Below are the abstracts for the CAAA Digital Archaeology Conference 2025, 2-3 October. The conference is free and will be held online via Zoom. You can register here. More information about our keynote address by Brett Leavy is available here. The full schedule with times for speakers is available here.
Please note, talks marked with a ^ will not be uploaded to our YouTube channel at the end of the conference. Affiliations shown are those of the first author. Titles marked with * are lightning talks.
Daeyoun Cho, Sanggu Han
Jeonbuk National University
A QGIS-Based Visibility Analysis of the Jeju-ro Beacon Network in the Joseon Dynasty of Korea
During the Joseon Dynasty in Korea, the beacon network evolved into more than a basic military signaling system. It became a political infrastructure that enabled centralized power to control local areas through the five major beacon stations that covered the entire country. This study focused on the Jeju-ro (Jeju Road) section of Jeollabuk-do and conducted a visibility analysis using QGIS. This section linked coastal beacon towers with inland ones, creating a central relay network for sending surveillance information from the southern coast to the capital.
The study used a digital elevation model (DEM) based on South Korea’s public spatial information services to calculate the visibility range of each beacon tower and analyze the mutual visibility between beacons. Several key findings were derived through this process. First, overall continuity was maintained by the beacon network, but interrupted visibility was exhibited by some sections, and in specific areas, overlapping surveillance ranges were observed. Notably, beacon towers were installed in areas with poor visibility. This suggests that political symbols and the way power is shown were more important than just how well communication worked. Therefore, the beacon network in this region can be understood as a complex system that combines national power and spatial strategy, beyond a simple signal transmission system.
This study takes a fresh look at the structure and meaning of the beacon network using QGIS-based visibility analysis. The results may provide essential data for comparative historical research on premodern communication networks in East Asia.
Hannah Chow
The University of Sydney
Over the City Walls: A UAV Survey Analysis of Extramural Features at Kerkenes Dağı^
Kerkenes Dağı is a central Anatolian, Iron Age city destroyed in ca 547 BC. Excavations and geophysical analysis began over forty years ago and illuminated understandings of the urban structure within the city walls. Beyond the walls, however, very little is known. Whilst Iron Age features are visible in satellite imagery on the surface, until now no comprehensive study has been undertaken to map the features in the extramural landscape. In 2023, a UAV survey, with an RGB camera and minimum 2 cm spatial resolution, was undertaken over 4000 hectares encompassing both the city and portions of the broader hinterland. A section of the northern UAV survey was selected for detailed investigations with boundaries demarcated by ridgelines and the city gates. Within this area, 200 polygons were randomly selected identifying areas to be mapped. Through various visualisation methods, including False Normalised Difference Vegetation Index (NDVI) and Digital Terrain Model (DTM) with shaded relief, features were identified and mapped. Results of mapping and interpreting these features indicated that there was considerable activity within the broader hinterland. Field systems were the most prominent, including both contemporary and ancient agriculture. However, there was also considerable evidence for transport pathways, tumuli and stone structures.
Tess Cornwall
University of York
Living like a Viking: A Digital Living Museum Reconstruction of Coppergate
This dissertation project explores the creation of an interactive digital reconstruction of 10th-century Coppergate, York, to see how this type of resource could help communicate Viking Age civilian life and challenge common stereotypes of Viking culture. The reconstruction is based on excavation data from 16–22 Coppergate, including building footprints, artefacts, and environmental evidence. It was built in Unreal Engine 5.6 as an explorable “living museum” environment. I created custom 3D models of Coppergate-style houses and a Viking comb in Blender, and used adapted Viking-themed assets to help dress the scenes and represent domestic activities. Decisions on materials, layout, and content were made using archaeological reports, GIS data, and historical references, with adjustments where evidence was incomplete.
While formal user testing was not possible within the project timeframe, the reconstruction was evaluated through an interview with the digital heritage consultancy Heritage 360, focusing on accuracy, accessibility, and interpretive choices. Their feedback, alongside my own reflections, informed the project’s conclusions. The results show that digital reconstructions like this can be an engaging way to present archaeological information, support public engagement with the Viking past, and complement physical heritage sites, while also highlighting the challenges of balancing evidence and interpretation in a 3D environment.
