Measurement of ambient carbonyls in a grid study in the Pearl River Delta region

The ambient carbonyl air concentrations in the PRD region were measured by active and passive DNPH sampling methods in July 2008, September 2008 and February 2009. The PRD region was divided into 100 grid cells and 84 of them, which were on the land area, were involved in this study. The aim of the study was to provide field air concentration data to help investigating the effect of carbonyls on the ozone pollution. The application of passive DNPH-coated sampler enabled us to conduct massive ambient carbonyl air concentration measurement since it was of low cost, convenient to carry around and easy to handle during field measurement. The effective uptake rates of carbonyls by passive sampling were obtained during the sampling period with parallel active sampling of the sites (Luhu in Guangzhou and Tuen Wan in Hong Kong). Twenty-four-hour averaged passive sampling was carried out at each of the 84 grid cells. Despite the large deviations of the effective passive uptake rates obtained, the calculated carbonyls air concentrations could still give an idea in how the carbonyls were spatially distributed over the PRD region. Active sampling was conducted in ten of the 84 grid cells to obtain two samples on each sampling day, one sample from 5am to 9am and the second sample from 9am to 2pm. The carbonyls air concentrations were usually higher for the second samples obtained, this indicated significant secondary sources for carbonyls over the PRD region. More intensified field measurements were carried out in Luhu and Tuen Wan, which diurnal carbonyl air concentration patterns were investigated in three consecutive days in the targeted sampling period. The reactivity-scaled contributions of carbonyls based on their reactivity with OH radicals indicated that, formaldehyde (18%-52%), acetaldehyde (10%-17%) and isovaleraldehyde (1%-32%) were important to the local photochemistry in Luhu and formaldehyde (22%-59%), acetaldehyde (13%-16%) and methylglyoxal (3%-11%) were important in local photochemistry in Tuen Wan. Better correlations were observed between carbonyl species in winter than summer in the Luhu site, suggesting the species shared more common source in winter. The opposite happened in the Tuen Wan site, perhaps suggesting that the carbonyl speices shared more common sources in summer than in winter at this site. Among the three sampling campaigns, glyoxal and methylglyoxal were always correlated (R2 = 0.63, 0.85, 0.82 and 0.62, for summer and winter in Luhu and Tuen Wan accordingly) and so as acetone and 2-butanone (R2 = 0.35, 0.84, 0.81 and 0.90, for summer and winter in Luhu and Tuen Wan accordingly). Maximum incremental reactivity (MIR) was used to investigate the ozone production contribution of the carbonyls, and it was found out that formaldehyde and methylglyoxal were important in ozone production among the species in summer for both Luhu and Tuen Wan; while methylglyoxal was important in winter for both Luhu and Tuen Wan. The ozone production potential of some volatile organic compounds (VOCs) was compared with that of carbonyls using the data obtained during the 1st sampling campaign in the eight active sampling sites. Toluene, carbon monoxide and m-xylene contributed the most to the ozone production by VOCs in both the morning and early afternoon session. VOCs contributed more to ozone production than carbonyls in the morning session (VOCs were collected at 5am) and the percentage contribution of ozone production by carbonyl increased significantly (up to more than 50% in some sites) in the early afternoon session (VOCs were collected at 10am). There is a trend that the carbonyl contribution to ozone production would increase with sunlight radiation but this needs further investigation with VOCs sampling in the afternoon