Urban stormwater from simulated rainfall on three different landuses in Queensland State, Australia (residential, industrial, commercial) was analysed for heavy metals and physico-chemical parameters such as Dissolved Organic Carbon (DOC) and Total Suspended Solids (TSS). Rainfall events were simulated using a specially designed rainfall simulator for paved surfaces. Event mean concentration samples were separated into five different particle sizes and analysed individually for eight metal elements (Zn, Fe, Cr, Cd, Cu, Al, Mn and Pb). Multivariate data analysis was carried out for the data thus generated. It was found that DOC and TSS influence the distribution of the metals in the different particle size classes. Zn was correlated with DOC at all three sites. Similarly, Pb, Fe and Al were correlated with TSS at all sites. The distribution of Cu was found to vary between the three sites, whilst Cd concentrations were too low to assess any relationships with other parameters. No correlation between Electrical Conductivity (EC), pH and heavy metals was found at the three sites. The identification of physico-chemical parameters influencing the distribution process kinetics of heavy metals in urban stormwater will significantly enhance the efficiency of urban stormwater management systems.