What is the relationship between the coronavirus crisis and air pollution in Tunisia?

Abstract Since the beginning of 2020, the COVID-19 pandemic has generated horror and panic around the world. Nevertheless, this terrible crisis is having a positive side effect: it is lowering pollution levels. The outbreak of the coronavirus has caused many governments to impose measures to slow the spread of the virus within populations, such as limiting population displacement, requesting social distancing and the isolation of individuals at home, and reducing industrial activity. In this work, we investigated the effects of governmental measures taken to limit the spread of COVID-19 on the concentrations of air pollutants over four Tunisian cities (Tunis, Sousse, Sfax, and Tataouine). Data on the average daily levels of nitrogen dioxide, sulfur dioxide, carbon monoxide, and particulate matter during January, February, March, and April of 2020 were collected, treated, and analyzed for each city. Curves of average monthly pollutant concentrations from 1 January to 30 April for each city investigated showed that measures taken to reduce the spread of the virus had a substantial impact on emission levels: there were tremendous drops of 51% in NO2 and 52% in SO2 over Sfax City during March compared to those during January, while nitrogen dioxide and sulfur dioxide levels dropped by about 38% and 42%, respectively, over Tunis City and by around 20% for Sousse. During the four months investigated, almost all of the pollutant concentrations showed a significant drop from mid-March. On 12 March, the Tunisian government imposed some individual and collective measures to protect the population from the virus, such as social distancing, limiting transportation, shutting down schools and universities, and reducing industrial activity. A general lockdown was brought in later. Thus, restricting human and industrial activities appeared to affect the air quality in Tunisia, leading to a marked improvement in the air quality index. Graphic abstract


Introduction
As of 30 April 2020, the novel coronavirus SARS-CoV-2, which originated in China in late December 2019, was reported to have infected 3,266,132 people around the world and killed almost 234,000 since the start of the COVID-19 (COronaVIrus Disease 2019) pandemic (Worldmeters 2020). This pandemic has affected almost all countries, although it is most prevalent in America, followed by Europe, Asia, and the Middle East. For instance, by 30 April 2020, there had been at least 994 confirmed cases of the novel coronavirus in Tunisia, including 41 deaths (Worldmeters 2020).
The World Health Organization (WHO) has warned countries that the COVID-19 pandemic requires urgent action. Moreover, they have been preparing for the worstcase scenario as the total number of virus infections has increased. This scenario, as reported by a former chair of the Global Health Council and a long-term collaborator with the World Health Organization (WHO) (Spinney 2020), is that the outbreak goes global and the disease eventually becomes endemic, meaning that it circulates permanently in the human population.
Due to the recent rapid emergence of the global COVID-19 pandemic, the World Health Organization has proposed emergency measures that aim to reduce virus transmission rates and contamination. These measures consist of limiting population movement and travel, strict social distancing, the isolation of individuals at home, and even, if necessary, a general population lockdown (Lai 2020). While this suppression of human activity has many negative effects, it is a critically important way of limiting the spread and impact of COVID-19 (Coccia 2020). On the other hand, the marked reduction in human activity caused by these measures appears to have coincided with an improvement in the air quality in cities, so this substantial change in human industrial character. We collected, treated, and analyzed freely available air monitoring data for the selected cities to assess the impact of the coronavirus outbreak on urban air pollution.

Investigated cities
Tunisia is a country in the Maghreb region of North Africa that covers 163,610 square kilometers and had a population of 11.582 million in 2018 (World Bank 2020). Four Tunisian cities that were affected by COVID-19 early during the pandemic were selected for this study. The four cities were chosen based mainly on location, economic activity, and population. The cities were the capital Tunis in the north of Tunisia, Sousse and Sfax, both of which are in the center of the country, and Tataouine, which is in southern Tunisia. The map shown in Fig. 1 indicates the locations of the investigated cities.
The four studied cities have different characteristics. Tunis, Sousse, and Sfax are highly populated urban regions with considerable industrial activity (Belhedi 2007). According to the 2014 population census, Tunis (the capital of Tunisia) has the most inhabitants (1,056,247) of any Tunisian city, followed by Sfax and Sousse, which are ranked the second and the fourth largest cities in Tunisia in terms of population, with 955,421 and 674,971 inhabitants, respectively (National Statistics Institute 2014). Another factor taken into consideration in this study was the date of the first confirmed COVID-19 case in the city. The selected cities were infected early during the pandemic (during the period 9-18 March) (Tunisian Ministry of Health 2020). For each city, we collected air pollution data and analyzed pollutant trends before and after the actions taken by the government on 12 March 2020 to contain the spread of the virus. A four-month period was examined: January, February, March, and April 2020. Table 1 lists relevant characteristics of the investigated cities and the dates of the first confirmed infections in those cities.
The main sources of industrial activity in Tunis are ICT, the ecoindustry, aeronautics, and pharmaceuticals. There is also a power plant in Rades, a suburban region of Tunis. Sousse, which is characterized by its mechanical and electronic industries, also has a power plant. Sfax, which is the second most important city in Tunisia from an economic perspective, is known for its chemical industry. The two power plants in Rades and Sousse are owned by the Tunisian  Electricity and Gas Company (STEG), a public company that is charged with the generation and distribution of electricity. As reported in the 2018 STEG Annual Report, electricity is mostly produced by fossil fuel-powered units (94%) in Tunisia.

