Assessing urban stormwater quality by investigation and characterisation of pollutants is a prerequisite for its effective management, for reuse and safe discharge. The stochastic nature of rainfall, dry weather periods, topology, human activities and climatic conditions generate and wash-off pollutants differently from event to event. This study investigated the major physico-chemical pollutants in stormwater runoff collected from an urban catchment over a period of two years. The aim of this study was to explore the use of UV spectroscopy to identify the first flush. In this study, the variation of pollutants during the passage of a rain event and the relationships among the measured pollutants was analysed to help broaden the application of UV spectroscopy beyond the detection of organic matter. Correlation analysis and principal component analysis (PCA) were performed to identify the possible relationship among measured pollutants. Although correlation analysis revealed some relationships between pollutants, in general they were not strong enough and was not helpful. PCA biplots suggested a few groups and revealed that the two components model could explain nearly 72% of the variability between pollutants. Pollutants in the group that included dissolved organic carbon (DOC) behaved in a similar manner. UV spectroscopy was applied to identify the first flush by comparing the recorded spectrum of consecutive samples that were collected in an event. Analysis of the spectra was able to isolate the point when first flush ends for DOC and pollutants that behave similar to it.