Volume 40, N. 2

May-August 2017

Three-Dimensional Geostatistical Estimation of Soil Units: A Case Study From Capitão Pocinho, Pará, Brazil

Case Study

Volume 40, N. 2, May-August 2017 | DOWNLOAD PDF (5 downloads)

Abstract

Lithology volumetric estimation was performed from samples of drills using three-dimensional geostatistical techniques. Monte Carlo simulations were performed to test the accuracy and influencing factors. Next, procedures were applied to a case study related to the environment. The simulation results show that estimates obtained via block Kriging can be relatively accurate. The case study in this research was challenging, given the quantitative limitations of information collected in the area. However, block Kriging proved to be consistent and feasible, and observed the model’s assumptions. It also offers the advantage that, after obtaining data from the drills, all necessary computational tools are available for free. A computer code was developed in the R program for mapping 3D and calculation of volumes of lithology. These results open precedents for future research using other geostatistical techniques, such as Indicator Simulation, Spatio-temporal Analysis, or even co-Kriging, correlating the lithologies with variables related to environmental contamination by pollutants. In the study area, the lithological profiles of piezometer wells revealed that the vadose zone was composed predominantly of sandy-clay material. The soil classified as sandy clay may have been associated with free aquifers. This sandy-clay material totaled around 2.67% of the study area, with a volume of approximately 596 240 m³. The classified material, such as silty clayed sand and sand, which seemed to predominate in the saturated zone, represented about 14.31%and 13.1%, respectively. In this case, the volume estimated for each lithology was around 3 189 280 m³ and 2 916 384 m³, respectively. The highest prevalence was sandy clay with 61.8% of the region’s volume, which is about 13 795 264 m³. Finally, the silty sand represented about 8.2% of the total, with an estimated volume of 1 836 672 m³.

Keywords: Lithology, Monte Carlo simulation, Block Kriging,


Submitted on November 28, 2016.
Final Acceptance on June 24, 2017.
Discussion open until December 29, 2017.
DOI: 10.28927/SR.402187