Joshua Emmitt, Richard Ng
Auckland Museum
Establishing photogrammetry in practice at Auckland Museum
3D modelling is now a standard part of heritage practitioners’ toolkits. While visualisation remains one of the key outputs for 3D modelling, it is now used to aid with conservation, displays, and exhibitions. Prior to 2023, Auckland Museum had used 3D modelling in collaborative projects for its faunal assemblages, but it was still practised on an ad-hoc basis. This paper will report on the establishment of photogrammetry as a 3D modelling solution at Auckland Museum, including expanding the range of objects it is used on, and using it for conservation, display, and exhibition purposes. Work has also been done to train people in the use of photogrammetry to de-centralise the knowledge, and to make sure it can be another piece in people’s toolkits opposed to locked behind specialist knowledge.
Calum Farrar
Griffith University
Aura and Presence with digital heritage: The usefulness of Augmented Reality*
Other than born digital data, most digital archaeological and heritage data is a translation, reconstruction or reproduction of something real into a digital form. This transition between states and how it, as a process, effects both the original and simulacra has been a topic of debate for millennia, but resurfaced in the last century in the context of mass reproduction brought on by new technologies. Walter Benjamin’s establishment of objects having an ‘aura’ has been fundamental discussions on the issue. Although his thoughts on how the aura manifested or diminished due to technological intervention and reproduction have become somewhat outdated since the early 20th century, they remain keenly relevant to digital archaeological practices. This lightning talk highlights some recent observations on how the use of Augmented Reality can overcome many of the issues associated with achieving aura and presence in a digital format normally found when using fully virtual environments.
Kristine Hardy, Mathieu Leclerc
Australian National University
The Use of Linked Open Data Location Identifiers for Pacific Archaeology.
Linked Open Data (LOD) relies on persistent unique resource identifiers (URI) for objects including locations. Spatial identifiers are essential for archaeological databases and the location identifiers used in archaeological LOD are Wikidata, Geonames, Getty Thesaurus of Geographic Names (GTGN) and Pleiades.
Islands are important locational units for Pacific studies including our database for Pacific ceramic fabrics, Kuden. The Pacific Island Database (PID) of Nunn et al. 2016 contains 1778 Pacific islands and the Pacific Archaeological Radiocarbon Database uses it as a location source. PID (as represented by the paper Supplementary Table) has no island URI. Having URI for Pacific islands, sub-regions of larger islands, and archaeological sites, would promote linking of the growing number of different datasets related to Pacific archaeology, irrespective of whether the data is LOD. We examined how well Pacific archaeological sites and islands are represented in the common LOD location identifiers. Pacific ‘archaeological sites’ (Q839954) are less represented than European sites in Wikidata and the other databases. An exception being the sites in regions linked to the USA. Vanuatu, Samoa, Fiji and Tonga have one or zero ‘archaeological sites’ (July 2025). Wikidata, Geonames, and GTGN have at least 99%, 95% and 5% respectively, of the islands in the island Pacific Island Database. As part of the process of creating a LOD version of Kuden, we are adding Wikidata and Geoname identifiers to our site and island names. Kuden will now feature a mapping table for the Wikidata, Geonames and PID IDs.
Ian Johnson
The University of Sydney
From ‘data banks’ to AI : are databases redundant?
Our evolving software environment enables increasingly complex actions with simple commands, risking loss of our understanding of the taphonomy of our data. In the 1970s there was brief enthusiasm in archaeology for “data banks” which would “answer all our questions”, but reality of scale and standards soon set in; even today FAIR and LOD offer only the promise of distributed data integration, raising issues of semantics, ontologies, paradata and reproducibility.
So is AI the next great step in the creation of a global ‘data bank’? Ask a question on anything and an AI will assemble an answer. However we have little idea where its knowledge came from, how the answer was generated or its ‘hallucinations’; we have no control, so we should have no confidence; we cannot entrust our data to AI’s black box. We need open-source formats, good para/metadata, stable public servers, and control over how our data are modelled and stored. In short, we need transparent data handling. The days of the database are not numbered!
That said, AIs offer many advantages. They excel at coding, software revision, bug detection, documentation and smart help. They can detect patterns, suggest hypotheses, summarise and execute complex repetitive processing tasks. They will (soon) be able to build effective data models for a specific data system from a verbal description.
