This research is focused on the definition of methods for the use of Light Detection And
Range - LiDAR data in structural geological mapping. In the Amazon Region, in particular in
the banded iron formations of the Mineral Province of Carajás - PMC, the vegetal cover
compromises the extraction of structural information from optical remote sensing data, since it
masks the geomorphological and structural features that can correspond the geological
structures . This thesis addressed the application of digital image processing techniques of
LiDAR data, from i) shaded relief model - MRS do, which used a configuration with eight
lighting modes with different views ranging from 45º between them, to generation of alignment
maps; and ii) laplacian filtration in shaded relief model - FLMRS in multidirectional visions. In
both resulting maps the same structural trends, NE-SW, NW-SE and secondarily N-S and E-W
were observed. However, the map generated from the FLMRS in multidirectional visions
presented a better geometry of the spatial distribution of the lineaments. The results obtained
from the analysis of LiDAR data were compared with structural data collected and analyzed
from the study of iron ore deformation and associated volcanic rocks at the Serra Sul de Carajás
mine - S11D. The structural analysis showed a single episode reflecting shortening in the E-W
direction, with the installation of two fault systems, one NE-SW direction and the new NW-SE
direction. This deformation is responsible for the geometry of the rocks of the S11D plateau,
forming a sinformal / antiformal sequence with NE and SE trim along with these fault systems
and with axial planes dipping at high angles to NW and NE forming folds with complex
interference patterns. Lamination in iron ore still preserves primary structures and there is no
evidence of pervasive milonization in these rocks. It proposes a deformational model for the
rocks of the S11D Plateau related to transpression controlled by the Carajás Fault during
regional sinistral movements. The comparative analysis of the structural maps generated with
airborne LiDAR data and traditional manual methods were performed from the comparative of
the two maps. The calculation and quantification of the dispersion of the structural contour
lines in each area from the different methods showed a correlation coefficient that varied
between 0.91 and 0.93, suggesting a good degree of similarity between the mapped structures,
although there are variations in the orientation and abundance in the structural contour lines. In
general, the results demonstrate the effectiveness of the airborne LiDAR data to extract detailed and accurate structural information on tropical terrain and can be used to complement the fieldbased
structural mapping.
|
