Mapping sources of atmospheric pollution: integrating spatial and cluster bibliometrics.

The intensified degradation of regional air quality in recent years has prompted an increase in research teams to pursue various aspects of air pollution research including source apportionment (SA) resulting in rapidly growing output of this literature, a rapid shift in research emphasis, and an extensive inter-institutional and international collaboration. To explore these recent trends in air pollution research, we present a spatial and cluster bibliometric evaluation of atmospheric SA research, ascertaining the relationship among the main research bodies and knowledge clusters. The evolution trend of the knowledge clusters was depicted with an emphasis on the recent surge of research in Asia. We retrieved relevant articles from the Web of Science and Scopus data set with the key words "air" and "source apportionment". In this field, the annual output of peer-reviewed papers has increased dramatically since 2005. Initially, air SA research was concentrated in developed countries from Europe and North America, but more recently it has been receiving substantial contributions from developing countries, most notably China and India. Meanwhile, there has been a methodological shift from single methods, such as source diffusion models or receptor models, to the integration of online high-resolution data monitoring with multiple models. In recent years, the main research objectives for atmospheric pollution research have changed from being focused predominantly on coarse particles to more focus on fine and ultrafine particles. Our bibliometric analysis yielded 666 knowledge nodes, forming 26 major co-citation clusters. Focusing on recently emerged knowledge clusters 7, 22, and 23, we identified four research themes based on the research frontier analysis: (i) practical research conducted to inform pollution-reducing policy, (ii) assessment of pollution-related health risks of human activities, (iii) analysis of pollutant composition and formation, and (iv) improvement of SA methods. The study provides an efficient and economical means to understand the history and the prospects for air SA research through a temporal and spatial analysis on the relationships among the main research bodies and knowledge clusters. [ABSTRACT FROM AUTHOR]