In this paper we present some preliminary results of an investigation concerning the Sabina Tiberina area (Rieti, Italy). The area studied, which is in the amministrative territory of Magliano Sabina, belongs to the Sabina region on the Tiber river area, where a large stratigraphy shows human presence from Protostoric to Medioeval period. The aim of this study is an environmental and historical reconstruction of the territory in the time, by the morphological characterization of the selected area, the identification and location of all archaeological sites and through the study and design of a GIS for the management of all collected, elaborated and interpreted data sets. The GIS allows to produce 2D and 3D thematic maps including different types of information and to create specific DTM for zones of particular interest. These detailed plottings are produced through the kinematic use of GPS. A data bases containing all the historical archaeological knowledges has been also realized and will be presented in this work. It allows to carry out multiple queries and to manage a large amount of information.

In recent years the traditional drawing methods for recording a site or excavation have been supplemented by, in some cases supplanted by, newer techniques such as GIS and digital photography. It is the opinion of the author that while both these are quite obviously very useful tools, neither offer a total solution to the task of recording a site or excavation in three dimensions since the former typically deals with "low-density" data and the latter, while capturing "high-density" data, is effectively restricted to two dimensions. The author’s current doctoral research project seeks to find a recording technique which combines GIS with high-density data (and hence the "photorealism" of digital photography) acquired by range-finding devices, that is to say, by 3D surface modelling. We will discuss and evaluate various range-finding technologies and devices which are currently available, and show the preliminary results of our computer modelling of excavations. Widespread acceptance of such a technique by the archaeological community can only be expected if it is a significant improvement over existing methods and so we will evaluate the proposed technique by five criteria, viz. speed, precision, suitability for use in the field, ease of use and cost. Finally, we will outline the remaining portion of the doctoral project and invite your comment on this effort to realise a better recording system for the field archaeologist.

The needs of research and CRM are frequently at odds with one another. In the first case maximum accuracy of all records and open access to complete data sets is essential. In the second case differential access to data for the general public and resource managers is a regrettable necessity for protecting archeological resources. The potentials offered researchers and resource managers by the availability of cheap, powerful, and portable computers and GPS units is offset by the potential for abuses of these same tools to exploit archeological sites for personal gain. The legal and moral issues raised for a free society by this dilemma are not amenable to a technical fix. However, careful design of a GIS database offers an interim solution to the problem by simultaneously providing accurate information to the research and management community, while rendering sensitive information in a protected form for the broader audience. Keywords: GPS, GIS, Database, Accuracy, Security.

Digital image processing is a usual technique in archaeology. Archaeological images range from the microscopic to the macroscopic, and a diverse toolbox of computer techniques are available to process these data. However, the very nature of images as archaeological data has not been evaluated. Images are not primary data, but a transformation of empirical reality, translated into a language of luminance contrast. Images are then the result of a goal.oriented modification. But how this modification can alter the reliability of the analysis? Ver few studies have been published about this topic. Our goal in this paper is to integrate different archaeological applications of microscopy (use-wear in lithic tools, and pottery archaeometry) to define the observational category we are dealing with: texture. If the texture is the complex set of surface properties in an artifact, how we can describe it? What kind of archaeological, historical information we get from the analysis of texture? A related problem is that of image sampling. Digital image techniques have been applied in disciplines where the assumption of surface homogeneity is valid. But the modified surface of an archaeological artifact is always discontinuous. Where you should look at for describing texture? Different images can be obtained from the same artifact, and all those image may be different. Statistical sampling is then a basic problem in archaeological image processing, and very few studies have been made about. We are exploring the use of Neural Networks and related approaches to deal with this problem. Therefore, this paper deals with the use of microscopic images as a way to describe textures, and with the statistical analysis of quantitatively described textures. CONTENTS: 1) the concept of Texture, 2) Observing Texture: the creation of images, 3) Describing Texture: measuring images, 4) The variability of textures: statistical sampling within an artifact and between artifacts, 5) Explaining texture: experimentation and statistical modelling.

In a paper presented at the 1998 CAA conference Jackson and Baxter considered some aspects of variable selection in archaeometric data analysis. It was argued that there would often be merit in using just a subset of what is often a large number of measured variables, as this often offers analytical, presentational and interpretive advantages. From the particular perspective of scientific studies of ancient glass it was noted that variable selection could be dictated by a variety of criteria, and that criteria driven by purely statistical considerations were not necessarily the best to use. The present paper will review, briefly, and expand on some of the reasons for being concerned with variable selection in a more general setting than that adopted in the earlier paper. A formal, and statistical, approach to variable selection - developed in the statistical literature but not previously applied to archaeological problems - will be described and illustrated on some glass data sets. Notwithstanding our previously expressed scepticism about such approaches, the methodology appears to perform quite well. The potential for considerable reduction in the number of variables used is demonstrated clearly and, in the context studied, the variables selected usually make sense. This does not obviate the need for the exercise of archaeological common sense in applying statistical analysis to complex multi-variable data sets, but does hold out hope that the statistical methodology described may be generally useful. If time permits other areas of potential application will be discussed.

The scientific analysis of the chemical composition of ceramics naturally leads to multivariate data that is often explored using techniques such as principal components analysis and cluster analysis. The general aim is to identify groups of chemically similar artefacts. Separate groupings could be assumed to indicate, for example, distinct origins or processing recipes of the raw materials (clay) used to make the artefacts. In this paper we aim to show how, with the aid of powerful statistical software such as S-Plus, traditional exploratory multivariate analysis can easily be used alongside, or in combination with, a technique designed specifically for grouping ceramics by chemical composition (Beier and Mommsen, 1994). This latter technique involves grouping together artefacts whose chemical compositions are "close" with respect to a mathematical measure of dissimilarity. The measure used takes into account uncertainty of measurement and the possibility of constant shifts in the data due, for example, to dilution of the clay or instrumental variation. We shall demonstrate the ease of implementation in S-Plus of this methodology and show how graphical exploratory techniques can also be used to create an integrated approach to the grouping of ceramic chemical compositional data. The methodology will be applied to real data of this type.

The Structural Documentation System for Buildings supports the systematic recording of basic and detailed characteristics of monuments and listed buildings, the classification of historical, structural, typological and morphological data and the documentation of alterations. More specifically the system supports the development of scientific knowledge base for the detailed description and the complete and systematic documentation of morphological and structural characteristics (history, typology, morphology, structural typology, material, construction pathology) of listed buildings as described in historical, archaeological, conservation and restoration studies. In addition, the system provides to the user views for comparing, validating and controlling alterations. The system offers automatic connection with cartographic documentation and computer-aided design systems, as well as administrative documentation systems. The requirement analysis of MAISTOR was carried out in close co-operation with the 7th Ephorate of Modern Monuments, 13th Ephorate of Byzantine and Post-Byzantine Antiquities, Institute of Mediteranean Studies, and the Sitia Development Organisation SA. This paper presents an outline of the architecture and functionality of the above system. Also it addresses the notions of spatial and temporal data, cad data, conservation and restoration data, complex objects and documentation sources and introduces a data model, which is -to a high degree- extensible by further analysis without the need for restructuring of the original model. Furthermore, the model allows the creation of dynamically relevant views under different primary aspects.

The Directorate of Cultural Heritage of Norway is the owner of several big and smaller national registries of sites, monuments and buildings. We want to present some of the work in progress to establish these databases on a GIS-platform. Among the databases is the Registry of Arhaeological remains, where the The Foundation for Nature Research and Cultural Heritage Research is responsible for updating and developing into av GIS-based database in collaboration with the Diretorate. The represent tive from the Directorate will give a brief presentation of the contents and legal implications the database will have, who is going to use it and for what purposes it is designed, while the research Foundation will give a demonstration of the new database. The interface between this Registry and the big databases owned by the archaeological museums will also be shown.

