palinology: A.
Miola, G. Valentini; Excavations: P. Michelini; VR reconstructions: A. Palombini, S. Pescarin,
V. Vassallo, CNR ITABC.

References
http://www.tenutacatron.it/fondazionecassamarca/catron/ciat/index.htm
GHEDINI, F. BONDESAN, A. 2002
La tenuta di Ca' Tron: ambiente e storia nella terra dei dogi. Fondaz. Cassamarca, Cierre 2002
BONDESAN, A. BONETTO, J. BUSANA, M.S. KIRSCHNER, P. MIOLA, A. MOZZI, P.
PESCARIN, S. VILLANI, M.C.
20,000 years of landscape evolution at Ca' Tron (Venice, Italy): palaeoenviroment,
archaeology, VR webGIS. In: Proceedings of “Hidden Landscapes of Mediterranean Europe:
Cultural and methodological biases in pre- and protostoric landscape studies” Collegio S.
Chiara, University of Siena, May 25-27, 2007, in print
[I.C. and V.V.]
Reconstructing Ancient Landscape S. Pescarin

.B.2 Via Flaminia Roman Landscape

Location: North of Rome, Italy;
Time: 21st century AD and 2nd AD;
Period: Summer
Location: North of Rome, Italy;
Time: 21st century AD and 2nd AD;
Period: Summer

The Project
The Virtual Museum of Ancient Via Flaminia, begun in 2005 by the Virtual
Heritage Lab of the ITABC-CNR, aimed to reach a virtual reconstruction of the
Via Flaminia archaeological landscape (about 2nd century A.D.). The goal was
the creation of a Multiuser Virtual Reality museum installation for the Roman
National Museum in Rome. The VR system is a MUD (Multiuser Domain)
planned for 4 users with single displays and a HD stereo display 1024 x 768.
The visitors share interactively the same virtual space, interacting through their
joined actions and creating a virtual performance which is visible to the other
visitors in stereoscopy (FORTE ET ALII 2007).
The reconstruction of the landscape was conceived of with two uses in mind.
On the one hand it was targeted to be used in the Virtual Reality application and
also to build computer-graphics movies, to be included in the VR multi-user
system, developed using a story telling technique. This application is now
installed in a specific room of the museum and it is characterized by four
computers, connected in a network through a local server (Fig. A2.1). On the
other, it was conceived as a basis for a 3D WebGis system, to allow people to
browse the Via Flaminia landscape and reach its information content through
the web.
The two platforms have different features and requirements. The museum VR
system implies a limited time for users, and a cyclic and structured application,
but at the same time it allows a wide-band of data-flow; whereas the web
system allows users to decide their fruition timing and features, but is more
limited in terms of data-flow.
The ancient landscape reconstruction has been carried out for use on both
systems, aiming for at least two levels of detail: the first related to the whole
territory around the Via Flaminia inside the Roman County area (about 20x40
km); the second, more detailed, for the Virtual Reality high-resolution
application, and limited to the areas of archaeological interest along the path.
At the same time, the double-sided work of reconstruction (monuments and
landscape) implied different strategies. For the monuments, a top-down strategy
Reconstructing Ancient Landscape S. Pescarin

was performed, starting from the hi-res reconstruction (obtained from Laser
scanning), and simplifying data in order to reach a web-optimized version. For
the ecological aspects of landscape, on the other hand, a bottom-up process was
carried out: the simple GIS information layers (landscape features, ecosystems
distribution and so on) were a starting point to obtain virtual reconstructions of
natural items (plants, rivers, etc.) in the landscape context, at varying resolution.
In both situations, the reconstruction needed a complete and reliable definition
of the ecological features for the whole territory studied, corresponding to about
135.000 hectares. As a matter of fact, while for the Virtual Reality application –
focused on a few hi-res monumental areas – vegetation was no more than a
frame for architectural context in the limited areas where users moved, theVR
webGis system required knowledge of the whole browsable territory (Fig.
A2.2).
Figure A2.1 MNV_avatar_sala.jpg

Figure A2.2 flaminia_reconstruction.tif

Pre-processing and post-processing

The available data for an historical approach to landscape reconstruction are
unfortunately quite incomplete. We have various resolution digital elevation
models (DEM), but they represent the actual terrain morphology, which slightly
differs from the ancient one, as does the hydrology. We also have nature studies
and literary texts which can give us information on the presence or absence of
species of vegetation, as well as distribution data on archaeological findings,
which may lead to indirect information about the distribution and typology of
human occupation (villas, necropolis, monuments, etc.).
Such data needs a certain amount of preliminary processing before it is suitable
for an historical landscape reconstruction.
Concerning the terrain model, the first step is the collection of reliable data on
the terrain height in antiquity, for the areas where such information is available
(from archaeological excavations, the presence of clearly datable items such as
monuments, roads, etc.). At the same time we delete the morphological items
whose modern origin is certain (banks, earthworks, hill-edge cutting, etc.).
Such specific areas can be used to be connected and/or interpolated with
modern observed DEMs, to get closer to the ancient landscape situation.
This kind of operation, particularly over extensive areas, is naturally not
exhaustive for the features of an ancient landscape, but its reliability is
proportional to the amount of available archaeological documentation and then
of ancient human presence in the territory, to give the most reliable possible
representation of it. When dealing with such a topic, we should recall the
Reconstructing Ancient Landscape S. Pescarin

concept of uncertainty management: narrative or virtual reconstruction do not

allow any “lack” of information. For this reason we are forced to operate
choices among the various possible options, whenever a doubt arises, so that the
fundamental issue is that choices should be transparent and plausible.
Once we have obtained a digital model of the ancient terrain, we have proceed
to the next step, preparing thematic maps of all the ancient features of the
territory: the reconstruction of the ancient hydrography, through the comparison
of historical maps and analysis of aerial photographs (Fig. A2.3), the map of
ancient roads, and the map of archaeological occurrences. The latter was
divided into different themes according to archaeological typology: the
thematic map of the villas was one of the most important parts of the work that
followed, and it is very important to have a reliable scheme of such buildings'
spatial position. To this end, to avoid considering the features of the same villa
as different sites, all the occurrences lying at less than 150 meters from each
other, were considered as individual instances (Fig. A2.4).
Figure A2.3 flaminia_hydrography.tif

Figure A2.4 flaminia_ecosystem.tif

Advanced elaboration
The next step represents the fundamental and original part of the landscape
reconstruction process. Starting from the Digital Terrain Model, a new slope
map was created and then simplified so as to reach a five slope categories map.
Geological and agronomic scientific literature offered a wide range of
documentation about possible cultivation on the terrain slope, so that it was
possible to assign to each slope category a list of the vegetation potentially
present (chapter 4). Such a map was then refined by taking into account a
lithostratigraphic soil map.
It was thus possible to obtain a new map concerning the ancient use of the soils,
including different cultivation areas (vineyards and olives, cereals, and,
particularly, fruits), as well as grazing areas, dense and light woods, bushes and
areas of river shores, normally not cultivated and rich in a dense spontaneous
vegetation (Fig. A2.5). Once all the areas potentially suitable for the various
agricultural uses had been selected, we determined which ones were probably
really cultivated among them. We hypothesized that such areas could be sought
for among the ones closest to living structures and therefore easily accessible.
Study of the ways to determine the most easily accessible areas for food
production purposes in ancient societies is by no means a new subject, having
been carried on since the dawn of spatial analysis in archaeology, well before
the spread of PC use. An important step in such a direction was Site Catchment
Reconstructing Ancient Landscape S. Pescarin

Analysis (Appendix B) (HIGGS VITA-FINZI 1970).

In the Roman context, we have many studies on the possible range of
agricultural activities (GAFFNEY STANCIC 1991). On the basis of previous work
and statistical analysis on specific site density, we chose to adopt 1357 mt ray
areas, around the villas (for a more detailed explanation of the process see
PALOMBINI VASSALLO 2007), inside such areas they are primarily considered the
ones easily reachable in terms of slope and distance (cost analysis).
All these data and other analysis factors (the ancient hydrography, the
reconstruction of possible ancient centuriatio, the correlations of fields to roads
and streams, and so on) can easily be integrated into the frame of GIS
calculation formulas, to obtain thematic maps of the final hypothetical
reconstruction of land use through a common GIS software (we used GRASS
6.0).
The first result of such an elaboration is highly bitty and fragmentary, so that
another step has to be taken to give the map a minimal threshold in shape
regularity and the size of cultivated areas. Once again it has to be stressed that
what we're going to reach is a plausible reconstruction of an historical
landscape, and not the reproduction of a real situation, whatever the word real
may mean in such a context. In fact, the real situation of the landscape, even if
accessible to our knowledge, would be a constantly changing context.
Nevertheless, this process, thanks to its scientific plausibility and procedural
transparency, can be considered an ideal meeting point between the ecological a
cognitive reality of a landscape, reproducing both the landscape’s scientific
model and (through the next step of terrain generation) the landscape’s
emotional impact on the observer.

Terrain generation
The reconstruction process performed so far has, for its output, a series of 2D
maps (raster or vector) containing information on ancient landscape features
and various areas. Specific Terrain generator software is used to shift from this
level to the real 3d landscape reconstruction.
Such software (we used Visual Nature Studio) helps in creating simple or
mixed ecosystems, using starting data as a basis, and generating single plants
according to specific graphic libraries which can be customized through
pictures of vegetables.
All ecological parameters (density, dimensions, the presence of species) can be
customized as well. The next step is the insertion in the landscape of
architectural models of monuments.
The final result (Fig. A2.6) is a real, highly impressive landscape, which can be
rendered at various resolution and complexity levels, according to the required
use (local or web-based).
Reconstructing Ancient Landscape S. Pescarin

Figure A2.5 - flaminia_landuse.tif

Figure A2.6 – flaminia_results.tif

Acknowledgements
Production: CNR, Istituto per le Tecnologie Applicate ai Beni Culturali di Roma. In
collaboration with: Arcus S.p.A., Soprintendenza Archeologica di Roma, Museo Nazionale
Romano, Complesso delle Terme di Diocleziano; Scientific Director: M. Forte; Project of the
multiuser system of virtual reality: M. Forte, E. Pietroni, C. Rufa; Coordination of the historical
and archaeological content: V. Vassallo; Coordination of the project communication: Eva
Pietroni; Coordination of the archaeological landscape reconstruction: Sofia Pescarin; Texts and
scripts: A. Palombini, E. Pietroni, V. Vassallo; Architectural modelling, photomodelling and
texturing: N. dell'Unto, M. Di Ioia, F. Galeazzi, A. Moro, L. Vico, S. Borghini, R. Carlani;
Landscape modelling: S. Pescarin, E. Pietroni; Virtual characters modelling: N. Dell'Unto, M.
Di Ioia; 2D Graphics: A. Moro; Photos: B. Trabassi, L. Vico; Video recording: B Trabassi; 3D
rendering movies: N. Dell'Unto, M. Di Ioia, F. Galeazzi, A. Palombini, S. Pescarin, No Real, I.
Tumino; Video editing: E. Pietroni, B. Trabassi, I. Tumino, No Real; GIS and VR Web GIS
elaboration: C.Camporesi, A.Palombini, S.Pescarin; Original music: G. Mangiocavallo;
Historical research: G. Messineo,V. Vassallo, L. Vico; Laser scanning: M. Di Ioia, F. Galeazzi,
A. Moro, E. Pietroni, B. Trabassi, L. Vico; Territorial reliefs: F. Galeazzi, M. Mascellani, A.
Palombini; Territorial reliefs in the area of Grottarossa: A. Angelini, R. Gabrielli, D. Peloso;
Resources implementation inside the application of virtual reality: E. Pietroni; Programming
and developing of the application of multiuser virtual reality: C. Rufa; Programming and
developing of the web application: C. Camporesi; Funding: Arcus S.p.a.

References
FORTE, M. 2007
La villa di Livia: un percorso di ricerca di archeologia virtuale, L'Erma di Bretschneider,
Roma.
FORTE, M. PESCARIN, S. PIETRONI E. RUFA, C. 2006
Multiuser interaction in an archaeological landscape: the Flaminia project. In Proceedings
of the 2nd Int. Conference on Remote Sensing in archaeology “From Space to Place”,
Rome 4-7 Dec. 2006, BAR Int. Series. 189-196.
FORTE, M. PESCARIN, S. PIETRONI, E. RUFA C. 2007
Multiuser Virtual Reality museum application for Cultural Heritage: the Ancient via
Flaminia project. In: D. Arnold, F. Niccolucci, A. Chalmers (Eds.), Proceedings of the 8th
International Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST
(2007) Project Papers .
VASSALLO, V. PALOMBINI A. 2009
Between Virtual Reality and Spatial archaeology: The Via Flaminia landscape
reconstruction. In CAA2008 proceedings.
web: http://www.vhlab.itabc.cnr.it/flaminia

[A. P.]
Reconstructing Ancient Landscape S. Pescarin

.B.3 Appia Archaeological Landscape

Location: Rome, Italy; Time: 21st; Period: Summer

In 2002, the Institute of Technologies applied to Cultural Heritage has signed a

research agreement with Roman Superintendency with the goal of creating in
four years a spatial bidimiensional and threedimanesional archive of the
monuments of Appia archaeological park, Rome (Italy).

Sources, methodology and techniques

The Appia Antica project (the archaeological park of Roman Via Appia, located
in Rome, Italy) was a good chance to experiment different technologies and
approaches, allowing a continuous comparison with different mapping and VR
techniques. It was developed between 2003 and 2006 by the interdisciplinary
team of CNR ITABC, in collaboration with Rome Archaeological
Superintendence. Maurizio Forte and Paolo Salonia have been the scientific
coordinators, while I was the responsible for 3D modelling, laser scanning,
virtual storytelling and the application The narrative Museum of the
Archaeological Park of Ancient Appia Antica. The work was characterized by
an intense activity of archaeological and architectonic survey with the goal of
acquiring data for a real-time spatial interactive system. All the techniques used
(DGPS, Total Laser Station, 3D Scanner Laser, 3D Stereo Photogrammetry,
Photo- Modelling Techniques) were used in integration.
All the data acquired were post-processed: part of them (topography) were
overlaid in a GIS project, while 3d models (architecture), obtained from
scanner-laser and photo-modelling or reconstructed from archaeological
sources, were geo-located and connected with external multimedia databases.
For our activity it is very important to preserve the link between the fieldwork,
survey activities and VR communication. That's why our efforts have been
oriented to experiment a common protocol to manage data, without loosing
information during the successive phases of elaboration and communication.
The software used: Arcview, Terravista (GIS elaboration), Photomodeler
(photomodeling), Cyclone and RapidForm (laser scanner data elaboration) 3D
Studio Max, Photoshop (optimization and design of the landscape), Virtools-
DEV (real time engine), OpenScenegraph for the 3D webGIS .

