Urban air quality and meteorology on opposite sides of the Alps: The Lyon and Torino case studies

Highlights

• Focus on the role of meteorology on local pollutant dispersion.

• Air quality simulations in Torino (Italy) and Lyon (France).

• These cities are on opposite sides of Alps and have similar pollutant emissions.

• Estimate of a burden load of pollution in Torino of 20–40%.

• Evidence of the need of site-specific policies and mitigation for the Po valley.

Abstract

Several European urban areas are characterised by low air quality due to high local emission per unit surface. A further key feature can be related to the pollutant load due to adverse local meteo-climatic conditions. This study aims to compare the two urban agglomerations of Torino and Lyon – located on opposite sides of the Alps and characterised by similar size and population – to enlighten the role of meteorology on local pollutant dispersion. The assessment of air quality has been developed by monitoring network data, emissions analysis and the SIRANE urban dispersion model. Although the two agglomerations have similar NO_X and PM10 emissions, the simulation results show higher ground level concentrations in Torino. To quantify the effect of meteorology on this excess of concentrations, we run simulations in Torino imposing the meteorological conditions of Lyon and vice versa. This implies an overall reduction of ground level concentrations in the city centre of Torino between 20% and 40% (analogously, Lyon concentrations increase by a similar amount). These results show the peculiar difficulties faced by Po valley's cities in maintaining pollution levels below regulatory thresholds and highlight the need of systemic policies and site-specific mitigation to reduce air pollution health risks.

Introduction

Several studies highlight the positive association between the long-term exposure to air pollution and the increased risk of morbidity and mortality (IARC 2016). The climate crisis and the growing number of people living in metropolitan areas (UN, 2018) increase the need for progress in the assessment and management of urban air quality. The challenge regards the development of integrated approaches to assess overall human exposures in indoor and outdoor work and life environments using state-of-the-art reference technologies, IoT tools, real-time measurements and computational models (Bo et al. 2017).

Many areas in Europe are characterised by low air quality (WHO 2016). As shown in Fig. 1 this includes almost all major urban agglomerations (ESA 2019), and particularly several megacities (urban and suburban areas spreading over hundreds of kilometres): Ile de France, South East England, Rhine-Ruhr and Benelux regions and Po valley. This scenario led the European Commission to renewed actions such as the recent enforcements to protect citizens due to the excess of Nitrogen Dioxide (NO₂) and Particulate Matter (PM) in 18 of 28 countries (EU, 2018). Despite the application of enduring policies aiming to reduce direct pollutant emissions, concerns for European ambient air pollution persist (EEA 2018). A main factor determining these critical pollution levels is the high population density, which implies high pollutant emission per surface area. Another key feature is instead related to peculiar meteo-climatic conditions, that can induce the stagnation of pollutant at the regional and/or at the local scale. The extensive literature on the influence of meteorology on air quality focuses on the observation of circulation patterns, precipitation, solar radiation and surface air temperature, pressure and humidity (Kalabokas et al. 2008; Pearce et al. 2011a; Wang et al. 2017; Wise and Comrie 2005). Indeed, the concentration of air pollutants (NO₂, PM and O₃) follow seasonal variations due to i) the role of photochemistry in the formation of secondary pollution and ii) the ventilation conditions governing the long-range or regional transport of pollutants (Otero et al. 2016; Pearce et al. 2011b; Zhang et al. 2012). Particularly adverse conditions are actually present in the Po valley, in Northern Italy (Bigi and Ghermandi 2014; Diémoz et al. 2019b).

This megacity region is burdened by intense and diffuse wintertime daily and annual NO₂ and PM10 concentrations due to different factors. Primary emissions, which are common in the most densely urbanised territories, are coupled with the formation of secondary pollutants favoured by the high persistence in air of gaseous precursors and urban ventilation (Buccolieri et al. 2015). This is the result of the combination of atmospheric stability, precipitation regimes and the particular ventilation forced by the proximity of the Alps. The high air pollution in the large urban and suburban agglomerations of Torino (Turin) and Milano (Milan) derives both from local emissions and a relevant background pollution, defined as the contribution from all pollutants sources located outside these agglomerations. Thus, the definition of effective policies to improve air quality requires the identification of main contributors and their localisation. To that purpose, previous studies in the region have focused on the apportionment of sources and distribution of concentrations in urban agglomerations and side valleys (Amato et al. 2016; Diémoz et al. 2019a).

To further explore the influence of the specific meteo-climatic condition of Po valley on urban air quality, this study aims to compare the air pollution scenarios related to two urban agglomerations: Torino and Lyon. These two cities are very similar in terms of size and population but are located on the opposite sides of the Alps and are therefore subjected to very different meteorological regimes.

The comparison is based on the simulation of air pollution over a whole year (2014) using the SIRANE model, with a spatial resolution of streets and buildings (10 m) and an hourly time-step. This model has been developed in last 20 years by the Atmosphere, Impact & Risk research group of the Laboratoire de Mécanique des Fluides et Acoustique (LMFA-AIR) of the Ecole Centrale de Lyon. Currently it represents the only operational urban air-pollutants dispersion model based on a street network approach which is applied over real cases (Coudon et al. 2018; Soulhac et al. 2017, 2011). It has been validated against wind tunnel experiments (Carpentieri et al. 2012; Garbero 2008; Salem et al. 2015) and in-field studies. The city of Lyon represents the main target domain for validations, performed so far for the years 2008 and 2014 (Nguyen et al. 2018; Soulhac et al. 2017, Soulhac et al., 2012). In the last decade, SIRANE has been applied continuously in various European urban areas, with some pilot studies developed in Italy (Biemmi et al. 2010; Castagnetti et al. 2008; Garbero et al. 2010; Giambini et al. 2010; Pognant et al. 2018; Pognant et al., 2017).

In what follows, the context and main characteristics of the two case studies are first described (§2). The input data for the two scenarios, developed by means of an original bottom up and top down approach, are presented (§3). The “transboundary” comparison aims to analyse meteorological indicators (§4), local emissions (§5) and modelled concentrations of NO₂ and PM10 (§6).