A multi-analytical approach to chemical analysis of inhalable urban atmospheric particulate matter (PM), integrating particle induced X-ray emission, inductively coupled plasma mass spectrometry/atomic emission spectroscopy, chromatography and thermal-optical transmission methods, allows comparison between hourly (Streaker) and 24-hour (High volume sampler) data and consequently improved PM chemical characterization and source identification. In a traffic hot spot monitoring site in Madrid (Spain) the hourly data reveal metallic emissions (Zn, Cu, Cr, Fe) and resuspended mineral dust (Ca, Al, Si) to be closely associated with traffic flow. These pollutants build up during the day, emphasizing evening rush hour peaks, but decrease (especially their coarser fraction PM2.5-10) after nocturnal road washing. Positive matrix factorization (PMF) analysis of a large Streaker database additionally reveals two other mineral dust components (siliceous and sodic), marine aerosol, and minor, transient events which we attribute to biomass burning (K-rich) and industrial (incinerator?) Zn, Pb plumes. Chemical data on 24-hour filters allows the measurement of secondary inorganic compounds and carbon concentrations and offers PMF analysis based on a limited number of samples but using fuller range of trace elements which, in the case of Madrid, identifies the continuing minor presence of a coal combustion source traced by As, Se, Ge and Organic Carbon. This coal component is more evident in the city air after the change to the winter heating season in November. Trace element data also allow use of discrimination diagrams such as V/Rb vs. La/Ce and ternary plots to illustrate variations in atmospheric chemistry (such as the effect of Ce-emissions from catalytic converters), with Madrid being an example of a city with little industrial pollution, recently reduced coal emissions, but serious atmospheric contamination by traffic emissions.