|
Shuyun Xie, York University (Canada)
Qiuming Cheng, York University (Canada)
Zhengyu Bao, China University of Geosciences (China)
Changming Wang, York University (Canada)
Xianzhong Ke, China University of Geosciences (China)
Ping Fan, China University of Geosciences (China)
|
|
The separation of anomalies from the background is an important part of geochemical data analysis since such identifications can have a profound influence and distort the final analysis results. 86,298 element concentrations of W in Nanling region, South China, is used as one example to separate anomalies from geochemical background. Compared with traditional anomaly contrast parameters, both local singularity theory and the method of the fences method for identifying outliers have been utilized to delineate the ranges of geochemical anomalies of W. To different degrees, all the results have shown that the highly anomalous area of W in the study is closely associated with the outcrop of granites. However, the results obtained by these three methods are not the same. Since the traditional anomaly contrast method does only take into account of the relative concentration of W, the anomalies delineated by anomaly contrast parameter are sparse in space, which is not so easy to provide enough information for the source of the anomaly. The fences method is not only simple but it is very useful to avoid the effects of extreme outliers of concentrations and thus a better relatively anomalous delineation has been obtained. By local multifractal singularity, the relationship between anomalous area and both the magma distribution and geophysical anomalies have been characterized well since local singularity theory considers both the absolute concentrations and the relative distribution patterns in space.
|