Air pollution data collection
In this study, we collected air quality index (AQI) data and air pollutant concentrations from the Air Matters monitoring platform, a global air quality service provider (www. air-matte rs.com) that forecasts and provides real-time information on the air quality index, allergenic pollen levels, and weather data. For each investigated city, we downloaded the average daily AQI values and concentrations of the six major air pollutants for the four-month period from 1 January to 30 April 2020. The pollutants considered were fine particulate matter (PM 2.5 ), coarse particulate matter (PM 10 ), nitrogen dioxide (NO 2 ), ground-level ozone (O 3 ), sulfur dioxide (SO 2 ), and carbon monoxide (CO). The air quality index (AQI) relates daily pollutant concentrations to health concerns for sensitive groups and the public (EPA 2014). Values below 100 are considered satisfactory. Values above 100 are considered unhealthy-initially for certain sensitive groups of people, and then for everyone as the AQI values increase further. The AQI can range from 0 to 500, and this range is divided into six classes, as shown in Fig. 2.

Data analysis
For each investigated city, we used the data collected daily to determine the average monthly air quality index and the average monthly concentrations of the six air pollutants for January, February, March, and April 2020.
We assumed that the air quality in January and February 2020 (before the pandemic hit Tunisia) was normal. Therefore, changes in air quality during March and April 2020 were considered to be caused primarily by the actions of the country aimed at containing the spread of COVID-19. We examined Tunisian air pollution for the four first months of 2020 and assessed whether there were linear trends in AQI and pollutant concentration values because of the protection and lockdown measures imposed by the government due to the COVID-19 pandemic. We then compared pre-and post-COVID-19 air quality for the affected cities.

Results
Plotted average monthly values of PM 2.5 , PM 10 , NO 2 , O 3 , SO 2 , CO, and AQI for the four chosen Tunisian cities are displayed in Figs. 3, 4, 5, 6, 7, 8, and 9, respectively. We compared the levels of the different air pollutants in January and February to those in March and April, when the COVID-19 protection measures were imposed or in place.
The mean concentrations of almost all the pollutants showed significant reductions in every city following the enactment of COVID-19 protection measures: the curves shown in the figures present substantial differences in pollutant levels between the period before early March 2020 and the period after it.
When the concentrations of PM 2.5 during the four months of interest were analyzed (see Fig. 3), we found that there was a slight drop in March compared to January for all of the cities considered except for Tataouine. The observed drops in March as compared to January for Tunis, Sousse, and Sfax were about 20%, 7%, and 23%, respectively.
No significant change in PM 10 concentration was observed for Tunis, Sousse, and Sfax, while a significant increase was noted for Tataouine (see Fig. 4).
For NO 2 , the concentration curves for the four cities (see Fig. 5) exhibited decreases in March and April as compared to January and February. Nitrogen dioxide concentrations in March were 37%, 22%, and 51% lower than those in January for Tunis, Sousse, and Sfax, respectively, whereas the NO 2 concentration over Tataouine remained constant across the period of interest. As depicted in Figs. 7 and 8, very similar behavior was observed for the levels of SO 2 and CO, as decreased concentrations were observed shortly after the restrictions for almost all of the cities. The measured concentration of SO 2 in Tunis, Sousse, and Sfax declined by 42%, 18%, and 52%, respectively, in March as compared to those in January. The by 6%, 3%, and 17% in April as compared to January for Tunis, Sousse, and Sfax. When compared to the previously mentioned pollutants, ozone showed a late response to the protective measures against COVID-19, which is why the curves shown in Fig. 6 do not follow the same trend as seen for the other pollutants. levels of carbon monoxide declined by around 26%, 4%, and 19% for Tunis, Sousse, and Sfax, respectively.
The protection measures enacted by the government appeared to cause a drop in air pollution, as indicated by the air quality index (see Fig. 9). The AQI dropped by 17%, 5%, and 15% in March as compared to January and

