key: cord-0978227-7173mcdx
authors: Liu, Qian; Harris, Jackson T.; Chiu, Long S.; Sun, Donglian; Houser, Paul R.; Yu, Manzhu; Duffy, Daniel Q.; Little, Michael M.; Yang, Chaowei
title: Spatiotemporal impacts of COVID-19 on air pollution in California, USA
date: 2020-08-10
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2020.141592
sha: 5fef624c6ad009eb9f04fe084d9168914f062dce
doc_id: 978227
cord_uid: 7173mcdx

Abstract Various recent studies have shown that societal efforts to mitigate (e.g. “lockdown”) the outbreak of the 2019 coronavirus disease (COVID-19) caused non-negligible impacts on the environment, especially air quality. To examine if interventional policies due to COVID-19 have had a similar impact in the US state of California, this paper investigates the spatiotemporal patterns and changes in air pollution before, during and after the lockdown of the state, comparing the air quality measurements in 2020 with historical averages from 2015 to 2019. Through time series analysis, a sudden drop and uptick of air pollution are found around the dates when shutdown and reopening were ordered, respectively. The spatial patterns of nitrogen dioxide (NO2) tropospheric vertical column density (TVCD) show a decreasing trend over the locations of major powerplants and an increasing trend over residential areas near interactions of national highways. Ground-based observations around California show a 38%, 49%, and 31% drop in the concentration of NO2, carbon monoxide (CO) and particulate matter 2.5 (PM2.5) during the lockdown (March 19–May 7) compared to before (January 26–March 18) in 2020. These are 16%, 25% and 19% sharper than the means of the previous five years in the same periods, respectively. Our study offers evidence of the environmental impact introduced by COVID-19, and insight into related economic influences.

The unexpected outbreak of the 2019 coronavirus disease has greatly impacted both economies and environments (Liu et al., 2020; Yang et al., 2020; Saadat et al., 2020) due to societal efforts and policies to mitigate or "lockdown" the disease by local and national governments-including the shutting-down of non-essential industries and the restriction of public transportation. Currently, the spread of coronavirus has been initially controlled in many regions of the world and some countries have chosen to reopen. Thus, evaluations of the impacts of COVID-19 on the environment and economy as well as study into infection and death rates are increasingly urgent and necessary to inform decision makers at all levels.

Air pollution has crucial influences on both nature and human health (Landrigan, 2017) .

For example, PM 2.5 describes fine inhalable particles with diameters that are generally 2.5 micrometers and smaller (EPA, 2019) . Some PM 2.5 can be directly emitted from various sources including power plants, motor vehicles, airplanes, residential wood burning, forest fires, agricultural burning, volcanic eruptions and dust storms, while others are formed when gases and particles interact with one another in the atmosphere (Jeong et al., 2019) . Therefore, other air pollutants such as sulphur dioxide and nitrogen oxides can influence the concentration of PM 2.5 (Manousakas et al., 2017) . CO is a colorless, odorless, tasteless, and toxic air pollutant that is produced in the incomplete combustion of carbon-containing fuels, such as gasoline, natural gas, oil, coal, and wood (The National Academies Press, 2002) . The largest anthropogenic source of CO in the United States is vehicle emissions. Indoor fuel-burning appliances such as clothes dryers, water heaters, furnaces or boilers, fireplaces (both gas and wood burning), gas stoves and ovens, motor vehicles, grills, generators, power tools, lawn equipment, wood stoves and tobacco smoke are also emission sources of CO (Wu et al., 2019) . NO 2 is the greatest concerned component of nitrogen oxide, which comes from fossil-burning sources such as vehicles, power plants, industrial emissions and off-road sources such as construction, lawn and gardening equipment (EPA, 2011).

We use the Nitrogen Dioxide Product (OMNO2d) of the Ozone Monitoring Instrument (OMI) aboard NASA's Earth Observing System's (EOS) Aura satellite to calculate the mean NO 2 tropospheric vertical column density (TVCD) in the pre, peri and post periods of CA for both J o u r n a l P r e -p r o o f 2019). OMI data is adopted to analyze the spatial patterns of COVID-19 impact on NO 2 emission.

