Towards Geostatistical Learning for the Geosciences: A Case Study in Improving the Spatial Awareness of Spectral Clustering

Mathematical Geosciences - The particularities of geosystems and geoscience data must be understood before any development or implementation of statistical learning algorithms. Without such...     

Band Selection and Dimension Estimation for Hyperspectral Imagery—a New Approach Based on Invasive Weed Optimization

Currently, hyperspectral images have potential applications in many scientific areas due to the high spectral resolution. Extracting suitable and adequate bands/features from high dimensional data is a crucial task to classify such data. To overcome this issue, dimension reduction techniques have direct effects to improve the efficiency of classifiers on hyperspectral images. One common approach for decreasing the dimensionality is the feature/band selection by considering the optimum dimensionality of the hyperspectral imagery. In this paper, a new method was proposed to select optimal band for classification application, based on a metaheuristic Invasive Weed Optimization (IWO) algorithm. In this regard, the K-nearest neighbour (K-NN) technique was used as the classifier. Moreover, as a by-product of our band selection method, a new method was proposed to estimate an optimum dimension of the reduced hyperspectral images for better classification. Experimental results over three real-world hyperspectral datasets clearly showed that the proposed IWO-based band selection algorithm of this study led to the significant progress in selecting suitable bands for classification applications and estimation of optimum dimensionality of these datasets. In this regard, the overall accuracy (OA) of classification of the proposed IWO-based band selection algorithm was 92.02, 93.57, and 89.72 % for each dataset, respectively. Moreover, results reveal the superiority of the proposed IWO-based band selection algorithm against the other algorithms including GA, SA, ACO, and PSO for band selection purpose.     

Application of plurigaussian simulation to delineate the layout of alteration domains in Sungun copper deposit

An accurate estimation of mineral grades in ore deposits with heterogeneous spatial variations requires defining geological domains that differentiate the types of mineralogy, alteration and lithology. Deterministic models define the layout of the domains based on the interpretation of the drill holes and do not take into account the uncertainty in areas with fewer data. Plurigaussian simulation (PGS) can be an alternative to generate multiple numerical models of the ore body, with the aim of assessing the uncertainty in the domain boundaries and improving the geological controls in the characterization of quantitative attributes. This study addresses the application of PGS to Sungun porphyry copper deposit (Iran), in order to simulate the layout of four hypogene alteration zones: potassic, phyllic, propylitic and argillic. The aim of this study is to construct numerical models in which the alteration structures reflect the evolution observed in the geology.     

Thresholds of land cover to control runoff and soil loss

This research focused on the determination of land cover thresholds that have a significant impact on runoff generation and soil loss at the pedon scale. For this purpose, six erosion micro-plots were set up on grassland and shrubland types of rangeland in the northeast of Iran, and the amounts of vegetation cover, litter, runoff and soil loss on them were measured. A factorial statistical analysis was carried out on the completely randomized design using land cover and rainfall factors. The results show that the effect of rainfall on soil loss and runoff was greater than that of land cover. Also, the effect of land cover on soil loss was greater than that on runoff generation. Furthermore, two specific thresholds were identified: the first was from 10 to 30% of landcover and the second from 50 to 70%.     

Simulation of the lately injected dykes in an Iranian porphyry copper deposit using the plurigaussian model

Ore deposits are usually composed of rock units or facies with different grade distributions and complex spatial structures. Being able to simulate the spatial layout of these facies are essential to have a comprehensive mining plan and an accurate resources and reserves evaluation. Modelers are faced with a set of challenges when creating the facies model such as: reproducing the facies proportions and spatial continuity as well as the topological contacts between facies, capturing post depositional overprinting, and honoring the data obtained from drill holes. Plurigaussian simulation (PGS) is a geostatistical approach that allows covering these challenges. This study addresses the application of PGS to Sungun porphyry copper deposit (Iran), in order to simulate the layout of three facies: mineralized porphyry and skarn and non-mineralized dykes. The aim of this study is to construct numerical models in which the dyke structures reflect the evolution observed in the geology.