There is a worldwide growing concern about soil pollution by a wide range of contaminants due to their increased accumulation caused by expeditious industrial and urban development in recent decades. This issue is of special interest because of the danger toxic metals may pose to food quality and human health when they enter the food chain. The present study was undertaken to determine the impact of industrial activities and traffic emissions on metal concentrations in soil samples and the waste product phosphogypsum. The major goal of the work was to assess the influence of different pollution sources resulting in changes in the composition of soil samples collected from Łódź city (central Poland, urban area) and the outskirts of Gdańsk city (northern Poland, rural area affected by industry). In context, knowledge of the natural (background) values in soils is of critical importance in order to evaluate human activity contribution. For that reason we have investigated soil samples taken from potentially uncontaminated sites as well (Lagiewniki Forest, central Poland). Analysis of metals was performed with inductively coupled plasma mass spectrometry, ICP-TOF-MS (Cd, Cr, Cu, Ni and Pb) and a mercury analyzer (Hg). The observed differences in studied metal concentrations in soil samples were a consequence of various degrees of anthropogenic activities on study areas. In general, the concentration of the measured metals tends to increase with the decline of the distance from the pollution source. The study compared relatively low levels of metals in urban soil collected from park areas in the center of Łódź with samples taken in the close vicinity of a phosphogypsum dump from a rural site affected by industrial activity. Total content of toxic metals turned out to be lower than the concentrations reported in the literature data for non-polluted and contaminated soils. Additionally, the application of OptiMass 8000 software allowed a comparison of full mass spectra of all investigated samples, which makes it possible to evaluate the qualitative variations among samples collected from affected forest, urban or rural-industrial areas.