摘要

Air pollution episodes of unknown origins are often detected by online equipment for air quality monitoring in industrial parks in China. The number of monitors available to provide observation data, as well as the source information, is often very limited. In such case, the identification of a potential source area is more practical than the precise back-calculation of the real source. The potential source area which can be deduced from the observation data from limited monitors was concerned in this paper. In order to do the source area identification, two inverse methods, a direct method and a statistical sampling method, were applied with a Gaussian puff model as the forward modeling method. The characteristic of the potential source area was illustrated by case studies. Both synthetic and real cases were presented. The distribution of the source locations and its variation with the other unknown source parameters were mainly focused in the case study. As a screening method, source area identification can be applied not only when the number of effective monitors is limited but also when an ideal number of monitors are available as long as the source information is almost uncertain.