This paper explores how we are using AI to enhance the Heurist database builder (HeuristNetwork.org) and near-term evolutionary perspectives, while stressing the use of AI as an invaluable assistant, not a substitute for well-structured databases and transparent analysis. Don’t trust AI with your data!
Thomas Keep
The University of Melbourne
Do You See What I See? Digital methods on the Marzuolo Archaeological Project*
This paper outlines in brief the digital methods used on the Marzuolo Archaeological Project, with a particular focus on the recent addition of photometric stereo and reflectance transformation imaging. The MAP has now concluded excavations of a rural Roman minor centre from the late Republican and early Imperial period, which uncovered significant assemblages of terra sigillata and a well-preserved blacksmithing and woodworking toolkit. Digital recording through GIS and photogrammetric modelling have been integral to the recording workflow since its inception. In more recent years, object photogrammetry, photometric stereo imaging, and reflectance transformation imaging have been conducted on key artefacts from the assemblage. This paper will discuss the benefits these techniques offer to the project and the associated benefits and shortcomings of each method.
Ilhong Ko, Woo Jin Shim
Seoul National University
Re-populating the Bangucheon Stream World Heritage Landscape with GIS Data
The “Petroglyphs along the Bangucheon Stream” in South Korea were inscribed on the World Heritage List in 2025. The site comprises two rock panels rich in carvings, located at the northern and southern ends of a three-kilometer stretch of the Bangucheon waterway. Uniquely, the petroglyphs were produced over a long span of time by Neolithic hunter-gatherers, Bronze Age farmers, and literate individuals of the Silla Kingdom. Yet, a review of the nomination dossier and Statement of Outstanding Universal Value (OUV) reveals little attention to the broader Neolithic, Bronze Age, and Silla communities who created the carvings or may have visited the panels. This presentation seeks to enrich the understanding of the site by “re-populating” its heritage landscape through the visualization of contemporaneous sites, drawing on the “Buried Cultural Heritage Spatial Data Set” provided by the Korea Heritage Service’s “National Heritage GIS.” In addition, we apply a new route-modelling algorithm developed by the authors that builds cost-surfaces based on “hillslope position unit movement suitability (HPMS)” rather than slope gradient. By reconstructing potential routes linking Neolithic, Bronze Age, and Silla sites—some of which may have intersected with the Bangucheon petroglyphs—we highlight the importance of human movement in evaluating the site’s cultural and historical significance.
John McCarthy
Flinders University
Digital strategies for engagement with First Nations Underwater Cultural Heritage^
This talk explores the digital and analogue strategies used to engage both the public and Traditional Owners with submerged landscape archaeology, focusing on Australia. Maritime underwater heritage is among the most invisible forms of cultural heritage, a challenge that is especially acute for ancient terrestrial sites now drowned by sea-level rise prior to its stabilisation in the early Holocene. Buried beneath marine sediments and hidden from view, these landscapes remain largely absent from public awareness, media representation, coastal development planning, and marine science agendas.
The confirmed discovery of sub-tidal archaeological sites in Western Australia in 2019 has begun to shift this landscape, though progress remains slow in the face of industrial pressures and climate change. Importantly, many Traditional Owner groups have long recognised the presence of these drowned sites, emphasising enduring connections to Sea Country.
Our work combines digital innovation with field survey to raise the profile of submerged heritage nationally and challenge the cognitive and cultural inertia that has left it overlooked. Approaches range from 3D recording and animation to immersive experiences in Virtual and Augmented Reality. Together, these methods not only support underwater prospection but also open new pathways for making this invisible heritage visible in support of its ongoing management in Australia and beyond.
