To explore the pollution characteristics and source of soil heavy metal in a coal mine area near the Yellow River in Shandong, the geo-accumulation index method and improved Nemerow pollution index method were used to evaluate the pollution characteristics of soil heavy metal. The absolute principal component-multiple linear regression model (APCS-MLR) was used to quantitatively analyze the source of soil heavy metal, and the spatial distribution of Hg and Cd were analyzed using the Kriging spatial difference method in ArcGIS. The result accuracy of the APCS-MLR model was further verified. The results showed that:The measured contents of soil heavy metal Cu, Zn, Pb, Cr, Cd, Ni, As, and Hg all exceeded the normal site, among which, Hg and Cd exceeded the background values of soil elements in Shandong. The coefficient of variation (CV) of Hg was higher than 0.500, indicating significant spatial heterogeneity. Moreover, the correlation between Hg and other heavy metals was generally low, and the possibility of the same pollution source was small. The results of the geo-accumulation index and improved Nemerow pollution index showed that the overall soil heavy metal pollution was at a moderate level, among which the Hg pollution level was the highest, and its maximum value was at a slanted-heavy pollution level; Cu, Cd, and As in soil caused local pollution, which were at a slanted-light pollution level. Soil heavy metal pollution was closely related to mining activities, rehabilitation, and engineering construction in the coal mine area. The two major pollution sources of soil heavy metal in the research area were the compound source of the parent material and industrial and mining transportation sources (known source 1) and the compound source of atmospheric sedimentation and coal production (known source 2), the contribution rates of which were 76.705% and 16.171%, respectively. The results of the APCS-MLR model were shown to be reliable by analyzing the content distribution of Hg and Cd using the Kriging space difference mode. This research can provide scientific basis for the precise control and improvement of soil heavy metal pollution, ensuring the safety of food and agricultural products and improving the quality of the ecological environment in the coal mine area in the Shandong section of the Yellow River Basin.
Keywords: Kriging space difference; absolute principal component score-multiple linear regression receptor model (APCS-MLR); pollution characteristics; soil heavy metal; source analysis.