The objective of this study is to ascertain the spatial distribution of Cu, Pb and Zn in order to determine the degree of contamination in urban soils from Talcahuano (Chile) and to identify the influence of possible contamination sources. A total of 420 samples were collected from the study area based on the following criteria: 140 topsoil samples (TS) (0–10 cm), 140 subsoil samples (SS) (10–20 cm) and 140 deep soil samples (DS) (150 cm). The soils were characterized for their physical characteristics such as grain size distribution, pH, organic matter content etc. and the concentrations of Cu, Pb and Zn were analyzed by Atomic Absorption Photospectrometry following Aqua Regia digestion. Correlations combined with spatial analysis were implemented in order to distinguish the sources of the trace metals and whether they are geogenic or anthropogenic of origin. Several simple and robust statistical methods were applied to the data sets in order to evaluate useful and robust background values. The degree of contamination along with the geoaccumulation index, enrichment factors and contamination factors were also evaluated. The median concentrations obtained for the studied trace metals includes: Cu 23.1 mg kg− 1, Pb 10.2 mg kg− 1 and Zn 56.7 mg kg− 1. In general, the concentrations of Cu, Pb and Zn decrease with depth however, in certain sites the subsoil samples (SS) levels show higher concentrations than topsoil samples (TS). A possible explanation could be related to the uncontrolled clandestine landfill sites using both construction material debris and/or industrial solid wastes. Correlation analysis suggests that Cu, Pb and Zn are contributed by external sources. The spatial distribution of Cu, Pb and Zn in topsoil samples (TS) displays a spatial pattern extending along major roadway environments and emission sources. Estimated background values determined with the iterative 2σ-technique yields 43.7 mg kg− 1 for Cu, 17.5 mg kg− 1 for Pb and 91.7 mg kg− 1 for Zn respectively. The geochemical index, enrichment factor and the contamination factor all register a moderate to high contamination level in some of the soil samples.