Abstract: Soil heavy metal pollution and complex sources have posed the serious risks on the ecological system in recent years. In this study, the positive matrix factorization (PMF) model was established to explore the spatial distribution and sources of heavy metals in soil. The composite polluted soil was taken from the Yuanyang floodplain of the lower Yellow River. 81 surface soil (0-20 cm) samples were collected to identify the soil heavy metal, such as Pb, Cu, Cd, Hg Cr, Ni, Zn and As. A systematic analysis was then made to determine the pollution content and spatial distribution of heavy metals. The geographical accumulation index and potential ecological risk index were used to assess the pollution degree of heavy metals and their ecology risk. The contents of Pb, Cu, Cd, Hg, Cr, Ni, Zn, and As in the soil of the study area were 27.77, 27.05, 0.31, 0.24, 73.55, 25.38, 88.25, and 29.00 mg/kg, respectively. Among them, Hg pollution was the most severe with an average content nearly 7 times of the soil background value, indicating a strong ecological risk, followed by Cd pollution with a strong ecological risk. An island -like pattern was observed in the spatial distribution of heavy metal concentration in the Yuanyang Beach District. The Cu, Ni, Pb, and Cd pollution were widely polluted with the high value in the central Jiangzhuang Township of the study area. An overlapping space was found on the high value distribution of Zn, Cr, As, and Hg in the surface soil, indicating the point source pollution. The geographical accumulation index also showed that the Hg pollution was the most serious, followed by As and Cd. The intensity of potential ecological risk was ranked in the descending order of Hg>Cd>As>Pb>Cu>Ni>Cr>Zn. The heavy metal of Hg presented the very strong ecological hazards, while Cd was the strong ecological hazards, and the rest were belonged to the slight ecological hazards. The potential ecological risk index (RI) was 459.31 on average among the eight heavy metals, indicating the level of strong ecological hazards. The contribution rates of Hg and Cd were 61.66% and 27.71%, respectively. The spatial interpolation was performed on the comprehensive potential ecological risk index of heavy metals. The most severe pollution of heavy metals was distributed in Xibianqiao North Township and middle Jiangzhuang Township, indicating a strong ecological risk level. The Hg and Cd pollution were relatively sereve and posed the greatest ecological harm to the environment. The cluster and source analysis showed that the heavy metal pollution was attributed to the industrial, transportation, natural, coal-burning, and agricultural sources. Agricultural sources were dominated in the soil heavy metal pollution, with the relative contribution rate of 23.5%. By contrast, the relative contribution rate of industrial, transportation, natural and coal-burning sources were 16.2%, 22.1%, 19.4% and 18.9%, respectively. The finding can provide a strong reference to identify the pollution sources of polymetallic composite polluted soil. The scientific basis and data support can greatly contribute to the "scientific pollution control" and "precise pollution control" of the soil for the better Ecological protection and high-quality development in the Yellow River Basin.