Bradley McDonald
Ōtākou Whakaihu Waka University of Otago
Going with the flow: A pilot experimental investigation into the in-water behaviour of pre-contact Māori fishing lures
In the early human settlement of Aotearoa New Zealand (thirteenth to fifteenth centuries) Māori ancestors made lures with stone shanks for fishing. Little is known of how the form and materials of lures impacted their in-water behaviour, and how this behaviour impacted their use. Given the important role of fish in traditional coastal Māori diets, understanding fishing methods and technologies contributes to our understanding of past human-environmental relationships. This presentation presents the methods and results of a novel project using experimental methods to test the functional behaviour of traditional Māori stone lure shanks. Replicas of archaeological lure shank examples from Waihemo/Shag River Mouth were created using 3D scanning and hand carving methods. Replica performance was measured in artificially generated water currents at controlled speeds. Footage from testing was then analysed in Kinovea, a motion tracking software, enabling the assessment of the in-water behaviour of these lure designs and the implications of these behaviours on past fishing methods. Several key aspects of shank design, which alter the lures’ behaviour, were identified as a result. Behavioural variation was observed in the required speed, angular deviation, and movement type, confirming the importance of lure design to fishing strategies, including possible fish species targeting. This project has served as a proof-of-concept study, confirming that pre-European Māori fishing devices can reveal past human-environmental relationships when tested through digital and experimental methods. This research method is relevant to international contexts, particularly in the Pacific.
Sébastien Plutniak
CITERES. CNRS
A Map to Navigate the Data Archipelago: the “Open-archeOcsean” Catalogue of Open-source Datasets for Pacific and Southeast Asia Archaeology. Situation, Limits, and Prospects^
Archaeological datasets are sparsely published and distributed, and efforts to collect and assemble them have been an inherent part of this scientific practice for decades. Contrary to other parts of the globe, Southeast Asia and Pacific archaeology does not benefit from large data infrastructures (e.g. “Ariadne” in Europe, “tDAR” in the USA, “ADS” in the United Kingdom, “DANS” in the Netherlands, etc.), resulting in higher dispersion of datasets and research efforts. In this context, an online dataset catalogue has been recently created and coined “Open-archeOcsean”, in reference to the European project “OCSEAN. Oceanic and Southeast Asian Navigators” where it was developed. “Open-archeOcsean” is an interactive online catalogue of open-source datasets for Pacific and Southeast Asia Archaeology, designed in compliance with the FAIR principles. As a collaborative initiative, “Open-archeOcsean” is an evolving tool, and contributions from users are welcome and will be beneficial to the whole archaeological community and beyond. Even if it draws on open-source software and is distributed under open licenses, this presentation will pay particular attention to its design and the actual limitations of some options adopted.
Michael Rampe, Katherine Moitino
Rampe Realistic Imaging Pty Ltd
Progress and Lessons from the Digitisation of a Maritime Museum Fleet
Further to the paper presented at the Auckland international CAA conference titled “Towards digitising a fleet”, this paper will present an updated case study on the ongoing work and innovations in this area over the past 18 months.
As high fidelity 3D digitisation in large state and national museums becomes more and more normalised, the uses and reasons for this activity are also becoming more appreciated and understood which in turn, drives further investment and uptake for this sector.
The maritime museum sector has some unique challenges when dealing with management of some of their items. Firstly, they are very very large. The largest accessioned museum item in Australia right now is the HMAS Vampire which is docked in Darling Harbour, Sydney. Secondly, they are actively degrading by the way they are stored, usually floating in corrosive salt water and open to a variety of animals and plant life constantly attacking from below and the vicious cycles of the sun and weather attacking from above.
The Australian National Maritime Museum in particular houses 53 historic vessels, 12 of which are floating. As part of the management of this fleet, a program of digitisation has been underway for the last few years and continuing opportunistically as access and need is available.
This case study will explain and discuss the approach taken for a variety of vessels since the previous talk as well as talk through lessons learnt, new processes invented and horizon uses of the already legacy data including gaussian splatting.
Michael Rampe, Benjamin Wharton
Rampe Realistic Imaging Pty Ltd
Unidentified Darling Harbour Boat 1 – Managing the Digital Journey
In 2018, when excavating waterfront land for the new Sydney Metro station, an unexpected assemblage of timber planking was uncovered. The timbers belonged to the wreck of an early Australian-built vessel, which at the time, was a rare, if not unique find in Australia. As is usual with modern finds such as these, detailed digital records were created along with the preservation journey.
When the find was removed piece by piece from the ground, georeferenced site photogrammetry scans were done for documentation. As the timbers first entered treatment in 2020, they were digitised with structured light meticulously to create a full record of the individual timbers and again in 2022, photogrammetry and further structured light was used on a few small tricky items. In 2024, after the curing process was complete, key timbers were scanned again to allow for morphometric analysis on timber warping and shrinking during the process.
