Abstract :
[en] Bernissart : Back to the crime scene
Thierry MARTIN1,*, Olivier KAUFMANN1
1 University of Mons
Faculty of Engineering, Mining and Geology Department
Place du Parc, 20
7000, Mons, BELGIUM
*Corresponding author: thierry.martin @umons.ac.be, +32 65 37 46 13
Keywords: Bernissart, Iguanodon, Geological modelling, 3D.
Abstract
In Bernissart, a large number of Iguanodons skeletons, many of them complete and articulated, were unearthed between 1878 and 1882. The discovery of these skeletons was made in a coal mine when an exploration gallery encountered a sinkhole affecting the coal seams. In order to reconstruct the geometry of the Bernissart sinkhole (a.k.a 'le cran'), its neighbouring environment and the excavation works of the skeletons within the sinkhole, we designed a workflow to processes the available data and build a 3D model.
For a start, we inventoried the documents available at the RBINS and other relevant sources. These documents were scanned or photographed. A special attention was given to minimize distortions of plans and cross-sections. The documents were then georeferenced and relevant information in these documents where vectorised using semi-automatic procedures. The vertical axis is specified as depth below the Sainte-Barbe mine shaft.
From maps and cross-sections of the coal works, the geometry of coal seams such as 'Luronne', 'Présidente' or 'Veine du Fond' seams is then reconstructed in the vicinity of the Iguanodon and South sinkholes. Outer limits of the sinkholes were also modelled. This reconstruction shows in detail the relationships between the geometry of the seams and the position of the sinkholes and sets the general geological context of the bonebeds.
The next part of our work is dedicated to process the descriptions made during the paleontological excavation, especially sketches of the plaster casts of the skeletons. These sketches have to be vectorised, oriented and scaled. Georeferencing is much more difficult with this data. Available information from cross-section and general setting plans must be included in the reflexion. Moreover, the spatial information are often incomplete and cross-checks or even hypotheses have to be made in order to replace these data in a 3D framework.