When intensive surface field survey took off as a major landscape research tool in the 1960s and 70s, it was considered a rapid and relatively simple means to reconstruct settlement system dynamics for large areas and over long periods of time. The developmental trajectory of surface survey has seen a necessary reduction of scale in the area surveyed and the confidence in maps produced. This is the result of the increasing recognition that surface artefact collections are as taphonomically complex as excavation assemblages. Fortunately the arrival of GIS has offered a powerful aid for the analysis of surface finds at the on-site and off-site level, just at the time when the sheer quantity of the available data from the new phase of ‘total surface survey’, and its highly-problematic character, have raised a challenge to surface surveyors as to whether a reliable overview of ancient landscapes is realistically-achievable. This paper will introduce a newly-developed methodology for micro- and macro-scale analysis of continuous artefact surfaces utilising GIS, currently being deployed on the Boeotia Survey, Central Greece.

The Stadtarchäologie Wien is in the fortunate position, of being able to look back on a long history of research. Since the 16th century archaeological remains have repeatedly been found in Vienna, particularly in the city centre. Since the end of the 19th century a large number of excavations have been both detailed in writing, and recorded with sketches. Of importance here are over 1000 sheets of notes, that were written between 1895 and 1915. The interpretation of these old finds and features was only practically possible through the use of EDV (electronic data processing). Therefore, in the last few years, a comprehensive GIS plan had to be drawn up and brought into use by the Stadtarchäologie Wien (Vienna Archaeological GIS). Firstly the sites of archaeological discovery had to be located on the present-day map of Vienna. This task posed a problem, because the cityscape has changed so much since the end of the 19th century, especially after the demolition of the city’s defences in 1860. Because of this, details from the land register had to be used instead. The city of Vienna and the Republic of Austria can draw from their large collection of digital and analog plans. The "Franziszeischer Kataster", a register, which was commissioned by the imperial government around 1819, for the purpose of taxing the inhabitants of the monarchy. This plan is very exact and was recently digitized, transformed and overlaid with the modern-day digital city map of Vienna. This basis enabled the registration of finding from the old excavations. The next step was to draw up and plan the framework for the database. This database was created in ACCESS and linked up to the CAD-programm AutoCAD-Map. In the foreseeable future it will also be possible to call up a light-version of the GIS-application on the internet.

The purpose of this paper is to present the experience of an archaeological museum about the management of museums collections. We begin the use of computers for managing collections about 10 years ago with a system developed in 4th Dimension where we could already see important ideas of how we should look to documentation and information in museums. With the development of international standards, communication protocols and controlled vocabulary we decided the development of a new system, in the activities of Project GEIRA that is an inactive of two Universities located in the North of Portugal and it’s main objective is the use of Internet and CD-ROM to present information about Portuguese cultural heritage of the North of the country. This new system, InPatrimonium, developed by a Portuguese company, besides the use of international standards allows the innovative use of exploiting the relations between information about the collections and the archaeological sites where they have been found. We will also analyze particular problems of small and medium size museums in their relation with technology and how we tried, with Project GEIRA, the improvement of the use of technology. We will present some of the main activities of GEIRA and their results.

The plans generated from surveys form the backbone of much archaeological research. They create a basis for future fieldwork and for understanding the site within its landscape in a way that is not possible on the ground. However, the robustness and applicability of past surveys are often questionable in terms of original motivation and priorities, with subsequent questions of technique, scale and depiction. Despite this, previous surveys are constantly used as the foundation for subsequent archaeological work without considering matters of accuracy. As such surveys are reproduced and re-used for later work, but without consideration for their true applicability. This paper demonstrates the potential inaccuracy present within archaeological survey in terms of both the original methods, representation and scale, and their reproduction and re-use for the purposes of subsequent research. The site considered consists of earthworks that were surveyed in the 1930s at an unknown scale. Since then, this resulting plan has been used as the foundation for excavations during the 1980s and 1990s, particularly since much of the site had been destroyed in the intermediate years. The site was recently surveyed at high resolution and at centimetre accuracy using Global Positioning System (GPS) equipment to generate an Digital Elevation Model (DEM). This model provided the basis for two seasons of excavation that highlighted the positions of features identified on the earlier survey. However, when the earlier survey was scaled and digitised, the aspect ration of the plan was found to be incorrect and no accurate fit could be obtained. This research demonstrates the hazards of using earlier surveys as the foundation for future research. The errors encountered within this work highlight the magnitude of error that can have a significant effect on the positioning of later excavation work.

by Otto Cichocki and Bernhard Knibbe Within the long term scientific project "The Synchronization of Civilizations in the Eastern Mediterranean in the Second Millennium BC", carried out by the Institute of Egyptology, University of Vienna and sponsored by the Academy of Science, Vienna and he Grant for the support of scientific research (FWF) the dendrochronological investigations of wooden artefacts (construction elements of buildings etc., coffins, panels, furniture, objects of art) intend to establish floating chronologies for different wood. They will determine the relative range in time of objects relevant to a relative chronology of certain historical events and further on allow the construction of standards for absolute dating. As most of the wooden objects are on display in museums and of major cultural importance it is impossible to use classical sampling techniques such as drilling cores or cutting disks. Since no alternative products were available, the first task of this project was the development of new approaches towards non- respectively less destructive sampling methods. Two contact free devices are already in use. If the situation allows measuring directly on the object surface by a step-motor-driven video camera, data can be processed and analyzed on location, thus allowing for immediate error correction. A similar appoach utilizes modern digital scanning devices for the measurement of unpainted wood surfaces with satisfying contrast of ring borders. The images are semiautomatically analyzed by a plugin module to the new PAST32 software package. For painted objects do not allow major preparation work on their surfaces a method is in development using inspection of specially preparated small boreholes (diameter 4,5 mm) with help of an endoscope, high resolution video cameras and video frame-grabbr cards for data acquisition. Again the data analysis is done by a PAST32 plugin module.

This paper summarizes a newly developing program for digital archiving of archaeological materials. The program uses a Minolta Vivid 700 non-contact, color, 3D digitizer to create accurate wire-frame models with detailed surface texture and color. The digitizing process requires multiple scans of each object. Using sophisticated 3D modeling software known as Polyworks, the scans are then aligned, merged, and edited to create a complete model, accurate to within 0.5-1.0 mm. Collections of 3D models will be archived on a dedicated server with associated relational databases using Oracle 8i. Each database contains the digital data for the object model along with other fields covering an array of data on object attributes an contextual information. These databases will be accessible via the Internet and Internet2, so that researchers anywhere my browse the databases and analyze their contents. All database entries will be linked relationally and presented in tree hierarches (or directories) according to taxonomic fields for ease of searching the database. Two archiving examples will be highlighted in this presentation: an archive of lithic artifacts, and an archive of hominid cranial endocasts. The portability of the Viid 700 means that it can be taken to distant artifact repositories where specific collections can be scanned and added to the databases. Digital databases will streamline information management, storage, and retrieval processes. Furthermore, such digital databases will significantly improve data analyses, modeling, interpretations, access, and curation in archaeology. While archives of this sort are not yet available, we have no doubt that 3D scanning, digital-model analysis, and digital curation will be essential components of archaeological and preservation management programs in the not-too-distant future.

The application of GIS to archaeological research has primarily facilitated landscape analysis up to the present moment. Can research concepts developed for landscape analysis with GIS be converted for "sitescape" analysis? Exactly what sort of analyses can be performed for intrasite research utilising a GIS? A preliminary set of questions arising from this type of research is presented for the site of Akrotiri on Thera. The relatively well preserved state of this site offers a number of reseach options. This paper presents results of the analysis of the site's plans, using MapInfo in order to determine aspects of spatial organisation within the site.

I briefly review the history of quantitative methods in American archaeology and assess their present status. I then identify topics that seem likely to be most active in the near future and express my opinions about the directions that I think will be most productive over a somewhat longer term. What topics, what methods, and what general approaches should be given special attention and encouragement? What roles should quantitative methods play in the archaeology (or archaeologies) of the 21st century?