Now, in 2025, as the timbers are cured and await their next step, the journey continues with a full digital audit to reprocess and adapt the digital material built over 7 years which is resulting in a full digital catalogue of timbers as well as a 1:5 3D printing endeavour to assist with the careful process of reconstruction.
This paper will provide an overview of these components and some of the challenges in managing this significant data set developed including commentary on the closing gap of data legacy with the speed of technology improvement.
Amin Ranjbari, Andrea Jalandoni, Atiyeh Ghorbani
Independent Researcher
A Non-Uniform Processing Method (NUPM) for Large Photogrammetry Datasets: Case Study of Shams-ol-Emareh, Golestan Palace (UNESCO World Heritage Site), Iran
Cultural heritage documentation benefits from high-quality 3D models that are geometrically and aesthetically accurate. Photogrammetry is used worldwide, yet increases in image resolution, image overlaps, and a combination of drone and terrestrial imagery have resulted in large datasets that raise new challenges. Conventional processing of large datasets requires expensive computational systems with high-capacity RAM. It also leads to large output files that are difficult to store; cloud share; or use in virtual reality, augmented reality, and gaming engines. In this study, we propose a non-uniform processing method (NUPM) to handle the 3D reconstruction of the Shams-ol-Emareh building of the Golestan Palace UNESCO World Heritage Site in Iran. We processed large datasets of photogrammetry on a consumer-grade computer to produce low-size point clouds and meshes with efficient texture size and resolution without sacrificing quality. The workflow first fragmented the object based on importance and roughness, processed each fragment separately, and then joined the fragments together. Non-uniform processing also meant that points, triangles, and pixels with a low level of importance were deleted from all parts of the object. The result was a point cloud, mesh, and texture where the space between points as well as the size of triangles and pixels were variable and non-uniform. In some cases, the number of points in the point cloud and triangles in the mesh were respectively reduced by more than 90% and 97%, leading to usable output sizes without any loss in data quality.
Aja Reyes
Guam Rock Art Study
The Guam Rock Art Study: Digital archaeology and creative/artistic expression
The Guam Rock Art Study (GRAS) is an ongoing passion project started in 2020 by a diverse team of professionals to survey previously reported rock art sites on Guam, the southernmost and largest island in the Mariana Islands Archipelago. As of 2024, the GRAS has surveyed 12 sites out of 16 known sites (additional sites were recently reported to the GRAS) using non-intrusive technologies and methods of digital archaeology. This included 360° recordings, laser scanning, structure-from-motion photogrammetry, Dstretch and was one of the first test cases for chiaroscuro photogrammetry. The findings of the GRAS study are preliminary and have led to more questions for the GRAS to consider. Still, the GRAS has allowed the most comprehensive assessment of Guam rock art to date. Furthermore, it has led to the interest of rock art recording in other Mariana islands and contributed to the repatriation rock art taken for the B.P. Bishop Museum collection in Honolulu, Hawaii a century ago.
In this presentation, I share how the GRAS has influenced my creative work in abstract figuration, conversely, how my art practice has improved my capacity to observe, identify and describe different motifs found in Guam rock art. I will compare art pieces created before and after joining the GRAS team (from 2021 to 2025), and GRAS examples of improved post-processing effort. Additionally, I explore how multi-disciplinary perspectives contribute to the discussions and interpretations of rock art, and how creative arts provide a way to visualize and communicate abstract gestating ideas.
Sayumphu Ros
CamTech University
Data Ethics in Aerial Archaeology at Angkor
“Aerial digital archaeology has become indispensable for investigating Angkor, where technologies such as LiDAR, drones, and satellite imaging generate unprecedented datasets for mapping and analysis. Yet, these advances are not neutral. This paper interrogates the ethical, legal, and political entanglements of aerial digital archaeology at Angkor.
Using Actor–Network Theory (ANT) as a methodological lens, the study conceptualizes technologies, international legal frameworks, Cambodian institutions, and local communities as co-constitutive actors in the archaeological process. The analysis traces how aerial datasets, often collected through international projects, are mediated, circulated, and appropriated, revealing structural asymmetries in access, control, and authorship. These patterns reproduce forms of data colonialism where knowledge extraction mirrors historical colonial encounters.