The archaeological landscape as a digital ecosystems

The priority of the archaeological information consists in the contextualization
of data and of cultural itineraries, in the inter-relation between human life and
Reconstructing Ancient Landscape S. Pescarin

sites, landscape and archaeological objects, the life of the past and the life of the
present. The more information enhances difference and connectivity, the more
the symbolic associations grow up and also the possibility for visitors to
assimilate and elaborate cultural contents. We need to introduce “life” in the
environment, activities, characters, behaviours, narrative contents to allow to
comprehend ancient people mind, in relation with cultural models of the past
and of present. How did ancient people live in the territory? What kind of
activities did they develop? Which symbolic values did they attribute to the
places? What kind of cultural message did ancient people perceived from object
and places? Which relations can we establish between the past and the present
culture?

Results: The narrative museum of Appia archaeological park

The aim of this project has been to utilize all GIS, topographical, architectonic
data to construct digital spatial models, and associate them to a communicative
system dedicated to the public and educational use.
A fly through the 3D model of the territory of Rome helps us to contextualize
the Appia Park before analysing and observing in detail the environment. Some
tools allow us to modify the visualization, changing the textures of the
geometry and loading cartographic layers, highlighting areas of interest, taking
measures on the territory. From the holistic level it is possible to choose two
area of interest, the Caffarella valley and the via Appia Antica and explore them
in a successive, more detailed level of representation. When we choose to walk
inside the park we are full immersed in the context, every element of the
landscape is described in 3D and we can decide to interact directly with
historical and cultural contents through complex behaviours, narrative
metaphors, characters and avatars. While walking through the park we can meet
ancient travellers, famous characters of the past, animals, talking sacred trees,
gods that tell us some tales about the park, legends, myths, stories (fig. A3.1 -2).

Fig.A3.1 Appia1.tif - Fig.A3.2 Appia2.tif

When we get near the archaeological structures we can listen to an audio
explanation of the monument we are looking at, we can open movies, tales,
animations, iconography, and so on; all the metadata are associated inside the
three-dimensional space. Sound is spatially integrated in the 3D space, so to
communicate the specificity of different places also through the suggestions
produced by sounds, noises, voices.
In the real time visualization and exploration of the whole landscape a
simplified version of the object and monuments, obtained by photomodelling
Reconstructing Ancient Landscape S. Pescarin

techniques, are used, while the high resolution version obtained by laser
scanner are employed for a monographic analysis of the monument.
For the Nimphaeum of Egeria and S. Urbano Church we can explore and
compare the models of the monuments as they appear today and their
hypothetical original aspect (fig.A3.3). The Roman road has been reconstructed
from the fourth mile to Casal Rotondo, also in this case monuments have been
acquired by laser scanner or photomodelling techniques.
Fig.A3.3 – Appia3.tif
An important part of the archaeological monuments along via Appia Antica
have been restored, and recomposed by Luigi Canina (1795-1856), an architect
exponent of the neoclassic and antiquarian culture. During his long studies he
made a lot of prints, designs of archaeological views, prospects of building,
detail of decorations. These views show funerary monuments as they were in
the 19th century but also as Canina imagined they could appear in the ancient
time. We have represented one of his views in three-dimensions, so we can
make our avatar walk or run inside Canina’s print, exploring and visiting
monuments as he imagined them (fig. A3.4). The narrative museum of the
archaeological park of Appia Antica has been shown to the public during some
international exhibitions of virtual archaeology: “Immaginare Roma Antica”
(Trajan Market Museum in Rome, Sept.15 – Nov.15, 2005) and “Archeovirtual”
(Paestum, Mediterranean Exchange of Archaeological Tourism, November
2006) obtaining a good success. Public had the possibility to explore and
interact directly with the virtual landscape and cultural contents wearing
stereoscopic glasses to enhance the impression of immersion inside the 3d
environment.
Figure A3.4 – Appia4.tif
[E. P.]

Aknowledgments
GIS and documentation: scientific dir: M. Forte, P. Salonia; Surveys: M. Mascellani, E.
Pietroni, F. Galeazzi, A. Palombini, S. Pescarin; Photomodelling: M. di Ioia, N. Dell’Unto, F.
Galeazzi, E. Pietroni; GIS and webGIS: S. Pescarin, A. Palombini, C.Camporesi; Photos:B.
Trabassi; Direction: Soprintendenza Archeologica Comunale di Roma, L. Sasso D’Elia and A.
Mucci. VR System: Computer graphics: E. Pietroni, N. Dell’Unto, F. Galeazzi, L. Vico, L.
Gomez, Scanner laser post-processing: E. Pietroni, 3d content implementation in VR engine: E.
Pietroni, programming: C. Rufa, 3d landscape: S. Pescarin, E. Pietroni

References
FORTE, M. PESCARIN S., PIETRONI E. 2005
Reconstructing Ancient Landscape S. Pescarin

.B.4 Robotics, Virtual Environments and Landscape Virtual

Communities
Location: Rome, Pisa (Italy), Luxor, Medinet Madi (Egypt) and Khor Rori
(Oman).
A FIRB (Funds for the Investments of Basic Research) project, Integrated
Technologies of robotics and virtual environment in archaeology, financed by
the Italian Ministry of the University and Scientific Research, gives us the
opportunity to experiment and realize a multi-user domain on the web
addressed to a multidisciplinary scientific community. The project, in
collaboration with the Department of Archaeology of the University of Pisa and
with Scuola S. Anna of Pisa, focuses on three archaeological sites: the Teban
tomb 14 in the necropolis of Gurna-Luxor, the Temple A of Middle Kingdom in
Fayum Medinet Madi, both in Egypt, and the ancient settlement of Khor Rori,
in Oman. Maurizio Forte was the scientific coordinator of the CNR unit, I was
the responsible of 3D modelling and VR Multiuser domain project.

Sources, methodology and techniques.

Typically the scientific communities do not consider VR environment an
operative tool for archaeological research and, in particular on the web, there
are few examples of 3D e-learning and e-communication. It is not yet common
to share interpretations, hypotheses and data in the same virtual domain.
We think it could be very important to promote the use of the “virtual” also
within scientific communities of experts, researchers, managers of cultural
heritage. How technologies, and in particular virtual reality, can develop the
organization and visualization of cultural contents? How can they support us in
the interpretation, simulation of hypothesis and in the creation of integrated
informative systems? Which are the most efficacious ways to interact with
models and metadata? How can we share and exchange data?
Archaeological research produces during the phases of acquisition, post-
processing and communication, huge quantities of spatial data, disseminated in
different archives (with various formats, ontologies and typologies). The
consequence is that the process of interpretation and communication becomes
partially compromised, or even obstructed, by the inaccessibility of all data
inside the same informative system. Moreover in most cases virtual reality
applications are developed as final processing and 3D output of a previous long
term work of research and study. They are not conceived as environment for
simulation, sharing of information, testing hypothesis, editing of digital
Reconstructing Ancient Landscape S. Pescarin

ecosystem in order to experiment different solutions.

The three archaeological sites involved in the projects present very different
characteristics and interpretative aspects. The tomb TT14 is a small and narrow
space with a very complex stratigraphy; the Temple A is a well preserved
architectonic context; Khor Rori is a typical example of archaeological
landscape correlated with environmental studies. We have expressly
distinguished and emphasized these features of all the archaeological contexts
in order to apply different methods of communications and virtual re-
contextualization. This variety of conditions of the archaeological contexts has
required the use of different integrated technologies of data acquisition,
elaboration and representation: scanner laser, computer vision and topographic
relief for TT14 (fig.A4.1); GPS, total laser station, GIS, remote sensing,
photogrammetry, computer vision, 3D panorama for the settlement and the
landscape of Khor Rori, 3D computer graphics on the base of topographical
relief for the Temple A of Medinet Madi. A collection of metadata,
interpretative layers, multimedia contents are linked to models and integrated in
the three-dimensional space.

Figure A4.1: Firb1.tif

Results
All these data converge in a virtual scenario in the web where the scientific
community can meet and interact in real time, exchange and test hypothesis,
share data and make simulations in the 3D space146. This virtual space is an
editable and dynamic environment in continuous evolution.
In fact 3D models are not closed and no longer accessible from users, as in most
part of VR applications, on the contrary, they are open to continuous possible
re-elaboration; they can be disassembled and recomposed according to different
combinations and solutions. The models can be also exported from the
application in “obj” format and re-used for different purposes. Beside the 3D
models of architectures and archaeological structures, obtained from
topographical reliefs, the VR application introduces other kinds of ontologies
such as the “Virtuoteca” (fig.A4.2), an imaginary cyberspace, like a library,
where users can find digital contents, papers, multimedia related to the
archaeological site and studies.

Figure A4.2: Firb2.tif

These are the main functions and tools of the applications (fig.A4.3-4):

146http:// www.vhlab.itabc.cnr.it/FIRB/Release/Home.html
Reconstructing Ancient Landscape S. Pescarin

1) set project path; 2) load object (in .nmo format); 3) load scene, (in .txt
format), 4) delete object, 5) load texture, 6) change the camera; 7) change
textures on models; 8) tools of lightning; 9) tools of measurements; 10) move
object, 11) hide/show objects; 12 link metadata to 3D models; 13) chat; 14)
walk and move in the space through the avatar; 15) export3D models; 16) save
scene.
Figure A4.3: firb3.tif
Every new version of the virtual environment can be saved and uploaded on the
web as a new “space” of the MuD (only users with full rights), so that many
different informative worlds can coexist and be compared in real time. The
possibility to load, share and interact with data in the same spatial virtual
environment can increase the level of learning and scientific communication; in
this way the information become hyper-real and contextualized.
The application is developed in Virtools DEV, a real time rendering engine, and
Virtools Mutiuser Pack.
It can be very usefull for education, collaborative sessions of learning, design
and planning of conceptual maps. A possible scenario is the virtual classroom
where the teacher can interact in 3D with the students, discussing about key
features of the archaeological sites, interpretations, hypotheses and general
overviews.

Figure A4.4: firb4.tif

Acknowledgements
A special thank to my colleagues of Virtual Heritage Lab of CNR ITABC; Claudio Rufa
(Softlogic, Rome), Nicolò dell'Unto from UC Merced, the informatics and archaeological teams
of Scuola S. Anna and University of Pisa (M. Bergamasco, E. Bresciani and F. Silvano, M.
Betrò, A. Avanzini, and their collaborators), A. Giammarusti.

References
Pietroni E., Forte M (2006). The Museum's mind: a cybermap for cultural exhibition, Proceedings
VAST 2008, Nicosia, Cipro 30 ottobre – 4 november “The evolution of Information
Communication Technology in Cultural Heritage”, edited by M. Ioannides, D. Arnold, F.
Nicolucci, K. Mania: 70 -73.
Forte M., Pescarin S., Pietroni E. (2006). Transparency, interaction, communication and open
source in Virtual Archaeology, in proceedings FROM SPACE TO PLACE: II
International Conference on Remote Sensing in Archaeology, 4-7 Dicembre 2006 CNR,
Roma, edizione BAR, pp.535-539.

[E. P.]
Reconstructing Ancient Landscape S. Pescarin

.B.5 Virtual Rome

Location: Rome, Italy;
Time: 2nd AD;
Period: Summer

Virtual Rome is an open project whose aim is the interactive exploration and
dynamic editing of 3d cultural information regarding landscape reconstruction
and the territory of Rome. One of the result of the project in fact has been the
identification of a method and the reconstruction of the potential landscape of
Rome during Imperial times (2nd AD) (fig. A5.1). Virtual Rome has also
developed an Open Source VR webGIS application, based on geospecific data,
3d models and multimedia contents, and a VR webLAB Content Management
System dedicated to interactive and shared reconstruction process.
In the web browser final users can interact dynamically in the reconstructed
space and activate different behaviours in order to enhance their knowledge of
the territory. The VR webLAB is developed as to involve different researchers
in the complex activity of landscape reconstruction, inside a cooperative
environment. The VR webGIS, on the other side is built to enable visitors to
explore archaeological landscape, trying to understand how should have been in
the past, where can most important and best preserved monuments be found and
how eventually to visit them, need. The project, started in 2006, has been
directed by the Virtual Heritage Lab of CNR ITABC in Rome (IT), in
cooperation with CINECA Supercomputing Center in Bologna (IT), thanks to
the sponsorship of Seat Yellow Pages and Roman Chamber of Commerce. It is
based on previous works, developed in 2004 and 2005, for the Appia Project
and Esaro Cultural District project (FORTE PESCARIN PIETRONI 2005; PESCARIN ET
ALII 2005). After an analysis of different open source projects and 3d graphic
toolkits, such as Virtual Terrain Project, OpenSceneGraph, OpenSG, etc. , we
decided to base the work on OpenSceneGraph, the only one that in that period
offering paging support for terrains and on-line publication capabilities, through
the .net plug in (KUEHNE MARTZ 2007). It was specifically developed a plug-in,
OSG4WEB, for Internet Explorer and Mozilla Firefox.
Figure A5.1 virtualrome_villa_XX_II.jpg

For Virtual Rome project the following characteristics and behaviours have
been developed:
• paged geospatial dataset support
• coordinate and projection handling (both in input and in output)
• large 3d terrain dataset management
• 3d models integration (modelled with software such as 3D Studio Max,
Reconstructing Ancient Landscape S. Pescarin

etc)
• natural elements, such as vegetation, integration vector layers
integration
• on-line 3d data publication and interaction, possibly embedded into a
web browser
• Fly and walk navigation tools
• Behaviours integration:
• terrains, models switching
• vector information loading
• models loading
• picking and loading external pages or multimedia contents
• overview map
• environment integration
The project followed three main directions:
1.the definition of a digital pipeline which could enable a share
collaborative work,
2.the creation of digital contents (study, interpretation and reconstruction of
the landscape and main archaeological sites)
3.the development of the software.
Digital contents include:
I archaeological and Roman potential landscapes reconstruction;
II 3d models creation, optimization and integration in 3d scene;
III vegetation creation and integration;
IV multimedia Contents creation and integration in the 3d space (fig. A5.4).
Software and developed tools are:
• OSG4WEB plug-in;
• front-end web site, based on JavaScript, PhP. and Ajax programming,
which embed the plug-in (front-end section) (fig. A5.2),
• CMS which integrates the plug-in and enables scene creation and
project publishing (back-end section) (fig. A5.3)

Figure A5.2 virtualrome_frontend.jpg

Figure A5.3 virtualrome_backend.tif
Figure A5.4 virtualrome_metadata.jpg

Digital Pipeline for landscape reconstruction and on-line publication

Virtual Rome project has identified a methodological approach which allowed,
first of all, the definition of the archaeological modern landscape and, secondly,
the reconstruction of the ancient landscape. The latter is based on the same GIS
data used for the first, modified in accordance with archaeological data and
with the integration of all relevant paleo-environmental and historical
Reconstructing Ancient Landscape S. Pescarin

information. Figure A5.5 and table A5.1 describes various activities, software
and formats defined.

