Abstract: Soils around coal mining areas are severely polluted by heavy metals, due to the impact of frequent and sustained mining activities. In fragile arid ecosystems, anthropogenic activities easily lead to ecological degradation, and the restoration is difficult once it is polluted, so accurate and quantitative assessment of soil heavy metal contents is crucial to evaluate the ecological risks caused by the contaminants in coal-oriented industrial area. In this study, a typical arid mining area, the Eastern Junggar Basin of Xinjiang was selected as the research area. To examine the soil heavy metal pollution characteristics, the contents of heavy metals of Zn, Cu, Pb, Cr, Hg and As were measured from 47 surface soil samples (0-20 cm) around the coal mining region. The basic statistical analysis, geo-accumulation index (Igeo), potential ecological risk assessment (PER) and positive matrix factorization model (PMF) were used to identify the statistical characteristics, pollution degree and possible sources of soil heavy metals in surrounding soils of the Eastern Junggar Coalfield. The results showed that: 1) Zn, Cu, Pb, Cr, Hg and As content in study area exceeded their background values of Xinjiang soils by 2.1%, 14.9%, 4.3%, 68.1%, 68.1%, 95.8%, respectively. However, the average value of As exceeded the both national standard I and II, that was 2.1 and 1.3 times of national standard I and II values, respectively; 2) The geo-accumulation index and potential ecological risk index of Hg showed strong pollution (Igeo>3-4) and extremely high risk level (Igeo>4-5), while Zn, Cu and Pb were in a state of no pollution or slight risk level. The comprehensive ecological risk index (RI) of the study area ranged from 50.09 to 1038.47with the mean value of 180.22, 21% and 11% of soil samples indicated high (RI>300-600) and strong risk levels (RI≥600), respectively . 3) The results of PMF model explained that Hg content was mainly related to coal combustion; the accumulation of Pb (53%), Zn (24%) and Cu (19%) originated from vehicle emissions. As, Pb and Hg in soils were derived from the atmospheric deposition. Industrial discharge was the possible source of Cr and As. The accumulation of Zn and Cu was mainly attributed to natural factors. The contribution rates of five possible sources (coal combustion, vehicle emissions, atmospheric deposition, industrial discharge and natural factors) were 20.79%, 16.83%, 16.83%, 27.72% and 17.82%, respectively. The results of this research revealed that positive matrix factorization model applied in this study effectively reflected the possible sources of heavy metal contamination in soils in a quantitative way, and suggested that potential ecological risks may be reduced by controlling emissions and waste materials during mining and other industrial activities.