Optimal grid resolution and canvas size
Heterogeneity
and uneven distribution are part of the nature of archaeological data. We have
proposed a method to define a common grid system and optimal grid resolution to
help archaeologists map and compare representations of their observations. More in ArchaeDyn
WG4 final report (Eng).
Representation/reliability/confidence
We have
developed a method to evaluate heterogeneity in archaeological inventories. It
is called a “confidence map” and is produced by combining representation and
reliability maps with map algebra. Read more in Oštir et al. 2007 (Eng).
Angle (orientation) histogram
A
historical field system can be documented in the form of digitized lines
representing field boundaries. We have designed a tool to summarize information
about the predominant direction(s) of the historical field boundaries. More in ArchaeDyn
WG4 final report (Eng).
Digital terrain model generation from a lidar point cloud
In
cooperation with the Slovenian Forestry Institute and the Jožef Stefan
Institute we have developed an algorithm for generation of a DTM from aerial
laser scanning point clouds. The REIN (Repetitive Interpolation) algorithm uses
data redundancy and spatial statistics for filtering out non-ground points. It
gives very good results in steep forested terrain and in low-penetration
conditions. Read more in Kobler et al. 2007 (Eng) and Kobler and Ogrinc 2007 (Eng).
Airborne lidar and historic environment records
The utility of lidar for finding new sites and defining historic
environments has been broadly recognised but systematic application and
evaluation is still in its infancy. With colleagues from the University of
Birmingham, UK, we have assessed the potential of lidar to enhance existing
records of the historic environment, using a case study from a river valley in
the agricultural midlands of England. Lidar data were compared with the
existing inventory of sites for the study area and with a selected sample of
vertical aerial photography. Read more in Challis et al. 2008 (Eng).
Application of laser scanning in the study of past
cultural landscapes – Kobarid case study
A Digital
elevation model, produced with the REIN (Repetitive Interpolation) technique,
for the area around Kobarid uncovered a previously unseen collection of remains
of past human activities and natural processes in the area, from the earliest
hill forts, moats and trenches, and First World War structures to the more
recent modifications of the landscape. The analysis has revealed details of
known archaeological sites and indicated locations of previously unknown ones.
Links
to posters (Si).
Read more
in (Si): Kokalj et al. 2008: Uporaba laserskega skeniranja za opazovanje
preteklih pokrajin - primer okolice Kobarida, in: Perko, D. et al. (ed.) Geografski
informacijski sistemi v Sloveniji 2007-2008, Ljubljana: 321-329.
Advanced lidar data visualisations for past cultural
features recovery in vegetated areas
Relief
visualisation is becoming an important part of spatial data interpretation, paticularly
in archaeological interpretations. Its main advantage is that it can simplify
and enhance the interpretation of features. A range of lidar data
visualisations, both traditional and novel have been compared and evaluated. These
techniques include: analytical hillshading, microrelief topography, derivatives
of hillshading from different directions (hill-shadings range, mean of
hill-shadings, hill-shadings PCA), composite of hill-shaded relief and
elevation differentiation, composite of hill-shaded relief and nDSM, composite
of DOF and nDSM, and sky view factor. The performance of this set of advanced visualisations
has been studied for different types of historical landscapes and
archaeological features (link to AARG poster, LIEPPEC posters).
Analysis of historical land use data
Historical
maps often contain important data for analysing long term changes and for
various other spatial analyses of land use, landscapes, urban development,
influences of the economic development, changes of toponyms, etc. An essential but difficult step in using old
data is the homogenization and transformation into modern cartographic
projections. The basic procedure for different study areas and sources is
similar on the whole, but differing cartography techniques, non-standardized
map legends, deterioration of the material and poor knowledge of projections
used makes each case quite unique. We have transformed several different
datasets produced for a range of application by differing mapping authorities,
of medium and large scale, going back two centuries. An extensive quality
analysis of the derived maps was performed followed by thematic interpretation.
