CAA-UK 2015, University of Bradford

Posted on July 22, 2015

0


Here are the videos from CAAUK in March. A bit about the conference-

CAA-UK aims to encourage communication between UK-based archaeologists, mathematicians and computer scientists in order to stimulate research and promote best practice in computational and mathematical approaches to the past. The conference was held at Bradford University’s dedicated Norcroft Conference Centre.

And the talks:

Developing an integrated digital data workflow for the 100 Minories project

John Layt
L – P : Archaeology

The 100 Minories project is a commercial excavation by LP archaeology of a site just to the north of the Tower of London. The project has sought to integrate a number of existing digital archaeology techniques into the daily commercial archaeology work-flow using open source tools and an on-site digital archaeologist. This includes the use of tablets, wifi-enabled digital cameras, photogrammetry, TotalStation, and traditional written context sheets and drawn plans. A core tenet of the project is that all data generated should be provided on-line as Open Data as soon as is practical. The main output of this work will be a set of apps and QGIS plugins to support the Museum of London Single Context Recording system, including the digitising of single context plans into the project GIS and the merging of this data into the ARK online database. This talk will demonstrate the work flow used on the 100 Minories project and the open source software used and developed to support this, and discuss the potential cost savings, improved data quality and community involvement that may result.

Plan, Features, Sections: Using NLP to remove ambiguity from Grey Literature

Michael Charno
Archaeology Data Service

Ever-increasing amounts of data are available within data repositories in individual institutions, national infrastructures and international services. The EC Infrastructures funded ARIADNE project is working to bring together archaeological research data from across Europe, for use and reuse in new research. One of the main challenges of this objective is the disparate data types and structures archaeologists produce, and all in a variety of languages. ARIADNE is building infrastructure to bring together, manage and provide access to these datasets. The project is embracing Linked Open Data, Natural Language Processing (NLP), and deploying Web Services and new tools to provide enhanced access to researchers.

Exploring sustainable publication and the web: a case-study from ARK perspective

Michael Johnson, J. Andrew Dufton, Elizabeth Fentress
L – P : Archaeology

Within the past decade, new platforms for creating and managing archaeological data during fieldwork have seen both widespread adoption and increasing intricacy. Although effective data management is now a necessary component of most field projects, the lives of these complex systems after the project has ended are less certain. Institutional concerns including basic software requirements, system upgrades, and web security can derail digital publication and result in largely static, albeit sustainable, archives, downloadable CSV files, simple HTML pages, and, occasionally, PDFs. These formats are only one piece of the puzzle, a necessary aspect of sustainability that, when used alone, can limit the functionality of these data for future researchers. How can we ensure the open, online publication of archaeological data while also maintaining the complexity of original field systems? What steps can we take to address institutional concerns while publishing information in interactive formats that facilitate later use and reuse? What is the future for data available in older systems as software continues to develop and new approaches become viable? This paper considers some of these questions specifically in reference to the Archaeological Recording Kit (ARK). Developed at LP : Archaeology this software is designed to manage the complex data generated during archaeological excavations. The generated interface goes beyond presentation in traditional formats, but the dynamic nature of the software poses many problems for sustainability. How should these emerging issues should shape the ongoing development of the software.

Multidisciplinary research of Iron Age sites and landscapes of Slovenia

Matija Crešnar, Branko Mušic, Dimitrij Mlekuž,
University of Ljubljana

In the recent years multidisciplinary research has made an enormous step forward when it comes to dealing with archaeological heritage in Slovenia but also throughout Europe and elsewhere. Projects including aerial imagery, lidar scanning, geophysics and beyond that, but also integrated studies, are increasingly covering vast areas and are producing enormous amount of data. Landscape archaeology is thus becoming one of the most fast-developing fields within archaeology, within which we can observe at least two different ways forward. One of them is heading towards the extensive collection of data, where we are encountering deficits in thorough data analysis, whereas the goals of the other studies are accuracy and precision when it comes to identifying buried archaeological structures. Although trying to understand the whole landscape in all its depth, it is easy to forget that we are dealing with a palimpsest of imprints, which we have to be understood as separate time-slices to open the gates to individual phases of its formation. Taking in focus just one particular site, the hillfort Poštela near Maribor with its extensive cemeteries, every technique and analytical method used has added specific information; studied independently but even more in intertwining with others, delivering new layers of information. We have started with aerial photography and historical analysis, added new layers of data with lidar scanning, data on natural settings like geomorphology for instance and an array of geophysical methods integrated in a multi-method approach (GPR, magnetics, resistivity, susceptibility, low frequency EM …). However, we have always included the intermediate steps, i.e. “ground-truthing”, where we use different methods of low invasive investigation (core-drilling, testtrenching), to guide our further steps. Not only gathering enormous amounts of data “for later analysis”, but collecting data “cum grano salis”. As the broader understanding of archaeological sites is the main final goal of our research we also combine the collected data with a series of GIS analyses to explore different viewpoints of the site and its position in the wider environment. Other material specific analytical methods as multi-detector computed tomography for scanning of urns, containing cremated human remains and grave goods, microscopy for the analysis of fabric used for pottery, petrography of stone debris and soil micromorphology etc., were also used for further in-depth investigations of specific research questions. However, it the end it is not the sheer quantity, but the quality which matters…

