An example of cluster analysis applied to a large geologic data set: Aerial radiometric data from Copper Mountain, Wyoming

Abstract

One ob/eetive of the aerial radiometric surveys flown as part of the U.S. Department of Energy's National Uranium Resource Evaluation (NURE) program was to ascertain the spatial distribution of near-surface radioelement abundances on a regional scale. Some method for identifying groups of observations with similar ~-ray spectral signatures and radioelement concentration values was therefore required. It is shown in this paper that cluster analysis can identify such groups with or without a priori knowledge of the geology of an area. An approach that combines principal components analysis with convergent k-means cluster analysis is used to classify 6991 observations (eaeh observation comprising three radiometric variables) ]'rom the Precambrian rocks of the Copper Mountain, Wyoming area, This method is compared with a convergent k-means analysis that utitizes available geologic knowledge. Both methods identify four clusters. Three of the clusters represent background values for the Precambrian rocks of the area, and the fourth represents outliers {anoma-lously high 214Bi). A segmentation of the data corresponding to "geologic reality" as inter. preted by other methods has been achieved by perceptive quantitative analysis of aeriaI radiometric data. The techniques employed are composites of classical elustering methods designed to handle the special problems presented by large data sets.

The potential in fluvial geomorphology of a new mineral identification technology (QEM*SEM)

“…As a test of the consistency of the three groups of sites and also to ascertain the significance of the differences of the values of the mineral percentages among the groups, a 3 cluster analysis was performed on the fine fraction (the BMDP K-means cluster algorithm using standardised data : for an example of k-means clustering using principal components see Pirkle et al, 1984). The three clusters given by this analysis had sample 6, the diatreme knoll sample, in one cluster, samples 1 and 2, the weathered bedrock fragment and colluvial slope deposit in another cluster group, and the remaining samples were in the third cluster .…”

Earthquake hazard potential along the Panjal Thrust and the Zanskar Thrust, NW Himalaya

“…In order to describe the relative tectonic activity, we evaluate a data mining technique known as K-means clustering to nd homogeneous clusters based on selected characteristics of geomorphic indices, earthquake data and eld observations (Clare and Cohen, 2001;Pirkle et al, 1984). On the basis of K-clustering grouping the basins having same values were grouped under similar classes of tectonic activity.…”

Utilização da classificação não-supervisionada na integração e interpretação de dados geofísicos em Cuba centro-oriental

“…Considerando a eficácia desta técnica na classificação não-supervisionada de dados (PARKS, 1966;PIRKLE et al, 1984;CRÓSTA, 1993), na área de estudo, Cuba centro-oriental, este método foi utilizado segundo duas abordagens e visando dois objetivos principais: (i) a análise tectono-estrutural, empregando como amostras de treinamento dados aeromagnéticos (reduzidos ao pólo) e gravimétricos (anomalia Bouguer); (ii) a relação do mapa geológico da área com os dados gamaespectrométricos aéreos (CT, K, eTh, eU) e a segunda e terceira principais componentes (PC2 e PC3) derivadas desses dados.…”