The paper adopts locations of major power plants, national highways and wildfires to address the potential causes on the spatial patterns of air-pollution emission in CA.

The Locations of Major Power Plants are derived from Wikipedia: List of Power Stations in California (https://en.wikipedia.org/wiki/List_of_power_stations_in_California). Information on different kinds of power stations are provided, including their locations and capacities. Natural gas and coal stations with capacity larger than 500 Megawatt are utilized in the study because NO 2 is mainly emitted from the combustion of fossil fuels such as coal and natural gas (Paraschiv and Paraschiv, 2019) . 

where P is percentage change between different periods, 1 is the mean concentration of former period, 2 is the average of latter period.

We further explore the spatial patterns of NO 2 TVCD over CA using OMI data as follows:

a. The average NO 2 TVCD in pre-, peri-and post-periods of 2020 and 2015-2019 are calculated for each pixel within CA based on Eq. (5):

where TVCD is the average NO 2 TVCD of the i th pixel in each period; start date and end date correspond to the first and last date of each period; n is the number of days in the period. More 

As shown in Fig. 3 , the concentrations of air pollutants are normalized to the means of the pre-period. Note that the NO 2 and CO data in 2020 are generally lower than in previous years without normalization, probably reflecting in part the effects of clean power plans to limit air pollution emissions from future and existing fossil-fueled power plants (Burtraw et al., 2015) and, more widely, the popularization of zero-emission vehicles in CA (McConnell et al., 2019) .

At the beginning of March (2020) There is a 51% drop in CO concentration during the peri-period of 2020 compared to the same period of 2015-2019, which is more significant than that of NO 2 and PM 2.5 (46% and 25% respectively). As previously illustrated, statistical results also show a sharper decline in CO concentration than NO 2 and PM 2.5 when comparing between peri-period and pre-period in 2020 (49% vs 38% and 31%). Therefore, CO has a larger decrease than the other two air pollutants.

And the concentration of PM 2.5 has the gentlest decline in these three air pollutants. As shown in Fig. 4(j) , the anomalies of NO 2 TVCD decrease significantly in the peri-period compared to the pre-period over region A and rebound in the post-period according to Fig. 4 (k) .

Still, this uptick is not large enough to remedy the lockdown reductions, which can be observed in Fig. 4(l) . There are many power plants concentrated around region A; NO 2 TVCD is dominated by emissions from power plants in this area. Due to the scale-back of non-essential industries in response to COVID-19 crisis, NO 2 declines in region A during the lockdown and recovers partially after reopening when they gradually go back to work. California Independent System Operator (CAISO) reported a 6.7% reduction of energy loads in peak hours of weekdays during the lockdown. Higher loads were observed by the end of May during the implementation of gradual reopening policies (CAISO, 2020).

Over regions B and C, NO 2 anomalies increase dramatically in the peri-period (Fig. 4(j) ) and

then drop in the post-period (Fig. 4(k) ). The entirety of region C and some of region B show negative values in the comparison between post-and peri-period, however, the post-period is still higher than the pre-period, as can be observed in Fig. 4 (l).

On one hand, fuel-burning from everyday domestic activities are an important source of NO 2 emissions in residential areas (Lee et al., 2002) , such as heaters and stoves (Kousa et al., 2001) .

Both B and C are populous regions and transportation hubs that are located at intersections of national highways including the city of Barstow, Napa and Woodland. It can be intuited that citizens spent more time staying in residential areas during the peri-period due to the stay-athome orders and social distancing policies and would likely produce more NO 2 in their daily lives.