Steps in the reconstruction of Virtual Rome Landscapes

1. Geoimagery, vector and DTM data acquisition GRASS GIS, QGIS
2. Data re-projection in the same coordinate system GRASS GIS
3. Colour calibration GRASS GIS and Er Mapper
4. Archaeological Landscape: Terrain Generation OSGdem
5. Vector thematic layers (Interpreted landscape) GRASS GIS and Quantum GIS
6. Database annotation DBF and MySql
7. Export of shape files in OSG format VTP ENVIRO extended version or direct integration
8. Geo-archaeological maps integration in the GIS and DB GRASS GIS
9. Creation of Roman Land Use Map GRASS GIS (or other GIS software)

10. Virtual Ecosystem generation Visual Nature Studio (or other terrain and
ecosystem generator)
11. GIS and vegetation Export Visual Nature Studio and GIS
12. Ancient Landscape: Terrain Generation OSGdem
13. Vegetation integration OSGdem
14. 3d models integration OSG4WEB, Python and MySQL
15. Publication over the web for end user OSG4WEB and JavaScript
16. Updating of the process through cooperative environment 3d CMS

Table A5.1

Figure A5.5 pipeline_vrome.tif

Digital Contents
To reconstruct the archaeological landscape, it was necessary to build an initial
GIS archive, containing all spatial information, such as the aerial photographs
(20 cm resolution) provided by Nuova Telespazio, and other satellite images
with different resolutions in integration to not covered areas. The used
coordinate system was WGS84 UTM32N. The entire geoimages dataset was
quite big (35 GB) and in order to handle lower and higher resolution images we
preferred to maintain separately all photographs, after blending and colour
calibration, rather then producing one large mosaic.
The available DTM (10 mt resolution) was used with the geoimages dataset to
build the 3d terrain with OSGdem. The final result was a 3d model of the entire
Roman area, divided, geometry and texture, in various Level of Details (LOD)
and in tiles. During the exploration, the user doesn't have the feeling of the
Reconstructing Ancient Landscape S. Pescarin

amount of data that are loading, because of the hierarchy of the model.
Vector thematic layers regarding the archaeological landscapes were also added
to the GIS, such as location of archaeological sites, Roman roads, river system,
etc. The majority of this information came from archaeological excavations and
surveys, aerial photo interpretation, historical maps and geoarchaeological
observations. A modified version of the open source tool VTP (Builder and
Enviro) was used to export vector layers in the *.osg and *.ive file format. Data
in this formats may be in fact published on line, and, in our case, they have
been published in the front-end, though the 3d plug-in and also thanks to the
back-end interaction.
Potential Roman Landscape was than reconstructed and published as well as a
further “switchable” layer. It was obtained thanks to the soil map and land use
map of Rome developed by Rome City Council. This map was then modified in
order to define different (Roman) Classes, and Land Uses (aLU: chapter 4)
(ARNOLDUS-HUYZENDVELD 2003, VOLPE AND ARNOLDUS-HUYZENDVELD 2005). Each
dominant soil use of the various areas was reconsidered, comparing it with
Roman geomorphology and known Roman archaeological sites or paleo-
botanical analysis. In this way the territory around Rome was divided into
different potential environmental classes, accordingly with different land
capacity values. A new map was thus obtained, representing the different
potential ecosystems of Roman period in that specific part of Rome (fig. A5.6).
Each ecosystem has been better defined by specific species, taken from a digital
library of Roman vegetation appositely created, and also by characteristics,
identified by sub-categories, such as top and bottom vegetation, density, rules
of nature, etc. Visual Nature Studio by 3D Nature was used for this specific
project. At the end of the process a new geoimage was generated and exported.
Different view points have been rendered and movies produced. The entire
coverage was also exported in GIS to generate the 3d terrain models in osg for
real time on line purposes. 3d models have been added in the ecosystem
generator, in the CMS or in the 3D graphic software.
The interpretation and reconstruction process of this project is not finish. New
data and new information can be always found, new hypothesis formulated,
new areas investigated. Nevertheless, although a work in progress, the VR
webGIS has been available since the beginning of the project on line (the front-
end) at www.virtualrome.net (fig. A5.7).
Figure A5.6 virtualrome_ecosystem.tif

Figure A5.7 virtualrome_archeovirtual2008.tif

[S. P.]
Reconstructing Ancient Landscape S. Pescarin

Acknowledgements
Scientific Direction: S. Pescarin (CNR ITABC) and M. Forte (CNR and Univ. California -
Merced); Software Design: C. Camporesi (CNR ITABC), L. Calori (CINECA); Programmers:
C. Camporesi, B. Fanini (CNR ITABC); L. Calori, S. Imboden, A. Negri, T. Diamanti
(CINECA); Web programmers & graphic: C. Camporesi, M. Pescarin, C. Albano;
Landscape Reconstruction: S. Pescarin, A. Palombini, V. Vassallo; 3D Modelling: F. Galeazzi,
M. di Ioia, A. Moro, L. Vico (ITABC); F. Delli Ponti (CINECA). Foundings and sponsorship:
Seat Yellow Pages, Roman Chamber of Commerce. Cooperatiion: Rome Dept. X - IV° U.O.
and Antonia Arnoldus-Huyzendveld.

References
PESCARIN, S. CALORI, L. CAMPORESI, C. DI IOIA, M. FORTE, M. GALEAZZI, F.
IMBODEN, S. MORO, A. PALOMBINI, A. VASSALLO, V. VICO, L. 2008
Back to II AD. A VR on-line experience with Virtual Rome Project. In: M. Ashley, S. Hermon,
A. Proenca, and K. Rodriguez-Echavarria (Eds), proceedings of the 9th International
Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST (2008).
CALORI, L. CAMPORESI, C. PESCARIN, S. 2009
Virtual Rome: a FOSS approach to WEB3D. In Web3d International Symposium 2009.
CALORI, L. CAMPORESI, C., PESCARIN, S., FORTE M. 2008
Virtual Rome. In Siggreph 2008- Poster
web: http://www.virtualrome.net (this version of the site is currently visible mainly with
Mozilla Firefox)
Reconstructing Ancient Landscape S. Pescarin
Reconstructing Ancient Landscape S. Pescarin

Appendix B
Sources and resources

(A. Palombini, V. Vassallo)

.B.1 Spatial repositories

Global Land Cover Facility

GLCF is a centre for land cover science with a focus on research using remotely
sensed satellite data and products to assess land cover change for local to global
systems. The GLCF provides earth science data and products helpful for the
study and understanding of global environmental systems. It develops and
distributes remotely sensed satellite data and products Every primary data and
products available at the GLCF are free to anyone via FTP (and on-line
accessible through the Earth Science Data Interface, ESDI): they are satellite
imageries and products derived from that, such as vector products.
http://glcf.umiacs.umd.edu

Military Geographical Institute (Istituto Geografico Militare: IT)

This is the organization preserving the largest aero photographic collection of
Italy, providing geo-topo-cartographic support to the Units and Commands of
the Italian Army as Cartographic Organization of the State. Of fundamental
importance are the photos taken before or during the Second World War.
http://www.igmi.org/

Italian Aerial Photo Archive (aereofototeca)

Since 1973, it is a centre for the collection and study of the Italian territory
aero-photographic material and it has acquired over the years a heritage of over
two million images.
http://immagini.iccd.beniculturali.it

Italian State Archives

The Italian Archive of Cultural Heritage.
http://www.archivi.beniculturali.it/
Access with search engines and on-line inventories:
http://www.archivi-sias.it/index.asp
Reconstructing Ancient Landscape S. Pescarin

Mapping Mediterranean Lands project

Part of the Digital Library for International Research and under the aegis of the
Council of American Overseas Research Centrers, it aims to identify, catalogue,
conserve, and make electronically accessible early and unique or rare maps
from archaeological excavation and exploration held in the collections of
American overseas research centres around the Mediterranean Sea. The multi-
phase project resulted in fully searchable catalogued records of over 4,000
single maps and maps in books through an on-line catalogue.
http://www.aiys.org/aodl/public/medmaps/Digital_Maps_MEDMAPS_project.
htm

OpenStreetMap
A free editable map of the whole world, made by users. Creating and providing free geographic
data allows to view, edit and use them in a collaborative way from anywhere on Earth. The
project was started to overcome the problem of legal or technical restrictions on the use of
maps, and to enable to use them in creative and productive ways.
http://www.openstreetmap.org/

Remote Sensing Data: various sources and resources

1. EROS Data Gateway: http://edcimswww.cr.usgs.gov/pub/imswelcome/plain.html
2. Astronaut Photography:
http://eol.jsc.nasa.gov/default.htm
3. MrSid GeoCover Landsat TM images:
https://zulu.ssc.nasa.gov/mrsid/
4. Terraserver:
http://terraserver-usa.com
5. Astronaut Photography:
http://eol.jsc.nasa.gov/default.htm
6. NGA Raster Roam:
http://geoengine.nima.mil/geospatial/SW_TOOLS/NIMAMUSE/webinter/rast_roam.html
7. GeoData:
http://gos2.geodata.gov/wps/portal/gos
8. Landsat:
http://www.landsat.org
9. Tropical Rain Forest Information Center (TRFIC):
http://www.trfic.msu.edu/index.html
10. African Data Dissemination Service: http://edcsnw4.cr.usgs.gov/adds/index.php
11. Centre for Remote Imaging, Sensing and Processing (CRISP):
http://www.crisp.nus.edu.sg/crisp.html
12. ESA Earth Observation Earthnet Online:
http://earth.esa.int/
13. SPOT Vegetation products:
http://free.vgt.vito.be/
14. Free data for Canada:
http://geogratis.cgdi.gc.ca/clf/en
Reconstructing Ancient Landscape S. Pescarin

15. Shuttle Radar Topography Mission (SRTM): ftp://edcsgs9.cr.usgs.gov/pub/data/srtm/

and GLCF
16. ASTER Protected Area Archive: http://asterweb.jpl.nasa.gov/APAA/default.htm

Rome: Map Resources

Project developed by the Dalton School and thought as a collection of Rome
resources for the community and a useful research tool to find interesting
information. It is divided in several categories: literature, archaeology and maps
of Roman Empire, etc.
http://intranet.dalton.org/groups/rome/RMAPS.html

United States Geological Survey (USGS)

American national centre for collecting, managing and analysing data aimed at
providing scientific understanding about natural resource conditions, issues and
problems. It offers a variety of maps and digital data (maps, aerial photographs,
satellite images, remote sensing resources etc.).
http://edcsns17.cr.usgs.gov/EarthExplorer/
[V. V.]

.B.2 Sources
The reconstruction of the archaeological landscape is a very complex process including many
kinds of data and activities in a virtual ecosystem following a multidisciplinary approach. The
integration of different data, makes it possible to interpret the functions, aspects and uses of
structures and places and propose a reliable reconstruction. Very important for the study of the
ancient landscapes and for the organization of the settlements are: ancient writers (Vitruvius,
Cato, Varro, Columella, Pliny the Elder, etc.), inscriptions of monumental roads, topographical
and thematic maps, etc.

Archaeological Maps
Fundamental to landscape reconstruction. The use of archaeological maps allows to analyse in a
wider way sites distribution within a territory and report survived archaeological elements or
even contribute in sites and remains recognition. Thanks to archaeological maps it is possible to
identify ancient roads, linear structures, ancient buildings etc. They contribute in the evaluation
of preventive archaeology. These maps can be used to find sites position and orientation. They
can be found in every official archaeological offices of different regions or counties, or, in some
cases, they may be recovered on specific texts147.

147Such as the Forma Italiae volumes, founded by G. Lugli and it is currently directed by P.
Sommella, published by Leo S. Olschki, Florence.
Reconstructing Ancient Landscape S. Pescarin

Geological, Lithostratigraphic and Geomorphological Maps

Geological maps show the distribution of geological features. They are usually built on top of a
base map, printed with light colours, to not interfere. Geology is represented by colours, lines,
and special symbols unique to geologic maps. They describe superficial rocks and soils, deepen
rocks and soils, age of the rocks, etc.

http://geology.wr.usgs.gov/wgmt/aboutmaps.html

http://www.nature.nps.gov/geology/usgsnps/gmap/gmap2.html

Land evaluation
“the process of assessment of land performance when used for specified
purposes" involving the execution and interpretation of surveys and studies of
landforms, soils, vegetation, climate and other aspects of land in order to
identify and make a comparison of promising kinds of land use in terms
applicable to the objectives of the evaluation” (FAO 1976).

Soil Maps
Soil maps present soil classification and distribution in a specific area, through a geographically
referenced soil database generated at a given resolution by using field and laboratory
observation methods coupled with environmental data through quantitative relationships148.
There is an International Soil Classification Systems that serves to correlate experiences on
similar soils all over the world. A World Reference Base for Soil Classification was published
by International Union of Soil Sciences in cooperation with the International Soil Reference and
Information Centre (ISRIC). The classification is based mainly on soil morphology. (FAO
1988)

Greek and Roman sources (regarding the landscape)

• Agricultural landscape of the Agro Pontino: Dionysius of
Halicarnassus 5.6.2-4; Livy, 2.14.6; (Roman and Greek). Archytas, De
Agricultura (from Diogenes Laërtius 8.82)
• Gardens as farms of poor people: Pliny, 19,52; Urban gardens
disappearance: Pliny Naturalis Historia, 19,59
• Landscape: Tacitus, Germania 5,1-3; Aboriginal people myth and
forest-marshy landscape . Sallust, De coniuratione Catilinae 6,1;
Archaic landscape: Lucretius, 5.1241, 5.1248, 5.1365; Cicero, De re
publica 2.6.11; Hesiod, Teogonia. v. 1013; Conflict between city and
countryside: Cicero, De officiis, 1, 150-151; Inexhaustible natural
resources: Aeschylus, Agamemnon, 958-962; Inexhaustible natural
148Working Group on Digital soil Mapping, WG-DSM: http://www.digitalsoilmapping.org
Reconstructing Ancient Landscape S. Pescarin

resources: Aulus Gellius, Noctes Acticae 2,22,27; Limits of man against

nature. Varro, De Re Rustica, 2, 1-2-3; 1,4,3-5; 1,12,1-4; Natural
beauties of Italy.: Q. Curzius Rufus, Strabo 2,5,26; Propertius, Elegies,
Book III, XXII elegy; Not propagandist story about natural beauties of
Italy. Pliny the Elder, Nat. Hist., Book III
• Natural resources: tuft galleries. Pliny Nat. Hist., 36, 108; 36, 124;
agricultural resources are not inexhaustible Sophocles, Antigone 332-
40:43; Not inexhaustible natural resources: Pliny the Elder, Naturalis
Historia, 2, 207-8; 33, 1-3; Exploitation of natural resources, Seneca,
Controversiae, 2,1,11; Inexhaustible natural resources. Virgil,
Georgicon, 2, 165-6; Xenophon, De Vectigalibus, 4, 25-29
• Pollution, Vitruvius, 8,4,2; Pliny the Elder, Naturalis Historia 18, 2-3;
Pollution by lead in water pipes, Vitruvius 8,6,10-11; Air pollution due
to the smoke of the chimneys: Seneca, Ad Lucilium de providentia,
104, 6
• Roman Landscape: Cato, Beginning of the De Agricultura 1,1-3;
Cicero, De lege agraria; Columella 1,3.1-2; Virgil, Georgicon, Book
II, 109-135; Virgil, Aeneis. 3,701
• Rome: crowd and traffic. Juvenal 3,25; 3, 254-62; Everyday life
scenes. Horace Epistulae. 2,2,65-80; Martial, Epigrammata 12,57 1-17;
Noise Seneca, Epistulae, 56, 1-4; Marshes of the first settlement is
crossed by swimming and sailing. Chronica urbis Romae, 18; Rome
originally was settled in a landscape of hilltop settlements that overlook
watercourses and swampy areas or subjected to flooding and suitable for
grazing. Livy, 1.4.6; 1.7.4
• Urban landscape. Against the excess of buildings; on the increase of
villas and decrease of gardens: Horace, Carmina 2, 15; Buildings
height in the city: Strabo, 5,3,7; Building rules: Ulpian, Digesta
43,8,2,8; Directions on where to build: Vitruvius 1,4,1-5; Urban
development: Seneca, Epistulae, 89,21
• Viability. Etruscan and Roman road system (Tiber)
Polybius, 3.22.11
• Water resources: water sacrality and prohibition of intervention:
Tacitus, Annales, 1, 79; Vitruvius 1,2,7; Attica hydrogeological
instability: Plato, Crizia, 110e – 111d; Creation of pipes to improve
marshes: Vitruvius, 1,4,11-12; Channel dug by Scaurus between Parma
and Piacenza to reclaim the marshes. Strabo, 5,1,11; Ostia connected to
Rome by channels for transporting salt. Virgil, Aeneis. 8.87 ss; 9.314 ss
• Woods. Sacred woods in imperial times in Rome Pliny 15, 77; 16, 235;
Decrease in sacred forests due to the construction expansion. Varro, De
Lingua Latina, 5, 49; Deforestation dedicated to pasture. Lucretius 41;
Reconstructing Ancient Landscape S. Pescarin

