Dialogues dhistoire ancienne [606574]

Dialogues d'histoire ancienne
The GIS in Cultural Resource Management
Madame Hélène Simoni, Monique Clavel-Lévêque
Citer ce document / Cite this document :
Simoni Hélène, Clavel-Lévêque Monique. The GIS in Cultural Resource Management . In: Dialogues d'histoire ancienne, vol.
25, n°1, 1999. pp. 222-233;
doi : https://doi.org/10.3406/dha.1999.2597
https://www.persee.fr/doc/dha_0755-7256_1999_num_25_1_2597
Fichier pdf généré le 16/05/2018

222 Chronique. Paysages et cadastres de l'Antiquité
III. ARCHÉOLOGIE SPATIALE ET OUTILS D'ANALYSE
1. The GIS in Cultural Resource Management
It appears rather peculiar to review two books that were published in
1990 and 1995 whereas ever since a relatively large number of edited volumes
on Geographical Information Systems (GIS) in archaeology have appeared88.
But as we are departing from the 1990s it might be worth looking back and
reminding some earlier stages of the application of GIS in Archaeology with an
emphasis placed on Cultural Resource Management (CRM).
The two books that were chosen to be reviewed here comprise a
landmark of GIS in Archaeology and provide information about the Geographic
Information Systems in relation to archaeological work. They represent the two
successive phases that GIS in Archaeology had passed :
1)Until the beginning of 1990 : At that time GIS was known to an
extremely limited number of pioneers who experimented via computerized
applications, especially in north America. Furthermore, they had the additional
task to introduce the fundamental elements and characteristics of GIS to their
colleagues and persuade them that they were still doing archaeology.
2)1990-1995 : By the mid 1990s GIS had become well known as many
people had heard or read about it and many researchers aware of the potentials
of this new tool in archaeology took over the responsibility to introduce it in
Europe and adapt it to the needs and trends of archaeological research here.
A third phase will be described later and will act as conclusion of this
paper.
Since the circulation of the two books other books have come out as well,
but these two books comprise a standard source of reference and are very
helpful to those who without having a previous contact with GIS in
Archaeology would like to learn something about it. What is common in the
88. Allen, K.M.S., S.W.Green & E.B.W. Zubrow (eds) 1990 : Interpreting Space in Archaeology : GIS and
Archaeology. London : Taylor & Francis (I.S.) and Lock, G.R. & Z. Stancic (eds) 1995 : Archaeology and
Geographical Information Systems : A European Perspective. London : Taylor & Francis (E.P.)-
DHA 25/1, 1999

Chronique. Paysages et cadastres de l'Antiquité 223
two books and could not be round elsewhere until the time of their publication
was the very structure of these two volumes. Although I will focus on the
treatment of issues regarding CRM in each one of the two books, both of them
contain papers on all aspects of GIS applications in archaeology, handle
theoretical issues, explain the function of the different systems involved, offer
methodological approaches and present various archaeological projects that
have made use of the new technology. What is more important is that both
books are readable, pleasant and easily understood even by those accustomed
to the traditional methods. This quality that cannot be taken for granted in
other volumes on computer applications makes the books vivid and attractive
along with the abundant black-white and colour pictures that are included. For
this reason both volumes can capture the readers' attention and convince them
to try to apply modern technology in their projects and experiment a new
approach.
Besides each book demonstrates the state-of-the-art in two different
regions, i.e. N. America (Interpreting Space) and Europe (Archaeology and
GIS) and are therefore representative of the different trends. Harris and Lock,
contributors in both volumes, name the « Interpreting Space » (I.S.) a watershed
that stands alone in importance because it records a heavy imbalance in early
applications in favour of the USA (E.P., p. 349). They notice the shift that
computer practitioners in Europe showed towards GIS following that book.
Their contribution in I.S. (1990) acknowledged the low rate of awareness
of British archaeologists on GIS and how influential the N. American
experience had been in the UK. They consider the likelihood of all those
services and offices at a national and regional level to beneficially adopt the GIS
techniques, given the already existent, very well documented sites and
monuments record in Britain. Integrating it with the GIS technology, it
facilitates the enquiries from planning authorities who want to assess the
impact on the cultural resourse arising from any permission for building
activities.
Differences in GIS applications between N. America and Europe are
guided by differences in the patterning of the cultural landscapes. Harris and
Lock describe the European archaeological record as spatially and temporally»
rich, denser and more complex than the N. American one (E.P., p. 353). That is
why CRM in Europe is so much more connected to Landscape Archaeology.
On the other hand the American experience with « huge tracts of federally
controlled lands » (Kwamme, E.P., p. 3) has led to the development and
DHA 25/1, 1999