Drawing on Science and Technology Studies (STS), the research further identifies how gaps in governance frameworks perpetuate inequities. While instruments such as UNESCO conventions or national heritage laws nominally address cultural heritage, they remain fragmented and insufficient for regulating emerging digital practices.
This paper therefore advances a model of ethical aerial digital archaeology centered on epistemic justice. It calls for long-term stewardship agreements, equitable data-sharing practices, and the institutional empowerment of Cambodian scholars and stakeholders. Ultimately, the project reframes Angkor not only as a site of archaeological inquiry but also as a crucible for debates on heritage sovereignty, digital ethics, and the decolonization of archaeology in the Global South.”
Sofia Samper Carro, Xavier Roda Gilabert, Javier Sanchez Martinez, Susana Vega Bolivar, Jorge Martinez-Moreno, Rafael Mora Torcal
Australian National University
The SUGAR project: Sweetening archaeological data management from fieldwork to laboratory
Archaeological fieldwork and data collection are fundamental to the interpretation of past human behavior. Nevertheless, the lack of standardized protocols for archaeological data management poses significant challenges for information preservation, accessibility, and scientific collaboration. Current practices often rely on private physical or digital formats, limiting data availability and hindering the principles of open science. Data standardization is a significant gap in Australasian archaeological research, hampering universal approaches to key questions in the region.
The SUGAR project addresses the need for a standardized solution for managing archaeological data that can be implemented in Australasian contexts. SUGAR provides an integrated, user-friendly solution for the entire archaeological research workflow, from fieldwork to laboratory analysis, data storage and dissemination.
SUGAR integrates open access tagging technologies such as Datamatrix with the use of a total station, personal digital assistants, and the management of three-dimensional data of contexts and artifacts recorded during excavation. The software offers QGIS archaeographic applications (horizontal and vertical plots, stratigraphic analysis, documentation of objects and structures, etc.). Thanks to its flexibility, it creates a dynamic interface between information retrieval and interpretation, stimulating a dialog between data and the development of hypotheses that can be tested as fieldwork progresses.
Elizabeth West
Independent Researcher
Shopping in Colonial Parramatta
I developed a blog using a number of digital methods to illustrate the possibilities with publication of academic material for a wider audience. The project drew from a key primary source store records from Rowland Hassall’s store in Parramatta in 1803-4. From this analysis was undertaken of purchasing patterns. Deeper investigation was undertaken of clothing as the predominant item in sales, and further sources investigated related to advertising and dress. Further research undertaken on a selected number of individuals to illustrate their story, social position and purchasing power. Digital methods used include Rstudio, PDF and text analysis tools, Voyant Tools, data analysis and cleansing using Openrefine, pivot tables and charting using Excel and Googlesheets, QGIS mapping and site development using WordPress. The site can be viewed here.
Bradley Young, Paul Penzo-Kajewski, Rebekah Kurpiel, Prishanta Gunawardhana, Keir Strickland
La Trobe University
LiDAR and Medieval Urban Landscapes in Sri Lanka
This paper presents preliminary results from drone-mounted LiDAR survey conducted within the archaeological reserve of Polonnaruwa, Sri Lanka, and its surrounding landscape. Polonnaruwa is a medieval urban centre and a UNESCO World Heritage Site located within the Dry Zone of northern Sri Lanka. The data presented in this paper was collected as part of the ARC funded Polonnaruwa project ( DP190100485) and was acquired using a comparatively affordable “off the shelf” package of a DJI Matrice 350 RTK drone equipped with a Zenmuse L2 LiDAR sensor.
Preliminary results highlight the potential of such “off the shelf” drone mounted LiDAR to map urban and sub-urban archaeological landscapes within a tropical environment. Although still a work in progress, the preliminary LiDAR data already appears to provide new perspectives on the city’s spatial layout and organisation, despite over a century of archaeological research at the site. LiDAR surveys at additional archaeological sites within Polonnaruwa’s hinterland further highlight the potential of such techniques to identify and map a range of archaeological features, including comparatively subtle landscape modifications that have been missed by terrestrial survey.
Finally, this paper explores the capacity for LiDAR data to support site monitoring, detect looting activities, and inform cultural heritage management strategies. This paper explores issues of point cloud classification and processing, DEM production, data visualisation, and feature detection using “off the shelf” products such as those applied here.