Deforestation in ancient time (Greece): Theocritus, 17, 9-10;

Deforestation in some areas made with criterion. Cassiodorus Variae, 5,
16, 2; Etruria wood for house girders. Livy, 28.45.13; Larici road cut:
from the north crossed the Po up to Ancona, but didn't arrive in Rome:
Vitruvius 2, 9, 16: Laws or measures to counter the excessive
deforestation and protect forests and water. Plato, Leges, 824a; Pliny
17,1; Pitch extracted from Sila forests (known as Calabrian). Used in
medicine and to treat the jars: Pliny 14, 127; 24, 37-40; Ravenna built
entirely of wood, Strabo 5, 1, 6; Search for minerals causes the
demolition of the mountains and natural upheavals. Pliny the Elder,
Naturalis Historia, 33, 1; Sila wood (poplar, beech, pine, oak, ash) used
for construction and shipbuilding. Dionysius of Halicarnassus,
Antiquitates Romanae, 20, 15; Villas and palaces of Rome constructed
with Etrurian wood. Strabo 5, 2, 5; Wood uses in Rome. Pliny 12,5

[ V. V.]

.B.3 Geographic spatial analysis

Spatial Analysis
During the second half of 20th century, archaeologists and environmental scholars developed
many models for spatial analysis (HODDER AND ORTON 1990), that is to say, to better understand
the criteria of human organization in relation to settlement distribution in the landscape. Today,
many of these models are easily developed by GIS software. Here below, we'll describe some of
the most diffused operations on raster and vector files, to apply spatial analysis models.

Buffering
Buffering is the operation of defining a circle with a given ray (or more concentric circles)
around points, defining areas referred to sites. This operation has a really high number of
applications. From an ecological/economic point of view the most interesting can be considered
the site catchment analysis.

Cost analysis
It is a study of the cost of paths, from a starting point to different destinations around it. It can
be calculated taking into account the linear distance together with slope and other factors
affecting each grid cell
Reconstructing Ancient Landscape S. Pescarin

Delaunay triangulation
The Delaunay triangulation is the operation to collect all the points on a surface by lines
forming a series of triangles not overlapping each other, the mathematical definition is that “a
Delaunay triangulation for a set P of points in the plane is a triangulation DT(P) such that no
point in P is inside the circumcircle of any triangle in DT(P).” It is used in many applications:
is the basis of 3D modelling, being the first step to create polygonal surfaces from vector points,
and can be used as well to calculate the Clarke-Evans index of randomness/organization in
settlement pattern through the nearest-neighbour analysis (see below). Delauney triangles also
have many important geometrical properties and can be used in a wide range of algorithms.

Nearest neighbour analysis

It is a method used to study the characteristics of a distribution of objects/sites and to compare
different distributions or parts of them, identifying different settlement logics in an area. For
each point the distance to the nearest point is calculated. The average of these distances
(observed value) is compared with the value that it would have if the distribution of points was
random (expected value). If the observed value is similar to that expected it means that the
distribution of points is similar to a random scatter. If the observed value is significantly lower
than the expected, it means that the points are grouped, because the distances between them
tend to be low. In the opposite case the observed highest average means that items are regularly
distributed in space, since they tend to have the greatest possible distance between them.

Site Catchment analysis

Site-catchment analysis can provide valuable information on ancient survival strategies and
social organization. Developed by Eric Higgs and Claudio Vita-Finzi during the late 1960s
(HIGGS VITA-FINZI 1970), the purpose of site catchment analysis was to reconstruct economical
aspects of an archaeological sites. The size of the areas may be based on the sources of material
found in the site, or on deductions such as the distance that could be covered within a day's
journey. It was noted that in ancient societies, most of the settlements are located in territories
located within a radius of 2-3 km from the site. Using site catchment analysis, it is possible to
analyse settlement choices of a group. A group, in fact, tends to minimize the energy needed to
survive, choosing a settlement, permanent or seasonal, located in such a way to facilitate the
exploitation of the certain environment, according to given rays of action. This technique can be
used to rebuild the productive development of a region. The relationship between sites and
environment can also be analysed also for predictive purposes: developing models on site
selection and settlement patterns improve analysis.

Thiessen polygons
This method assumes the sites as set of points in a plan without dimensions, hierarchies or
characterizations. It divides the space geometrically in areas belonging to each point. Polygons
are composed by a series of lines equidistant from the two closest points and perpendicular to
the lines which joins the same points. The result is an ideal model of the zones of influence for
individual points: it does not evaluate the geomorphological differences of the areas and
Reconstructing Ancient Landscape S. Pescarin

considers centres of the same type and size. To create models closer to reality, it is possible to
get some corrections taking into account other factors, or improve the model by the use of
polygons "weighed", according to the importance of the site (economic, demographic, etc.).

Viewshed analysis
It is the analysis of the part of a territory that is visible from one or more given points. Taking
into account the terrain's morphology, it is useful to select part of the landscape particularly
suitable for strategic settlements and to study inter-visibility among sites.

[A.P.]

.B.4 SOFTWARE

Software useful in archaeology

BASP: The Bonn Archaeological Software Package, since the late 70s provides a set of tools
for managing archaeological records through database systems, statistical analysis, mapping
facilities and low-level remote sensing algorithms. OS: Windows; http://www.uni-
koeln.de/~al001/basp.html

MeshLab: a software developed by the ISTI-CNR (Pisa, Italy), for the advanced elaboration of
3d Scanner data. It is a mesh processing system, for user assisted editing, cleaning, filtering and
rendering of large unstructured 3D triangular meshes (typical 3D scanning meshes). It relies on
the gpl VCG library (vcg.sf.net); OS: Windows, Linux, MacOS.
http://meshlab.sourceforge.net

ARC3D webservice: It is a web service allowing to create a 3D model from a series of digital
pictures uploaded by the users. OS: Any; http://www.arc3d.be

Blender: is a 3D object creating suite, for 3D modelling and advanced editing. OS: Windows,
Linux, MacOS; http://www.blender.org

3D Studio Max: is a three-dimensional vector graphics and animation. It has strong modelling
capabilities and is mostly used by video game developers, TV studios and architectural
visualization. OS: Win. http://usa.autodesk.com/adsk/servlet/index?
siteID=123112&id=5659302

Maya: is a powerful, integrated 3D modelling, animation, visual effects, and rendering

solution. Is based on an open architecture and for this reason is used in different fields OS: Win,
MacOS (cinema, television, game development, and design projects).
http://usa.autodesk.com/adsk/servlet/

Cinema 4D: is a commercial, cross-platform, high-end 3-D graphics application. It can operate
with solid modelling or 3D mesh surfaces and is very popular due to its simple interface and
Reconstructing Ancient Landscape S. Pescarin

speed. It is mainly devoted to post-movie for the creation of special effects, but is also
appreciated in the world of graphics and animation, thanks to integration with the most widely
used software. OS: Win, MacOS, Linux http://www.maxon.net/

AutoCad: is the first CAD software (computer aided design) developed for PC. It is mainly
used to produce 2D/3D drawings in engineering, architectural, mechanical, etc. The product is a
vector and it also allows to create three-dimensional models of geometric objects in vector
mode. OS: Windows. http://usa.autodesk.com/adsk/servlet/index?siteID=123112&id=2704278

Game engines
A List of game engines, authoring tools and 3d toolkits is available at:
http://en.wikipedia.org/wiki/List_of_game_engines
Delta 3D: It's an open source game engine suitable for a wide range of modeling and simulation
applications. OS: Win, Linux http://www.delta3d.org/

OGRE: is currently one of the most powerful open source rendering engine. OS: Win, MacOS,
Linux http://www.ogre3d.org

Unity 3D: is a multiplatform game development tool, with a high compatibility both in terms of
file formats and hardware applications. OS: Win, MacOS http://unity3d.com

VirTools Dev: is a development and deployment platform with an innovative approach to

interactive 3D content creation. It facilitates prototyping and robust development up to large-
scale and the immersivity. OS: Windows. http://a2.media.3ds.com/products/3dvia/3dvia-
virtools.

GIS, webGIS, Remote Sensing

A lists of available mapping software is available at: MapTools
(http://maptools.org/) or GFoss (http://www.gfoss.it/drupal/software).
ARK (The Archaeological Recording Kit) is a web-based ‘toolkit’ for the collection, storage
and dissemination of archaeological data. It includes data-editing, data-creation, data-viewing
and data-sharing tools, all of which are delivered using a web-based front-end. OS: Any.
http://ark.lparchaeology.com

ENVI: it is a powerful software joining GIS and image processing/analysis features. OS: Win
http://www.ittvis.com/

ErMapper: It is the a rich software for remote sensing and advanced multispectral image
analysis OS: Win http://www.ermapper.com

ESRI ArcGIS is a leader services company providing Geographic Information System and
geodatabase management applications. Its products, as ArcGis, are affordable, easy-to-learn
desktop mapping tool. OS: Win http://www.esri.com
Reconstructing Ancient Landscape S. Pescarin

GRASS: is the most famous and powerful Open Source GIS suite. It allows all kind of GIS
analysis, on raster and vector data, as well as advanced geoimage processing. The referring site
includes links to OSGeo4W Project. OS: Windows, Linux, MacOS. http://grass.itc.it

Idrisi: one of the earliest GIS software, still preserves the feature of very impressive and
pleasant graphic outputs. OS: Win http://www.clarklabs.org

MapInfo: one of the earliest GIS software, for geo-spatial data visualization and analysis. OS:
Win http://www.mapinfo.com

OpenEV: Is an open source graphic library and application for viewing and analyzing raster
and vector geospatial data. OS: Windows, Linux. http://openev.sourceforge.net

OSGeo4Win: Project is a binary distribution of a broad set of open source geospatial software
for Win32 environments (Windows XP, Vista, etc). It includes many of the most useful
geographic libraries and applications. OS: Win (GDAL/OGR, GRASS, MapServer, OpenEV,
uDig, QGIS etc.) http://trac.osgeo.org/osgeo4w/

QGis: is a simple and very easy-to-use Open Source GIS allowing to visualize, manage, edit,
analyse data, and compose printable maps. OS: Windows, Linux, MacOS. http://www.qgis.org

Terrain Generators
For a complete list of terrain generators the most updated document is:
http://www.tec.army.mil/research/software/TD/tvd/survey/vendors_web_alpha.
html
OSGdem: it is a part of the OSG (OpeneSceneGraph) package for terrain generation. It is a
command-line tool which allow to create a paged terrain starting from DTM and geoimage data.
OS: Windows, Linux, MacOS. http://www.openscenegraph.org

Presagis Tools: Presagis has acquired and developed some of the earliest suites for terrain and
3d environment generation, such as TerraVista and Multigen Creator. The software still keep an
intuitive interface and realistic immersive outputs. OS: Win, MacOS,
Linux.http://www.presagis.com

Visual Nature Studio: VNS is a terrain generator which allows to create highly realistic
geocoded landscapes from GIS data. It has a wide range of output formats in terms of
raster/vector layers, movies and pictures, and terrain/3d environments. OS: Win
http://www.3dnature.com

VTP: Open Source suite for viewing, creating, editing 3D environments, to be browsed or
exported for web publishing. OS: Windows, Linux. http://www.vterrain.org

Terrain (2d- 3d) web-viewer

MapServer is a webGIS, an Open Source platform for publishing on the web spatial data and
Reconstructing Ancient Landscape S. Pescarin

interactive mapping applications to the web in a 2d space. OS: Windows, Linux, MacOS. http://
mapserver.org

OSG4WEB is the result of a CINECA and ITABC-CNR effort to provide a framework for in-
browser OpenGL-based application wrapping. The projects started in 2004 to fill the need of
web-enabled 3d terrain and geo-spatial data browsing in a pre-Google-Earth time. The plug-in
allows run-time loading of different application cores at the web page opening, allowing the
same installed plug-in to brows pages that require different application codes. OS: Win, Linux,
MacOS. http://3d.cineca.it/storage/demo_vrome_ajax/osg4web.html

3d Via Player is the free on line viewer developed by VirTools Dev for real time on line
experience. http://dl.3dvia.com/software/3dvia-player/

Unity web player is the free on line viewer developed by Unit 3D engine. It enables to view
3D content created with Unity directly in the web browser. http://unity3d.com/unity-web-
player-2.x.html

[A.P.]

.B.5 Projects

Geo-spatial 3d viewers
Google Earth is a stand alone software that allows to explore the 3d World at multiple levels,
from Space to the streets, identifying maps, directions and various kind of points of interest. It
is freely distributed by Google and requires a Microsoft Windows operating system, Mac OS X
or Linux and currently has developed a version for iPhone and iPod Touch. Satellite images and
photographs of the Earth are displayed with a very high detail (res. 15 mt) and, in some cases,
with a resolution less than one meter. http://earth.google.it/

Google Maps allows to view 2d maps in the web browser. On the base of the location, it is
possible to view standard or customized maps and information. Interesting the “StreetView”
application recently automatically available after a certain scale. http://maps.google.com/

Google Plug-In and JavaScript API let embed Google Earth, a true 3D digital globe, into the
browsers. Using the API it is possible to draw markers and lines, drape images over the terrain,
add 3D models, or load KML files, allowing to build sophisticated 3D map applications. The
Google Earth API is a free service, available for any web site that is free to consumers.
http://code.google.com/apis/earth/

Google O3D is an open-source web API for creating rich, interactive 3D graphics applications
(games, ads, 3D model viewers, product demos, virtual worlds) in a browser window. It
provides a browser plug-in that adds graphics capabilities inside standard web browsers, and a
sample COLLADA Converter, which can be used to import files in the COLLADA format, an
open standard for 3D assets that is supported by popular content creation applications

Google SketchUp is a very simple software to learn and allows to create, edit and share 3D
Reconstructing Ancient Landscape S. Pescarin

models. The program is part of the same family of GoogleEarth, which ensures the exchange of
information between them in a very simple way. It allows to import and export different
formats (DXF, DWG and 3DS, JPG, TIFF, etc.).
http://www.google.com/sketchup/download/gsupwitthankyou.html.