Read more in Podobnikar and Kokalj 2008, Quality of historical land use data in
Triglav National Park, and Kiraly et al. 2008, Georeferencing historical maps -
methods and experiences, both published in (Eng.) Csaplovics et al. (ed.) Spatial
information systems for transnational environmental management of protected
areas and regions in the central European space : selected results and outputs
of the Interreg IIIB project SISTEMaPARC, (Fernerkundung und angewandte
Geoinformatik, Bd. 4), Berlin: Rhombos-Verlag.
Application of Internet GIS tools for heritage management
An internet
based mapping application for Archaeological Sites and Monuments Records of
Slovenia (ARKAS) has been designed and implemented at the Scientific Research
Centre of the Slovenian Academy of Sciences and Arts, with knowledge and
experience contributed by the ModeLTER team. The idea behind the project was that
internet based databases that include mapping capabilities can provide a useful
and efficient way of storing and disseminating data and results to the public
and researchers. Read more in Kokalj et al. 2007 (Eng.)
Pixel-based change detection on Landsat imagery
Change
detection methodologies for the use of optical remote sensing imagery (Landsat)
and their pitfalls when multitemporal analyses are performed with pixel-based
(raster algebra) techniques have been analysed. We note that a certain amount
of noise persists in the imagery, regardless of the care taken in preprocessing,
and this can drastically influence comparisons of imagery. Essentially, this
noise behaves as a detected change and the quantitative evaluation might fail
because of its effect on the identified change pattern. A multi-resolution
change detection approach is proposed to tackle this problem. Taking into
account the neighbourhood and change information derived by joining different
spatial scales, the multi-resolution approach effectively reduces the amount of
false changes. More in Veljanovski 2008 (Si).
Concept for a harmonized cross-border land information
system
The
consortium of Austrian and Slovenian research organizations and local
authorities has demonstrated the feasibility of harmonizating regional
cross-border land cover/land use information systems. The developed object
oriented data model has been tested in different study areas in both countries.
The most suitable processing methods for the classification of various land
cover types have been extensively evaluated (land cover types incl.
settlements, forests, alpine areas, agriculture, water, and non-alpine nature
areas). More in Granica et al. 2009 (Eng).
Observation of natural disasters with remote sensing
The ModeLTER
team has been actively involved in providing support for post damage assessment
from satellite imagery after major natural disasters in Slovenia. A range of
satellite images from different sensors have been processed and analysed in
studies of landslides (i.e. Oštir et al. 2007 (Eng); Oštir et al. 2008
(Si), Primerjava uporabe metodologije PSInSAR in DInSAR za opazovanje premikov
površja - primer severozahodne Slovenije, in: Perko, D. et al. (ed.) Geografski
informacijski sistemi v Sloveniji 2007-2008, Ljubljana: 63-71), floods after
torrential rains (i.e. Kokalj et al. 2007 (Eng.); Oštir et al. 2007 (Eng.)), and in general (Lamovec et
al. (Si) 2010, Opazovanje naravnih nesreč s satelitskim daljinskim zaznavanjem,
in: Gomboc et al. (Ed.) Slovenia and space: yesterday, today and tomorrow,
Ljubljana: 59-61).
Dynamic road pricing
Members of
the team have developed a dynamic road pricing model and defined variables for
the impact of road transport on the environment, which are appropriate for
modelling and monitoring with remote sensing techniques. Notably, this project
focused on environmental noise modelling. Manual, semiautomatic and automatic
methods for defining the parameters (structural forms, relief morphology and
break lines, and acoustic impedance of ground) were selected and tested, and
guidelines for their expected accuracies and potential to be monitored with
remote sensing were presented. Also developed and tested was a new method to link
the measured spatial distribution of noise impacts on the environment to kilometre
road segments, based on the study of noise impacts on humans. More in Kokalj 2010 (Si).
