Exploring the n Dimensions of Archaeology/L’archéologie sous toutes ses dimensions.

Date/Time: 
Thursday, May 16, 2019 - 9:00am to 2:40pm
(CST)
Room: 
Ste-Foy
Organizer(s): 
  • Nicolas Cadieux, McGill Applied Remote Sensing Lab
Session Description (300 word max): 

As archaeologists, we are trained to think first and foremost in terms of time dimension.

Yet, continual methodological and technological advances are bringing new dimensions into perspective. The democratization of GIS through OpenSource software, the accessibility of precise GNSS, and progress in photogrammetry, remote sensing, drones and ground-based devices like laser scanners, have allowed 3D space, n dimensional datasets and n dimensional hyperspectral images (HIS) to complement the time dimension. In a science where paper maps and charts still lay on every desk, we will explore how these new dimensions offered by the desktop computer through spatial modeling, point clouds and hyperspectral data cubes can add depth to our knowledge of archaeological sites.

En tant qu’archéologue, la dimension temporelle prend toute son importance.

Toutefois, les progrès du point de vue méthodologique et technologique nous forcent à considérer de nouvelles dimensions. La démocratisation des systèmes d’information géographique (SIG) via les logiciels libres, l’accessibilité des systèmes de positionnement par satellite de haute précision (GNSS), les progrès en photogrammétrie, télédétection, au niveau des aéronefs télépilotés et des relevés au laser 3D permettent d’enrichir l’espace-temps avec la dimension spatiale 3D et les jeux de données et d’imagerie hyperspectrale multidimensionnels. Dans le but de mieux comprendre les sites archéologiques, nous explorons les multiples dimensions offertes notamment par la modélisation spatiale, les nuages de points et les données hyperspectrales.

Presentations
09:10 AM: Archaeological survey and the precision of GNSS receivers
Presentation format:
Author(s):
  • Simon Paquin - Université de Montréal
  • Samuel Seuru - Université de Montréal
  • Ariane Burke - Université de Montréal
  • François Girard - Université de Montréal

Archaeological surveys use a variety of GNSS receivers (Global Navigation Satellite System) to record the location of artefacts and artefact clusters. Manufacturer’s generally state the accuracy and precision of their devices but do not consider the impact of precision on the ability of the devices to maintain the spatial integrity of clusters of points at scales archaeologists typically work at. Few studies have addressed this issue before now. Yet, this information is critical to assessing how confidently we can interpret “clusters” of artefacts on surveyed archaeological sites and at what scale. With this question in mind, we designed an experiment testing different GNSS receivers of varying cost and advertised accuracy. The experiment was carried out in order to compare the performance of the devices in field-like conditions. We test instrument precision and quantify the shape deformation of clusters of points at 4 distinct spatial scales relevant to archaeological interpretation.  Here we set out how the experiment was implemented and present the results of our comparison.

09:40 AM: Avantages et réflexions sur l’application de la photogrammétrie en bioarchéologie
Presentation format:
Author(s):
  • Anthony Rochon - Université de Montréal
  • Alexandre Bisson-Larrivée - Université de Montréal
  • Diane Martin-Moya - Université de Montréal
  • Isabelle Ribot - Université de Montréal

En sciences humaines, les techniques numériques ont pris une place importante dans les recherches bioarchéologiques au cours des dernières années. L’imagerie 3D, dont la photogrammétrie, est de plus en plus utilisée pour optimiser la collecte de données virtuelles pour affiner les recherches et la diffusion vers le public. Dans le cadre d’une  approche préliminaire à la photogrammétrie réalisée durant le premier cycle d’étude universitaire, un premier inventaire anthropologique virtuel a été développé à l’Université de Montréal.

Ces archives numériques ont pour but d’évaluer les possibilités d’intégrer cette méthode dans la formation académique en bioarchéologie. L’acquisition de ces données s’inscrit dans le cadre d’une recherche doctorale en phylogénétique (Diane Martin-Moya sous la direction d’Isabelle Ribot) et des activités de l’équipe ArchéoScience, ArchéoSocial (As2) (anthropologie, Université de Montréal). Cette expérience a permis de tester différents protocoles afin d’optimiser l’acquisition de données sur des collections ostéologiques issues du Laboratoire de bioarchéologie humaine destinées à la ré-inhumation.