Nasa World Wind. World Wind is a free open source virtual globe developed by NASA. The
program overlays NASA and USGS satellite imagery, aerial photography, topographic maps
and available GIS data on 3D models of the Earth and other planets. Apart from the Earth there
are several worlds in World Wind: Moon, Mars, Venus, Jupiter and imagery of stars and
galaxies. http://worldwind.arc.nasa.gov

Microsoft Virtual Earth allows to visualize geographic and location-based information by

combining on-line maps with single users' data. The platform is an integrated set of services and
helps to visualize data and provide immersive end-user experiences. The Virtual Earth platform
consists of three services: Virtual Earth Map Control that includes immersive imagery; Virtual
Earth Web Services; MapPoint Web Service, a programmable Web service hosted by Microsoft
to integrate location-based services into software applications and business processes.
http://www.microsoft.com/virtualearth/

Skyline, founded in 1997, is a leading provider of 3D earth visualization software and services.
It offers 3D geo-spatial applications, with support for real-time fusion and streaming of massive
data sets, open standards and a full API.
http://www.skylinesoft.com/SkylineGlobe/corporate/home/index.aspx?

[A.P.]

Second Life projects:

The Akragas doors, dedicated to the Archaeological Park of Agrigento Temples Valley, aims
at illustrating the defensive structure of the city in the 4th century BC, through the creation of
immersive stereoscopic images and a reconstruction in Second Life. Here visitors can actively
participate, also dressing as a soldier, thus entering into history.

Villa of Livia in Second Life at the UCM Virtual Heritage Island: UC Merced is
experimenting digital learning and research in Second Life. In these experiments of virtual
reconstruction of different landscapes (for example the reconstruction of the ancient Villa of
Livia) is possible to test feedback, reactions, participation, and informational transmission
between users and environments.

Digital Humanities Island is a Sim in Second Life developed by King's College of London,
under the direction of Prof. Beacham. It has been opened in 2007 and it is currently used for
educational activities on ancient theater and Roman architecture.

Active Worlds projects:

Quest Atlantis. QA is an international learning and teaching project that uses a 3D multi-user
environment to immerse children, ages 9-15, in educational tasks. QA combines strategies used
Reconstructing Ancient Landscape S. Pescarin

in the commercial gaming environment with lessons from educational research on learning and
motivation. It allows users to travel to virtual places to perform educational activities, talk with
other users and mentors, and build virtual personae. http://www.questatlantis.org

Other geospatial projects:

Bologna 3D was developed from maps with the buildings elevation and aerial photos. The 3d
model of the city is explorable in Internet through a flight simulator (by Skyline).
http://sit.comune.bologna.it/sit/volo3D/bo.FLY

CyArk is a non-profit entity whose mission is to digitally preserve cultural heritage sites
through collecting, archiving and providing open access to data created by laser scanning,
digital modelling, and other state-of-the-art technologies. CyArk has developed the Total
Process for Digital Preservation. http://archive.cyark.org/

Exhibits 3D, based on VirTools Dev engine, provides a dynamic virtualization system of
historic and cultural reality to organizations, institutions, galleries, museums, both public and
private, for the dissemination and promotion of cultural heritage on-line. In a section it is
possible to visit the virtual gallery of available projects. http://www.exhibits.it

Fasti on line. It is a webGIS database of archaeological excavations since the year 2000. It was
created by AIAC (International Association for Classical Archaeology) and supported by PHI
(Packard Humanities Institute). http://www.fastionline.org/index.php

Florence on Earth (IT). This project, GoogleEarth-based, regards the visualisation of one
century urban archaeological researches in Florence, referred to various periods: Roman, Late
Roman, Early Medieval and Medieval. The archaeological finds have been located in
GoogleEarth (trough the its API) and sub-divided into categories (roads, buildings etc.).
http://florenceonearth.com/

Itinera Time Machine (IT). It is a project developed by Digital Archaeology Laboratory of

the University of Foggia, about the archaeological site of Faragola (Ascoli Satriano, FG) and
reconstructed using the scientific documentation derived from the field survey. The Time
Machine allows the user, trough an avatar, to travel in the time and visit both the different
virtual reconstructions and the entire stratigraphic process.
http://www.archeologia.unifg.it/ric/lab/Lad/Farlad.asp

NuME project (IT) aims at creating a digital museum of the historical city of Bologna, with a
specific interest in the reconstruction of the Medieval period, based on historical sources,
archive documents and historical maps. Visitors and scholars may explore the city in different
chronological periods and analyse relating historical sources.
http://3d.cineca.it/3d/Nume/nume_3d.php

Pagine Gialle Visual (IT) by Seat Yellow Pages integrates into a Shockwave-based
multimedia tool the Italian territory. 103 cities can be explored and queried with high spatial
graphic detail (resolution of 20 cm ), accuracy and richness of contents.
http://www.visual.paginegialle.it/3d
Reconstructing Ancient Landscape S. Pescarin

Sardegna 3d (IT) is an application which enable to explore Sardinia in three dimensions,

offering a complete geographic information of the island (cultural, historical and environmental
heritage are constantly update on the base of a regional GIS), through a tool designed to
strengthen the government and management of the territory and to promote sustainable
development. http://www.sardegna3d.it/

Virtual Archaeologist Educational Environment. This project is a virtual, immersive, multi-

user, spatially oriented, 3D educational environment that simulates the real world conditions of
an archaeology excavation (developed by the Archaeology Technologies Laboratory of the
North Dakota State University). This environment allows students in archaeology to enter in the
role of an archaeologist, encouraging them to evaluate and solve scientific problems. The
project permits students to think like archaeologists, creating an active and educational space
that promote exploration and problem-solving. http://fishhook.cs.ndsu.nodak.edu/

XVR project and Piazza dei Miracoli (Pisa, IT). XVR (eXtreme Virtual Reality) is an
innovative development environment dedicated to virtual reality and augmented reality
applications, created by VR Media, a spin-off company of “S. Anna School of advanced
studies” in Pisa. The website allows multilevel access with guided tours for the simple user.
http://www.vrmedia.it/Xvr.htm, http://piazza.opapisa.it

Virtual Rome, coordinated and directed by CNR ITABC in collaboration with CINECA, aims
at studying and reconstructing the archaeological and potential landscape of ancient Rome (21st
century - 2nd century AD) and at enabling its distributed and interactive visualization through a
web-based Virtual Reality application, based on Open Source libraries, on Remote Sensing and
GIS data and on 3d models and through a OSG based plug-in, OSG4WEB. Detailed exploration
is possible in three areas: via Flaminia, via Appia and Roman Imperial Fora. The project has
also developed a 3D CMS, with the goal of creating an on-line cooperative laboratory for
landscape interactive reconstruction. http://www.virtualrome.net
[V. V.]
Reconstructing Ancient Landscape S. Pescarin

Appendix C
Bibliographic references
AA.VV. 1999
Open Sources: Voices from the Open Source Revolution. Cambridge, Massachusetts,
O’Reilly Media.
AA.VV. 2008
Suvey of terrain visualization software. U.S. Army Topographic Engineering Center
Topography, Imagery and Geospatial Research Division Data Representation Branch, (703)
428-6838. 23 January 2008 (http://www.tec.army.mil/TD/tvd/survey).
ADAMS, W. Y. 1987
Three Perspectives on the Past: The Historian, the Art Historian, and the Prehistorian. In:
Hдgg, T. (ed.), Nubian Culture Past and Present. 285–291. Stockholm.
ADKINS, L. – ADKINS, R. A. 1989
Archaeological Illustration. Cambridge University Press, Cambridge.
AGUGIARO, G. – KOLBE, T.H. 2009
Definition of a transition surface with the purpose of integration between laser scanner 3d
model and a low resolution DTM. In: Remondino, F. –El-Hakim, S. – Gonzo, L. (eds.),
Proceedings of the 3rd ISPRS International Workshop 3D-ARCH 2009. Int. Arch. of
Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XXXVIII-5/W1.
AKENINE-MÖLLER, T. – HAINES, E. 2002
Real-Time Rendering Second Edition. 477–479. A K Peters Natick, Massachussets US.
ANNUNZIATO, M. – BONINI, E. – PIERUCCI, P. – PIETRONI, E. 2008
Cultural mirrors: an epistemological approach to artificial life for cultural heritage
communication. In: DMACH 2008, Digital Media and its Applications in Cultural
Heritage, Univerity of Petra, Jordan, 3-6 November '08.
ANTINUCCI, F. 2004
Comunicare nel Museo. Laterza, Milano.
ARISTOTLE
Prior Analytics. Hugh Tredennick (trans.), In: Aristotle, Vol 1, 181–531. Loeb Classical
Library, William Heinemann, London, UK, 1938
ARNOLDUS-HUYZENDVELD, A. 2003
I Suoli di Roma. Comune di Roma, Dipartimento X – II U.O. Sviluppo Sostenibile. Roma.
ARNOLDUS-HUYZENDVELD, A. 2007
Le trasformazioni dell'ambiente naturale della pianura grossetana. In: Citter, C. – Arnoldus-
Huyzendveld, A. (eds), Archeologia urbana a Grosseto. Origine e sviluppo di una città
medievale nella Toscana delle città deboli. Le ricerche 1997 - 2005. Vol. I , Firenze, all'Insegna
del Giglio.
ARNOLDUS-HUYZENDVEL,A. – POZZUTO, E. P. 2009
Una lettura storica del paesaggio attuale: il territorio di Castel di Pietra tra Antichità e
Medioevo. In: Citter, C. (ed.), Dieci anni di ricerche a Castel di Pietra. Edizione degli scavi
1997-2007. 15–39
Reconstructing Ancient Landscape S. Pescarin

ARNOLDUS-HUYZENDVEL, A. – VOLPE, R. 2004

Geomorfologia e Suoli. In: Gioia P. –Volpe R. (eds.), Centocelle I, Roma S.D.O. le indagini
archeologiche. 177–183. Sovraintendenza ai Beni Culturali del Comune di Roma, Rubbettino.
ARON, J.L. – PATZ, J. 2001
Ecosystem change and public health: a global perspective. JHU Press.
ARTIOLA, J.F. – PEPPER, I. – BRUSSEAU, M. 2004
Environmental monitoring and characterization. Academic Press.
ASHBEE, P. 1972
Field Archaeology: Its Origins and Development. In: Fowler, P.J. (ed.), Archaeology and
the Landscape. 38–74. London.
ASHMORE, W. – KNAPP, A. B. (eds.) 1999
Archaeologies of Landscape. Malden and Oxford.
ATTEMA, P. A. J. 2000
Landscape Archaeology and Livy: warfare, colonial expansion and town and country in
Central Italy of the 7th the 4th c. BC. In Bulletin Antieke Beschaving, 75. 115-126.
ATTEMA, P. A. J. – BURGERS, G.-J. – VAN JOOLEN, E. – VAN LEUSEN, M. (EDS.) 2002
New developments in Italian landscape archaeology. Theory and methodology of field survey,
land evaluation and landscape perception, pottery production and distribution, BAR
International Series 1091. Oxford. Archaeopress
ATTEMA, P. – VAN JOOLEN, E. – VAN LEUSEN, M. – BURGERS, G.-J. – MATER, B.
(EDS.) 2002
New Developments in Italian Landscape Archaeology. In: Proceedings of a 3-day
conference held at the University of Groningen, April 13-15, 2000: theory and methodology
of field survey, land evaluation and land perception, pottery production and distribution.
BAR International Series 1091. Oxford. Archaeopress
BAIGRIE, B. S. (ed.) 1996
Picturing Knowledge. Toronto, University of Toronto Press.
BARAB, S. – DODGE, T. – TUZUN, H. – JOB-SLUDER, K. – JACKSON, C. – ARICI, A. –
JOB-SLUDER, L. – CARTEAUX JR., R. – GILBERSTON, J. – HEISELT, C. 2007
The Quest Atlantis Project: a socially-responsive play space for learning. In: Shelton, B. E.
– Wiley, D. (eds.), The Educational Design and Use of Simulation Computer Games.
Rotterdam, The Netherlands, Sense Publishers.
BARD, K. A. – FATTOVICH, R. – MANZO, A. – PERLINGIERI, C. 2007
Archaeological investigations at Bieta Giyorgis (Aksum), Ethiopia. Journal of Field
Archaeology 24, 387–403.
BARKER, G. 1985
Prehistoric farming in Europe. Cambridge University Press, Cambridge.
BARKER, G.1991
Approaches to archaeological survey. In: Barker, G. –Lloyd, J. (eds.), Roman Landscapes:
Archaeological Survey in the Mediterranean Region. 1–9. London.
BATESON, G. 1972
Steps To An Ecology of Mind. San Francisco, Chandler Press.
Reconstructing Ancient Landscape S. Pescarin

BATESON, G. 1979
Mind and Nature: A Necessary Unity. Bantam Books.
BEEK, K. J. – BENNEMA, J. 1972
Land evaluation for agricultural land use planning. An ecological methodology.
Department of Soil Science and Geology. Agricultural University, Wageningen.
BELL, J. 1994
Interpretation and testability in theories about prehistoric thinking. In: Renfrew, C. – Zubrow,
E. B. W. (eds.), The Ancient Mind: Elements of Cognitive Archaeology. 15–22. Cambridge.
BENKLER, Y. 2005
Sharing nicely: on shareable goods and the emergence of sharing as a modality of economic
production. Yale Law Journal 114, 273–358.
BERNARDI, M. (ed.) 1992
Archeologia del paesaggio. Firenze.
BEWLEY, R. H. – RACZKOWSKI, W. 2002
Aerial Archaeology: Developing Future Practice. IOS Press.
BIETTI SESTIERI, A. M. 1981
Economy and Society in Italy between the late Bronze Age and the early Iron Age. In:
Barker, G. –Hodges, R. (eds.), Papers in Italian Archaeology: Archaeology and Italian
Society. Prehistoric, Roman and Medieval Studies. 115-131. BAR International Series 102.
Oxford. Archaeopress.
BINFORD, L. 1968
New Perspectives in Archaeology. Aldine Transaction.
BLASI, C. – CARRANZA, M. L. – FRONDONI, R. – ROSATI, L. 2000
Ecosystem classification and mapping: a proposal for Italian landscapes. Applied Vegetation
Science 3, 233–242.
BONDESAN, A. – BONETTO, J. – BUSANA, M. S. – KIRSCHNER, P. – MIOLA, A. –
MOZZI, P. – PESCARIN, S. – VILLANI, M. C.
20,000 years of landscape evolution at Ca' Tron (Venice, Italy): palaeoenviroment,
archaeology, VR webGIS. In: Proceedings of “Hidden Landscapes of Mediterranean Europe:
Cultural and methodological biases in pre- and protostoric landscape studies” Collegio S.
Chiara, University of Siena, May 25-27, 2007. in print.
BOURGEOIS, J. 2005
Aerial photography and archaeology 2003: a century of information ; papers presented
during the conference held at the Ghent University, December 10th - 12th, 2003. Academia
Press.
BRIDGES, E. M. – DAVIDSON, D. A. 1982
Principles and applications of soil geography. Longman, London and New York
BRÜCKNER, H. 1986
Man’s impact on the evolution of the physical environment in the Mediterranean region in
historical times. Geojournal 13, 7–17.
BUOL, S.W. – HOLE, F. D. – MCCRACKEN, R. J. 1973
Soil genesis and classification. Ames Iowa State University Press.
Reconstructing Ancient Landscape S. Pescarin