224 Chronique. Paysages et cadastres de l'Antiquité
extensive use of spatial statistics and the building of predictive models in order
to prevent any catastrophic actions on potential archaeological sites.
Counting the papers that are directly linked to the issue of C.R.M., a
balance is observed between the two books : Chapters 3, 4, 5, 23, 20 in I.S. and
chapters 2, 3, 4, 26, 27 in the E.P. However, the different focus is obvious. Four
contributions out of five of the Americans link cultural resource management
with predictive modelling. The one left cannot be considered American at all. It
has been written by two British scholars and describes the potential use of GIS
in their country (p. 33-53). The European articles see the connection of GIS and
CRM, from the organisational aspect of it. What is done or can be done to
organize an integration of various archaeological data within a GIS. Legislation
is considered as well as internal regulation among archaeologists and other
users, as to how such data can be used and exchanged for the benefit of the
heritage. The creation of databases seems to be of high priority. The final
chapter hosts a critique on environmental determinism.
Although both books attempt to categorise the chapters most
applications presented are overlapping. The American volume contains more
theoretical chapters, describes GIS in theoretical and operational terms,
presents the GIS hardware and software, discuss GIS data sources and their
integration and provides a list of GIS acronyms and a list of federal agencies
who use and maintain GIS. The reason is obvious, as it was the first to come out
and the contributors had the additional task to present and explain things that
had never been written and assembled together for the purpose of one
collective publication before.
In this paper an attempt will be made to separate the most relating to the
issue of CRM and comment on the applications proposed by some of the most
distinguished GIS practitioners in Europe and North America. The
management of cultural heritage whether at an international or regional level
seems to be the key for the development of many archaeological projects
worldwide. The shortage of money spent on culture has never allowed the
archaeological units and offices to prepare a generous budget for their works.
As a result a very small portion of the budget involved pure research and the
bulk of money was guided towards actions necessary for the protection of
archaeology and history, usually after a serious threat had occurred. The most
obvious threat has been the building activity in rural or urban landscapes that
cannot stop because it is important for the local prosperity and progress.
DHA 25/1, 1999

Chronique. Paysages et cadastres de l'Antiquité 225
Cultural Resource Managers have the difficult task to reconcile the need
for protection of historical heritage and the need for development of a
landscape and its inhabitants. Therefore, excellent knowledge of the landscape
regarding its physical condition, evolution, economy, cultural background, as
well as the relation and the link of its inhabitants to it are essential. As a matter
of fact, any manager needs a huge data bank to store endless papers
containing all pieces of information that are available to the moment and
enough free space for the new ones.
Additionally, conflicts concerning the extent of an archaeological
research in a private field, the duration of the works, the delay of any building
constructions are frequent, as such works mean loss of money and time.
Complicated bureaucracy, bad organisation of the offered services combined
with picturesque faces of romantic archaeologists who work all day under the
strong sun with dust all over when contrasted to well-dressed executives of
planning enterprises equipped with their own laptops lose much of their
prestige that would otherwise suit them. Good communication and contact
with the public and the local planners means that you speak their language.
All the raw information that is eventually collected through the research
and the communication with the public or other services seeks for some sort
of manipulation and processing in order to be usable and meaningful. How
can somebody manipulate piles of manuscripts written in all sorts of formats
and containing so diverse information ? How quickly can someone access the
culminated information and choose only these parts that are useful ?
Obviously, coordinated CRM is significant and prerequisite for the
preservation of our own heritage and the development of our society. The
exciting changes that have taken place recently in the magnificent world of
computers are to be used very extensively and intensively in the near future by
archaeologists and that makes the application of Geographic Information
Systems, but also the information about it essential. As GIS is still a relatively
recent innovation in the field of archaeology many offerings in international
gatherings and collective volumes as the ones in question deal with the theory
and those questions that have received an answer in the propositions of a GIS
application. Green (I.S., p. 3-8) describes I.S. as an examination of the
interpretation of behaviour and material culture over space. To a certain extent he
views archaeology as a discipline involved in sampling space in order to
understand human behaviour. It seems that « space » and « behaviour » are two
words so natural and so abstract that have always been used and combined
DHA 25/1, 1999

