3D geostatistical modeling of Lascaux hill from ERT data
Introduction
The Lascaux Cave, located in the Dordogne department of France, is considered to be one of the most important prehistoric caves in the world. It contains some of the best-known Upper Paleolithic art, and is included in the list of UNESCO World Heritage Sites (World Heritage Committee, 1979).
The certification of the rock art by Henri Breuil in the same year (1941) of its discovery led to its classification among the ‘historical monuments’, i.e. the National Heritage Sites of France (Breuil, 1941). The Lascaux bestiary is composed of about 600 depictions (predominantly horses, deer, aurochs, and ibex) in addition to the famous hominoid depiction (Aujoulat, 2004). According to Leroi-Gourhan (1965), the cave paintings are from the Solutrean period. The study of the movable art gathered by Abbot Glory led Leroi-Gourhan and Allain (1979) to link the cave to the Magdalenian II period. Since its discovery, several issues related to the preservation of the paintings and engravings led the authorities to close the cave to public in 1963. Since 1964, a daily monitoring of micro-climatic conditions (e.g., temperature, CO2, humidity measurements) has been performed and scientists are still using the sensors deployed on site to maintain the cave in stable environmental conditions (e.g., Kervran and Fleury, 2013). Aiming to scientists can understand more about the geological structure and due to the strong protection policy a nondestructive geophysical survey based on electrical resistivity has been carried out. Therefore, we can estimate 3D limits between major geological formations around the Lascaux cave. The cave is located within an unsaturated epikarstic environment, which usually consists of a particularly weathered zone formed by limestone that lies immediately beneath the soil. The epikarst is a potential perched aquifer with a leaky capillary barrier at the bottom (Mangin, 1975). Moreover, this shallowest part gradually gives the way to the main body of the vadose zone that is mostly unweathered bedrock at depth. Both features constitute the upper part of the karst aquifer, which is typically 3 to 10 m deep.
We choose the electrical resistivity tomography (ERT) method to acquire the 2D resistivity as the entry data for the 3D modeling. This method is adapted to karstic heterogeneous environment and it has the major advantage of being non-destructive. The ERT can be used to provide conclusive mapping of karstic cavities (e.g., Chalikakis et al., 2011, Deceuster et al., 2006, Kaufmann et al., 2012, Sirieix et al., 2014). Several research works have supported its efficiency in identifying surface structure in a karstic environment (e.g., Valois et al., 2010, Gautam et al., 2000, Carrière et al., 2013, Martinez-Moreno et al., 2014). Moreover, the ERT is also a relevant tool to provide a reliable mapping of the structure of a karstic system up to 100 m depth (Šumanovac and Weisser, 2001).
The application of geostatistics to geophysical data has different purposes: for example, it is possible to rebuild a 2D data from the 1D data using geostatistical method. Sainato and Losinno (2006) used Vertical Electrical Sounding (VES) to map the spatial variability of an aquifer's resistivity, using a geostatistical method based on 35 VES surveys. Dewashish et al. (2007) used geostatistical methods “to refine and improve the VES interpretation with quantitative information on geology using borehole data”. Riss et al. (2011) have developped a methodology to convert traditional vertical electrical soundings into 2D resistivity models to propose a geologic model of the Saïss basin (Morocoo). Cassiani et al. (1998) used co-kriging procedure from the inclusion of seismic data and sonic log data to estimate hydraulic conductivity without postulated à-priori relationship. Tronicke and Holliger (2005) used geostatistical method to establish a conditional simulation from hydrogeophysical data. The simulation shows its applicability for the hydrogeophysical characterization of aquifers and the potential of flexible simulation techniques such as simulated annealing to generate aquifer models combining different data and a priori information.
Guekie Simo et al. (2013) have also experienced geostatistical method to develop an urban hydrogeological model. Their work showed the potential of geostatistics to transform the lithological data into hydraulic interpretation and the possibility to develop hydrogeological model on an urban scale area. Wang et al. (2013) applied a geostatistical approach to study the soil saturated hydraulic conductivities of the Loessial Plateau (China) at a regional scale. The spatial distribution of the hydraulic conductivities reflected soil hydraulic properties and the combined effects of soil texture, vegetation, topography and human activities.
Moreover, it is not always possible to arrange the geophysical survey in a 3D layout in the field sounding due to the accessibility, irregularity of site, length of survey line etc. In order to find a balance between the resolution and the coverage of ERT surveys on the site of Lascaux, we choose geostatistical modeling to create the 3D resistivity model from 2D resistivity data. Furthermore, to optimize the horizontal and vertical density of measurements, the survey grid has been designed by taking into account the latest geological settings.
