Taiyuan, the capital of Shanxi province, China, is one of the most polluted cities in the world. To characterize the ambient particulate pollution, samples of particulates with aerodynamic diameter less than 10 µm (PM10) were collected during a 6-day campaign. Individual particles were analyzed by Scanning Electron Microscope with Energy-Dispersive Spectrometer (SEM-EDS) to determine their chemical composition. Meanwhile, photomicrographs were obtained from SEM to aid in particles’ source identification. The lumped data from SEM-EDS were subjected to hierarchical cluster analysis (HCA) to sort out particle types chemically. HCA combined with SEM photomicrographs allowed us to identify 20 different particle types, namely (in order of particle frequency), soil/fly ash particles, coal-burning, sulfur-rich, and iron-rich particles, gypsum, syngenite, quartz, cement, silicon sulfide, siliconferro alloy, calcium-rich particles, ferrochromium alloy, ammonium sulfate and chloride, iron-zinc, ammonium chloride, molybdenum-rich, potassium sulfate, dolomite, lead sulfate, and copper-rich particles. Their possible origins and pathways are suggested. The majority of the particles seem to originate from coal combustion, which conforms to Taiyuan’s industrial structure.

[1] In this work, a photochemical dispersion model, CAMx (Comprehensive Air quality Model with eXtensions) was used to simulate a high ozone episode observed in the Po basin during the 2003 FORMAT (Formaldehyde as a Tracer of Oxidation in the Troposphere) campaign. The study focuses on formaldehyde and ozone, and a budget analysis was set up for interpreting the importance of different processes, namely emission, chemistry, transport and deposition, for three different areas (urban, downwind, suburban) around the Milan metropolitan region. In addition, a sensitivity study was carried out based on 11 different VOC emission scenarios. The results of the budget study show that the strongest O 3 production rate (4 ppbv/hour) occurs in the downwind area of the city of Milan, and that accumulated O 3 is transported back to Milan city during nighttime. More than 80% of the HCHO concentration over the Milan metropolitan region is secondary, i.e., photochemically produced from other VOCs. The sensitivity study shows that the emissions of isoprene are not, on average, a controlling factor for the peak concentrations of O 3 and HCHO over the model domain because of very few oak trees in this region. Although the paraffinic (PAR) emissions are fairly large, a 20% reduction of PAR yields only 1.7% of HCHO reduction and 2.7% reduction of the O 3 peak. The largest reduction of O 3 levels can be obtained by reduced xylene (XYL) emissions. A 20% reduction of the total anthropogenic VOC emissions leads to 15.5% (20.3 ppbv) reduction of O 3 peak levels over the Milan metropolitan region.

Abstract. As part of the EU project FORMAT (Formaldehyde as a Tracer of Oxidation in the Troposphere), a field campaign was carried out in the vicinity of Milan during the summer of 2002. Results from a 3-D regional chemical transport model (NILU RCTM) were used to interpret the observations focusing primarily on HCHO and ozone. The performance of the model was assessed by comparing model results with ground based and aircraft measurements. The model results show good agreement with surface measurements, and the model is able to reproduce the photochemical episodes during fair weather days. The comparison indicates that the model can represent well the HCHO concentrations as well as their temporal and spatial variability. The relationship between HCHO and (O 3 ×H 2 O) was used to validate the model ability to predict the HCHO concentrations. Further analysis revealed the importance of the representativeness of different instruments: in-situ concentrations might be locally enhanced by emissions, while long path measurements over a forest can be influenced by rapid formation of HCHO from isoprene. The model is able to capture the plume from the city of Milan and the modelled levels agree generally well with the aircraft measurements, although the wind fields used in the model can lead to a displacement of the ozone plume. During the campaign period, O 3 levels were seldom higher than 80 ppb, the peak surface ozone maxima reached 90 ppb. Those relatively low values can be explained by low emissions during the August vacation and unstable weather conditions in this period. The modelled O 3 / NO z slope at Alzate of 5.1 agrees well with the measured slope of 4.9.