Henges have long been interpreted as centres of prehistoric ritual for growing communities in the Later Neolithic. During fieldwork a hengiform site in Co. Sligo Ireland was noted to have acoustic properties similar to a Greek Amphitheatre. It is possible this henge was designed so that principal participants (actors) could be heard by the wider community (audience). This has implications for the study of prehistoric social structure as well as the origins of theatre. Since the site has suffered erosion it is difficult to test this possibility directly. An interdisciplinary project was designed to create an accurate digital terrain model of the site and test the acoustic properties of that model with different assumptions about degradation over time. While the 3D modelling of archaeological sites is fairly commonplace, the application of acoustics software involves a broader reach for computing in archaeology.

Modern excavations create huge amounts of digital information. Whether it is the on-site recording of the archaeology, specialist databases created during post-excavation or publication standard interpretative maps and plans, digital information has the potential to be created at every stage from assessment to publication. Within the discipline there has been an increasing awareness that this vulnerable data is as much a part of the primary site archive as artefacts and paper records. It has equally been recognised that data creators and traditional archives are under-resourced and ill prepared to accession and effectively archive digital information. The Digital Archiving Pilot Project for Excavation Records (DAPPER) arose from these concerns. DAPPER was a collaborative venture between the Archaeology Data Service (ADS), English Heritage (EH), the Museum of London Archaeology Service (MoLAS) and the Oxford Archaeological Unit. The Pilot Project aimed to prove the concept of digital excavation archives in order to inform the development of best practice in this emerging field. DAPPER provided access to two full digital excavation archives and quantified the efforts and costs involved in their production and dissemination The pilot project also wanted to encourage the re-use of these digital repositories and explored the use of the Internet to enhance access to archaeological data. DAPPER has created the world's first on-line digital excavation archive; this paper will critically assess the project's success and offer a future vision for archaeological digital project archives.

From a landscape archaeology perspective that takes advantage of GIS investigative tools, the debate regarding the definition of reachable and proximity areas related to sites and settlements is still currently underway. The adoption of models and mathematical functions that take into consideration the interaction between space and human beings is one of the more stimulating issues of GIS application in archaeology. It appears necessary to develop models that take into consideration the entire complex of variables that participate in the definition of human behaviour in the real world. This paper focuses on one of these variables, particularly relevant in a prehistoric context, and which constitutes human effort in walking within the landscape. The latter is calculated in terms of energy consumed related to movement on plane or sloping areas both upwards or downwards. In this study the model adopted calculates the cost of movement through space from a prehistoric settlement perspective which takes into consideration the effort to move away from a site but also return to it. For example with regard to a hillside settlements, a hunting place or working site located at the bottom of the valley is not necessarily more advantageous, in terms of cost, then one located at the top of the hill, since in each case a return journey ‘home’ is implicated. It is likely that return journeys will occur to and from the settlement throughout the course of daily activities. In order to obtain a ‘friction to movement’ surface, a mathematical function related to a linear path has been modified to be used on a raster grid. Subsequently, a ‘round trip’ criteria has been adopted to calculate cost distance from prehistoric settlements. In particular this methodology has been applied to a few Neolithic settlements in a study area of the Larino territory (Molise, Italy).

An attempt of a state of the art concerning artefact Analysis is given, based mainly on the results of the world-wide archaeological research since forty years. The report begins with a brief history of the concept of artefact since the XIX° century, in association with their archaeological knowing. Then a semiotic theory of artefact Analysis is proposed, based on the interaction between artefact and archaeologist, and developed around the different meanings of perception, authentication, naming, characterisation, encyclopaedic knowing and recognition. The artefact description is formalised, using item vocabulary, syntactic relationships, different intrinsic meanings, various measures or variables, and quantification of the attributes, for a statistical approach of similarities and dissimilarities between artefacts. The concept of type is then analysed, in all its complexity : real or virtual types, intrinsic or extrinsic types, and further the concept of artefact taxonomy. In the second part of the report, the use of the artefact in the archaeological constructs is listed and criticised. It concerns mainly artefact taxonomy (or clustering), "culture" identification, seriation, intrasite spatial analysis, raw material identification and artefact circulation from manufacturing sites, etc. The mathematical approaches of the previous structuring constructions are described, from elementary statistics or graphics to data Analysis, using factor analysis and clustering methods. The data processing environment (hardware, software and packages) is given. The main contributions of different scholars in the seventies and the eighties is summarised. These theoretical approaches are the same in Archaeology when applied from an artefact to a set of artefacts : for example, a unique artefact (a tool), a set of identical artefacts (the stones of a wall), a set of different artefacts (burnt stones, charcoals and ashes of a combustion system), a system of different sets of artefacts (dwelling structure), etc. It is then possible to integrate the artefact system analysis in a more global reconstitution, like for example the spatial analysis of the peopling of a territory by the means of a G.I.S tool, where the same structuring statistics are embedded in the software G.I.S. package. Nevertheless, the concept of artefact is different according to the different Archaeologies: Classical Archaeology from History, Prehistoric Archaeology from Natural Sciences, New Archaeology, Post-Processual Archaeology, Cognitive Archaeology. These different approaches have involved different methods and various interest and role in artefact analysis, which need to be analysed. In conclusion, if there is no archaeological reconstitution without artefact analysis, its cognitive potential is directly associated with the good use of Semiotic, Statistics and Computers applied to a cognitive interaction between an artefact and the archaeologists. Hopefully, archaeologists have never stopped meddling with things that they know nothing about ! Even and especially when they were not specialists of semiotic, statistics and computers.

Controlled terminology becomes more and more important for the electronic documentation of archeological object description and its consistent retrieval in wide area networks. Even though there is a vast literature and established practise in reating thesauri, the logical organization, the "metastructure" of a thesaurus is still rather unreflected, in particular in the more abstract layers. So-called "major" and "minor" facets, guide terms or node labels (ISO2788), are distributed intuitively or in reaction to the growth of the authority in order to group a set of sibling terms by an implicit property or aspect of documentation these terms xpress. As facets and node labels are not eligible terms for classification, but rather classify the terms themselves, we regard them as elements of one possible metastructure of a thesaurus. The Greek Ministry of Culture is developing a general purpose thesaurus for archeology in cooperation with ICS-FORTH. This thesaurus aims at classifying consistently traditional terms and at providing an indexing language for object documentation activit es of the Ministry. Based on contextual and phenomenal notions, an extensible, systematic metastructure for terminology in archeology is proposed, which seems to be more comprehensive for the expert than traditional approaches. This paper discusses the benefits of such an approach for the three tasks of a thesaurus: (1) to assist users to find correct terms, (2)to clarify the meaning of terms and (3) to provide an indexing language for objects of documentation. In particular, the proposed structure can provide a better understanding of the implicit criteria intrinsic to object classification terms and the relation between compound expressions and traditional terms. Finally it is argued, that more effective implementation scheme than ISO 2788 are needed to capture so rich semantic structures.

Graeme Earl (Department of Archaeology, University of Southampton) Sarah Poppy (GeoData Institute, University of Southampton) Nick Ryan (The Computing Laboratory, University of Kent at Canterbury) Mobile computing with context awareness has been shown to offer great potential for the recording of spatial data. In particular, a GPS and palm computer configuration provides an intuitive means for recording spatially referenced data without the rigidity imposed by traditional survey approaches. Past work by the Mobile Computing in the Field group, and others, has demonstrated the potential of this approach in a broad range of fieldwork contexts. Located in Rome's northern hinterland, the Tiber Valley, the town of Falerii Novi was founded in the 3rd century BC. Although currently under cultivation, a few excavated remains and survey results demonstrate this 30 hectare site to be a well-preserved example of Roman urbanism. In conjunction with a program of systematic field survey at Falerii Novi, software developed by the MCFE group has been tested in real-world, archaeological applications. This paper discusses the benefits gained from the methodology, in particular the freedom and speed improvements introduced by context awareness. Three applications are discussed in detail. Firstly, the use of palm computers to provide a note-taking facility to aid fieldworkers in the course of survey activities. Here information is collected and stored in a logical, consistent and spatially referenced format for easy integration within database and GIS applications. Secondly, the paper describes a fieldwalking strategy based entirely on context awareness, with a sample of surface observations generated very rapidly for the entire town area. Finally, these data are analysed and compared to data gathered through conventional means, and interpreted with reference to site layout and chronology.