Discussion
From mid-March to the end of April, levels of the pollutants PM 2.5 , NO 2 , SO 2 , and CO as well as AQI values were observed to be substantially lower than those in early January, when power plants and industrial factories were operating at normal levels. The appreciable drop in PM 2.5 concentrations can be attributed to the reduction in emissions from vehicular traffic and industrial activity caused by the measures brought in by the Tunisian government to suppress the spread of COVID-19. The increase in PM 2.5 concentrations in Tataouine following the introduction of these measures is thought to be due to the meteorological conditions during this time. Tataouine is a pre-Saharan city-a doorway to the desert in southern Tunisia. Storms from the Sahara during the spring months bring a lot of dust, which explains the increased PM 2.5 levels observed for March and April.
The remarkable increase in PM 10 for Tataouine during March and April (see Fig. 4) can also be explained by the weather conditions. In Tunis, Sousse, and Sfax, large particulates are generally produced through human activity. The relatively constant levels of PM 10 in these cities during the monitored period may be due to the long residence time of PM 10 in the atmosphere.
For NO 2 , SO 2 , and CO, decreased levels were seen for March and April than for January and February. This appears to have been due to the reduced activities of power plants and other industrial facilities and the restrictions on vehicular use that were imposed from March on to limit the spread of COVID-19. This impact of restrictions due to COVID-19 on pollutant concentrations has been noted in various studies performed in cities around the world (Otmani 2020;Shi 2020;Zhao 2020).
Urbanization and motorization are considered two important sources of congestion. They consume energy and therefore generate pollutant emissions. This explains why the capital Tunis exhibits the highest mean values of the pollutants SO 2 , NO 2 , CO, and PM 2.5 .
The only city that showed nonsignificant changes in NO 2 , SO 2 , and CO concentrations following the introduction of the measures aimed at tackling COVID-19 was Tataouine. The increased concentrations of O 3 in all of the cities in March and April can be attributable to the meteorological conditions, because O 3 levels tend to rise during spring and summer owing to increased solar radiation (in terms of its intensity and daily duration) (Escudero et al. 2019;Wang et al. 2020). Similar behavior of ozone was observed in Milan (Collivignarelli 2020).
The general air quality improved in March and April, as inferred from the trends in the air quality index (Fig. 9). This general shift in air pollutant concentrations would be expected given that the governmental measures relating to the COVID-19 pandemic were introduced on 12 March. The resulting partial lockdown of the Tunisian population and partial shutdown of industrial activity changed individual and collective behavior. The number of vehicles used dropped dramatically. Moreover, electricity production until 4 April was about 25% lower than that for the same period in 2019. As can be seen in Figs. 10 and 11 in Appendix A, the demand for electricity and natural gas was also significantly reduced during lockdown and shutdown (Ministère EMTE 2020).
Due to the pandemic, the Tunisian authorities progressively shut down transportation and travel in and out of the country. They also curtailed local business travel, limited local transport, and closed down schools, colleges, and universities to reduce the spread of the disease, as well as establishing numerous quarantine sites in infected areas. Those governmental measures taken in March (limiting transport and shutting down schools on 12 March) rapidly and markedly reduced human movement and activity. The social distancing and almost complete cessation of economic activity caused by the COVID-19 measures led to a significant drop in air pollution over the four investigated towns.
When analyzing the effects of the protection measures and lockdown on the air pollution in the studied cities, we observed that the impact varied depending on the level of industrial activity usually associated with the city.Tataouine, which has an economy that is largely based on agriculture, showed notably different trends in pollutant levels than the other more industrial cities. The concentrations of pollutants in the industrialized cities were more strongly influenced by the COVID-19 measures. Similarly, the air pollution in big cities that usually consume more electricity or suffer from substantial traffic was particularly strongly reduced when the lockdown was implemented.

Conclusion
Since 12 March 2020, the Tunisian government has taken a series of protection measures to limit the spread of the novel coronavirus among the Tunisian population. These unique circumstances are perfect for studying the impact of human behavior on air pollution in Tunisia. We carried out the present study to investigate the possible impacts of the COVID-19 human lockdown in Tunisia on air pollution. We utilized average daily air pollutant concentrations collected during air quality monitoring for four cities in Tunisia (Tunis, Sousse, Sfax, and Tataouine). Analysis of the trends in monthly average air pollutant concentrations during the first four months of 2020 showed major drops in PM 2.5 , NO 2 , SO 2 , and CO levels from 12 March, when the protection measures were first imposed. A comparison of air quality status between the four Tunisian cities before and after this date revealed that the concentrations of almost all of the pollutants dropped after the measures were introduced. The concentration of PM 2.5 reduced over each city except for Tataouine. The concentration of larger particulates (PM 10 ) did not change significantly in Tunis, Sousse, and Sfax over the period of interest. However, for Tataouine, the southernmost city, PM 10 increased during March and April because of the prevailing meteorological conditions. We also noted marked decreases in the concentrations of nitrogen dioxide, sulfur dioxide, and carbon monoxide in almost all the cities. The largest reduction was observed for the capital Tunis. On the other hand, notable increases in O 3 were seen in all the studied cities. The general air quality as assessed using the air quality index appeared to improve in almost all the cities following the introduction of the COVID-19 measures.
Thus, our study shows that the progressive limitation of industrial activity and vehicular traffic and the drop in electric power demand in Tunisia caused by the measures brought in to address the COVID-19 pandemic were accompanied by apparent decreases in air pollution and the air quality index in Tunisian cities. In other words, one side effect of this horrific virus is cleaner Tunisian air.
Funding There was no external funding for this work.

Conflict of interest The authors declare no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.

Appendix A
See Figs. 10 and 11.