BURGERS, G.-J. L. M. 1998

Constructing Messapian landscapes. Settlement dynamics, social organization and culture
contact in the margins of Graeco-Roman Italy. Gieben, Amsterdam.
CALORI, L. – CAMPORESI, C. – FORTE, M. – PESCARIN, S. 2008
Virtual Rome. In International Conference on Computer Graphics and Interactive
Techniques, SIGGRAPH 2008. Los Angeles, California, 11-15 August 2008, ACM pub.
CALORI, L. – CAMPORESI, C. – PESCARIN, S. 2009
Virtual Rome: a FOSS approach to WEB3D. In Web3d International Symposium 2009.
CALORI, L. – FORTE M. – PESCARIN S. 2005
Real-time interactive reconstruction of archaeological landscapes. An opensource approach:
from GIS to Virtual Reality. In: Workshop Italy- Canada 2005 3D Digital Imaging and
Modeling: Applications of Heritage, Industry, Medicine and Land. Padova.
CAMBI, F. – TERRENATO, N. 2004
Introduzione all'archeologia dei paesaggi. Carocci.
CAMPANA, S. – FRANCOVICH, R. 2006
Laser Scanner e GPS. Paesaggi archeologici e tecnologie digitali 1. Firenze.
CAMPANA, S. – PIRO, S. (eds.) 2008
Seing the Unseen. Geophysics and Landscape Archaeology. Taylor and Francis.
CAMPANA, S. – SORDINI, M. 2006
Laser Scanner e GPS in archeologia: geografia dei servizi e delle risorse in Internet. In:
Campana, S. – Francovich, R. (eds.), Laser Scanner e GPS. Paesaggi archeologici e
tecnologie digitali 1. Firenze: 327-333.
CAMPORESI, C. – PALOMBINI, A. – PESCARIN, S. 2007
La documentazione digitale: GIS e 3D WebGIS. In: Forte, M. (ed.), La Villa di Livia, un
percorso di ricerca di archeologia virtuale. 111–120. L'Erma di Bretschneider, Roma.
CARDARELLI, A. – CATTANI, M. – GIORDANI, N. – LABATE, D. – PELLEGRINI, S. 2000
Valutazione del rischio archeologico e programmazione degli interventi di trasformazione
urbana e territoriale. In: S. Gelichi (ed.), Dalla carta di rischio archeologico di Cesena alla
tutela preventiva in Europa. 31–40 e 97–102. Edizioni all’insegna del giglio, Roma.
CARDARELLI, A. – CATTANI, M. – LABATE, D. – PELLEGRINI, S. 2004
Archeologia e geomorfologia. Un approccio integrato applicato al territorio di Modena. In:
Mazzeri, C. (ed.), Per un Atlante Storico Ambientale Urbana. 65–77. Modena.
CARLSTEIN, T. 1982
Time resources society and Ecology. Vol I, Preindustrial societies. London
CARTER, J. R. 1988
Digital representations of topographic surfaces. Photogrammetric Engineering and Remote
Sensing 54, 1577–1580.
CAVALLI SFORZA, L. – CAVALLI SFORZA, F. 2007
Scienza e umanesimo oltre le due culture. In: Dionigi, I. (ed.), I Classici e la Scienza. Gli
antichi, i moderni, noi. 55–65. BUR, Milano.
CHEVALLIER, R. 1978
Preface. In: Dufournet, P.: Pour une archéologie du paysage, 5–7. Paris.
Reconstructing Ancient Landscape S. Pescarin

CHIUSOLI, A. 1985
Elementi di paesaggistica. Clueb, Bologna.
CHIUSOLI, A. 1999
La scienza del paesaggio. Clueb, Bologna.
CIGNONI, P. – SCOPIGNO, R. 2008
Sampled 3D Models for CH Applications: a viable and enabling new medium or just a
technological exercise? In: ACM Journal on Computing and Cultural Heritage, Vol. 1, No.1,
Art.2, June 2008
CLARK, S.C. 1996
Mediterranean ecology and the ecological synthesis of the field sites. In: Brandt and
Thornes (eds.), Mediterranean desertification and land use. John Wiley and Sons, England.
CLARKE, D. L. (ed.) 1977
Spatial Archaeology. London - New York - San Francisco
COHEN, J. – VARSHNEY, A. – MANOCHA, D. – TURK, G. – WEBER, H. – AGARWAL,
P. – BROOKS, F. – WRIGHT, W. 1996
Simplification Envelopes. In: Proceedings of Siggraph '96.
CONOLLY, J. – LAKE, M. 2006
Geographical Information Systems in Archaeology. Cambridge University Press.
CORTELLAZZO, M. – ZOLLI, P. 1999
DELI: Dizionario etimologico della Lingua Italiana, Zanichelli, Bologna.
CREMASCHI, M. – BARONI, C. 2000
Manuale di geoarcheologia, Laterza.
CREMASCHI, M. – RODOLFI, G. (eds.) 1991
Il suolo. Pedologia e scienza della terra e nella valutazione del territorio. La Nuova Italia
Scientifica..
D'AGOSTINO, B. 1992
Introduzione. In: Bernardi, M. (ed.), Archeologia del paesaggio, vol. I, 17–21. Firenze.
DE FELICE, G. – INTRONA, A. – MORO, A. – VOLPE, G. 2007
ITINERA Project – Information Technologies for Valuation, E-Learning and Research in
Archaeology. In: Arnold, D. –Chalmers, A. –Niccolucci F. (eds.), VAST2007 – Future
Technologies to Empower Heritage Professionals, EPOCH Publication. Short and Project
Papers from VAST2007. 123–118. Archaeolingua, Budapest.
DELL'UNTO, N. – FORTE, M. – PIETRONI, E. – PESCARIN, S. – RUFA, C. 2006
Dal laser scanner alla realtà virtuale: metodologie di ricostruzione per il paesaggio
archeologico. In: Campana, S. – Francovich, R. (eds.), Paesaggi Archeologici e Tecnologie
Digitali. Laser Scanner e GPS. 93–106. Firenze, All’Insegna del Giglio.
DEVEREUX, B. J. – ARNABLE, G. S. – CROW, P. – CLIFF, A. D. 2005
The potential of airborne lidar for detection of archaeological features under woodland
canopies. Antiquity 79, 648–660.
DI FIDIO, M. 1990
Architettura del paesaggio, Pirola.
DI GENNARO, F. – BARBINA, P. – DE FILIPPIS, M. – DELL'ERA, F. – FRATIANNI, G. –
Reconstructing Ancient Landscape S. Pescarin

TOGNINELLI, P. 2005
Il liberto Faonte, il notabile Marco Claudio Ponzio Ponziano Marcello e i loro vicini. In:
Santillo Frizell, B. –Klynne, A. (eds.), Roman villas around the Urbs. Interactionwith
landscape and environment. Proceedings of a conference held at the Swedish Institute in Rome,
September 17–18, 2004. The Swedish Institute in Rome. Projects and Seminars, 2, Rome.
DIMBLEY, G. W. (ed.) 2007
The domestication and exploitations of plants and animals. Duckworth, London.
DIONIGI, I. 2007
I Classici e la Scienza. Milano.
DONEUS, M. – BRIESE, C. 2006
Full-waveform airborne laser scanning as a tool for archaeological reconnaissance. In:
Campana, S. – Forte, M. (eds.), From Space to Place. Proceedings of the 2nd International
Conference on Remote Sensing in Archaeology. 99–106. BAR International Series 1568.
Oxford, Archaeopress.
DRAMSTAD, W. E. – OLSON, J. – FORMAN, R. 1996
Landscape Ecology Principles. Washington, DC, Island Press.
EDDINGTON, A. S. 1929/1935
The nature of the physical world. New York, MacMillan.
EDMUND, N. W. 2005
End the Biggest Educational and Intellectual Blunder in History. Scientific Method Publishing.
EL-HAKIM, S. F. 2001
A flexible approach to 3D reconstruction from single images. In: ACM Proceedings of
SIGGRAPH ’01, Technical Sketches, Los Angeles, California, 12th to 17th August 2001.186
FABECH, C. – RINGTVED, J. (eds.) 1998
Settlement and Landscape. Proceedings of a conference in Århus, Denmark, May 4-7, 1998.
Jutland Archaeological Society, Moesgård, Denmark.
FAO 1976
A framework for land evaluation. Soil Bulletin 32. 1976–1981. Rome, Food and
Agricultural Organisation, Rome.
FAO 1977
Guidelines for soil profile description (2nd edition), FAO pub., Rome.
FAO 1985
Guidelines: land evaluation for irrigated agriculture. In Soils Bulletin 55, Rome.
FAO 1988
World Soil Resources. In Report, 60, FAO, Rome.
FEDELI, P. 1990
La natura violata. Ecologia e mondo romano. Sellerio Editore Palermo
FISH, S. 1980
The authority of Interpretative Communities. Harvard University Press.
FLECHSIG, P. 1960
Gehirnphysiollogie und Willenstheorieni. En. trad. In: Von Bonin, G.: Some paper on the
cerebral cortex. 75-89. Thomas, Springfield 1960.
Reconstructing Ancient Landscape S. Pescarin

FONSECA, C.D. – ROUBIS, D. – SOGLIANI, F. 2007

Jure Vetere. Ricerche archeologiche nella prima fondazione monastica di Gioacchino da Fiore
(indagini 2001-2005). Rubettino, Cosenza.
FORNI, G. 1989
Questioni di storia agraria pre-romana: le quattro fasi dell’agricoltura etrusca. In: Secondo
Congresso Internazionale Etrusco III: 1501-15. Rome, ‘L’Erma’ di Bretschneider.
FORNI, G. 1990
Gli albori dell’agricoltura. Reda, Roma.
FORSER, K. – MAZZUCCO, K. 2002
Introduzione ad Aby Warburg e all'«Atlante della memoria», Milano 2002.
FORTE, M. 2000
About Virtual Archaeology: disorders, cognitive interactions and virtuality. In: Barcelo,
Forte M. – Sanders D. (eds), Virtual Reality in Archaeology. BAR International Series 843.
Oxford, Archaeopress.
FORTE, M. 2003A
Mindscape: ecological thinking, cyber-anthropology, and virtual archaeological landscapes. In:
Forte, M. – Williams, P. R. (eds.), The Reconstruction of Archaeological Landscapes
through Digital Technologies. Proceedings of the 1st Italy-United States Workshop (Boston,
Massachussets 2001), 95–108. BAR International Series 1151, Oxford, Archaeopress.
FORTE, M. 2003B
The Remote Sensing project for the archaeological landscape of Aksum (Ethiopia). In:
Forte, M. – Williams, P. R. (eds.), The Reconstruction of Archaeological Landscapes
through Digital Technologies, 81–93. BAR International Series, Oxford, Archaeopress.
FORTE, M. 2007
Introduzione. In: Forte, M. (ed.), La villa di Livia: un percorso di ricerca di archeologia
virtual. 1–36. L'Erma di Bretschneider, Roma.
FORTE, M. – PESCARIN, S. 2005
Integrated real-time visualisation of archaeological landscape. In: Piro, S. (ed.), Proceedings
of 6th International Conference on Archaeological Prospection (Rome Sept. 14-17,2005), 231–
236. ITABC CNR, Roma.
FORTE, M. – PESCARIN, S. 2006
The virtual museum of landscape. In: Virtual Museums and Archaeology, Archeologia e
Calcolatori Special Issue, N.17, Dec. 2006. All'insegna del Giglio.
FORTE, M. – PESCARIN, S.
Behaviors, Interactions and Affordance in Virtual Archaeology. In: Advanced ICT Methods
Guide, in print.
FORTE, M. – PESCARIN, S. – PIETRONI, E. 2005
The Appia Antica Project. In: Forte, M. (ed.), The reconstruction of Archaeological
Landscapes through Digital Technologies, Proceedings of the 2nd Italy-USA Workshop, Rome,
Nov. 3-5 2003, Berkeley, USA, May 2005. 79–92. BAR International Series 1379, Oxford,
Archaeopress.
FORTE, M. – PESCARIN, S. – PIETRONI E. – RUFA, C. 2006
Multiuser interaction in an archaeological landscape: the Flaminia project. In: Proceedings
Reconstructing Ancient Landscape S. Pescarin

of the 2nd Int. Conference on Remote Sensing in archaeology “From Space to Place”,
Rome 4-7 Dec. 2006. 189–196. BAR International Series, Oxford, Archaeopress.
FORTE, M. – PESCARIN, S. – PIETRONI, E. – RUFA C. 2007
Multiuser Virtual Reality museum application for Cultural Heritage: the Ancient via
Flaminia project. In: Arnold, D. –Niccolucci, F. –Chalmers A. (eds.), Proceedings of the 8th
International Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST
(2007) Project Papers. 90-98
FORTE, M. – PESCARIN, S. – PUJOL TOST, L. 2006
VR applications, new devices and museums: public's feedback and learning. A preliminary
report. In: VAST 2006 Proceedings, Short Presentations 64-69.
FORTE, M. – PIETRONI, E. 2008
Virtual Reality Web Collaborative Environments in Archaeology. In: VSMM 2008, 20-25
october 2008, Cyprus, Proceedings of the 14th International Conference on Virtual Systems
and Multimedia Vol III. 74–78.
FORTE, M. – PIETRONI, E. – DELL'UNTO N. 2008
3D Multiuser Domain and Virtual Ecosystems for Transmission and Communication of
Cultural Heritage. In: DMACH 2008, Digital Media and its Applications in Cultural
Heritage, College of Architecture and Arts, Univerity of Petra, Jordan, 3–6 November,
2008. University of Petra, Amman.
GABBA, E. – PASQUINUCCI, M. 1998
Strutture Agrarie e Allevamento Transumante nell’Italia Romana. 79–182. Pisa, Giardini.
GAFFNEY, V. –STANCIC, Z. 1991
GIS approaches to regional analysis: a case study of the island of Hvar. Ljubljana.
GESER, G. HAZAN, S. KARPP, C. KASTEREN, V. SINAZZÈ, A. STEEMSON, M. AND
WOOD, H. 2004
Digicult . Thematic Issue 5, Jan. 2004, Salzburg Research. Salzburg
GIACOMINI, V. – FENAROLI, L. 1958
La Flora. Conosci l’Italia. vol. II. Touring club italiano, Milano.
GIBSON, J. 1979
The Ecological Approach to Visual Perception. Lawrence Erlbaum Associates. Hillsdale NJ
GILLINGS, M. – MATTINGLY, D. – VAN DALEN, J. (eds.) 1999
Geographical Information Systems and Landscape Archaeology, Oxford.
GIOIA, P. – VOLPE, R. 2004
Centocelle I. Roma S.D.O. le indagini archeologiche . Soveria Mannelli.
GIORDANO, G. 1999
Pedologia, Torino: 312.
GAFFNEY, V. – STANCIC, Z. 1991
GIS approaches to regional analysis: a case study of the island of Hvar, Znanstveni institut
Filozofske fakultete, Ljubljana
GOBBETTI, E. – MARTON, F. 2005
Far Voxels - A Multiresolution Framework for Interactive Rendering of Huge Complex 3D
Models on Commodity Graphics Platforms. In: Proceedings of Acm Siggraph 2005.
Reconstructing Ancient Landscape S. Pescarin