Legacy Data – Open strategies for closed data

Graeme Attwood, Finnegan Pope-Carter
GSB (SUMO)

Archaeological geophysics surveys have been conducted by groups based in Bradford for in excess of 40 years. GSB Prospection (Part of the SUMO Group) combined with the School of Archaeological Sciences at the University of Bradford have an archive of reports both analogue and digital stretching back this far, many of which have been viewed by only a select few. Spurred on by a project to catalogue the late Professor Arnold Aspinall’s project archive and a need to update GSB’s internal database and archiving systems, GSB and postgraduate students in the school of Archaeological Sciences have embarked on a project to digitise, re-georeference and disseminate work that can be made available in the public domain. The project is built primarily on open source software and database systems and wherever possible will be made available to those seeking to replicate the work we are doing.

The Open aspects of the project break down into three key tasks, these are:
• Bringing historic data into a format that can be utilised on current systems.
• Batch Process historical word, CAD and data archives to ensure the contents are converted to and stored in freely available formats. Historic archives have been lost when proprietary formats
become obsolete.
• Salvage born digital data from archives only available in printed analogue formats. In salvaging archives that currently only exist in paper or legacy formats it safeguards the data for future use.
• Ensuring the location of surveys can be easily determined from datasets
• Compare digital mapping information with project metadata to verify georeferencing information
in historical CAD archives.
• If georeferencing information is incorrect automatically identify known features that correlate with open mapping data, to position the historic local grid within the world view.
• If no automated georeferencing is possible flag the site for review
• Make datasets available in the most appropriate way
• Categorise surveys to determine those that are commercially sensitive and those that have previously migrated into the public domain.
• If accurate georeferencing has been possible make data only previously available in printed form available on a google maps type service.
• If only an approximate georeferencing has been possible add a Pin to a map linking to a web viewable report.
• Undertake consultation to determine if the final display should contain data ‘tiers’ in order to protect uninvestigated archaeology.

The first tier being an approximate location pin with basic survey details, leading to a final tier of fully georeferenced data and interpretive plots that can be accessed on request.

What have the Romans ever done for us? Digital strategies for research syntheses & fieldwork reports

Tim Evans
Archaeology Data Service

The paper presents the ongoing work on the dissemination of the research data produced by the Roman Rural Settlement of Roman Britain project; a collaborative endeavor of the University of Reading, Cotswold Archaeology and the Archaeology Data Service. Although large-scale research syntheses incorporating the datasets and reports produced by research and commercial fieldwork are now relatively common, this project is arguably unique in planning for online dissemination from the outset. The paper will discuss the online interface (due for release in April 2015 in tandem with the traditional hard-copy monograph), which combines a database/ Web Mapping front-end, with digital versions of the unpublished reports identified by the research team; thus allowing the capacity for broad or detailed interrogations of these grey reports on detailed thematic and methodological criteria that was previously achievable. In addition the paper will also discuss the lessons learned from planning and building a model for simultaneous paper and digital publication during the lifetime of a project, particularly the respective merits of Web Mapping technologies, but also a noticeable shift in attitudes towards the free dissemination of data by the community as a whole. Finally, the paper also looks to the possibilities and challenges for incorporating the results of future fieldwork into such academic syntheses. It is suggested that by using online technologies, it should be possible to move away from the recent, yet essential, trend of retrospective syntheses of decades of unpublished and sometimes amorphous backlogs, towards a model that allows the participation of commercial units, community groups, researchers and curators in identifying and contributing to specific research themes. Thus offering the potential for a more holistic yet immediate culture of research.

Geospatial Geophysics: Processing GNSS located data in python

Finnegan Pope-Carter
GSB (SUMO) & University of Bradford

Geophysical data collection is increasingly moving from Grid based data collection to Real Time GNSS Located systems. While a number of commercial and proprietary systems exist for collecting such data few commercial geophysics softwares exist to facilitate the processing of such data. Python is a human readable Open Source programming language with bindings and modules available for nearly all standard geospatial data types. Some of these are available through Open Source GIS packages such as QGIS however the majority are more powerful and extensible when utilised directly through Python. This talk will initially discuss some of the widely available modules available for processing and manipulating and presenting geospatial data. These include:

•PyProj
•GDAL
•Shapely
•Fiona
•SimpleKML
•DXFWriter

We will then introduce work being done at the University of Bradford and GSB SUMO to build a python package for the processing of geophysical datasets. This includes:

•Manipulating GNSS NMEA Strings
•Projection of GNSS Data
•Image Processing

CBA East Midlands Boundaries Project

Catherine Wells
Ischus Limited

CBA-EM is the Council for British Archaeology local group for the East Midlands. Covering five counties (Derbyshire, Leicestershire, Lincolnshire, Nottinghamshire, and Rutland) since the 1940s, we are an educational charity working to enable the public to be involved with archaeology and to promote the appreciation and care of the historic environment for the benefit of the present and future generations. We achieve this by holding events, such as several day conferences during the winter months and visits to sites of interest during the summer. In recent years we have been investigating the possibility for a more active involvement of CBAEM in an archaeological project, though not something that would generate a large finds based archive. After much thought, discussion and canvassing of our members, the result is the
Boundaries Project. The English countryside is a multi-layered patchwork on which each generation has left its mark and this forms the basis of the project. We intend to discover more about the countryside by looking at the ancient boundaries that define it. Ancient boundaries, marked by a vast variety of walls, ditches, banks, hedges and trees, were important and defined the landscape for centuries, some dating back into prehistory, and by studying them we intend to add a systematic record of the patterns of their individual elements to the knowledge we have of the past. Volunteers will start looking at parish boundaries marked on historic maps and then go out and record their form and condition to create a 21st century domesday GIS record with linked drawings and photographs. A CBE-EM sub committee was tasked with defining the data to be collected and designing the method by which it would be recorded and stored. Archaeological, historical, administrative, and computer related experience and skills were provided by relevant expert members of the sub committee. We conducted a pilot study during 2014 and are intending to launch the project at our AGM in March. This paper will outline the project, the technical aspects and the software development involved in achieving its realisation, along with lessons learned in the pilot study and planned future development.

Digitised Diseases: 3D digital documentation of bone change in cases of chronic disease

Andrew S. Wilson, Tom Sparrow, Andrew Holland, Becky Storm, Emma L. Brown, Pawel Eliasz, David Keenan, Carina Phillips, Natasha Powers, Jo Buckberry, Chris Gaffney, Hassan Ugail, Keith Manchester University of Bradford

This talk reviews ‘Digitised Diseases’, an important legacy project based at the University of Bradford which received funds from Jisc’s mass digitisation programme and was partnered with the Royal College of Surgeons of England and MOLA. The project used 3D digital documentation to record examples of chronic diseases manifest in the human skeleton. From the outset we had to design a hierarchical classification structure that was comprehensible to both clinicians and palaeopathologists, recognising that this would serve as an important framework for presenting our 3D content. We created photo-realistic 3D digital models of type specimens of diseased bone from archaeological and historical medical collections in Bradford, London and York. These models are now available as an open access resource that enables users to view and manipulate models online and also download higher resolution textured models for use with freely available viewers on a range of platforms. From the outset we used regular project bulletins via a range of social media to target potential users from the bioarchaeological community, with considerable global interest. We continue to update this resource which went live in December 2013. Usage now exceeds initial expectations with users in medicine, anatomy and the medical humanities as well as amongst the general public coming from over 125 different countries.

Adventures in Agriculture: Experimental modelling for economic analysis

Helen Goodchild
University of York

The modelling of agriculture is a complex discipline, and it is therefore not surprising that the multitude of techniques currently applied to modern, very detailed datasets, have not been explored in more depth in terms of their applicability to the past. Methods such as Ecological Niche Modelling, Agro-Ecological Zoning, Habitat Suitability Modelling, and more, reflect the diversity of approach taken by geographers to either measure or predict human productive output. This paper will present some recent experiments (some successful, some not-so successful) in applying modern agronomic and climate modelling techniques, and discuss the potentials and limitations of using these methods to investigate economic strategies in relation to agricultural and pastoral practices in the past.

Developing a method for a spatial correspondence analysis

Martin Sterry
University of Leicester

Multivariate analyses, in particular correspondence analysis, (CA) have become a standard exploratory tool for analysing and interpreting variance in archaeological assemblages. Notable examples include those involving artefacts (e.g. Cool and Baxter 1999; Pitts 2010), archaeobotanical assemblages (van der Veen 2007) and faunal assemblages (Manning et al. 2013). Although it has been noted that CA “might be of considerable relevance to spatial analysis” (Wheatley and Gillings 2002: 146), as yet there has been no truly successful integration. This makes it difficult to isolate spatial influences on assemblage composition from the CA scatterplots (except through pre-determined regionalisations) even though these may be of equal or greater importance to other determinants such as, cultural, socio-economic and temporal aspects. This paper will present a novel method for visualising CA in ArcGIS by transforming the resultant scatter graphs of the CA into colour maps within which the similarity and difference between assemblages directly corresponds to the similarity and difference of the colours used to display them. Utilising a dataset of faunal assemblages from Late Iron Age to Late Roman central England, the paper will demonstrate how the method is applied and how it can be used to draw out spatial and temporal trends of different animal husbandry strategies. Further, the paper will explore the potential of this analysis for using the data to define past zones (and how they change over time).

Please give some feedback about the videos e.g. quality, changes you would like, etc. Thank you. To see more videos like these please go to the YouTube channel Recording Archaeology- http://www.youtube.com/channel/UC08QKQO1qs6OPQs9l1kMQPg

Posted in: Uncategorized