One of the basic assumptions behind archaeological interpretations of the results of field walking surveys is, that the recorded materials represent the ancient landscape. In some areas this is clearly not true. In the current case study the author uses historic elevation data of a part of the Pontine marshes (Lazio province, Italy) in comparison to a more recent DEM, to assess the influence of earth moving activities since the late 1920's. Small-scale changes can thus be related to the results of a detailed field walking survey conducted in the same area by the RPC Project team in 1998-9.

This proposal comes from a self-confessed computer illiterati, who is fascinated by the way things move from the cultural present to the past. It will consider computers and software as artefacts which can be - and in deed are - being conserve as museum exhibits. Some things are too new to be on display, but nevertheless out-of-date with regard to their original context. The paper will draw on research into companies and institutions and interviews with those who are selecting what should be preserved of the computer age.

Digital technologies may be very helpful in the graphic documentation of paintings, not only for the inherently more detailed and faster method of drawing, but, in particular, for the 3D representation of data (structures and materials) in real time, that is, the possibility to have many dynamic views of the same model in order to visualize and interpret features, shapes, and state of conservation of structures and materials. Moreover, it is very important to highlight that paintings or mosaics are inteligible as 3D objects rather than 2D objects, since they represent solid objects, after all. Then, using GIS technologies for collecting data, it is possible to process the relative information in detail. So we have chosen to consider the walls and the polychrome mosaic vault of the "Cristo Sole" Mausoleum (III cent. a.D.) in the Vatican Necropolis. The mosaic presents a small tessera decoration (of which only a fraction is still in place), on the walls add on the highly depressed vault; where the tesserae are missing, the mosaic preparatory surface still enables us to interpret the original decoration.. The main goal of our project is the virtual detailed 3D reconstruction of a micro-model of the above mentioned decorations through the use of digital techniques, in order to have a final "cognitive" representation. To obtain this, we have implemented new tools for the acquisition of detailed 3d surfaces. The general positoning (georeferencing) was achieved thanks to a Leica TCR 1103 total station; this instrument also enabled us to obtain vast amounts of surface data for the 3D model thanks to its reflectorless Laser EDM (Electronic Distance Measuring) range finder. Where more detail was necessary, further surface data was gathered using a Microscribe 3D Mechanical Arm Digitizer coupled to a laptop computer. The project involves these main approaches: digital acquisition of surfaces, digital documentation of mosaics and paintings, 3D virtual reconstruction of models, 3d virtual communication of the information by VRML metaphors through Internet.

The Archaeological Expedition at Aksum (Ethiopia) of the Istituto Universitario Orientale, Naples (Italy), and Boston University, Boston (USA) resumed investigations at Bieta Giyorgis, a hill to the NW of Aksum in May-June 1993. Six field seasons have been directed by Rodolfo Fattovich (IUO) and Kathryn A. Bard (BU), in 1993, 1994, 1995, 1996, 1997 and 1998. The general goal of the IUO and BU Expedition is to study the development of complex societies in Tigrai from late prehistoric (3rd-2nd millennia BC) to medieval times (14th century AD). Earlier investigations have focused on the rise of complex societies in the western Sudanese lowlands (Kassala, Sudan) in the 3rd-2nd millennia BC, and the pre-Aksumite state on the Tigrean plateau (Yeha, Tigrai) in the 1st millennium BC. The current stage of the project is mainly interested in the rise and development of the Aksumite state (late 1st millennium BC- early 1st millennium AD). Particular emphasis was also given to the study of the involvement of Aksum in the trade network from the Mediterranean to the Indian Ocean. Excavations at Bieta Giyorgis were aimed at testing the hypothesis, based on traditional Ethiopian sources, that the hill was an area of early development at Aksum. An important goal of this project is to investigate the origins and urban development of Aksum within its environmental setting. The project includes research in archaeology, paleoethnobotany, archaeozoology, ethnoarchaeology, history, geology and geomorphology, digital technologies as well as systematic mapping and conservation. In particular the paper will present the preliminary results of GIS and Remote Sensing applications concerning the area of Aksum, comparing in 2D and in 3D digital data such as: aerial photos (1:10.000), satellite images (Landsat TM, SPOT XS), cartography, landscape documentation.

The concept of metadata – or ‘data about data’ – is rapidly taking root in the museum and archaeological world. Having originated as an extension of traditional library practice into an electronic environment, metadata description is now seen as a way of facilitating access to the rapidly proliferating information sets being generated by archaeological fieldwork and research. In Britain, for example, the Dublin Core standard – which was designed with an eye to simple resource discovery – is being implemented by the Archaeology Data Service. In this paper I shall explore some of the assumptions and philosophy that underlie standards such as Dublin Core, in order to consider how satisfactorily bibliographic practices can be applied to the description of archaeological data sets. To what extent, for instance, does the metadata record itself – by definition, selective and rigidly framed – embody unacknowledged biases, restrict multivocal interpretation and discourage, rather than encourage, research? After this, I shall turn to examine how current metadata systems might be extended into the fields of digital preservation and data management. This will include a review of several projects in progress in north America and Britain. The paper will draw throughout on the author’s experiences in devising standards for the deposition of archaeological archives with the Museum of London (http://www.museumoflondon.org.uk/archguide.htm) and in designing access systems for the London Archaeological Archive and Research Centre (a Heritage Lottery-funded project).

This report will illustrate how some Swedish archaeological surveys are performed, using modern methods for determining the positions of the objects. The development of the DGPS-system has a high degree of accuracy and makes it possible to locate objects in the range of 0,6 to 1 metre. It enables you to use free differential correction broadcasts from government-established navigation beacon reference stations around the world. The area surveyed is situated in the Dannemora and Film parishes in the Östhammar district. In earlier times this region was a very important centre for mining. Numerous remains of this early industrial activity have been recorded here and also some foundations of old manors. Furthermore many prehistoric remains have been recorded, as graves and burial fields from the Bronze- and Iron Age, mounds of fire cracked stones, ancient field systems etc

A Project of the Department of Bible and Ancient Near Eastern Studies of the Ben-Gurion University; Sorin Hermon and Patrice Kaminsky Ben-Gurion University of the Negev, Israel. The purpose of the lecture is to present a new project of the Archaeological Division of our university, its aim being the building of a large data-base of main archaeological sites in Israel, starting with the archaeological projects of our university. The database includes a virtual reconstruction of main architectural features and archaeological sections, the information regarding the excavation of the sites, with emphasis of main features in each site and a virtual reconstruction of the context (3D spatial distribution) of the selected artifacts. The selection of the sites is done according to their coordinates, or navigating through the map of Israel. As soon as a site is selected, a zooming option enables the user to navigate through the site and choose desired features to be shown, either in 2D (plans) or 3D model. Upon availability, maps, photos or inventory lists related to the selected features are presented. Several programs are used: 1. AutoCad for mapping, zooming (through layers) and presentation of 2D architectural plans and 3D models of sections and main architectural features. 2. PhotoShop for adjusting the relevant pictures related to the features of the archaeological site and for the preparation of the plates, presenting selected artifacts. 3. Microsoft Access for building the database of the inventory of the archaeological artifacts. 4. EndNote for the preparation of the relevant bibliography. 5. CorrelDraw for the preparation of vector lines of the architectural plan drawings. The importance of the project is in: 1. Presentation of a large database concerning the main archaeological projects in Israel. 2. The initiative to combine traditional presentation of archaeological projects (2D) with 3D visualization 3. Accessibility to a large amount of information (photos, sections, inventory lists, etc.) concerning the excavations and their findings, embedded in the visualization model.

