br Introduction br Outdoor air quality represents a main
Outdoor air quality represents a main challenge for the scientific community. In fact, the European Environment Agency (2017) has re-cently confirmed that key air quality standards for the protection of human health, including particulate T-5224 (PM), are currently not met in various air quality monitoring stations in the EU (European Envi-ronmental Agency, 2016). This is a main concern since airborne parti-cles represent the preferred carrier of carcinogenic compounds in the lung (Reche et al., 2012; Roller, 2009; Schmid et al., 2009; Schmid and Stoeger, 2016).
A proper evaluation of the exposure to airborne particles, and of the related risk, of people in urban areas is quite complex since Environ-mental Protection Agencies (EPAs) of the European Countries just pro-vide daily average mass-based concentration (PM10, PM2.5) at a limited number of outdoor fixed sampling points (FSPs) (European Par-liament and Council of the European Union, 2008; U.S. Environmental Protection Agency, 2006). However, such data are not effectively repre-sentative of the real outdoor exposure due to (i) the spatial variation of particle concentrations within the cities (Kumar et al., 2013; Moreno et al., 2015; Rizza et al., 2017; Spicer et al., 2009) and (ii) the long sam-pling period (e.g. PM10 is expressed as 24-h average value). Moreover, the current air quality standards do not consider particle metrics repre-sentative of ultrafine (diameter b 100 nm) and sub-micron particles, such as number and surface area concentrations (Moreno et al., 2015; Rizza et al., 2017; Schmid and Stoeger, 2016; Stoeger et al., 2006). Thus, such standards do not allow to evaluate properly the lung cancer risk of the population in urban areas.
A promising approach to assess the risk received by people in urban areas could be carried out dispersing the “risk of lung cancer emitted” by the sources. To this end (i) the lung cancer risk emitted by each source should be evaluated and (ii) the simulation of the consequent disper-sion of such risk should be performed. The latter aspect was already in-vestigated in our previous papers where the risk of people at a receptor sites of incinerators and in urban street canyons was estimated (Scungio et al., 2016; Scungio et al., 2018b). On the contrary, the first aspect (eval-uation of the lung cancer risk emitted by source) still remains an open question. Indeed, to date, the emissions of the different pollutants can be estimated or measured separately (pollutant by pollutant) and no quantitative information of their overall lung cancer risk potential can be easily carried out.
The estimate of the pollutants emitted by the different sources of a city is typically performed through emission inventories: in particular “bottom-up” approaches can be adopted to estimate the overall emis-sion of the city by summing the emissions of each specific source on the basis of indirect estimates (European Environmental Agency, 2013). Such estimates are performed by multiplying the “emission fac-tor” of the source by the corresponding “activity data” (Wing, 2006). In-deed, the emission factor, typically obtained from inventory databases (European Environmental Agency, 2013), represents the average rate of emission of a pollutant per unit of activity data for a given sector (e.g. mass of pollutants emitted per km, per hour, per emitted flow rate, per fuel mass etc.), where the activity data is a measure of the scale of activity causing the emissions. Therefore, the “bottom-up” in-ventory collects the activity data at a fine spatial scale (e.g. point sources, roads, households) and thereafter aggregate them at the re-quired spatial resolution (e.g. city scale). Emission inventories devel-oped for local and urban applications mostly rely on such bottom-up approach (instead of the “top-down” inventory approach) as it is also able to locate the emission sources within the city then allowing the simulation of the pollutant dispersion emitted by each source (López-Aparicio et al., 2017; Thunis et al., 2016; Trombetti et al., 2018).