Suspended particulate matter (SPM) is an important multi-source component of air pollution that has been linked to a number of adverse health and environmental effects resulting from factors which influence its characteristics, composition and distribution in ambient atmosphere. The focus of the present work was to characterize urban ambient particulate matter in terms of its metal content and then to evaluate its distribution as a function of meteorological conditions, followed by apportionment of major sources of urban SPM in metropolitan Islamabad, Pakistan. For SPM monitoring high volume air samplers were used to collect the particulate samples (n=429) from July 2003 to June 2005. Using Flame Atomic Absorption Spectrophotometry (F AAS) seventeen metals (Ca, Na, Fe, K, Zn, Mg, Cu, Pb, Sb, Sr, Mn, Co, Ni, Cr, Li, Cd and Ag) were estimated in SPM samples collected on glass fibre and cellulose filters using high volume sampling at two stations in Islamabad. The microwave extraction and wet digestion method based on HNO3/HClO4 was used for metal analysis. Regular meteorological data were obtained from National Agriculture Research Council (NARC) on daily basis.
Average SPM levels, for the entire study period were estimated at 165.1 µg/m3 , compared with annual values of 163.8 - 164.3 µg/m3. Among the trace metals, maximum mean contribution was noted for Ca (5.004 µg/m3) followed by Na (4.082 µg/m3) Fe (2.411 µg/m3) K (2.125 µg/m3) Zn (1.814 µg/m3) and Mg (1.162 µg/m3) Other metals showed sub- µg/m3 levels. Minimum concentration levels were found for Cd (0.005 µg/m3) and Ag (0.004 µg/m3). Overall, the decreasing order of metal concentration was: Ca> Na > Fe > K > Zn > Mg > Cu > Pb > Sb > Sr> Mn > Co > Ni > Cr > Li > Cd > Ag.
A Mastersizer was used for size fractionation of particulate matter on % volume basis; these included PM<1.0, PM1.0-2.5, PM2.5-5.0, PM50-10, PM10-15, PM15-25, PM25-50, PM50-100 and PM>100. The PM5.0-10 fraction was found to be the most abundant fraction (21.17 %) followed by PM10-15 (17.48 %), PM15-25 (16.37 %) and PM2.5-5.0 (14.46 %) while the smallest fraction (PM<1.0) and giant particles (PM>100) showed minimal levels at both sites. The overall particle size distribution pattern was found to follow the decreasing order: PM5.0-10 > PM10-15 > PM15-25> PM2.5-5.0 > PM1.0-2.5 > PM25-50 > PM50-100 > PM<1.0 > PM>100.
Pearson correlation analysis was conducted to determine the dependence of metals on particulate size fractions in SPM. Significant correlations were observed for Fe-Mn-Mg-Ca, Cu-Sb-Sr-Co, and Pb-Cd. Fine particulate fractions exhibited a strong mutual association, while an inverse relationship was observed with larger fractions. Most of the metals showed, in general, significant positive correlation with fine particle fractions but negative with coarse fractions. The metal and particulate size distribution in the SPM was examined in relation to their dependence on selected meteorological parameters, such as temperature, relative humidity, wind speed, sunshine, rainfall and pan evaporation. Significant positive correlations evidenced the temperature dependence of some metals (Fe, Mg, K and Mn) and SPM distribution. Particulates up to PM10-15 were found to be strongly correlated with temperature and inversely with relative humidity while giant/large fractions showed an opposite behaviour. Multiple regression analysis duly supported by correlation study, showed positive regression gradients between most of the metal levels and fine particulate fractions, while negative gradients emerged for large/giant particulate fractions with significant correlation coefficients.
Study of seasonal variation in metal content exhibited high mean concentrations for Na (4.873 µg/m3) K (2.191 µg/m3) Cu (0.410 µg/m3) Pb (0.204 µg/m3) and Cd (0.006 µg/m3) during winter, while Ca, Fe, Mg and Mn peaked during autumn/spring. The particulate fraction PM5.0-10 emerged as a dominant fraction during the four seasons. Fine particulates showed higher levels in spring/autumn while coarse and giant particulate fractions dominated in winter. Average monthly levels of SPM, trace metals and particulate fractions verified the seasonal effects, also supported by day-to-day variations in terms of time series analysis. Some of the metals (Ca, Fe, Mg, Pb, Sb, Sr, Mn, Cr and Cd) showed enhanced levels during day time while others (Na, Zn, Cu, Co, Ni, Li and Ag) exhibited comparable diurnal and nocturnal distribution levels.
Variations in SPM metal levels and particulate fractions during rainy/non-rainy days and their wind speed dependence were also investigated in detail. Most of the metals revealed higher mean levels during non-rainy days. While Ca, Mg, Fe, Mn, K and Ag showed an increasing trend with wind speed (47-94 %), the Cu, Sb, Sr, Li and Cd levels showed a depleting effect (17-86 %). Enrichment factors of metals in airborne SPM were calculated to find out the extent of anthropogenic loading. Accordingly, Zn, Na, Cu, Pb, Cd, Co, Sb, Ag, Sr and Ni were found to be highly enriched in airborne particulate' matter. The study revealed that under the prevailing environmental conditions, most of the metals would show further increase in their levels in the future.
The source identification/apportionment was carried out by Principal Component Analysis (PCA) and Cluster Analysis (CA). Six main sources were resolved to this effect: industrial emissions, vehicular emission, electrochemical/metallurgical units, biomass burning, excavation activities/wind blown dust and fugitive emissions. Source categories with similar elemental profiles were obtained during the annual and seasonal study periods. The data pertaining to the present study were compared with the counterpart data reported for other parts the world. Comparative evaluation showed that the current status of the local ambient air quality is highly polluted in terms of metal and particulate burden which is higher than that reported for most urban, rural and background sites around the world. Recommendations for a future pollution abatement programme are forwarded.