The purpose of the experiment was to evaluate the potential existing in image processing techniques for the analysis of attributes of flint artifacts. The case study was a sample of one hundred sickle blades from the chalcolithic period of Southern Israel. During the experiment, several steps were evaluated, in order to obtain maximum information regarding the morphology of the analyzed artifacts: 1. The location, the angle and the intensity of light source, in order to catch the contour of the artifacts, in order to reduce information corrupted by background and shade. 2. In order to obtain information regarding the surfaces of the artifacts (which correspond to scars of previous removals on the dorsal face of the items), several sequences were taken, from different angles. 3. The problem of the influence of the ambient source of light was evaluated. 4. The optimal source of light 5. How images should be taken (distance from camera, sequence of frames, background, saving format of images). Following this experiment, the most suitable conditions for taking the images were defined, and several characteristics of the morphology of the flint artifacts were defined, for an appropriate comparison and classification of the attributes to be analyzed. The software used for the image processing was VTK, using TcL script, the images taken were prepared with PaintShop Pro, both softwares being in the public domain. The lecture will focus on the various stages of the experiment, the archaeological implications of it, its advantages and its limitations.

Excavation of the 70,000 year old Late Levantine Mousterian occupations at the Tor Faraj Rockshelter in South Jordan point to behavioral organizations resembling those of modern humans. Three dimensional pin-point plots of artifacts, features, and related evidence were collected with a laser theodolyte linked to a data storage unit. (Sokkia Set-6 Total Station). These positional data were downloaded to a PC on a daily basis and subsequently examined with several software programs for spatial analysis. This allowed for rapid evaluation of the contextual relationships of data exposed across successive living floors. Contextual relationships of artifacts and hearths show that the hearths were synchronously used and positioned to provide a "sleeping corridor" between the fires and the shelter's sun-warmed backwall. The shelter's setting, exposure, and opal phtolith data are consistent with a winter occupation. The artifact distributions also define concentric zones of "drop" and "toss" areas tethered to the hearths. Spatially sensitive wear-pattern studies of tools and artifacts in different areas of one occupation floor support interpretations of consistently placed discrete 'activity' areas. Significantly, these contextual patterns are similar to those reported for Late Paleolithic and ethnographic groups from various parts of the world.

Electronic publication has become an accepted part of archaeology. It is seen as relatively cheap and easy to produce and gives a rapid result. As well as being a widely accessible resource, but is it being used to its full potential? This paper will address the issues which have been created by electronic publication in archaeology; how technological capabilities rival multimedia possibilities, and speed and ease of publication have a bearing on the quality of presentation. Is it good enough to use the Internet for dissemination alone, or should we be focussing more upon the possibilities available through this alternative medium. The above points will be illustrated using an electronic excavation report of the excavations of Cottam B, East Yorkshire.

In this paper we will outline the design principles of an Internet-accessible metadata clearinghouse for spatio-temporal datasets conforming to the TimeMap data structuring standards. The clearinghouse allows the discovery of datasets and seamless querying of the data from multiple servers anywhere on the Internet, both through textual database search and through display of geographic data in the TimeMap viewer software (TMView). The paper will examine the strengths and weaknesses of the snapshot-transition model adopted by the TimeMap project, and the extent to which existing datasets can be regarded as conforming to the model and made accessible to TMView through appropriate metadata. The paper will draw on practical experience gained through the implementation of an SQL server-based metadata clearinghouse for the Electronic Cultural Atlas Initiative and the development of TimeMap-compatible datasets registered with the clearinghouse. It will illustrate both the web-based and standalone metadata editors and data upload toolkits developed for the project, and summarise options for GIS data serving on the Internet.

Archaeologists, especially those trying to apply mathematical-statistical methods in their work aim at a certain objectivity: do "science" rather than "arts". But: are we really objective, and how much? Such questions will be treated, in connection with one of the tools commonly used to visualise our results in geographical space, i.e., the distribution map. This paper aims at surveying factors which influence the objectivity of this type of representation, such as: scope - in geographical, chronological and thematical sense dimensions - how many variables are, or can be taken into consideration without loosing the meaning relevance - is the feature we are tracking, meaningful for something - and for what, considering a (historical) interpretation competence - how, and how much the content of the data represented depend on the practice, capabilities and information of the analyst. It will be also essayed to show different representations of the same theme in course of (research) time.

Substantial progress has been made in developing text databases of the historic environment in central government over the last decade in England. With the exception of the computerised mapping system (CMS) which provides mapping support for the statutory protection of monuments (known as Scheduled Ancient Monuments), this has not been matched by parallel investment in systems to support the graphic representation and spatial analysis of heritage records. This paper will set out our experience in procuring and delivering a major spatial information system to redress the balance. The Heritage Spatial Information Service is an ambitious project to integrate the key heritage datasets held by English Heritage, the principal central government heritage agency in England. Developed under the Private Finance Initiative, it is being implemented by IBM, using Intergraph’s GeoMedia Pro software as the GIS and ORACLE Spatial Data Cartridge. This will provide a managed service, under a five-year contract, on a scale that has rarely been attempted within the heritage sector. The HSIS will encompass heritage layers for Scheduled Ancient Monuments, Listed Buildings, Historic Parks and Gardens, Historic Battlefields, World Heritage Sites, and NewHIS (a large general database of archaeological monuments, including maritime records). When completed, it will contain over 1 million records of the English historic environment, and will be one of the largest spatially enabled resources of its kind. There have five major anticipated benefits in the development of the HSIS: Improving data quality: A key purpose in implementing the system was to improve the locational accuracy and representation of heritage objects. For example, the depiction of visible (and probable) extents of monuments, and constraint areas is currently represented in most systems as a single point. 1. Enabling better integration of data sets, through viewing in a common spatial environment. Currently, many of the national datasets sit on independent platforms, and to different standards. This will enable the major statutory and quasi-statutory records to be viewed against the totality of heritage records held in the National Monuments Record. Potentially, it could have a significant input to future conservation policy and selection of heritage sites for protection. 2. Automating changes in spatial information in text databases: When information in the ‘real world’ changes, for example the boundaries of administrative areas, the process of identifying and correcting records within a text database is slow, expensive and subject to human error. Close attention has been paid to using the spatial engine of the GIS to control the spatial information held in the external databases.3. Improving data retrieval: despite a great deal of effort on methods of indexing text databases, there are still a number of types of enquiry which cannot be easily resolved without effective tools for spatial capture and analysis (for example, identifying sites which may be affected along an irregular development corridor by generating a buffer zone in a GIS). 4. Improving understanding of data: In common with many other records management organisations, far more effort has been expended on gathering data together, organising it, and indexing it than on understanding what it means. The research potential for many such records remains largely uncharted territory. Although data quality may not always fully meet the needs of researchers, the scale of the datasets allows questions and manipulation which can rarely be performed within the scope, for example, of a typical PhD thesis. HSIS should become and integral tool for the field teams within English Heritage. It is also hoped that it will be used extensively by external researchers. There should be a cycle between ‘what we know’ as expressed in the heritage datasets and ‘what we want to know’ - our future research agendas. Currently, there is an uncomfortable gap between ‘current knowledge’ and ‘deposited knowledge’ (i.e. knowledge expressed in a commonly accessible databank). Although HSIS is set to deliver the anticipated benefits, a number of limitations and difficulties have been experienced in the course of implementing the project: - Incorporation of data from existing text databases using complex data models and processes within the GIS has imposed limitations on the method of querying datasets. - Careful thought has been required to maintain synchronicity between the GIS and the external data - The depiction of sites has been complicated by the substantial volumes of legacy data to be incorporated - The varying quality of mapbases used to capture data have resulted in positional errors, which are apparent when displayed against a common map background (which is being exacerbated by field information captured using GPS. - Some key datasets are not yet geo-referenced in a suitable way for incorporation. One of the central purposes of English Heritage’s National Monuments Record is to enable the dissemination of information and to foster physical and intellectual access to heritage information. To this end, in time, we will be aiming to make the information available and spatially retrievable, over the Internet. The paper will conclude by looking the potential for web-based dissemination of information, and the implications for encouraging wider enjoyment of the heritage in the new Millennium.