226 Chronique. Paysages et cadastres de l'Antiquité
somehow arbitrarily without avoiding theoretical problems. Despite the
negative connotation of the word « arbitrarily » which no one likes, it is true
that certain frustration arises when we have to answer the seven spatial
questions that Green poses on the same page. The questions pertain to the
definition of terms such as « activity areas », « site », « behaviour », « cultural
and natural environment ».
The fact that many interpretations and analyses of human behaviour
rely on statistical calculations of visible remains of past activity due to the lack
of any other evidence leads scholars to different approaches as to the degree of
confidence we should show. Especially the environmental factor seems to
influence a lot the interpretations and the extensive use of environmental
variables (e.g. visibility, slope, orientation, distance etc.) to interpret spatial
archaeological patterning is questioned as to which extent should be allowed.
The debate between van Leusen and Gaffney (E.P., p. 367-82) shows how two
GIS practitioners react to the idea of such an environmental determinism.
According to van Leusen environmental determinism cannot be avoided
since GIS is inherently geographically defined and its mostly environmental
information that is released in some GIS format (Digital Elevation Model,
geology, pedology, hydrology etc.) and can be mapped. On the other hand, he
suggests the creation of cognitive models using those cultural variables that
have some spatial component direct or indirect and try to test their unmappable
consequences, e.g. locational preferences can be seen as reflecting aspects of
past social-symbolic and economic behaviour (p. 370). Regarding CRM, the use
of environmental models depicts an existing reality that calls for some sort of
protection or development and may be used to predict site locations under
similar environmental conditions. The need to understand and explain the
symbolic dimensions of a cultural resource cames second.
Gaffney does not agree that GIS models based upon the physical
qualities of a site can work as reliable examples for the prediction of other site
locations since such models are deprived of any qualitative description of what
comprises a site and any reasoning for the cultural development and this
information is totally ignored. He emphasises the need for « contextual analyses
that incorporates all levels of human activity and discourse in an
interpreta tional framework and does not simply rely upon the environment » (p. 375).
The debate arises from the inadequacies of GIS, but as a matter of fact
one cannot blame the tool but the user and the inadequacies of GIS can
be overcome when there are preset goals and well established methodology.
DBA 25/1, 1999

Chronique. Paysages et cadastres de l'Antiquité 227
Surely, the development of better GIS software in the near future will solve
many temporary problems.
The tormenting procedure of human brain to accommodate and analyse
spatial data is tremendously facilitated by the use of maps. Whether « mapping
– like painting – precedes both written language and systems involving
number » (I.S., p. 10 as stated by Harley and Woodward) or not it is certainly a
necessary accompaniment right from the beginning of any project. A map is
maybe the only way to view a site wholly and a series of good maps can
substitute significantly for the physical presence and observation of a
landscape. However, the construction of one map, let atone of a set of accurate
maps is a laborious and expensive task but does not lack theoretical limitations
as the ones mentioned above.
As long as the definition of a site remains unclear so does the quality of
accuracy and the details. The questions expressed above are to be asked while
mapping as well. The interpretation of behaviour and material culture over
space depends on the visualization of it. People who are involved in landscape
management like cultural resource managers, recreation planners, engineers
etc. are the most frequent map users. They all face the traditional dilemma of
Marble, i.e. : « how to insert the maximum amount of information into a given
map without either making it intelligible to the reader or increasing its physical
size to an unmanageable level ». The inadequate access to all the spatial data
that even a traditional and conventional map contains concerns the inherent
weakness of a map to perform complex queries, especially those which contain
a quantitative component like multiple measurements of distance, direction,
area etc. (I.S., p. 11). Marble gives examples and how GIS contributes as a new
and modern tool that increases the access to spatial information greatly (I.S.,
p. 9-21).
M. Forte (E.P., p. 231-238) defines the aims of scientific visualisation
techniques as to process digital images to produce a simulation model thus
increasing the information in the images. The technique consists of the
integration of a reliable Digital Elevation Model and digital images from aerial
photographs. The result product is a three-dimensional landscape model which
includes archaeological sites ; a model where the researcher can « move,
navigate and explore like a real landscape ». This paper puts an emphasis on a
so called intra-site navigation of a Bronze Age site and the settlement evolution
from Prehistory to the Middle Ages. Yet, the final chapter on future directions
describes the implications of such a technique towards environmental and
DHA 25/1, 1999

