Principal Component Analysis (PCA) is a multivariate statistical technique used to reduce the dimensions of a data set so that trends in the data can be visualized more easily. In an effort to differentiate potential sources of key metals of interest in an urban area with multiple industrial facilities, PCA was conducted using metals concentrations measured in surface soil and sidewalk/street dust. The objective of this analysis is to identify key metals or combinations of metals that can be used to distinguish patterns of metals that are descriptive of natural and anthropogenic sources, and to distinguish among anthropogenic sources. Following data reduction, 15 metals were retained for analysis. Both Classical and Robust PCA were evaluated in this study and executed in the R open source statistical package (version 3.1.2). As metals concentrations were consistently higher in dust than in soil, PCA analyses for dust and soil were run separately. Results were plotted on maps to allow for geospatial evaluation of results. The first PC, describing the greatest variance (ranging from 40-63%), distinguished anthropogenic from natural background. The second and third PCs were also able to differentiate among non-natural sources, describing 16-28% of variance. The signature associated with one source (primarily influenced by Cr, Co, Mo, and Ni) was consistent with air emissions tests from the facility and could be clearly identified. The extent of metals consistent with this signature was limited to a small area of approximately 1.5 blocks in the immediate vicinity of the facility. In this study, PCA was used to effectively differentiate natural and anthropogenic sources of metal emissions in an urban area with historical and current sources, and made it possible to delineate the spatial area of impact from a specific individual source.