The insuperable limit of archaeological research consists on the fragmentation of the contextual set of any observation field, which has never been preserved intact during the time: the "sealed contexts". The reconstruction nowadays, is possible with a stochastic proposal that, for growing approximations, gives us a full framework of pertinent data and able to link information and contexts otherwise desegregating in the textual or on sheets based lecture. In this framework, we decided that our application, born on a MURST project, could be helpful during the analytical and documentation phases of excavations, for the corpus and heterogeneity of the data, and for the techniques and the methodology of recording and analyses of an archaeological excavation. We decided therefore, to implement and to test a new methodology and also a new strategy for each particular characteristic of the examined situation and at the end to catch what are the most appropriate types of queries and analyses on a historical monumental complex. The project, started in May-June 1999 at the University "La Sapienza" of Rome, under the scientific direction of Prof. Andrea Carandini, focused on three cases study: Pompeii, Veio and Palatino. Until now, we developed the application of this model on the first case study (Pompeii), with a multidimensional GIS system, able to help the field research and the analytical aspects of the archaeologists, analysing the single operative moments. In such way, this different moments will help the archaeologists themselves, creating a kind of feedback, as a progressive skilling of the information and working on different and deepest analytical levels, producing information with a retroactive cognitive value. Briefly, the system consists on the elaboration of a topographic and topological database with inside all the spatial and topological information of the monuments, with the descriptive data and attributes of single contexts, at a Macroscale, and of the single unit and/or action stratigraphical, with topological databases, at a Microscale.

We intended in this communication to present the results obtained in a simulation in which a fictitious society of hunter-gatherers and farmers was produced, animated by agents, by introducing in it the most basic rules as being born, to eat and to die. This type of simulations is designated by complex adaptive systems (CAS) and they're based on a system of rules where interact agents that adapt, complexifying the system in where they're inserted. The individual behaviours are going solemnity-being organised surpassing, individually, an entire series of obstacles that are opposed to them, demonstrating that nature is not a previsible, nor a completely certain entity. These behaviours are governed by general beginnings, which supply tracks for the resolution of problems, to us fit us to isolate these general beginnings and to translate them in rules computer sciences that formed the our intelligent agents' characteristics. Following the maximum processualist nomotetic. The main characteristic of this system, is the emergency concept. This phenomena happens when it is witnessed that a group of simple beginnings develops complex phenomenon. What is observed in our system is that the strategy of hunt-gathering comes as the most favourable, but also as the most unstable in comparison to the agricultural activity.

In this communication it's intended to present an objects and buildings group built tridimensionaly and that one finds inserted in the project Archaeology in 3D. Archaeology 3D seeks the three-dimensional reconstruction of archaeological data through geometric elements, definitions of textures, computation of colours, inclusion of environmental and atmospheric effects and the illumination simulation. It intends to create a group of resources in terms of photos, videos, and virtual worlds that allow an approach to the past and a help to the investigation. If possible we intended to reach the concept of computerised station. The rebuilding in 3 dimensions, provides a new one to look at and an important progress for the visual understanding of the archaeological elements. The study acquires the necessary tridimensionality for a global understanding of the object. The presentation of archaeological researches using three dimensions can create the atmosphere the most approximate possible of the visual world in its more original form. The objective of the virtual reconstruction of archaeological vestiges is the obtaining of a realistic reproduction, in way to get an approach of the Real just as it was imagined and built in the past. It allows to contextualize the object's in its space reality, and to analyse subjects stratigrafics. To associate old and new information of the excavation, that is, the information is going been added in successive layers that can be correlated and crossed to each other.

Frederic F Leymarie, David Cooper, Martha Sharp Joukowsky, Benjamin Kimia, David Laidlaw, David Mumford, Eileen Vote: The Brown University SHARP Lab, USA

The SHARP* Lab was recently established (1999), with a grant from the United States National Science Foundation, by Brown University Departments of Engineering, Applied Mathematics, Computer Science and The Center for Old World Archaeology and Art and Department of Anthropology. It is a significant interdisciplinary effort for scientific research with a direct application to important problems in the analysis of archaeological finds and artifacts. We will present the concepts that underlie a 3D shape language, and an interactive, mixed-initiative system, along with machine and decision-directed Bayesian surface-estimation software for the recovery of 3D free-form object and selected scene structure from one or more images and video. This work will have impact by providing new practical tools. It will also provide an effective testbed for 3D shape reconstruction and recognition, more descriptive local and global models for working with 3D shapes, a better understanding of human/decision-making-machine-interaction for free-form geometric modeling and for extracting 3D geometry from one or more images and video, as well as associated computational complexity issues. As applied to the field of archaeology, this technology will provide, specifically, new ways to analyze and reconstruct pottery, compare objects from different sites and reconstruct sculpture and architecture. *SHARP: SHape analysis, digital Archaeology, Photogrammetry

In spite of their severe limitations, viewsheds are very much present in any GIS application in archaeology (see Gillings forthcoming). A lot of work is needed to improve existing routines, however, these improvements will not be forthcoming because of our lack of knowledge about the visual aspects of landscape. There is much to we need to learn about, and much to be discovered. Here I present some simple developments based on an existing viewshed routine. They are part of a general effort towards obtaining new insights on the visual aspect of landscapes. The concepts of a visual and total landscape (Llobera 1999) are introduced, the former being an extension of what traditionally has been know as a cumulative viewshed (Wheatley 1992)

Els Vilars is and Iron Age site located in the north-eastern part of the municipality of Arbeca in Catalonia, Spain. Starting from the previous proposals, the present phase of the project can be defined through the following objectives: A) The reconstruction of the fortress This will be carried out diachronically, including 4 phases: Vilars 0, I, II and III - IV, given that the upper levels do not allow individualised restoration. A virtual reconstruction of the elements of the post-depositional processes that they have affected the enclosure after its abandonment will also be incorporated, with the aim of illustrating the causes of its current state. The reconstruction will include two levels of detail. Firstly at a macro level, affecting the principal urbanistic elements (defences, street plan, habited spaces). Secondly, at a micro level, permitting a virtual visit to the interior of some of the houses with enough elements to differentiate the functional use of the space. B) The reconstruction of the paleo-topography. This will be centred on the area of the immediate surroundings of the site (a radius of 5 km.), given that the scope of current research does not permit longer-range reconstruction. C) The reconstruction of the environment. The lack of specific intervention outwith the space for the immediate use of the inhabitants of Vilars constitutes the most awkward aspect for the researcher when it comes to situating the different elements that appear. Nevertheless, the existing information permits a reconstruction of the types of the crops, livestock and the hunting resources, as well as the mineral and forestry resources exploited. D) The user interface. An interactive system are being developed to allow arqueologists to interact using real time direct manipulation interfaces to add arqueological information to the arqueological database.

The Northwest Coast of North America is a deeply indented and rugged fjordland archipelago. Ethnographic and archaeological information and environmental constraints suggest that water transport has long been the dominant means of mobility. With this starting assumption, it is possible to confidently represent archaeological sites as being joined by a schematic mobility network of least-path distances across the water. Knowing both site locations and the routes which join them opens up interesting avenues of investigation, including investigation of landscapes of optimality, history and habit. This paper presents results of the network application of the location-allocation spatial interaction model in two western Canadian study areas: Nuu-Chah-Nulth territory on west coastal Vancouver Island and the Gwaii Haanas National Park Reserve and Haida Heritage Site on the southern Queen Charlotte Islands.