228 Chronique. Paysages et cadastres de l'Antiquité
cultural resource management and the creation of archaeological parks. The
data that can be acquired by such a model include slope, aspect, visibility from
a specific viewpoint and increase knowledge of archaeology and topography.
Wheatley (E.P., p. 171-185) introduces the cumulative viewshed analysis
(CVA), « a GIS based method for investigating intervisibility and its
archaeological application ». This function can be carried out after a digital model of the
elevation (DEM) of a landscape has been made as well as an additional map of
the locations of the relating sites. Any errors on these two maps will follow the
analysis and the final results. In any case the application of CVA will allow
the scholars to study the visibility of the landscape from a target point that may
represent a site or an individual monument. So it is possible to generate a
map that will show which other sites /monuments are visible from the
target point. Wheatley has used this operation at an attempt to understand the
spatial relationship among archaeological monuments namely the Neolithic
long barrows found in southern England. The idea of wider applications is
mentioned but not very well advanced in his paper. The importance of
investigating intervisibility will have very positive implications for those who
not only want to find ancient relationships of monuments, but who would like
to control a group of sites or monuments in a territory, to draw boundary
lines for better monitoring, to provide the best projection of a whole set of
monuments located round a central focus.
The creation of a data base that will contain all those pieces of
information spatial and non-spatial, quantitative, temporal is a major concern
not the least important and tiring than the design of a map. As has been
explained above, a database serves as a constant reference for its users and the
way that data are stored, the speed with which somebody can find the useful
ones, the ease to be updated are factors for an effective use of it. The choices
that the designer is obliged to make are not less than the ones concerning
maps. The inclusion of data in a data bank presupposes their transcription into
a codai language, a language that does not mean anything to the rest of the
world, but is easily deciphered by the users. Again choices must be made as to
the degree of detail and precision. The resolution of a map as well as the
resolution of a database reflect the objectives of the designer. Too many details
will be confusing, too broad information will be futile, too little will be
inadequate, too much will be unreadable.
The classification system of features must be organized so that all the
data in the database can communicate to each other and if possible different
DHA 25/1, 1999

Chronique. Paysages et cadastres de l'Antiquité 229
databases can do so, as well. GIS can provide this facility and combined with
the use of world wide gridding systems becomes a benefit. Arroyo-Bishop and
Lantada Zarzosa stress the « spirit of collaboration in multidisciplinary
studies » that prevails in archaeology and calls « intrinsic the need to structure
and formalise the fundamental way we record and store data » if archaeology
wants to enter the world of computerisation without the coexistence of myriads
of databases each one utilised by a restricted number of people who can
understand them. The need to adjust the traditional way of recording finds for
intra-site or inter-site analyses in order to become adequate for modern
computer applications is presented here and satisfied with the recording of
features as entities (archaeological, spatial, temporal) that represent the three
basic elements of an archaeological/historical study which are object, space and
time.
What is most useful in the GIS applications, the database со operates with
the maps and new maps are generated whereas the old ones are automatically
corrected, once the database is updated and vice versa. Certain manipulation of
the map (e.g. : interpolation, distance etc.) can create new maps and add to the
database information that we would never be able to have otherwise.
The general presentation of a database structure and its advantages
when in computerised form and integrated in a GIS application becomes
more specific in Massagrande's paper (E.P., p. 55-65). He handles data from
non-systematic field surveys which unlike the nature of a systematic
excavation suffer from many biases introduced by the random nature of
collection. He refers to the disadvantages of such a survey which are
outweighted however, by the low cost of its realisation and the wide coverage
of area. In his chapter, he introduces a methodology to gather and exploit
information that would otherwise be regarded as largely useless. He is obliged
to suggest a standard definition of what a site is, and a standard classification
of the sites in one area based on archaeological criteria, keeping in mind the
fact of the involvement of so many different people in so many different sites,
otherwise the creation of a database structure would face additional problems
especially in the process of analyzing the site distribution and making
comparisons or other statistical calculations. The database was structured and
all data were exported to a GIS for further processing.
In his paper, Jackson (I.S., p. 274-283) gives a detailed account of the
structure of historic era site databases, the particular problems of it and its
integration with GIS. The project was still in progress at the time he wrote the
DHA 25/1, 1999