GOLDBERG, P. – MACPHAIL, R. 2006

Practical and theoretical geoarchaeology, Wiley-Blackwell.
GRANELLI, A. – SARNO, L. 2007
Immagini e linguaggi del digitale. Le nuove frontiere della mente. Milano.
GREGORY, R. L. 1973
The Counfounded eye. In: Gregory R. L. – Gombrich E.H.: Illusion in nature and art. 49–96.
Duckworth, London.
GROGER, G. – KOLBE, T. – CZERWINSKI, A. – NAGEL, C. 2008
OpenGIS City Geography Markup Language (CityGML) Encoding Standard, Version 1.0.0,
International OGC Standard. Open Geospatial Consortium, Doc. No. 08-007r1.
GRUN, A. 1978
Accuracy, reliability and statistics in close-range photogrammetry. In: Proceedings of the
Symposium of Comm. V of the ISP. Stockholm, August 14–17.
GRÜN, A. 2000
Semi-automated approaches to site recording and modelling. International Archives of
Photogrammetry and Remote Sensing 33. (5/1), 309–318.
GUIDAZZOLI, A. – BONFIGLI, M. E. 1999
The creation of the Nu.M.E. Project. In: ACM SIGGRAPH 99, Conference abstracts and
applications. 31–32. ACM, Los Angeles, California.
GUIDAZZOLI, A. – PESCARIN, S. – FORTE, M. – FELICORI, M. – GROSSELE, R. 2004
From Gis to Landscape Virtual Museums. In: Forte, M. (ed.), Archaeological Landscapes
through Digital Technologies: Proceedings of the 2nd Italy-United States Workshop, Rome,
Italy, November 3-5, 2003 Berkeley, USA, May 2005. BAR International Series, Oxford,
Archaeopress
GUIDI, G. – FRISCHER, B. – DE SIMONE, M. – CIOCI, A – SPINETTI, A. – CAROSSO, L. –
MICOLI, L. – RUSSO, M. – GRASSO, T. 2005
Virtualizing Ancient Rome: 3D acquisition and modeling of a large plaster-of-Paris model
of imperial Rome. In: Beraldin, E-H., Gruen, W. (eds.), Videometrics VIII. 18-20 jan 2005,
S.Josè, California, SPIE, Vol. 5665. 119–133.
GUIDI, G. – REMONDINO, F. – MORLANDO, G. – DEL MASTIO, A. – UCCHEDDU, F. –
PELAGOTTI, A. 2007
Performance evaluation of a low cost active sensor for cultural heritage documentation. In:
Gruen/Kahmen (eds.), VIII Conference on Optical 3D Measurement Techniques. Vol.2. 59–69,
July 2007, Zurich, Switzerland.
HARRIS, D. – FAIRCHILD RUGGLES, D. 2007
Sites Unseen: Landscape and Vision. University of Pittsburgh Press.
HAZAN, S. 2004
Weaving Community Webs: a position paper. In: Geser, G. (ed.), Digicult . Thematic Issue
5. Jan. 2004, 47–10. Salzburg Research, Salzburg.
HERMON, S. – NICCOLUCCI, F. 2003
La logica fuzzy e le sue applicazioni alla ricerca archeologica. In: Archeologia e Calcolatori
XIV, Edizioni All'Insegna del Giglio, Firenze. 97–110.
HIGGS, E.S. (ed.), 1975
Reconstructing Ancient Landscape S. Pescarin

Palaeoeconomy, 111–175. Cambridge University Press, Cambridge.

HIGGS, E. S. – VITA FINZI, C. 1970
Prehistoric Economy in the Mount carmel Area of Palestin: Site Catchment Analysis. In:
Proceedings of the Prehistoric Society, XXXVI.1–37.
HODDER, I. (ed.) 2001
Archaeological Theory Today, Cambridge.
HUGUE B. N. – LOADER B.D. (eds.) 1999
Digital Democracy: Discourse and Decision Making in the Information Age. London-New
York, Routledge.
HUNT, C. – MALONE, C. – SEVINK, J. – STODDART, S. 1990
Environment, soils and early agriculture. Apennines Central Italy. World Archaeology, Vol
22, nr. 1, 34–44.
ISAGER, S. – SKYDSKAARD, J.E. 1992
Ancient Greek Agriculture. An Introduction. London and New York.
GOLVIN, J. C. 1999
Voyage en Egypte ancienne, Ed. Errance.
GOLVIN, J. C. 2007
Ancient Cities Brought to Life. Thalamos Publishing, Paris.
JONES, S. 1998
Cybersociety 2.0: revisiting computer-mediated community and technology. London, Sage
Publication.
KAMERMANS, H. 2000
Land evaluation as predictive modelling: a deductive approach. In: Lock, G. (ed.), Beyond
the Map. Archaeology and Spatial Technologies. NATO Science Series, Series A: Life
Sciences, vol. 321. 124–146. IOS Press, Ohmsha, Amsterdam.
KAREL, W. – KRAUS, K. 2006
Quality parameters of digital terrain models. In: Hohle, J. (ed.), Seminar on Automated
Quality Control of Digital Terrain Models. Aalborg, Denmark. EuroSDR.
KASSER, M. – EGELS, Y. 2002
Digital Photogrammetry. CRC Press.
KAUFFMAN, S.1995
At Home in the Universe. Oxford University Press.
KEITH, A. M. 2000
Engendering Rome: women in Latin epic. Cambridge University Press.
KENNETT, D. J. 1996
The Scientific Method in Archaeology. In: Fagan, B. M. (ed.), The Oxford Companion to
Archaeology. 629–631. Oxford University Press, New York.
KLINGEBIEL, A. A. – MONTGOMERY, P. H. 1961
Land-capability classification. Agricultural handbook 210. USDA. Soil Conserv. Serv.,
Washington, DC.
KOLLER, D. – TURITZIN, M. – LEVOY, M. – TARINI, M. – CROCCIA, G. 2004
Reconstructing Ancient Landscape S. Pescarin

Protected Interactive 3D Graphics Via Remote Rendering , Proceedings of ACM SIGGRAPH

2004.
KORZYBSKI, A. 1933
Science and Sanity. Lakeville: Institute of General Semantics.
KUEHNE, B. – MARTZ, P. 2007
OpenSceneGraph Reference Manual ver. 2.2. Skew Matrix Software and Blue Newt.

.B.6 LANGTON, C. 1989

.B.7 Artificial Life: proceedings of an interdisciplinary workshop on the Synthesis and

Simulation of Living Systems. Los Alamos, 1987. Addison-Wesley.
LASER, 2005
Il Sapere liberato. Il movimento dell'open source e la ricerca scientifica. Milano, Feltrinelli.
LAURINI, R. 1998
Spatial multi-database topological continuity and indexing: a step towards seamless GIS
data interoperability. International Journal on Geographical Information Science, Vol. 12,
N.4. 373–402.
LESSING, L. 2001
The Future of Ideas. The fate of the commons in a connected world. Vintage Book, New
York, Random House.
LEUG, C. – FISHER, D. 2003
From UseNet to CoWebs: interacting with social information spaces. Amsterdam, Springer.
LEVEAU, P. ET ALII (eds.) 1999
Environmental Reconstruction in Mediterranean Archaeology, Oxford.
LOADER, B. D. 1998
Cyberspace Divide: Equality, Agency and Policy in the Information Society. 3–16. London-
New York, Routledge.
LOCK, G. 2000
Beyond the map: archaeology and spatial technologies. North Atlantic Treaty Organization.
Scientific Affairs Division, IOS Press.
LOCK, G. 2003
Using Computers in Archaeology: Towards Virtual Pasts. Routledge.
LOCK, G. STANČIČ, Z. (EDS.) 1995
Archaeology and Geographical Information Systems: A European Perspective. London.
LOGOTHETIS, N.K. ET ALII, 1995
Shape representation in the inferior temporal cortex of monkey. Current Biology Vol 5.
552–565.
LOOMIS, R. S. – CONNOR D. J. 1992
Crop Ecology: Productivity and Management in Agricultural Systems.Cambridge
University Press, Cambridge.
LUEBKE, D. P. 2003
Reconstructing Ancient Landscape S. Pescarin

Level of Detail for 3D Graphics: application and theory. Morgan Kaufmann Publishers.
LUEBKE, D. P. – REDDY, M. – COHEN, J. D. – VARSHNEY, A. – WATSON, B. –
HUEBNER, R. 2002
Level of Detail for 3D Graphics. Morgan Kaufmann Publishers.
LUKE, R. – CLEMENT, A. – TERADA, R. – BORTOLUSSI, D. – BOOTH, C. – BROOKS,
D. – CHRIST, D. 2004
The Promise and Perils of a Participatory approach to developing an Open Source
community learning network. In: Proceedings Participatory Design Conference 2004,.
Toronto Canada.
MACEDONIO, G. – COSTA, A. – LONGO, A. 2005
A computer model for volcanic ash fallout and assessment of subsequent hazard. Computer
& Geosciences, 31, N.7. 837–845.
MANZOTTI, R. – TAGLIASCO, V. 2008
L'esperienza: più estesa del sistema nervoso. Sistemi Intelligenti, n.3, Il Mulino, Bologna
MATURANA, H. R. – VARELA, F. G. 1980
Autopoiesis and Cognition.. Dordrecht, Netherlands, Reidel.
MATURANA, H. R. – VARELA, F. G. 1999
L'albero della conoscenza. (Orig. El àrbol del conocimiento, 1984) Garzanti.
MELLET-D'HUART, D. 2006
A Model of (En)Action to approach Embodiment: A Cornerstone for the Design of Virtual
Environments for Learning. In: Win W. & Hedley N. (eds.), Journal of Virtual reality, special
issue on education. Vol 10, Numbers 3-4 / December, 2006. 253–269. Springer, London.
MERRY, P. 2000
Open Source Learning: a key to multiculturalism, citizenship, and the knowledge of society.
1–22. Engage, Utrecht. (www.engage.nu/interact/)
MOGENSES, P. H. – SHAPIRO, D. 1998
When Survival is an Issue: PD in Support to Landscape Architecture. Computer Supported
Cooperative Work 7. 187–203.
MOORE, I. D. – GRAYSON, R. B. – LADSON, A. R..1991
Digital terrain modeling: a review of hydrological, geomorphological, and biological
applications. Hydrological Processes 5, 3–30.
MOORE, P. D. – WEBB, J.A. 1978
An illustrated guide to pollen analysis. Hodder & Stoughton, London.
MUELLER, P. ET ALII 2006
Procedural Modelling of Buildings. In Proceedings of ACM SIGGRAPH 2006 / ACM
Transactions on Graphics
MUSCO, S. – PETRASSI, L. – PRACCHIA, S. 2001
Luoghi e paesaggi archeologici del suburbio orientale di Roma.. 149–251. Roma.
NIJBOER, A.J. 1998
From household production to workshops. Archaeological evidence for economic
transformation, pre-monetary exchange and urbanisation in central Italy from 800 to 400
BC. Donkel & Donkel, Drachten, The Netherlands.
Reconstructing Ancient Landscape S. Pescarin

ODIFREDDI, P. 2007
La Guerra dei due mondi. In: Dionigi I. (ed.), I Classici e la Scienza. 45–54. Milano.
PALOMBINI, A. – VASSALLO, V. 2007
Il Paesaggio archeologico, l'ambiente e il contesto. In: Forte M. (ed.), Villa di Livia, un
percorso di ricerca di archeologia virtuale. 54–68. L'Erma di Bretschneider, Roma.
PARCAK, S. H. 2009
Satellite Remote Sensing for Archaeology. Taylor & Francis.
PANIZZA, M. 2005
Manuale di geomorfologia applicata, Franco Angeli.
PEIRCE, C. S. 1903
Lectures on Pragmatism, Cambridge, MA, March 26 – May 17, 1903. In the version: Peirce
Edition Project (eds.), The Essential Peirce, Selected Philosophical Writings, Vol 2 (1893–
1913), 133–241. Indiana University Press, Bloomington, IN, 1998 .
PESCARIN, S. 2000
Historical and geo-archaeological research in urban context: the example of Bologna city. In:
VAST2000 Euroconference, proceedings of the Meeting of Virtual Archaeolog, Arezzo 24-25
Novembre 2000. BAR International Series, Oxford, Archaeopress.
PESCARIN, S. 2002
GIS contribution to urban history and to the reconstruction of ancient landscape. In: Burenhult,
G. (ed.), Archaeological Informatics: pushing the Envelope: CAA2001, Proceedings of the 29th
Conference, Gotland, April 2001. 125–128. BAR International Series 1016. Oxford,
Archaeopress.
PESCARIN, S. 2006
Open source in archeologia : nuove prospettive per la ricerca. In: Archeologia e Calcolatori,
Vol.17, 137–155. All'insegna del Giglio.
PESCARIN, S. 2007
Reconstructing archaeological landscape. Interpretation and integration in spatial and real-time
open system. In: Keeler, D. –Salisbury, R. (eds.), Space - Archaeology's Final Frontier? An
Intercontinental approach. Cambridge Scholars Publishing, Cambridge.
PESCARIN, S. 2007
Il paesaggio archeologico di Bononia GIS e Realtà Virtuale tra ricerca e comunicazione. In:
Cisalpina, Bologna, in print
PESCARIN, S. 2008B
Archeologia Preventiva. In: D'Andrea, D. – Guermandi, M. P. (eds), Strumenti per
l'archeologia preventiva : esperienze, normative, tecnologie.59–71. Budapest,
Archaeolingua.
PESCARIN, S. in print
Explicit theoretic pipeline: GIS analysis and data integration for archaeological landscape
reconstruction. In: CAA2008 Budapest.
PESCARIN, S. – CALORI, L. 2005
Verso il VR-WebGIS. Il caso del Distretto Culturale della Valle dell'Esaro: un sistema open-
source per le risorse culturali, turistiche e ambientali. In: MondoGIS N.51, Roma.
PESCARIN, S. – CALORI, L.– CAMPORESI, C. – DI IOIA, M. – FORTE, M. – GALEAZZI, F.
Reconstructing Ancient Landscape S. Pescarin

