RESUMO
The geographical detector is a new statistical method to detect spatial stratified heterogeneity and reveal the driving factors behind it. It can not only reveal the influence of a single factor on dependent variables but also evaluate the influence of two-factor interactions and does not need to consider linearity, while also avoiding the influence of multivariate collinearity. Without strong model assumptions, it solves the limitations of traditional methods in analyzing category variables. The research on the spatial differentiation of heavy metals in soil is increasingly widely used. This study collected 40 research reports on the spatial differentiation of soil heavy metals via geographical detector, combed the discrete methods of independent variables, research scale, dependent variables and types of independent variables, factor detection, exchange detection, risk detection, and ecological detection and put forward the problems that need to be clarified in the future application of this research. It is expected to provide support for the deep application of geo-detector in the field of spatial differentiation of soil heavy metals.
RESUMO
Geographic detectors can quickly detect spatial stratified heterogeneity and quantitatively reveal the intensity of driving factors of heavy metal content, which is of great significance for the prevention, control, and remediation of soil heavy metal pollution. In order to reveal the spatial differentiation and influencing factors of soil heavy metal content on the town-scale, 788 topsoil samples were collected from a town in the hinterland of Chengdu Plain. Soil heavy metal (Cd, Hg, As, Cu, Pb, Cr, Zn, and Ni) pollution risk assessments were carried out by using the geo-accumulation index method. Additionally, based on the geographic detector model, 15 factors such as soil properties, topography, soil forming factors, and distance were taken as independent variables, and the contents of each heavy metal element were taken as dependent variables to explore the spatial differentiation and influencing factors of heavy metal content in soils. The results showed that:the average contents of Hg, As, Pb, Cr, Cu, Ni, and Zn in the study area were 1.06-1.93 times the background value of Chengdu, and the content of Cd was lower than the background value; among them, Hg reached the light pollution level, and the other seven heavy metals were at the non-pollution level. The spatial distribution of eight heavy metals was significantly different, the correlation among the elements was significant, and a significant correlation was found between most heavy meals with soil properties; however, the correlation with distance factor and topographic factor was relatively weak. The factor detection showed that TP, TK, pH, TOC, elevation, and distance from the railway had the most significant explanatory power for the heavy metal contents. Interaction detection showed that the interaction between soil properties and other factors was the dominant factor for the spatial variation in heavy metals, and elevation, distance from residential area, distance from railways, and distance from industrial areas were also important factors. Risk detection showed that Hg had the most significant difference in the subregion of elevation and distance from railway, whereas the other seven heavy metals had the most significant difference in the sub-regions of influencing factors of soil properties. The spatial distribution of heavy metals varied significantly in soil at the town-scale, which was closely related to soil properties, topography, and human activities in the study area.