DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
(titres de publication, de périodique et noms de colloque inclus)
2018-10-22 - Colloque/Présentation - poster - Anglais - 1 page(s)

Wattier Marie-Laure , Vandycke Sara , Bergerat Françoise, Descamps Fanny , Tshibangu Katshidikaya , "Roughness description and characterisation of major fractures in Coniacian chalk at Wellington quarry (Arras, France)." in Réunion des Sciences de la Terre, Lille, France, 2018

  • Codes CREF : Géologie (DI1411), Mécanique des roches (DI1418)
  • Unités de recherche UMONS : Génie Minier (F408)
Texte intégral :

Abstract(s) :

(Anglais) Marie-Laure Wattier, Sara Vandycke, Françoise Bergerat, Fanny Descamps, Jean-Pierre Tshibangu Roughness description and characterisation of major fractures in Coniacian chalk at Wellington quarry (Arras, France). A dozen meters below the city of Arras lies a vast network of about 20km of tunnels initially resulting from the intense production of chalk by room-and-pillar method since the XVIth century. The network was further developed into warrens in the First World War. The architecture of the widespread complex lying under the city clearly benefits from an intelligent use of the main tectonic structures present within the Upper Cretaceous Coniacian chalk rock mass. The objective of our study was to establish a description of chalk fractures from the small-scale (sub-millimetric) up to the macroscopic scale, associating the analytical quantitative description of joints and faults in terms of roughness, together with qualitative observations of the geological features, within the related tectonic system. In the southern-most parts of the Wellington quarry complex, the primary networks of subvertical joints and faults, orientated 105-125°N and 160-180°N, allowed for relatively easy rock removal during quarry production and provided planar pillar walls for a naturally stable structure. So, natural fractures can be directly observed and characterised along the galleries. Samples were hence collected at 12 different spots. Once in the lab, small cylinders were cored from collected blocks. Series of cylinders were broken in half - by tensile (Brazilian) or shear method to be compared with natural fractures. On the natural surfaces, fault planes striations, ridges, twist hackles and plumose structures were observed. After visual classification, all samples were scanned with a high-precision laser to digitize the 109 fracture surface topographies (grid spacing of 172µm in X and Y directions and Z accuracy +/-30µm). Systematic computation of roughness indices was conducted using statistical parameters (average height and associated standard deviation Ra and σa, RMS parameter Z2) and fractal dimensions (semi-variogram, yardstick rule). Z2 and yardstick fractal dimension Dyard provide the most consistent results in comparison to visual observations. Lab-generated fractures display distinctively higher roughness indices than natural fractures – with for example, Z2 varying from 1.22 to 0.41 and Dyard varying from 1.0104 to 1.0039. Within natural samples, no clear trend differentiate joints from fault fractures. Lab-generated series of fractures present greater roughness for shear fractures with increasing confinement; the highest roughness values being seen on tensile fractures.