Beyond mere pollution source identification: Determination of land covers emitting soil heavy metals by combining PCA/APCS, GeoDetector and GIS analysis

Abstract

In this work, we propose a method that is not limited in the identification of the type of pollution source but it also suggests the land covers that emit heavy metals into the surrounding soils by introducing a three-stage procedure, as follows: (a) the Principal Component Analysis/Absolute Principal Component Scores technique is applied to the spatial distribution of soil heavy metal accumulations to identify the type of source that is responsible for soil heavy metal accumulation, (b) based on the spatial distribution of the principal component scores and on four selected driving factors (land cover, distance to mine or smelter, distance to road, and topographic elevation), the Geographical Detector model was used to identify the effect intensity of the driving factors on soil heavy metal accumulation and obtain the landscape type of pollutant sources, and (c) GIS analysis (buffer and overlap analysis) was performed on the principal component scores around the suspected land covers linked to the landscape type of pollutant sources to determine the land covers that, in fact, emit heavy metals into the surrounding soils. Based on the proposed approach, four mining and metallurgy land or land groups were determined to be the actual sources of soil heavy metal pollution in Daye city, Hubei Province, China. Lastly, a Multiple Linear Regression model with decay function was proposed to quantify the contributions of previously identified pollution sources to soil heavy metals accumulation. It was found that the HuangJin mountain quarry, the Tonglu mountain cooper mine (together with some related mineral processing and smelting enterprises), the Lion mountain mining and mineral processing base, and the large Oujia mountain mine are the four sources that contributed 3.2%, 34.3%, 8.3%, and 44% of the total soil heavy metal accumulations in the study area.