Cette présentation a pour objectif d’exposer : ii) les avantages d’intégrer ces compétences au sein d’une formation universitaire préliminaire en bioarchéologie; et iii) les retombées au sein du milieu académique ou professionnel au Québec. La création de modèles numériques présente de nombreux avantages scientifiques, pédagogiques et de diffusion vers le grand public, en plus de fournir un outil d’archivage d’objets archéologiques de haute qualité.

 

10:30 AM: In-field 3D documentation using close-range photogrammetry - The case of the Paleolithic site of Riparo Bombrini (Italy)
Presentation format:
Author(s):
  • Diane Martin-Moya - Département d’Anthropologie – Université de Montréal
  • Catherine  Brun - Département d’Anthropologie – Université de Montréal
  • Fabio  Negrino - Dipartimento di Antichità, Filosofia e Storia (DAFIST) – Università di Genova
  • Julien Riel-Salvatore - Département d’Anthropologie – Université de Montréal

Archaeological excavations are by nature destructive processes, since investigating and documenting past contexts requires paradoxically that archaeologists destroy their initial integrity. With the optimization and accessibility of imagery technologies, it is increasingly possible to digitize a collection of accurate and high-quality datasets that preserve snapshots of a site’s initial structure. In particular, there has been a growing recognition that using photogrammetry is a better alternative than using surface scanners in the field in order to generate photo-realistic 3D models. Photogrammetry is cost effective, replicable and user-friendly. Riparo Bombrini (Liguria, NW Italy) is a small but complex site as a result of successive phases of development of the cliffs at the base of which it is found since the 19th Century. This poses a series of challenges and opportunities for the use of 3D technology. Here, we use the 2018 field season at the site as a case study to test the viability of applying photogrammetry to record the day-to-day evolution of the excavation and to verify the use of adding such datasets to traditional field documentation strategies. It also provides an opportunity to assess the feasibility of applying such methods (acquisition and data processing) by archaeologists and students with no prior 3D modeling experience. Using Agisoft Photoscan, an semi-automated 3D software combined with spatial data, we generated preliminary virtual models that serve two distinct functions: 1) extract detailed information from the excavation (orthophoto) and to generate digital elevation models for GIS and statistical analysis; and 2) improve public outreach.

11:00 AM: Emerging Technologies using LIDAR and Elevation to Locate Palaeo-Coastal Archaeological Sites
Presentation format:
Author(s):
  • Alexandra Lausanne
  • Daryl  Fedje - Hakai Institute; University of Victoria
  • Quentin  Mackie - University of Victoria

LIDAR (Light Detection and Ranging) and LIDAR-derived products such as DEMs (Digital Elevation Models) have emerged in geoarchaeological research as a key tool for mapping landscapes and aiding in archaeological site discovery and interpretation. DEMs and bare-earth DTMs (Digital Terrain Models) can be used to improve the time and efficiency of field surveys. A case study on Quadra Island, British Columbia, utilized these products as important components in an integrated approach to uncovering previously unknown archaeological sites using an archaeological potential model. Palaeo-coastal archaeological sites that are now located at inland and elevated locations were able to be identified by combining: i) geomorphic interpretation of the landscape through LIDAR, (ii) GIS-based archaeological site potential mapping, and iii) local Relative Sea Level (RSL) history. On Quadra Island, the Pleistocene-Holocene RSL shows regression over the past 14,300 years due to post-glacial isostatic rebound. In a period of 50 years, the span of a human lifetime, people in the terminal Pleistocene would have seen the sea level fall up to five meters. By isolating two key elevations (10 m and 30 m above mean sea level), we were able to select two snapshots in time that may represent stillstands in the localized RSL regression. Subsequent archaeological findings suggest that this is a promising integrated methodology for archaeological prospection that can be applied to similar palaeo-coastal settings. 

11:30 AM: The Hitchhiker’s Guide to the Digital Landscape, Part 2: What’s on the Menu for Archaeologist?
Presentation format:
Author(s):
  • Nicolas Cadieux - Applied Remote Sensing Lab, Department of Geography, McGill University
  • Margaret Kalacska - Applied Remote Sensing Lab, Department of Geography, McGill University
  • Andre Costopoulos - Department of Anthropology, University of Alberta

Digital elevation models (DEMs) represent one of the first digital geographical information to be readily available and widely used by scientists. In archaeology, they are frequently employed because they represent a concrete, directly observable reality that humans can relate to: the surface of the planet. Moreover, DEMs allow a vast array of GIS-derived analyses (i.e. slope, shade, aspect, viewshed, path analysis, floods, drainage modelling) that all directly relate to how past and present human beings experience their environments. Since all these models and analyses rely on DEMs, we argue that digital elevation models should not be taken for granted. While they may all look good on the menu, it’s important to recognize that they come in many flavours!