On the other hand, transportation (which produces large amounts of NO 2 ) between different cities is reduced, which indicates decreasing NO 2 emissions along many parts of the national highways. Still, essential vehicular use likely increased within residential areas-especially those serving as transportation hubs-such as greater utilization of food and grocery deliveries (Sarmiento, 2020) and cargo transportation (Bates, 2020). After stepping into the post-period, people started to resume their normal lives and residential NO 2 levels drop compared with periperiod; levels are still higher than pre-period in region C due to the incompleteness of reopening. Furthermore, more wildfires occurred in region B in the post-period resulting in a higher discharge of NO 2 (Martin et al., 2006) than the pre-period (Fig. 4(l) ). Therefore, we do not observe a simple declining pattern in region C in Fig. 4(k) . Similar trends can be found in other residential and wildfire locations.

Note that the increase/decrease of NO 2 anomalies over one region does not mean a higher/lower absolute TVCD value; it indicates more/less NO 2 is emitted due to non-seasonal factors. NO 2 primarily pollutes the air from the burning of fossil fuel such as emissions from cars, trucks and buses, power plants, and off-road equipment. Given that most non-essential businesses are shutdown or limited in peri-and post-periods, COVID-19 related shutdown and J o u r n a l P r e -p r o o f reopening policies are the most likely reasons accounting for the change of NO 2 spatial patterns, especially when there are no major wildfires.

To combat the spread of COVID-19 pandemic, the CA government implemented a series of policies including the shutdown of non-essential businesses, mandating social distancing, and the natural factors such as wildfires, the restriction of non-essential industries and quarantine of people in residential areas are the most likely factor to account for these patterns. Although transportation was reduced between cities, it increased within residential communities, especially those serving as transportation hubs.

Although overall trends are similar in ground-based observations and satellite data, discrepancies still exist between the two data sources mainly due to the following reasons:

(1) Ground stations monitor the concentrations of air pollutants near the surface, whereas satellite data retrieves the vertical column density of NO 2 in troposphere;

(2) Ground-based observations are sparsely distributed; not every county in CA has available data.

(3) Ground-based NO 2 data reflects daily mean 1-hour maximum concentrations while satellite observations are retrieved at the moment when the sensor scanned across the area.

This study is an initial effort to understand the impact of COVID-19 mitigation efforts on air pollution and several related factors still have not been quantitatively considered. For example, the patterns of air pollution could also be influenced by climate and geographical changes, such as global warming (Williams et al., 2019) and vegetation (Solins et al., 2018) . Although these effects are partially accounted for by comparing with previous years and observing anomalies, they cannot be entirely eliminated from the results, leading to complex interaction of influential variables in some regions. Spontaneous reduction in human mobility before the lockdown announcement could also influence air pollution emissions (Chinazzi et al., 2020), especially those relating to inter-state and international travels. Detailed transportation volumes within residential areas also need to be further investigated and integrated into a comprehensive analysis.

Similar studies have been done by other researchers in other areas or scales. Liu et al. (2020) conducted a similar study in China and found that satellite measurements showed a 48% drop in J o u r n a l P r e -p r o o f NO 2 TVCD from the 20 days averaged before the lockdown to the 20 days averaged after. This decline is 21% larger than that from 2015 to 2019. The drop from the pre-to peri-period in California was 33% in 2020, which is 3% larger than that of 2015-2019. Compared to China, California has a relatively smaller decline and variation in NO 2 TVCD due to the COVID-19 mitigation policies. Berman and Ebisu (2020) assessed air quality during the COVID-19

pandemic NO 2 in the continental United States and discovered a 26% and a 5% reduction in NO 2 and PM 2.5 respectively in 2020 compared to the same period in 2017-2019. The declines are 46%

and 25% for NO 2 and PM 2.5 respectively during peri-periods between 2020 and previous years.

The drops in air pollutants are more significant in California than the US overall may be potentially due to the fact that air pollution before the pandemic is more severe (American Lung Association, 2020) and mitigation-policy stringency in CA is higher than most of other states (12 out of 54, Fig. A in NO 2 compared to the same period in 2019. These numbers are comparable to that of California (46%) in the peri-period.

Despite the effort made by this study and all the other research, more work needs to be done on the impact of COVID-19 mitigation efforts, including:

(1) Conducting similar research over other parts of the world especially those areas that have rarely been studied, such as Africa, and analyzing the impact of COVID-19 mitigation efforts on different income groups, e.g. low-income countries and high-income countries.

