Use of Moran’s I and robust statistics to separate geochemical anomalies in Jiurui area (Southeast China)
Authors:
Thanh NGUYEN TİEN
ABSTRACT
Separation of geochemical anomalies from background plays an important role in the study of exploration geochemistry. The limitations of commonly used methods are not taken into account spatial correlation, variability and the unsatisfactory of the statistical assumption of the normality of geochemical data. For solving these limitations, an indirect method for the separation of geochemical anomalies is proposed based on anomaly separation of local Moran’s Ii values using robust statistics in this study. The experiment was carried out using 1481 samples collected from Jiurui copper prospect (southeast China). The steps for the anomaly separation are (i) spatial association and variability were fi rst analyzied by means of Moran scatterplots at six spatial scales (2, 4, 6, 8, 10 and 12 km) using both raw data and Box-Cox transformed data; (ii) local Moran’s Ii was used to measure spatial autocorrelationat these six local scales; (iii) anomalous separation was fi nally performed using the MEDIAN ± 1.5*IQR (IQR: interquartile range) rule on local Moran’s Ii values. The results show that geochemical anomalies are mostly concentrated around known ore-deposits, according the objective reality and a strong correlation with known ore-deposits in Jiurui copper prospect.
http://dx.doi.org/ 10.19111/bulletinofmre.351376
Thanh NGUYEN TİEN, Tuyen VU DANH, Hung TRİNH LE HUNG, Hang NGUYEN THİ LE HANG
Indexed In · Volume 153 / 2016
SPATIAL CLUSTER AND OUTLIER IDENTIFICATION OF GEOCHEMICAL ASSOCIATION OF ELEMENTS: A CASE STUDY IN JUIRUI COPPER MINING AREA
Authors:
Thanh NGUYEN TİEN, Tuyen VU DANH, Hung TRİNH LE HUNG, Hang NGUYEN THİ LE HANG
ABSTRACT
Spatial clusters and spatial outliers play an important role in the study of the spatial distribution patterns of geochemical data. They characterize the fundamental properties of mineralization processes, the spatial distribution of mineral deposits, and ore element concentrations in mineral districts. In this study, a new method for the study of spatial distribution patterns of multivariate data is proposed based on a combination of robust Mahalanobis distance and local Moran’s Ii. In order to construct the spatial matrix, the Moran’s I spatial correlogram was first used to determine the range. The robust Mahalanobis distances were then computed for an association of elements. Finally, local Moran’s Ii statistics was used to measure the degree of spatial association and discover the spatial distribution patterns of associations of Cu, Au, Mo, Ag, Pb, Zn, As, and Sb elements including spatial clusters and spatial outliers. Spatial patterns were analyzed at six different spatial scales (2km, 4 km, 6 km, 8 km, 10 km and 12 km) for both the raw data and Box-Cox transformed data. The results show that identified spatial cluster and spatial outlier areas using local Moran’s Ii and the robust Mahalanobis accord the objective reality and have a good conformity with known deposits in the study area.
Keywords:
Spatial Outliers, Spatial Clusters, Spatial Variability, Local Moran Statistic, Geochemistry