Indeed, researchers navigate the world of geomatics without an awareness that there are multiple DEMs of variable quality to choose from. Since they tend to ignore the DEMs’ qualities and limitations, models are chosen not for their intrinsic qualities or level of precision.

In this study,  we evaluated the qualitative and quantitative characteristics of various SRTM-based DEMs.  This includes the yet-to-be independently tested NASADEM, coined by NASA as the “finest resolution, global, freely-available DEM products for the foreseeable future”. We also look at the CDEMs (Canadian Digital Elevation Model), ALOS World 3D - 30m,  Aster GDEM v.2 model and the TanDEM-X data (© DLR 2017) from the German Aerospace Centre (DLR).

01:40 PM: Documenter le passé, au présent, pour le futur !
Presentation format:
Author(s):
  • Richard Lapointe - Expertise laser 3D - iSCAN inc.

Technologies du XXIe siècles pour la documentation technique de sites, de monuments, d’immeubles, d’objets, d’œuvres d’Arts ou de vestiges patrimoniaux, le numérisation 3D et la photogrammétrie aident les professionnels à conserver, transformer, restaurer et mettre en valeur notre patrimoine. Des exemples concrets de différentes techniques employés sur des dizaines de sites, démontrent l’apport important que ces outils apportent pour la gestion, la représentation, la maintenance et la transformation de lieux, de sites, d’espaces, tous uniques, avec leurs intérêts propres.

Multidisciplinaires dans leurs bénéfices, nous tenterons, par un survol rapide, de démontrer les apports techniques d’applications sur des dossiers d’actualité, à la fois en archéologie, en architecture, en ingénierie, en gestion et en muséologie. Technologies du XXIe siècles pour la documentation technique de sites, de monuments, d’immeubles, d’objets, d’œuvres d’Arts ou de vestiges archéologiques, la numérisation 3D et la photogrammétrie aident les professionnels à relever, conserver, transformer, restaurer et mettre en valeur notre patrimoine.

Des exemples concrets de différentes techniques employés sur des dizaines de sites, démontrent l’apport important que ces outils apportent pour la gestion, la représentation, la maintenance et la transformation de lieux, de sites, d’espaces, tous uniques, avec leurs intérêts propres. Multidisciplinaires dans leurs bénéfices, nous tenterons, par un survol rapide, de démontrer les apports techniques d’applications sur des dossiers d’actualité, à la fois en archéologie, en architecture, en ingénierie, en gestion et en muséologie.

02:10 PM: hsi_3d_fusion.py: An OpenSource Hyperspectral Imagery – 3D point cloud fusion program.
Presentation format:
Author(s):
  • Nicolas Cadieux - Applied Remote Sensing Lab, Department of Geography, McGill University
  • Deep Inamdar - Applied Remote Sensing Lab, Department of Geography, McGill University
  • Margaret Kalacska - Applied Remote Sensing Lab, Department of Geography, McGill University
  • J. Pablo  Arroyo-Mora - Flight Research Laboratory, National Research Council of Canada, Aerospace Research Center, Ottawa
  • Oliver Lucanus - Applied Remote Sensing Lab, Department of Geography, McGill University

Hyperspectral imagers (HI) collect spectral information (amount of reflected solar radiation) within each pixel of an image (commonly referred to as a hyperspectral image or HSI). From this information, the physical and chemical properties of the materials within each pixel can be inferred. As such, these signatures can be used to detect anomalies such as man-made objects, disturbed soils, and other aspects of a study area that may not be visible to the human eye.

In a HSI, information is not collected with respect to the 3-D structure of the analyzed surface. By fusing HSI with a surface elevation point cloud, it is possible to introduce the 3-D structural information.

The presentation describes a data fusion algorithm to generate a 3D point cloud with integrated spectral information. The algorithm was applied two data sets that were characterized by distinct spatial resolutions. A high spatial resolution example (Mer Bleue Bog, Ottawa, ON) was generated by fusing a surface elevation point cloud (1.6 cm ground sampling distance) generated from UAV a based structure from motion - multiview stereo workflow, with a hyperspectral image that contains 288 bands/channels for each pixel (4 cm).

A lower spatial resolution example (Fort Senneville, Montréal, QC) was generated by fusing a satellite based false color image (1 m) with a LiDAR derived point cloud. These two examples are used to showcase and discuss the potential archaeology applications of the data fusion algorithm.