Levantamento GPR e geoelétrico para a caracterização 3D de feições cársticas do tipo paleocavernas colapsadas na borda Oeste da bacia potiguar (RN/CE)

Abstract

This thesis presents the results of applying the Ground Penetrating Radar method (GPR) together with the resistivity method in an outcrop located on the western edge of the Potiguar Basin, NE Brazil. In the study area, a system of numerous collapsed paleocaves outcrops, partially exposed on a roadway scarp. The purpose of this research was to evaluate the performance and reliability of the abovementioned geophysical methods for imaging this karstic features and their complex internal geometry. For this purpose, methodological adjustments on data acquisition, processing and visualization were made. The results are presented in the form of geophysical sections and volumetric blocks of resistivity and GPR data. Electrical and petrophysical properties of outcropping geological units were obtained based on geophysical data and porosity and permeability measurements of rock samples. The collapsed paleocaves were interpreted based on the identification of high resistivity zones and high amplitude GPR reflectors. The paleocaves occurs in the shallow subsurface (vadose zone) and are filled with tufas, breccias and speleothems, whose average porosity is about 35%. The lateral contacts with carbonate soil and host rocks are marked by zone with low to medium resistivity (3-1,250 Ωm) and GPR reflections ranging from low amplitude to transparent. The paleocaves developed in both modern tufas deposits and Cretaceous carbonate rocks, which have now become weathered, forming a light gray soil. . A preliminary evolutionary model for the partially outcropping karstic features on the western edge of the Potiguar Basin is proposed based on both geophysical data and geological mapping. The formation processes of paleocaves propagated in depth to reach the calciferous sandstone layer, whose top represents the lower limit of these karstic features. The high resolution of GPR data allowed even identify distinct radar facies within the paleocaves, which possibly could be related to different types of carbonate materials that filled the paleocaves before and after their collapse. For better identification and delineation of subsurface paleocaves, mathematical operators were applied to synthetic and real GPR data, such as instantaneous frequency, amplitude and geometric attributes. The resulting radar sections assisted in characterizing the pattern of response of collapsed paleocaves regarding bedrocks and allowed to highlight the numerous fractures and contacts of clasts, caused x by the collapse of the paleocaves roof and walls. The GPR attributes contributed for imaging the 3D geometry of the collapsed paleocave system more accurately. Specially, the digital volumes of resistivity and RS Energy attribute show details of the complex 3D distribution of collapsed paleocavernas system. In depths of around 10 m, the paleocaves are more widely spaced and eventually isolated, sometimes vertically connected through shafts. However, in the near surface, the paleocaves occur interconnected by pipelines or coalesced, forming a paleocave system of hundreds of meters long in an area of 12,000 m2.