Archaeologists are now a fast growing group of researchers with pressing needs to apply predictive models as decision-support tools. However, there are important issues to be resolved before predictive modeling can be widely adopted by the many researchers and planners who need to use it. Increasingly, GIS-format statewide archaeological databases are being created in the United States, but theoretical issues, data requirements, and methodological matters must be closely examined in an attempt to formulate the baseline requirements for predictive models that can be used as decision-support tools

To imagine the different construction states of historical buildings there are various ways. In the case of the Koerich castle situated in the western part of Luxembourg were planned photogrammetric studies to take the today´s state of the ruin and geophysical prospection to investigate the remains of not preserved constructions in the subsoil. In the quite small area of the castle courtyard (about 650 sqm) were applied three geophysical methods: geomagnetics, geoelectrics and ground-penetrating radar (GPR). The combination guarantees a high reliability of the results and allows detailed statements about location, depth and condition of the hidden remains. Unexpectedly geomagnetics didn´t yield useable results concerning the hidden foundations. Obviously the used sandstone hasn´t got any magnetic contrast to the surrounding soil. A very high magnetic anomaliy suspected of beeing the signals of a WWII arial bomb turned out to be caused by a foundation of a forgotten flagpole from the thirties. Geoelctrics and GPR yielded better results regarding the really aim of the prospection. GPR measurements were carried out with a profil distance of 0.5 m and resolution of 0.05 m inline using a SIR-10 equipment and a 500 MHz antenna. Profiles were layed out in two orthogonal directions. Geoelectrical dipole-dipole sections used the same profil lines but with distances of 1 m or 2 m. Geoelectrics can´t outstrip the spatial resolution of GPR but it´s necessary to localize the base of constructions which can´t be resolved by GPR. Areal GPR data usually are displayed by time slices or after a migration by depth slices. Geoelectrical sections are combined in vertical or horizontal sections of the courtyard. By combination of all geophysical results we obtained a detailed image of the contructions in the subsoil. Overlaying the images of geophysical data and including the information of building history and photogrammetry a base for 3D reconstruction of the castle will be arranged.

The facts, how ofteh could a phaenomenon occur in an environment, or which was the number of identical examples of a specific type observed in a layer etc. have always been of interest in archaeological research. Now the computer gives much help for analyses of this kind. But several experiences, made during past decades in our country on finding places with traces from the Roman Age, could give some useful advise, which questions should be asked and what answers could be expected, as among others: The numbers of sherds in a layer does not represent oe ipso the number of vessels, contented in this layer. The rules in the production and distribution of Terra Sigillata are only in a part valid also for the other kinds of pottery, distributed by trade, and mostly not for local produced pottery. One should, even in large projects, where the fields of interest are strictly separated, before the analysing the quantities in diferent layers, be well informed, how every peculiar layer has been built (by sudden destruction or long decay, firmly covered with a construction of the next phase or laying open for a long period etc.). For the analysing of complete buildings, parts of settlements or even settlements as a whole the present resp. absent types of a particular period, can procure good evidence to the function of this building/s. But again the character of the archaeological record should be carefully considered: by large building activities f. e. the complete destroyed material of previous phases could be also in our conditions moved out of the area, or further, even cultural layers originating from an other spot could be brought in to the complex as the material for equalisation of levels. The extend of fields, adaptable for the growth of cereals in the area of four towns, Emona, Celeia, Neviodunum and Poetovio, can be in the specific countryside measured enough reliable. The quantity of cereals necessary to an legion during each winter period in the first cent. is also enough well known. The author believes there could be some possibilities in these facts to get better evidence to Roman time economy and land use in today Slovenia. Graveyards, especially when complete known or even excavated, could be used with enough probability by the frequency of funeral gifts, the use of different types for such gifts etc. in conclusions about the way of funerals and other habits, ev. ethnic character of the respective group of population.

The absence of Upper Palaeolithic cave art in Britain may be the result of the accelerated weathering of limestone caves during the Holocene. To test this proposition, it is intended to place mineral-based stencils in three Mendip caves and record monthly changes in pigment colour and adherence, and compare to seasonal variations in local climate. Conventional photography has proved to be an inadequate medium for recording changes to the pigments. A high-resolution digital camera would facilitate digital image processing of the data and improve the accuracy and precision of the monitoring and evaluation. This paper reports on the proposed image analysis techniques which have been tested on pseudo-real data generated specifically for this project using a simple digital camera and stencils specifically drawn on cave walls. The analysis uses techniques such as edge detection to locate a colour chart. A circle detecting Hough transform is then applied to locate the circular stencils. Methods are employed to normalise the colour data to incorporate any environmental changes from one image capture session to another. After normalisation and colour data correction, the stencilled areas are segmented using k-means clustering and their colour and shape statistics are measured. These measurements can then be compared with the master image or data to record changes.

Paper approaches application of dynamic Geographical Information System (GIS) models at the reconstruction of palaeo floodplain dynamics. Traditional model of the Neolithic lakeside settlement on the Ljubljana Moor was challenged using dynamic GIS based model of stream migration. Implications of the modeled dynamic landscape related to the settlement patterns and perception of the landscape by the Neolithic men and women dwelling in it are discussed.

Nara National Cultural Properties Research Institute (Japanese abbreviation: Nabunken) has an important collection of aerial photographs. After the World War II, some private company began to take aerial photographs for the mapping, investigations etc. Private companies have not enough space to conserve them. Then Nabunken stock these old records of the surface of all over Japan. This collection does not include the aerial photos by the governmental office for national mapping, Total number of the photo is about 1 million 800 thousands and we continue to make database about it. The original film is 10 inch square and difficult to treat. Then we make a microfilm from the original one. About 900 thousands pictures have micronized until now. We watch a microfilm and determinate the place where it was taken for the database. The data entry system make a coordination data from the map image. The retrieval system was separated from the entry system. Now we make a new system with the digital map image and the digitised image of each aerial photo. This system is for the data entry and the data retrieval. And also, this system is intranet system. We continue the long way for data entry and digitising the aerial photos for this real GIS method.

The mountain areas of the Marche Region represent an extremely conservative area under several points of view, then very promising for the research. Particularly considerable appear the strong place names continuity, a phenomenon clearly emerging from the comparison between the toponyms recorded in the medieval written docu-ments from the XIIIth century and those - numerous - listed in the books of the Catasto Gregoriano (end of the XVIII and the beginnings of the XIX century). The place named stored there, one can locate by means of the 1:2000 maps attached to the books, are linked to detailed information on the land-owners, on the landuse, on the productivity of the fields and in a general way on the productive areas, on the presence of the woods, on the roads, and so on. The collection of such information allows to achieve, by the means of place names location, a wide map of the landuse from the XIII century. This is the basis of a multi-phase project aimed to a G.I.S. management of the above data and illustrated in this paper. The Author describes, besides the general characteristics of the project, the problems arising from the treatment of the data and some recent results concerning the space organisation in Roman and Medieval times.

The possibilities of applying a raster based GIS for interpreting the results of archaeological prospecting were tested on the Iron Age to Late Roman settlement site of Rodik – Ajdovscina, Slovenia. In this paper, results of our GIS testing on what appear to be different activity areas are presented. Different data sets were used for evaluation of archaeological potential by multivariate statistics (magnetometry, magnetic susceptibility, resistivity, surface distribution of ceramics and slag, magnetic, chemical and mineralogical analyses of small samples, etc.). Results of various prospections can be presented in a single image where relevant archaeological spatial units (activity areas) will be discernable by spatially defined classes. All surface units convey information which comes from combination/correlation between different input data sets. The results will vary in terms of quality, but they enable both a comparison of the individual data layers and the generation of new information based upon the level of correlation between them. The output will therefore be a composite image of differing, but normalised input data. This could lead to partial automation of the interpretation procedure.

