Spatial Interpolation For Geochemical Map

The analysis of geochemical data can't separate from its surroundings and the geochemical behavior of the element itself. On the one hand, this paper analyzes the regional metallogenic background of copper, such as the regional tetonic patters, geological history evolvement, rock formation and Cu, Ni deposit character, etc. in the area of Hami-jingerquan of eastern Tianshan. Then based on the geochemical behavior of the element itself, this paper analyzes the copper's character, such as physics and chemistry and crystal chemistry, the natural distribution and admeasure , the transferred form and occurrence state , the geological evolvement and circular histry enrichment approach the primary copper's deposit, etc. The geological signification and intrinsic rules are mostly discussed by the two aspects that mentioned above.Draw-out the information, such as unknown but valuable models or rules, from a mass of data, this is the basic idea of Data Mining. In the guide of this idea, this paper is composed of four aspect as following: firstly, understanding the geological signification and intrinsic rules of the deep penetrating geochemical data of Cu. Secondly, the necessary preprocessing is done and then analyzes the spatial character. Thirdly, factures the geochemistry plots by using many spatial interpolation toward the Cu data. Finally, the paper estimates the validity and optimality of many models of spatial interpolation in the study area. Then check up the feasibility and validity of interpolation model in the field and then the interpolation model is expressed by the form of 2.5 dimension.Based on the prevenient research, the paper summarizes many taches in the course of preprocessing of geochemical data,and uses more methods to preprocess such as missing data processing,redundant data processing,data transformation and unification of data reference frame,etc.Aiming at the shortage of prevenient research, the paper emphasizes two significant of data processing taches. One is the analyzing of spatial statistic character, another is that the spatial character is applied in spatial interpolation of the geochemical data.The analyzing of spatial character of geochemical data include the analyzing of spatial statistic character and the examination of spatial relation. The paper applies Histogram, QQpIot to analyzethe statistic character of deep penetrating geochemical data such as distribution, mean, median, variance, standard deviation, skewness and kurtosis, etc. Then, the paper applies voronoi plot, trend analysis and semivariogram/covariance cloud to analyze detail and investigate the spatial bizarre instance,spatial trend and spatial structure of spatial relation towards the deep penetrating geochemical data in the study area, which settles steady groundwork for determining neighborhood in the next processing of spatial interpolation.On the basis of determining interpolation neighborhood, this paper factures many geochemistry plots by using more spatial interpolation of the deep penetrating geochemical data in the study area. The methods include inverse distance weighted interpolation, global polynomial interpolation, local polynomial interpolation, radial basis function, simple Kriging and universal Kriging, etc. In order to determine which interpolation model is more applicable for the deep penetrating geochemical data, the paper compares the interpolation models. Ultimately according to the contrast of validity and optimality of the interpolation models, we reach the conclusion that the model of ordinary Kriging is the optimal interpolation model for the deep penetrating geochemical data in the study area, further more, we prove its feasibility and validity by practice and appliance in the fieldwork.