Section snippets
Geological context
The Lascaux cave develops within the Coniacien limestone C4b (Fig. 1). Its main entrance is at 182.89 m NGF (French ordnance datum), which is about 100 m above the Vézère River. According to Schoeller (1965), its geological framework is as follows with new annotations of geological formation (Guillot et al., 1979):
C4a – Marls and marly glauconitic limestones from base of the Coniancian (10 m thick);
C4b – Coniancian limestone (100 m thick);
C5a – Sandy Santonian limestone;
SP – Fine-grained, yellow
ERT measurement
ERT surveys in the site of the Lascaux cave for the 3D geostatistical modeling have been carried out in March 2013 (Fig. 1). The measurements were performed with the resistivity SyscalPro® Switch 96. Seventeen profiles have been carried out in four directions (Fig. 2); for each profile, three types of arrays (gradient and pole-dipole forward/reverse) were used with 96 electrodes spaced 1.5 m. We checked the quality factor that is a coefficient of variation of the apparent resistivity data (
Geostatistical modeling
The kriging step allows us to get the 3D resistivity model. A view of the 3D model from the south (Fig. 5) shows the main structures of the resistivity model. We can identify five structures according to their resistivity ranges and next interpret them with regard to the previous geological studies (Vouvé, 1967, Lopez, 2009, Bruxelles and Camus, 2014) and our current knowledge of the site:
- •
At the east of the site, there is a superficial formation with estimated median of resistivity at 271 Ω·m;
- •
In
Discussion
This section is devoted to a detailed presentation of the various geological formations. Two thresholding values are chosen in order to separate the specified formations – 60 Ω·m for the western/eastern outcrops and 275 Ω·m for the center limestone promontory. The thresholding values are chosen according to a previous statistical analysis (Xu, 2015, Xu et al., 2014) that has been carried out based on two-years Time-Lapse ERT survey (PS and PN in Fig. 2). By applying a Hierarchical Ascendant
Conclusion
Based on ERT data and using the kriging interpolation method, the geological structures of Lascaux site are rebuilt, including both the limestone promontory and the sandy clay formations around the limestone. The resistivity of the 3D model gives a general view of the spatial distribution of resistivity at the site of Lascaux. This model identifies the limits between the limestone at the center of the site and the sandy clayey formations to the east and west of the site. These observed limits
Acknowledgements
We are grateful to the French Ministry of Culture and Communication (Regional Direction of Cultural Affairs of Aquitaine) for their collaboration and financial support in this study. We also would like to thank Stéphane Perrin's study of tree settings in Lascaux, and Laurent Bruxelles & Hubert Camus' geologic research on Lascaux hill, for providing insights in the development of this paper. We'd like also to thank Jean-Christophe Portais for their helps in finding the old borehole data of
References (39)
- et al.
Combining electrical resistivity tomography and ground penetrating radar to study geological structuring of karst unsaturated zone
J. Appl. Geophys.
(2013) - et al.
A geostatistical framework for incorporating seismic tomography auxiliary data into hydraulic conductivity estimation
J. Hydrol.
(1998) - et al.
2D cross borehole resistivity tomographies below foundations as a tool to design proper remedial actions in covered karst
J. Appl. Geophys.
(2006) - et al.
Mapping of subsurface karst structure with gamma ray and electrical resistivity profiles: a case study from Pokhara valley, Central Nepal
J. Appl. Geophys.
(2000) - et al.
Reconstructing hydraulic conductivity field for hydrogeological modeling in an urban environment
Eng. Geol.
(2013) - et al.
Geotechnical modeling at the city scale using statistical and geostatistical tools: the Pessac case (France)
Eng. Geol.
(2009) - et al.
Integrated geophysical methods for studying the karst system of Gruta de las Maravillas (Aracena, Southwest Spain)
J. Appl. Geophys.
(2014) - et al.
Evaluation of resistivity and seismic methods for hydrogeological mapping in karst terrains
J. Appl. Geophys.
(2001) - et al.
Regional-scale variation and distribution patterns of soil saturated hydraulic conductivities in surface and subsurface layers in the loessial soils of China
J. Hydrol.
(2013) Lascaux: le Geste, L'espace et le Temps
(2004)
Une Altamira française: la caverne de Lascaux à Montignac (Dordogne)
Compte-rendus de l'Académie des Inscriptions et Belles Lettres
Grotte de Lascaux - Formes et Formations Anciennes de Surface et Sub-surface, Montignac, Dordogne. Inrap/Protée. Personnal communication
Contribution of geophysical methods to karst system exploration: an overview
Hydrogeol. J.
Geostatistics: modeling spatial uncertainty
Reducing ambiguities in vertical electrical sounding interpretations: a geostatistical application
J. Appl. Geophys.
Notice de la carte géologique de la France à 1/50 000 – Terrasson (N° 784)
Regionalized Variables and Quantitative Analysis of Spatial Data. In Display and Analysis of Spatial Data
An electrical resistivity imaging-based strategy to enable the site-scale schedule over palaeokarst features covered in the area Tournai (Belgium)
Eng. Geol.
Lascaux, Une Caverne Ouverte Puis Refermée. France Culture 18 June
Cited by (18)
Imaging hydrological dynamics in karst unsaturated zones by time-lapse electrical resistivity tomography
2024, Science of the Total EnvironmentHeat transfer in rock masses: Application to the Lascaux Cave (France)
2023, International Journal of Heat and Mass TransferDetection of undercover karst features by geophysics (ERT) Lascaux cave hill
2020, GeomorphologyCitation Excerpt :The spacing between profiles is large, and, as we know, the karst heterogeneity is so variable that information can be easily missed between the profiles. We reduced this spacing from 20 m in Xu et al. (2016) to 5–10 m between two parallel profiles. As Sirieix et al. (2014) showed, the smaller the electrode spacing, the better we can detect karstic features.