VOC reactivity in central California: comparing an air quality model to ground-based measurements

摘要

Volatile organic compound (VOC) reactivity in central California is examinedusing a photochemical air quality model (the Community Multiscale AirQuality model; CMAQ) and ground-based measurements to evaluate thecontribution of VOC to photochemical activity. We classify VOC into fourcategories: anthropogenic, biogenic, aldehyde, and other oxygenated VOC.Anthropogenic and biogenic VOC consist of primary emissions, while aldehydesand other oxygenated VOC include both primary anthropogenic emissions andsecondary products from primary VOC oxidation. To evaluate the modeltreatment of VOC chemistry, we compare calculated and modeled OH and VOCreactivities using the following metrics: 1) cumulative distributionfunctions of NO_x concentration and VOC reactivity (R_OH,VOC), 2)the relationship between R_OH,VOC and NO_x, 3) total OH reactivity(R_OH,total) and speciated contributions, and 4) the relationshipbetween speciated R_OH,VOC and NO_x. We find that the model predictsR_OH,total to within 25–40% at three sites representing urban(Sacramento), suburban (Granite Bay) and rural (Blodgett Forest) chemistry.However in the urban area of Fresno, the model under predicts NO_x andVOC emissions by a factor of 2–3. At all locations the model is consistentwith observations of the relative contributions of total VOC. In urbanareas, anthropogenic and biogenic R_OH,VOC are predicted fairly wellover a range of NO_x conditions. In suburban and rural locations,anthropogenic and other oxygenated R_OH,VOC relationships arereproduced, but calculated biogenic and aldehyde R_OH,VOC are oftenpoorly characterized by measurements, making evaluation of the model withavailable data unreliable. In central California, 30–50% of the modeledurban VOC reactivity is due to aldehydes and other oxygenated species, andthe total oxygenated R_OH,VOC is nearly equivalent to anthropogenic VOCreactivity. In rural vegetated regions, biogenic and aldehyde reactivitydominates. This indicates that more attention needs to be paid to theaccuracy of models and measurements of both primary emissions of oxygenatedVOC and secondary production of oxygenates, especially formaldehyde andother aldehydes, and that a more comprehensive set of oxygenated VOCmeasurements is required to include all of the important contributions toatmospheric reactivity.