Starting in 1997, CILEA (Consorzio Interuniversitario Lombardo per l'Elaborazione Automatica) and the Universita degli Studi di Milano, Dipartimento di Scienze dell’Antichita began work on a field work opportunities database that would be accessible over the Internet. Part of the EC project ArcheoNet (archweb.leidenuniv.nl/archeonet), the starting-point for the work was the obvious need for students as well as researchers looking for fieldwork projects to contact any European organisations promoting fieldwork and surveys directly rather than through various slow-moving and erratic intermediaries. An on-line database, using a web forms interface, was decided to offer the best means for bringing research projects to the attention of students from all over Europe. The first release of the project, then named EARP, was demonstrated during the CAA97. Since then, the site has gone through two reviews in order to increase performance and functionality. The first review ended in September 1998, and resulted in a faster and more efficient navigation and an easier database management (http://archeonet.cilea.it/archeosite). In November 1999 the site was adopted by another EC project (ArchTerra) and subjected to a second restructuring stage aimed at simplifying and automated the service. The purpose of the current paper is to present the design and features of the restructured service, emphasising the changes made since November.

The Munsell system has shown to be a inexpensive, quick, standardized method of reading and reporting the color of an archaeological object or assemblage. However, the system has limits. Changing lighting conditions and variations within the artifact’s fabric can alter the reading. Different people can see different shades of the same color, giving different readings for the same material. Given the possible variances, the Munsell designations are best used as a rough guide as to an item’s color. General color designations are unsuitable in cases when the quantification of color is deemed important. Efforts to use corlorimeters have shown to be useful in laboratory environments, but their application in the field has been often both cost prohibitive and difficult due to the inability to control varying lighting conditions. A field method for accurately measuring color (or at least controlling the variability) still eludes us. The method presented has been determined to be an effective means of quantifying, describing, and controlling the variance observed using the Munsell color system. In this method, Munsell values are transferred to the Y, x, y values of the chromaticity coordinate system. The resulting numerical values can then be manipulated using standard statistical analyses. In testing this method, 20 subjects were given a set of 12 artifacts to measure with the Munsell system. Transference of the Munsell readings to chromaticity coordinates enabled the author to quantifiably determine the variance and mean of subjects’ resoonses, as well as to determine color-based artifact groupings through cluster analysis. Applications for this method are numerous and broad, including the study of firing temperatures in ceramics, and the quantification of color variability amongst lithic artifacts. This method could also be used as a "quality control" measure when several members of a project are measuring color for the same body of material.

In a previous paper the authors discussed the use of a commercial package enabling web access to GIS, applying it to the study of an archaeological case. For the high cost of the package (tested in a limited-time demo version) a permanent version of the site was not created. However, the experiment was very promising and met the consensus of the audience to which it was presented. In the present paper the authors propose to use the public domain GIS GRASS and the GRASSLink package, freely available under GNU public licence, which enables web access to the GIS. The GIS may also be linked to the free RDBMS Postgres, which also has a web interface, in order to provide a completely web-accessible DBMS/GIS system at no cost. This will enable publication of GIS and database data, up to now made almost impossible by the high costs of web interfaces to most common GIS, thus allowing public access to archives both in alphanumeric and graphic form. Standardisation issues are also addressed, to document completely the data formats for user access and future archival, conforming to accepted standards for archaeological digital data conservation. The paper examines also the enhancement of data description using XML, which allows a unified approach to structured data, texts and spatial data. The process has been applied to an archaeological GIS derived from previous research, which has been transferred to the required format and published on the Internet, where it can be accessed using the above described packages. The consequences of this approach on archaeological research are briefly discussed. In our opinion, remotely accessing a GIS might encourage co-operative research work and increase data re-use and knowledge dissemination towards users with no database/GIS skill, who could access data also with a limited knowledge of the data structure by means of the standard, well-known interface of Internet browsers.

The application of fuzzy logic in archaeology has a great potential, to process uncertain or partially determined data, but so far has received little attention, except in some viewshed analysis studies. Recent developments in database theory have addressed this topic and can have interesting applications in archaeological databases and GIS, as the authors already noticed in a previous paper. As a matter of fact, many of the concepts and relations currently used in archaeological analysis have an inherently fuzzy nature and many of the attributes stored in databases have imprecise values. Ordinary databases and GIS do not manage adequately these features, so we propose to integrate the traditional database and GIS tools with fuzzy attributes and relations. In many cases, moreover, archaeological data derive from statistical analysis (for instance, ostheological data) and an inadequate treatment may produce misleading results. After discussing the main features of the proposed fuzzy logic approach, we consider the example of an Etruscan cemetery and introduce fuzzy attributes and fuzzy relationships in the database, involving also spatial features of the data. The assignment of values to these attributes is discussed in detail, showing how ostheological analysis may be correctly used for this purpose. This approach is then generalised, inserting general fuzzy attributes and relationships into the structure of a RDBMS and creating the necessary functions to manipulate them. We used for this the public domain Postgres RDBMS, GRASS GIS and PostGRASS, a set of functions to connect them, adding special functions of our own creation. So every researcher may use, at no cost, our improved database. Moreover, all the software has an easy-to-use graphical interface, developed in Tcl/Tk. The system has been applied to the above archaeological example, showing how it can be used to investigate this case study, for instance to analyse specific palaeodemography aspects.

The appearance of hillforts – the first monumental constructions in the Kras landscape – is dated to the beginnings of the Bronze Age (the first centuries of the 2nd millennium BC). Once constructed, the hillforts remained by far the most dominant type of settlement until the arrival of the Romans in the 1st century B.C. This paper focuses on the territorial organization of settlement and, consequently, on social organization as expressed in patterns of territoriality. Prior to the Late Bronze Age there is hardly any evidence for hierarchical organisation in the settlement pattern. The 2nd millennium settlement system appears to consist of a mosaic of individual setlement niches. It is posited that the basic social and settlement unit there was based on the lineage or extended family. The 1st millennium BC settlement pattern exhibits different features in settlement and community organization. The analysis of cemeteries has demonstrated the existence of settlements, occupied by multi-lineage communities from the beginning of the 1st millennium BC. On the other hand, historical sources record civitates in the Kras area at least for the 1st century A.D. Clearly, we are dealing here with a process of aggregation of traditional descent groups into larger social units. This process went hand-in-hand with the process of increased social differentiation between lineages, the appearance of central places and with aggregation of smaller territorial units into larger (community) territories. In an attempt to reveal the changing territoriality, we based our analysis on three archaeologically attested premises: 1) Potential centres of community territories have formal and quantifiable features in common, which can also serve as criteria for isolating central places. 2) In the Kras, the territorial organization of prehistoric settlement is to a great extent determined by geomorphology. 3) Settlement size (the surface of the walled area, implying the amount of the work invested in rampart construction) and the size of the population occupying it, exhibits a positive correlation. Based on Slapsak's (1988, 1995) studies of the Rundictes (a late prehistoric and early Roman community in the SE part of the Kras) a model was developed for identifying potential centres of territorial communities. It was based on a series of quantifiable and non-quantifiable criteria. These included relative hillfort size compared to that of neighbouring hillforts, the size of viewsheds and classes of visually controlled areas from the centre, the continuity into the Roman period, the presence of a cemetery, the complexity of the internal plan of the settlement, presence of the 2nd millennium occupation etc. 15 potential centres were isolated in this way and nearest neighbour analysis demonstrated a high level of uniformity in their distribution across the landscape (premise 1). In further analysis, an attempt was made to define communal territories by applying Thiessen polygons (drawn from the centres) and by considering geomorphological features in the landscape (premise 2). The final step comprised the summing up of all walled areas within individual community territories and applying a regression analysis between the areas of community territories and their respective sums of walled areas (premise 3). The result of the analysis demonstrated rather strong positive correlation between the variables. The community territories exhibit 2 major patterns - a 'large, single settlement' pattern and a 'multi-settlement' pattern. The first describes a situation with one very large settlement in the community territory (5 times larger or more, than the second ranked settlement), which was probably occupied by all or most of the population. The second pattern (the largest settlement is only 2-3 times larger than the second ranked settlement) exhibits a different hierarchy, in which the rank-size curve is much closer to the rank-size rule.