key: cord-0840319-ocaqgg5d
authors: Gualtieri, Giovanni; Brilli, Lorenzo; Carotenuto, Federico; Vagnoli, Carolina; Zaldei, Alessandro; Gioli, Beniamino
title: Quantifying road traffic impact on air quality in urban areas: a Covid19-induced lockdown analysis in Italy()
date: 2020-09-18
journal: Environ Pollut
DOI: 10.1016/j.envpol.2020.115682
sha: 120a26bf2af5a5849db64993f01f70860a2a5792
doc_id: 840319
cord_uid: ocaqgg5d

Covid19-induced lockdown measures caused modifications in atmospheric pollutant and greenhouse gas emissions. Urban road traffic was the most impacted, with 48‒60% average reduction in Italy. This offered an unprecedented opportunity to assess how a prolonged (∼2 months) and remarkable abatement of traffic emissions impacted on urban air quality. Six out of the eight most populated cities in Italy with different climatic conditions were analysed: Milan, Bologna, Florence, Rome, Naples, and Palermo. The selected scenario (24/02/2020‒30/04/2020) was compared to a meteorologically comparable scenario in 2019 (25/02/2019–02/05/2019). NO(2), O(3), PM(2.5) and PM(10) observations from 58 air quality and meteorological stations were used, while traffic mobility was derived from municipality-scale big data. NO(2) levels remarkably dropped over all urban areas (from ‒24.9% in Milan to ‒59.1% in Naples), to an extent roughly proportional but lower than traffic reduction. Conversely, O(3) concentrations remained unchanged or even increased (up to 13.7% in Palermo and 14.7% in Rome), likely because of the reduced O(3) titration triggered by lower NO emissions from vehicles, and lower NO(x) emissions over typical VOCs-limited environments such as urban areas, not compensated by comparable VOCs emissions reductions. PM(10) exhibited reductions up to 31.5% (Palermo) and increases up to 7.3% (Naples), while PM(2.5) showed reductions of ∼13–17% counterbalanced by increases up to ∼9%. Higher household heating usage (+16–19% in March), also driven by colder weather conditions than 2019 (‒0.2 to ‒0.8 °C) may partly explain primary PM emissions increase, while an increase in agriculture activities may account for the NH(3) emissions increase leading to secondary aerosol formation. This study confirmed the complex nature of atmospheric pollution even when a major emission source is clearly isolated and controlled, and the need for consistent decarbonisation efforts across all emission sectors to really improve air quality and public health. Main finding A 2-month urban traffic ban extended to the whole Italy only significantly reduced NO(2) levels, while O(3), PM(2.5) and PM(10) concentrations were affected to a minor extent.

In order to quantify the impact caused by road traffic on air quality in urban areas, a straight 166 comparison has been performed between data observed across a period of strongly modified traffic 167 mobility and a period of undisturbed ("baseline") mobility. The first scenario (2020) was set from 168 24/02/2020 to 30/04/2020 to meet the following criteria: (i) to include the very start of the Covid19- (Table S3) . Therefore, as expected, it experienced on average the highest 2020-to-2019 mobility reduction 268 among the cities analyzed here. As shown in Fig. 2 , in the period 24/02/2020-30/04/2020 traffic 269 mobility dropped up to 93%, overall averaging 60% (Table 1 ). In the boxplot of 2020-to-2019 270 change rate, traffic mobility distribution exhibited a median value of -69% (Fig. 3) . The 2020 271 period was slightly colder than 2019, as T decreased on average by 0.5 °C, and slightly less windy, 272 as WS reduced by 6.6% (Table 1) was observed (Fig. S2a) , higher at UB (-34.8%) than at UT (-24.9%) stations (Table 1) . By 275 contrast, O 3 increased at both types of station (Fig. S2b ) by an average of 11.4% (UB) and 12.7% 276 (SB), exhibiting an increase of 17% in the median value of the distributions (Fig. 3) . PM 10 remained 277 basically unchanged (Fig. S2c) , showing a 7.6% reduction (UT) and a 1.8% increase (UB), while 278 PM 2.5 (Fig. S2d ) exhibited a median increase of 17% at UT stations and a median decrease of 8% at 279 UB stations (Fig. 3) . (Table 1) and -56% in terms of median value (Fig. 4) . On average, 284 T reduced by 0.8 °C, WS by 0.4%, Rain by 36.4%, and CC by 3.1% (Table 1) . As in Milan, no 285 appreciable change was observed in the wind regime (Fig. S9) . A significant drop in NO 2 was 286 recorded (Fig. S3a) , higher at UT (-39.9%) than at UB (-25.1%) stations. O 3 remained basically 287 unchanged (Fig. S3b) , although exhibiting a slight average increase at SB stations (+8.2%).

Boxplots show that PM 10 remained substantially unchanged (Fig. S3c) , and that PM 2.5 slightly 289 increased (Fig. S3d) , with distribution's median value of 9% at both types of station (Fig. 4) . In 2020, Florence's traffic mobility reduced on average by 50% with respect to the same period 293 in 2019 (Table 1) , with a distribution's median reduction of 60% (Fig. 5) . Similarly to Bologna, the 294 year 2020 was slightly colder than 2019 (-0.6 °C), less windy (-4.2%), less cloudy (-3.4%), and 295 less rainy (-41.7%, Table 1 ): this significant Rain reduction did not reflect on Rad, that did not 296 appreciably change (-0.5%). The 2019-to-2020 wind rose comparison basically shows the same 297 prevailing wind directions, although in 2020 NE winds proved to be slightly stronger and more 298 frequent than in 2019 (Fig. S10) . As in Bologna, a clear NO 2 decrease was observed (Fig. S4a) , 299 most pronounced at UT (-43.9%) than at UB (-27.4%) stations (Table 1) . Again similarly to 300 Bologna, O 3 levels did not significantly change (Fig. S4b) , exhibiting an average reduction of 1.2% 301 at SB and 2.8% at UB stations. A decrease averaging 21.5% was observed for PM 10 at UT stations, 302 and 8.2% at UB stations (Table 1, Fig. S4c) . A comparable decrease (17%) was recorded on 303 average for PM 2.5 at UT stations, while no variation resulted at UB stations (Fig. S4d) . This 

The R Project for Statistical Computing

Atmospheric chemistry and physics: from air pollution to climate 605 change

Fluctuations in environmental pollutants and air quality during the lockdown in the USA and 608 China: two sides of COVID-19 pandemic

Effect of restricted emissions during 610 COVID-19 on air quality in India

Lockdown 612 caused by COVID-19 pandemic reduces air pollution in cities worldwide. EarthArXiv

Amplified ozone pollution in cities during the COVID-19 lockdown

Changes in air quality during the lockdown in Barcelona (Spain) one month into the 621 SARS-CoV-2 epidemic

COVID-19 mobility report. The effect of the COVID-19 lockdown on mobility in 623

A preliminary assessment of the impact of COVID-19 on environment-A 625 case study of China