230 Chronique. Paysages et cadastres de l'Antiquité
article, but the advantages had been obvious. The manual manipulation of data
and maps had become limited and they had created a system to help in
evaluating historic era sites for US National Register eligibility, thus assessing
their significance.
Farley, Limp and Lockhart (I.S., p. 141-164) integrate different types of
software for database management, exploratory data analysis, GIS and
remotely sensed image processing at three different spatial scales the multi-
state, the regional and the local in the U. S.A. In this way they prove how well a
system works at several scales provided that the users can make good use of
these differences in resolution. At the first level GIS is used for portraying
broad spatial patterns that can identify those rapidly developing areas where
threats to cultural resources can occur. At the second level, it concerns the
development of a state-wide geographic database to be used as a management
tool. At the lowest level, the relationships between environment and human
settlement location behaviour are examined with regard to two individual
project works.
Guillot and Leroy (E.P., p. 15-26) present the work of the Direction du
Patrimoine of France for the computerisation of the « Carte archéologique de la
France », the National Record for archaeology. The computerisation of it was
thought to be the most effective way of enhancing its function at a period of
rapid, rural and urban growth when the cost of archaeological surveys and
rescue excavations was growing as well. Having established in advance the
functional needs and the technical requirements, they selected a convenient
software that was to be used by fifty archaeologists, not skilled in majority in
sophisticated computing. The problems faced were not only connected to the
quality of the equipment and the limited training of the users. Soon, it was
realised how incomplete the archaeological record had been and how confusing
was the spatial definition of sites in a way that can be imported and handled by
a GIS. The experience gained by the application of the system at one case study
shows the need to balance the archaeological survey methods with the
geographic data of a region. Despite the limitations they confronted, related to
the limits of archaeological knowledge and of the tool itself, the preliminary
results demonstrated the potential for monitoring a developing area by creating
risk maps to be shared with the main agencies responsible for local
development.
Biswell, Cropper, Evans, Gaffney and Leach (E.P., p. 269-285) link the
application of GIS to the overall existing problems of limited financial support
DHA 25/1, 1999

Chronique. Paysages et cadastres de l'Antiquité 231
for the performance of necessary archaeological work with particular emphasis
on the excavation. As direct funding for site-based work has been curtailed to
33% at the staff of the 90's, nearly half the money required come from the land
developers. Of course, it is to the developers interest to pay as little as they
possible can, and to the curators' interest to investigate any archaeological
remains to the highest standard possible. At the end of the negotiation both
parties have to reconcile at the expense of the archaeological research,
methodology and data analysis. This has resulted in « a shirt in emphasis away
from the spatial and towards the more strictly chronological and functional » as
the latter are considered more economical. The authors propose the use of a GIS
that despite the limitations in its use and data input can handle spatial data and
they give an example of how « a combination of artefactual, environmental and
structural databases can contribute to our understanding of conceptual
arrangements within sites ».
It has already been mentioned how much gravity is given to the
predictive modelling for site-location in the USA. Seven chapters of
« Interpreting Space » bear witness of the successful function of this operation
in N. America. On the other hand, there is limited confidence or need of it in
Europe and a consequent absence in literature.
The main idea behind predictive modelling is the expectation to predict
areas sensitive to the presence of archaeological sites in advance of
development and plan the development phase of a terrain altering project, to avoid the
sensitive archaeological areas. A model is built inductively or deductively
through empirical observation or theoretical reasoning. But in practice there is a
simultaneous use of both practice and theory in the selection of those variables
whether environmental or cultural that will be used for the model.
The projects that are presented in the volume are built by people very
well aware of the limitations that this method may have and it seems that a
major concern for them is to test the effectiveness and suitability of each model
in a number of ways. There are various statistical techniques that can be applied
and they are mentioned in two papers (I.S., p. 22-32, 90-111). The complicated
terminology and sophisticated formulae should not scare the non expert users
as there are many available packages that will perform the necessary
calculations and give a meaningful result to be incorporated in a GIS analysis.
As with any other map deriving model, its reliability is heavily
dependent upon the raw data that are used, their resolution, the variety and
DHA 25/1, 1999

