Abstract: Developing regions often face critical air pollution problems due to the rapid expansion of emissions related to industrial and vehicular sources. Thus, assessment of air quality has become a relevant environmental issue to be addressed by policy-makers due to its high health risk for civilian population, singularly in large urban areas, where it has been associated with the worsening of respiratory, cardiovascular, and neurological diseases, especially in children and the elderly. A fundamental problem is whether the number and location of such stations are adequate to optimally cover the city and, therefore, the spatial representativeness of air quality stations is a crucial factor in monitoring networks for establishment of cost-effective strategies and public policies to mitigate the adverse effects of air pollution on well-being, health and quality of life. A systematic review (SR) was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. For this purpose, the development of the review involved definition of the research objective; database selection; identification of keywords and terms; definition of filters to be applied to define the research; data collection and treatment, and data extraction and evaluation. This systematic review was performed by searching relevant studies published in different electronic databases (Scopus, Web of sciences, Science Direct, Scielo, and PubMed) from 2013 until 2023. The search strategy was developed using the following keywords: “spatial distribution”; “PM2.5”; “PM10”; “Urban air pollution”; “air quality monitoring”; “ambient air monitoring”; “network assessment”; “Surface monitoring networks”; “spatial representativeness”; “spatial variations”; “Redundant stations”; “Optimization”. Boolean operators (“AND” and “OR”) were used to combine the above-mentioned search key terms. To ensure the consistency of documents, full texts of the studies were used for eligibility check and the following criteria were defined: i) inclusion: articles or book chapters; only studies written in English; fully text accessible; the focus of the studies is related to air monitoring network design or assessment; studies that reported the quantitative and qualitative methods to assess the representativeness area of an air quality monitoring station; ii) exclusion: studies not associated with the research objective; non-reviewed, master and Ph.D. thesis; studies that do not present the research method; studies that do not show results in the paper abstract; non-English language papers, conference abstracts, and news articles. For the selection of compatible articles, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were employed. This process included the following steps: the identification and elimination of duplicates; the exclusion of articles outside the scope by title, abstract, and full-text reading. No studies were excluded based on quality. All of the relevant data were extracted, including (i) paper details (title, authors, journal, year of publication, DOI, impact factor [InCites Journal Citation Reports – JCR], number of citations and institution location); (ii) information about the method (methodology applied, limitations, study area, and technical issues); and (iii) future research (recommendations by the authors and gaps found). The main characteristics of the most relevant works found in the literature regarding air quality network design were summarized. From those approaches considered, half of them deal with single pollutant-oriented techniques to select or assess the final location of sampling sites, which is not particularly useful for the present study since it is necessary taking into account all pollutants measured in the Air Quality Monitoring Network (AQMN). From the multi-pollutant approaches, a limitation observed is that experts’ opinions are involved in the selection of weights using linguistic variables whose values can be rather ambiguous (“few” or “near” or “some”) and may not be reproducible. Although their results may demand high computational requirements, air pollution models can provide valuable information on air quality based on knowledge of atmospheric emissions and processes found on relatively accessible data from the studied area.

Keywords: spatial representativeness; air quality monitoring network; criteria pollutants; network assessment; urban air pollution; optimization.