– IMBODEN, S. – MORO, A. – PALOMBINI, A. – VASSALLO, V. – VICO, L. 2008

Back to II AD. A VR on-line experience with Virtual Rome Project. In: Ashley, M. –Hermon,
S. – Proenca, A. –Rodriguez-Echavarria, K. (eds.), Proceedings of the 9th International
Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST (2008).
PESCARIN, S. – CALORI, L. – CAMPORESI, C. – FORTE, M. 2005
Interactive landscapes reconstruction: a Web 2D and 3D opensource solution. In: Mudge, M. –
Ryan, N. – Scopigno, R. (eds.), The 6th International Symposium on Virtual Reality,
Archaeology and Cultural Heritage, VAST’05 Short Presentations. 33–38. STAR, Pisa.
PESCARIN, S. – CALORI, L. – CAMPORESI, C. – FORTE, M. – GUIDAZZOLI, A. –
IMBODEN S. 2005b
Open Heritage: an Open Source approach to 3d real-time and web-based landscape
reconstruction. In VSMM 2005, Proceedings of XI International Conference on Virtual
Systems and Multimedia: Virtual Reality at work in the 21st century, Oct. 3-7 2005 Ghent
Belgium, (Thwaites Ed.) Budapest.313-320.
PESCARIN, S. – FORTE, M. – GUIDAZZOLI, A. – MAURI, M. – BONFIGLI, M. E. 2007
Bologna in età romana dal GIS alla realtà virtuale. In: Ut Natura Ars. Virtual Reality e
archeologia. Studi e Scavi. 115–122. University Press Bologna, Imola.
PESCARIN, S. – RUZZA, R. 2004
Un Museo Virtuale per valorizzare il paesaggio del Delta del Po. In: MondoGIS n. 42.
PICARRETA, F. 1987
Manuale di fotografia aerea: uso archeologico. L'Erma di Bretschneider.
PIGGOTT, S. 1978
Antiquity Depicted: Aspects of Archaeological Illustration. Thames and Hudson, London
and New York.
PIGNATTI, S. 1979
I piani di vegetazione in Italia. In: CNR Collana del Progetto Finalizzato “Promozione della
Qualità dell'Ambiente”. 411–428. Firenze, Estratto da Giornale Botanico Italiano 113.
POLANYI, M. 1990
La conoscenza personale. Verso una filosofia post-critica.248–252. Rusconi, Milano.
POPPER, K. 1934
The Logic of Scientific Discovery. 2.
PREECE, J. – MALONEY-KRIHMAR, D. – ABRAS, CH. 2003
History of Emergence of Online Communities. In: Wellman (ed.), Encyclopedia of
Community. London, Sage.
PRICE, T. D. 2000
Europe's first farmers. Cambridge University Press, U.K.
RAYMOND, E. S. 2001
The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental
Revolutionary. Sebastapol, CA, Paperback Ed. Rev. O’Reilly.
REILLY, P. 1990
Towards a virtual archaeology. In: Lockyear, K. –Rahtz, S. (eds.), Computer Applications in
Archaeology. 133–139. BAR International Series 565. Oxford. Archaeopress
Reconstructing Ancient Landscape S. Pescarin

REINGOLD, H. 1994
Virtual Community: finding connection in a computerized world. Ebury Vintage
(www.rheingold.com/vc/book/)
REMONDINO, F. – EL-HAKIM, S. 2006
Image-based 3D modelling: a review. The Photogrammetric Record, Vol.21 (115), 269–291.
REMONDINO, F. – EL-HAKIM, S. – GIRARDI, S. – RIZZI, A. – BENEDETTI, S. –
GONZO, L. 2009
3D Virtual reconstruction and visualization of complex architectures - The 3D-ARCH
project. In: Remondino, F. – El-Hakim, S. –Gonzo L. (eds.), Proceedings of the 3rd ISPRS
International Workshop 3D-ARCH 2009, International Archaeology of Photogrammetry,
Remote Sensing and Spatial Information Sciences, Vol. XXXVIII-5/W1.
RENFREW, C. 1994
Towards a cognitive archaeology. In: Renfrew, C. – Zubrow (eds.), The Ancient Mind:
Elements of Cognitive Archaeology 3–13. Cambridge.
RENFREW, C. 2003
What are we? Where do we come from? Figuring it out. The parallel visions of artists and
archaeologists. Thames and Hudson, London.
RENFREW, C. – BAHN, P. 1991
Archaeology, Theories, Methods and Practice. Thames and Hudson, London.
ROBERTS, B. K. 1987
Landscape Archaeology. In: Wagstaff, J. M. (ed.), Landscape and Culture. 77–95. Oxford
and New York.
ROSSITER, D. G. – VAN WAMBEKE, A. R. 1997
Automatic Land Evaluation System. ALES version 4.65 User’s Manual. Cornel University,
Department of Soil, Crop and Atmospheric Sciences, Ithaca, NY, USA.
ROUBIS, D. – SOGLIANI, F. – GABELLONE, F. – GNESI, D. 2008
Archaeological landscapes through GIS and Virtual Reality: A case study on the monastic
site of Jure Vetere
ROUSSOU, M. 2005
Can interactivity in virtual environments enable conceptual learning? In: 7th Virtual Reality
International Conference. 57–64. Laval, France.
RYAN, M. L. 2001
Narrative as Virtual Reality: Immersion and Interactivity in Literature and Electronic Media.
Baltimore and London, Johns Hopkins University press. HB. 399.
SANDERS, D. 2001
Persuade or Perish: Moving Virtual Heritage beyond Pretty Pictures of the Past. In: VSMM
2001, Seventh International Conference on Virtual Systems and Multimedia: 236
SANTORO, S. – MONTI, A. 2003
Carta del rischio archeologico della città di Durres. Durazzo.
SCHAFERSMAN, S. D. 1994
An Introduction to Science Scientific Thinking and the Scientific Method, Digital article:
www.freeinquiry.com/intro-to-sci.html
Reconstructing Ancient Landscape S. Pescarin

SCHOPENHAUER, A. 1819
Die Welt als Wille und Vorstellung. 3rd ed., Brockhaus, Leipzig 1859
SCHULZ, J. 1975
Michelangelo's unfinished works. Art Bulletin Vol. 58., 366–373.
SEVINK, J. 1985
Physiographic soil surveys and archaeology. In: Malone, C. – Stoddart, S. (eds.), Papers in
Italian Archaeology IV. 41–52. BAR International Series 243. Oxford. Archaeopress.
SHANKS, M. 2001
Culture/Archaeology. The Dispersion of a Discipline and its Objects. In: Hodder, I. (ed.),
Archaeological Theory Today. 284–306. Cambridge.
SMITH, M. – KOLLOCK, P. 1999
Communities in Cyberspace. London, Routledge.
SODDU, C. 1991.
Simulation tools for the dynamic evolution of town shape. In: Global change and local
problems. Oxford Polytechnic.
SOGLIANI, F. – GABELLONE, F. – DANESE, M. – GNESI, D. in print
Archaeological Landscapes through GIS (Cost Surface Analysis) and Virtual Reality. A case
study on the monastic site of Jure Vetere (Calabria – Italy). In: On the Road to Reconstructing
the Past. CAA 2008, Computer Applications and Quantitative Methods in Archaeology,
Proceedings of the 36th Conference, Budapest, Archaeolingua.
SOIL SURVEY STAFF 1975
Soil Taxonomy. A basic system of soil classification for making & interpreting soil surveys.
In: Agriculture handbook no. 436
SPURR, M. S. 1986
Arable cultivation in Roman Italy, c. 200 B.C. - c. A.D. 100. Journal of Roman studies.
Monographs, no. 3.
STALLMAN, R. 2002
Free as in Freedom. Richard Stallman's Crusade for Free Software. By Sam Williams
STALLMAN, R. – LAWRENCE 2002
Free Software, Free Society: Selected Essays of Richard M Stallman, GNU Press, Edited by
Joshua Gay, Boston.
STEINER, G. 2007
Le due culture: cause e forme di dicotomia. In: Ivano Dionigi (eds.), I Classici e la Scienza. 21–
32. Milano.
SURMAN, M. – REILLY, K. 2003
Appropriating the Internet for Social Change. Towards the Strategic Use of Networked
Technologies by Transnational Civil Society Organizations, Study commissoned by the
Social Science Research Council, Nov. 2003, Ed. SSRC,
www.ssrc.org/programs/itic/civ_soc_report/
SZABÓ, P. – MÜLLEROVÁ, J. in print
Ancient Woodland, Landscape Archaeology and GIS: Towards an Integrated Approach. In: On
the Road to Reconstructing the Past. CAA 2008, Computer Applications and Quantitative
Methods in Archaeology, Proceedings of the 36th Conference, Budapest, Archaeolingua.
Reconstructing Ancient Landscape S. Pescarin

TAYLOR, J. R. 1999
An Introduction to Error Analysis: The Study of Uncertainties. In: Physical Measurements.
128–129. University Science Books.
THOMAS, J. 2001
Archaeologies of Place and Landscape. In: Hodder, I. (ed.), Archaeological Theory Today,
165–186. Cambridge.
TILLEY, C. 1994
A Phenomenology of Landscape: Places, Paths and Monuments. Oxford.
TLAUKA, M. – WILSON, P. N. 1996
Orientation-free representations from navigation through a computer-simulated environment.
Environment and Behaviour. 25(8), 647–664.
TOMASELLI, R. 1973
Note illustrative della carta della vegetazione potenziale d'Italia (prima approssimazione).
Collana Verde 27, 1970: 1-64.
TORVALDS, L. – DIAMOND, D. 2001,
Rivoluzionario per caso: come ho creato Linux (solo per divertirmi). Milano, Garzanti.
TRAINA, G. 1990
Ambiente e paesaggi di Roma antica. NIS, Roma.
TURKLE, S. 1997
La vita sullo schermo (Life on the Screen, 1996). Apogeo, Milano.
UCKO, P. – DIMBLEBY, G. – TRINGHAM, R. 1972
Man, settlement and urbanism. London
UNESCO / FAO 1963
Bioclimatic map of the Mediterranean zone. Explanatory notes. Arid Zone Research XXI.
Paris
VAN ANDEL, T.H. – ZANGGER, E. – DEMITRACK, A. 1990
Land use and soil erosion in prehistoric and historical Greece. Journal of field Archaeology,
17, 379–396.
VAN DER LEEUW, S. 1994
L’Homme et la dégradation de l’environment. Actes des rencontres 20-21-22 octobre 1994.
115-141.APDCA, Sophia Antipolis.
VAN JOOLEN, E. 2003
Archaeological land evaluation. A reconstruction of the suitability of ancient landscapes for
various land uses in Italy focused on the first millennium BC. PhHD dissertation at
University of Groningen.
VAREL F., –THOMPSON, E. – ROSH, E. 1991
The Embodied Mind. Cognitive Science and Human Experience. MIT Press, Cambridge.
VASSALLO, V. – PALOMBINI, A. in print
Between Virtual Reality and Spatial archaeology: The Via Flaminia landscape
reconstruction. In: On the Road to Reconstructing the Past. CAA 2008, Computer Applications
and Quantitative Methods in Archaeology, Proceedings of the 36th Conference, Budapest,
Archaeolingua.
Reconstructing Ancient Landscape S. Pescarin

VASSALLO, V. – VICO, L. – MORO, A. 2006

The importance of the relief and the sources to interpret and communicate the Cultural
Heritage. In: From Space to Place 2006.
VERMEULEN, F. – DE DAPPER, M. 2000
Geoarchaeology of the landscapes of classical antiquity. International Colloquium Ghent,
23-24 October 1998. 49–70. Leiden, Stichting BABESCH, Supplement 5.
VITA-FINZI, C. 1969
The Mediterranean Valleys. Geological changes in historical times. Cambridge, Cambridge
University Press.
VITA-FINZI, C. – HIGGS, E. 1970
Prehistoric economy in the Mount Carmel area of Palestine: site catchment analysis. In:
Proceedings of the Prehistoric Society 36. 1-37
VOLPE, R. – ARNOLDUS-HUYZENDVELD, A. 2005
Interpretazione dei dati archeologici nella ricostruzione storica e ambientale del paesaggio
suburbano: l’area di Centocelle nel suburbio sudorientale. In: Roman villas around the Urbs.
Interaction with landscape and environment. Proceedings of a conference at the Swedish
Institute in Rome, September 17-18, 2004; Rome.
VORREN, K.-D. 1986
The impact of early agriculture on the vegetation of Northern Norway, a discussion of
anthropogenic indicators in biostratigraphical data. In: Behre, K.-E.: Anthropogenic
Indicators in Pollen Diagrams. Balkema, The Netherlands.
WAGSTAFF, J. M. 1987
Landscape and Culture. Oxford and New York.
WALSH, K. 1999
Mediterranean Landscape Archaeology and Environmental Reconstruction. In: Leveau, P. et
al. (eds.), Environmental Reconstruction in Mediterranean Landscape Archaeology. 1–8.
Oxford .
WEIBEL, R. – HELLER, M. 1991
A framework for digital terrain modelling. In Proceedings of the Fourth International
Symposium on Spatial Data Handling. 219–229. Zurich, 1.
WEST, D. M. – TRAVIS, L. E. 1991
From Society to Landscape: Alternative Metaphors for Artificial Intelligence. AI Magazine,
Vol. 12, No 2.
WHEATLEY, D. – GILLINGS, M. 2002
Spatial technology and archaeology: the archaeological applications of GIS. CRC Press
WHITE, K. D. 1970
Roman Farming. Thames and Hudson, London.
WIENER, N. 2001
Introduzione alla cibernetica. Universale Bollati Boringhieri, Torino.
WILCOX 1904
Farmer’s cyclopedia of Agriculture: a compendium of agricultural science and practice on
field.. New York.
Reconstructing Ancient Landscape S. Pescarin

WILLIAMS, S. 2002
Free as in Freedom: Richard Stallman’s Crusade for Free Software, Sebastapol. CA, O’Reilly
Media.
WILLIAMS, M. A. J. – DUNKERLEY, D. L. – DECKKER, P. DE – KERSHAW, A. P., –
STOKES, T. J. 1993
Quaternary environments. Edward Alnold, London.
WILSON, D. R. 2000
Air Photo Interpretation for Archaeologists. Tempus, Stroud.
WILSON, J. P. – GALLANT, J. C. 2000
Digital Terrain Analysis. In: Wilson, J. P. – Gallant, J. C. (eds.), Terrain Analysis:
Principles and Applications. 1–27.
WISEMAN, J. – EL-BAZ, F. 2007
Remote Sensing in Archaeology. Springer.
WRIGHT, H.E., JR. – KUTZBACH, J.E. – WEBB, T.III. – RUDDIMAN, W.F. – STREET-
PERROTT, F. A. – BARTLEIN, P.J. 1993
Global climates since the last glacial Maximum. University of Minnesota Press,
Minneapolis, 569 ss.
YOUNG, A. 1973
Rural land evaluation. In: Dawson, J.A. –Doornkamp, J.C. (eds.), Evaluating the human
environment. 5–33. Arnold, London.
ZEUNER, F. E. 1963
A history of domesticated animals. Hutchinson, London. 560 ss.
ZOHARY, D. – SPEGEL-ROY, P. 1975
Beginnings of fruit growing in the old world. Science, 187. 319–327.