Renewable energy sources and solar radiation modelling
Solar
radiation is a renewable and abundant energy source and therefore it is sensible
to take it into consideration when planning new buildings or renovating old ones.
We have analysed how the amount of solar radiation a certain part of terrain receives
is influenced by the shape of relief and other factors. We have also determined
the amount of received solar radiation for Kras region, Slovenia on a monthly
basis, and estimated an optimal combination of azimuth and inclination of solar
energy collectors. Read more in Zakšek et al. 2007 (Si) and a book (Si) Kras: trajnostni razvoj kraške pokrajine,
2008, Ljubljana, Založba ZRC: 302-337.
ModAgSpace: Lidar in the Languedoc
The plain
of Mauguio (France, Lunellois) is a study-area of the European Associated
Research Laboratory ModeLTER coordinated by L. Nuninger (UMR6565, Besançon) and
K. Ostir (ZRC SAZU, Ljubljana).
Lidar and hyperspectral surveys were conducted in 2007 in this area in order to
study changes in the level of the bay of Mauguio, to reconstruct a paleo-DEM of
the area, and better understand the various factors (importantly hydraulic)
which have affected development in this plain since pre-history.
This data is now being used to study the development of the agricultural
landscape, focused on long term trends in agrarian manuring patterns. The
ceramics-based archaeological data on agrarian manuring were acquired in the
1980's by F. Favory and C. Raynaud and are now being integrated with data on
field boundaries and systems and large scale changes in terrain morphology derived
from the lidar survey, making use of spatial analysis and terrain modeling
techniques. The aim of this project is to relate the time-space dynamics of
agricultural activities to the variations in the level and coastline of the bay
and the enhancement of the plain through hydrological management systems. More
in Poirier et al. 2010 "The Modagspace project: Lidar data
and landscape archaeology in southern France (Languedoc)", communication
au XXXVIIIe CAA (Computer Applications in Archaeology) Conference, Grenade (Espagne),
6-10 avril 2010. (Eng). (link to ppttt)
The LIEPPEC Project: Lidar around Mandeure-Mathay
The
villages of Mandeure and Mathay are located on the site of the Roman
town of Epomanduodurum and an earlier unnamed Iron Age settlement. The
substantial Roman town had shrunk significantly by the Late Antique Period and
was reduced to villages by the Early Medieval Period. The long term history of
this settlement and the surrounding rural landscape is the subject of a
research project led by Phillipe Barral (Université de Franche-Comté). Resistivity,
magnetometry and ground penetrating radar have been deployed over the past six
years to elucidate the plan of the Roman town of Epomanduodurum (modern
Mandeure-Mathay). The combined geophysical surveys have revealed a Gallo-Roman
sanctuary, temples, residential and artisinal quarters, and the urban road
network. The lidar model, based on data collected in 2009, is now being used to
study the topographic context of urban monuments and to support a detailed
study of the changing course of the Doubs river, which passes through the
town. Pierre Nouvel has been leading a campaign of intensive fieldwalking
around Mandeure to study the town's rural hinterland and the long term
development of this landscape. Lidar data was acquired in 2009 for an 81 km2
window around Mandeure-Mathay. In late 2009 field prospection began in the
forested areas surrounding Mandeure, based on the analysis of the lidar terrain
model. More in Thivet
et al. 2009 (Fr).
(link to Foire Comtoise Poster)
The LIEPPEC Project: Lidar around Besançon
Lidar data was collected for an 120
km2 area around Besançon in 2009 with support from the Regional
Council of the Franche-Comté. This data is now being studied within a new
archaeological research project focused on the rural landscape around Besançon.
A new campaign of field prospection, based on lidar data acquired in 2009 and
currently being analysed, is underway to develop new approaches to survey in
wooded and scrub landscapes. Particular emphasis is placed on the
interpretation of lidar data and the integration of lidar data with data
collected through field prospection. See some early results in Opitz et al. 2010 (Eng). (link to LAC Poster)