232 Chronique. Paysages et cadastres de l'Antiquité
independence of variables, the modules used to produce secondary variables
from digital elevation model, the size of the area on which the model is to be
applied. All the possible errors ( in surface modelling, operational, in feature
location) as well as a list of suggestions to be considered by the users are
mentioned in details by Marozas and Jack (I.S., p. 165-72). Similar stress on the
quality of data is put by Warren (I.S., p. 213) and Carmichael (I.S., p. 223).
However, the former lists the practical benefits when such models
applied to extensive, unsurveyed tracts of land, for archaeology who are given
not only images of the pattern of settlement in one area, but aise evidence of the
most important environmental determinants of site location. Land managers
are provided with expected distributions of the resources they are in charge
with protecting. Third, planners are provided with preliminary guides to the
places where cultural resources are least likely to be affected by future
construction projects (I.S., p. 202).
AltshuTs paper (I.S., p. 226-238) is very illustrative for those who want to
« manage large numbers of poorly documented resources and resources whose
location is not known ». His focus is not just the prediction of archaeological
sites located on broadly known spatial patterns but the so-called « red flags ».
They are sites located in unfavourable settings according to any typical
predictive model and may cause serious project delay when discovered. For his
analysis predictive models are not « an end product but dynamic analytical
trois » (p. 237) that save money and time and give new directions to
archaeological knowledge as well.
Similarly, Hasenstab and Resnick present a model whose goal is to
divide the project area into rive discrete strata in which varying levels of survey
intensity will be applied, according to the potentiality for containing
archaeological resources (I.S., p. 284-306).
As we have seen the new directions in CRM that are opened via GIS,
though some of them are still at the stage of experiment can make the difference
in a more effective way of dealing with CRM. The systems are user friendly and
their costs vary according to specific requirements. Besides, the final product
can show different aspects of the user's work some of which can be directly
communicated to the rest of the world, either to public and private agencies
who schedule the development of a region, or to the public who in this way
shapes a positive and respectful behaviour about archaeology even when their
immediate interests seem to be in conflict. The issue of accessibility and security
DHA 25/1, 1999

Chronique. Paysages et cadastres de l'Antiquité 233
of database and digitised information with regards to researchers,
governmental bodies, private agencies and the public will inevitably rise and
will need to be confronted with certain legislative steps such as the Freedom of
Information Act in the U.S.A. (I.S., p. 54-64). « How to look good and influence
people » is non just the title of an article by P. Miller (E.P., p. 319-333) but the
key to the expansion of Information Technology in CRM.
At the beginning of my review I mentioned the two earlier phases of the
development of GIS in archaeology and CRM. Since 1995 we have been
experiencing a third phase. GIS is famous and attractive. Everybody who
conducts a research wishes to include a GIS expert in the work group.
Although most people still cannot understand or have misunderstood how it
works, it is clear that the extensive use of GIS in other disciplines will give the
necessary push in CRM, as well. The cultural resource managers could borrow
some of the experience land managers have acquired over the years The biggest
challenge is the inventorying of antiquities and their compilation in State
Records as this is the only way to protect heritage. What has been happening to
a certain degree89 so far is gradually spreading at national and international
level all over in Europe.
Heleně Simoni
2. The "Cadastre Grid Software" to deduce possible Roman Cadastre grids
This paper describes a new GIS-based software, the 'Cadastre Grid
Software' that enables archeologists to manipulate archaeological lines, sites
and grids, in order to extract all the possible Roman cadastre grids, and select
the more probable one.
The use of modern tools, such as GIS and GPS, tend to be necessary to
almost every scientific work that involves cartography. The 'Cadastre Grid
Software' simulates the traditional methodology with modern tools. It offers to
the scientists the ability to process the archaeological data fast and accurately
using the same concept that archeologists are still using in order to extract the
possible Roman Cadastres.
89. For examples from several countries confer to : C.U. Larsen (ed.) 1992 : Sites and Monuments :
National Archaeological Records, Kopenhagen : The National Museum of Denmark, DKC.
DHA 25/1, 1999