Search Results

You are looking at 1 - 10 of 10 items for :

  • "cluster analysis" x
  • Earth and Environmental Sciences x
  • All content x
Clear All

quality data or link them to properties of the sampling location / related watershed. Statistical models mostly consist of multivariate analysis tools (hierarchical cluster analysis [HCA], principal component analysis, correlation analysis

Open access

location / related watershed. Statistical models mostly consist of multivariate analysis tools (hierarchical cluster analysis [HCA], principal component analysis, correlation analysis, linear discriminant analysis [LDA] –to mention only a few), which lead

Open access

Abstract

Analytical data of primary oxidized manganese ores were processed by statistical methods. Six hundred and twenty-one samples were measured (Mn, Fe, Si, and P); thus 2,426 assay data were available. The statistical pointer numbers, the distribution of the elements and the results of the correlational analysis showed the heterogeneity of the ore samples where the measured elements correlated weakly. The samples were grouped by the 4 elements to decrease the heterogeneity and the concentration of elements, and these relationships in the groups were examined. Very few and weak relationships were proved in the groups by the results of the correlational and regressional analysis. It is possible that not the heterogeneity of the samples but one or more syngenetic or postgenetic processes caused the absence of relationships. The multivariate statistical processes (principal component analysis, discriminance analysis) allow the determination of the background factors, namely which are the effects that produced the ore. Consequently — with high probability — the ore was formed by two processes. The most likely are hydrothermal and microbial ones (on the basis of geochemical results), but supergene enrichment processes can also be taken into consideration. Both hydrothermal and microbial processes played a significant role in the majority of the samples (81%), which are the ferruginous manganese ores. In the smaller group of samples (19%) the hydrothermal process predominates but the microbial one is also influential, namely for the low iron-bearing manganese ores of excellent quality.

Restricted access

Abstract

Miocene siltstone with variable sand content and bentonitic clay is the most abundant sediments encountered at the metro construction site at Rákóczi Square (Budapest). Core logs, drilling reports and records of laboratory analyses were studied to better understand the local geology and to prepare a database on engineering geologic properties of the materials. Using this database, geologic sections were prepared and geomathematical methods were used to obtain a better correlation of the strata in the area and a reconstruction of the geologic evolution of the area. The samples were divided into five groups based on physical properties. These five parameters allowed the use of multivariate statistical methods as cluster and discriminant analysis. As a result it was possible to identify several types of lithotypes, including two bentonitic clays with substantially different properties, one fat clay, one medium clay and one sandy, lean clay and siltstone group.

Restricted access
Central European Geology
Authors: Edit Borbás, József Kovács, Katalin Fehér, Gábor Vid, and István Gábor Hatvani

Abstract

Water was observed in the sediment of Baradla Cave, located in Northeast Hungary. In order to investigate its characteristics wells were drilled. Hydrochemical samples were taken directly from the wells and from the cave stream on several occasions between November 2009 and April 2010. In February 2010 there was an opportunity to observe how the chemical composition of the waters of the creeks and the sediments altered during the snow melt. Several chemical parameters of the samples were analyzed. Based on the results of the hydrochemical analyses cluster analysis was applied to define the relationship between the sampling points. Discriminant analysis was conducted to verify the classification. As a result of the classification, the water of the observation wells in the sediment proved to be distinct from the water of the cave's creek and the springs on the surface.

Research shows that there is no permanent connection between the water in the cave sediment and the water of the cave creek in the cave water system.

Restricted access

influential in accessing the information related to the considered topic. It also results in favorable access to topic-shared data. Ultimately, it can be said that the nature of clustering analysis is unsupervised classification. Basically, data or

Open access

influential in accessing the information related to the considered topic. It also results in favorable access to topic-shared data. Ultimately, it can be said that the nature of clustering analysis is unsupervised classification. Basically, data or

Open access

distances retain the intervals and ratios between the points as much as possible ( Scheidt and Caers 2007 ). Clustering of the realizations Cluster analysis can discover the inner organizations of a data set by searching

Open access

suitably selected multivariate method, such as principle component analysis or cluster analysis. Then, for example, the cluster centers can be selected as representative realizations for post-processing. The aim of this study is to introduce a novel

Open access

(NMDS, function “monoMDS”) from package “vegan” ( Minchin and Oksanen 2015 ), agglomerative hierarchical cluster analysis with p values acquired by multiscale bootstrap resampling (function “pvclust”) from package “pvclust” ( Suzuki and Shimodaira 2014

Open access