geostatical analysis of Pb and Zn concentration and combining this way with univariate, bivariate and multi-variate analysis methods in Sehezar Vally of Tonokabon

meysam yazdani¹ FIROOZ ALINIA² anahita zahedi³

1) MSc student, Department of Mining and Metallurgical Engineering of Bandarabbas Campus, Amirkabir University of Technology Email:
2) Assistant Professor, Department of Mining and Metallurgical Engineering, Amirkabir University of Technology Email:
3) MSc student, Department of Mining and Metallurgical Engineering of Bandarabbas Campus, Amirkabir University of Technology Email:

IOI Code : XYBD-ZKDBD
Publication : 4th International Conference On Research Science And Technology(4rstconf.com)
Abstract :
    Sehezar area in the southern city of Tonokabon in Mazandaran province located near the belt Tarom-Hashdtjin, which is why region is a granitiod plutonic could potentially mineralization is important. To identify anomalous elements of Pb, Zn of 50 stream sediment samples collected from the area and were analyzed by ICP-MS was used. First replace the censored values and out side of rows of data using box plots graph and Q-Q plot graph identify and using Doerffel procedure, dropped out of the row. To do this Doerffel procedure code written in MATLAB software environment and reduced our external data tier (which is part of the code of this study) and the data were normalized. Non-parametric Kolomogrov-Smironov test to check normality was used. then used of univariate analysis to draw the map of the anomalous area in the environment RockWorks software and then bivariate analysis (Pearson correlation coefficient) and multivariate analysis was performed on normal data in SPSS software. In multivariate analysis of factor analysis used and the number of 7 factor was identified. And finally was draw the map factors in the environment ArcGIS software. Since this two element are located in a factor (factor 4), as a result the fourth factor is our goal and then geostatical analysis for the elements Pb, Zn did. Variography studies shows that spherical model is the best model fitted. The range of spatial dependence for three Pb, approximately 2180 meters and for Zn is approximately 483 meters. Finally using ordinary Kriging geostatical method were developed exponentially in the GS+ software environment. The evaluation results which calculation of root mean square error (RMSE) and the mean average error (MAE) indicate acceptable accuracy is Kriging. In conclusion, the results of these methods were checked. The results were almost coincide and finally anomaly area for elements Pb, Zn introduced in the region.

Keywords : doerffel factor analysis threshold limit correlation coefficient variography kriging