(2) Including other potential factors that affect the patterns and trend of air pollution such as human mobilities, inner-city transportation, climate and geographical changes, to isolate the influence of COVID-19 mitigation efforts more accurately;

(3) Further investigation of the COVID-19 impacts after reopening is carried out.

(4) Investigating the impact of COVID-19 mitigation efforts on the California economy.

(5) Studying the impact of air pollution and other climate factors on the spread of COVID-19.

J o u r n a l P r e -p r o o f

NO2 pollution over India observed from spacethe impact of rapid economic growth, and a recent decline

Chemical characterisation of PM2. 5 emitted from motor vehicles powered by diesel, gasoline, natural gas and methanol fuel

Temporal and spatial variability of traffic-related PM2. 5 sources: Comparison of exhaust and non-exhaust emissions

Air pollution and health. The Lancet Public Health

Nitrous acid, nitrogen dioxide, and ozone concentrations in residential environments

Spatiotemporal Patterns of COVID-19 Impact on Human Activities and Environment in Mainland China Using Nighttime Light and Air Quality Data

California's Evolving Zero Emission Vehicle Program: Pulling New Technology into the Market

Assessment of PM2. 5 sources and their corresponding level of uncertainty in a coastal urban area using EPA PMF 5.0 enhanced diagnostics

Significant enhancements of nitrogen oxides, black carbon, and ozone in the North Atlantic lower free troposphere resulting from North American boreal wildfires

OMI/Aura NO2 Cloud-Screened Total and Tropospheric Column L3 Global Gridded 0.25 degree x 0.25 degree V3, NASA Goddard Space Flight Center

Analysis of traffic and industrial source contributions to ambient air pollution with nitrogen dioxide in two urban areas in Romania

Trends in OMI NO2 observations over the United States: effects of emission control technology and the economic recession

Environmental perspective of COVID-19. Science of The Total Environment

Assessing the distribution and growth rates of NOx emission sources by inverting a 10-year record of NO2 satellite columns

As Restaurants Across the Country Close Their Doors, Deliveries Pick Up

Trilateral association between SO2/NO2 emission, inequality in energy intensity, and economic growth: A case of Indian cities

Riparian canopy expansion in an urban landscape: Multiple drivers of vegetation change along headwater streams near Sacramento

Personal exposures to NO2 in the EXPOLIS-study: relation to residential indoor, outdoor and workplace concentrations in Basel, Helsinki and Prague

TIGER/Line Shapefile, 2016, nation

Identification of sources contributing to PM2. 5 and ozone at elevated sites in the western US by receptor analysis: Lassen Volcanic National Park, California, and Great Basin National Park

The ion chemistry, seasonal cycle, and sources of PM2. 5 and TSP aerosol in Shanghai

Severe air pollution events not avoided by reduced anthropogenic activities during COVID-19 outbreak. Resources, Conservation and Recycling

Climate change and wildfire in California

Observed impacts of anthropogenic climate change on wildfire in California

Exposure to air pollution and COVID-19 mortality in the United States

Study of CO Sources and Early-warning Concentration of Spontaneous Combustion at Air Return Corner in Fully Mechanized Mining Faces. Combustion Science and Technology

Big Earth data analytics: A survey

Big Spatiotemporal Data Analytics: A research and innovation frontier

Effects of meteorological conditions and air pollution on COVID-19 transmission: Evidence from 219 Chinese cities. Science of The Total Center for Climate Simulations; ground-based air pollution data were downloaded from EPA website

OMI NO 2 data are from GES DISC

CA are achieved from Wikipedia

List_of_power_stations_in_California; locations of major wildfires are downloaded from California Department of Forestry and Fire Protection

/; the locations of national highways in California are obtained from the official website of US Census

The study is funded by NSF I/UCRC program and Covid-19 rapid response program and NASA

According to the experiments and analysis results, this study has come to the following conclusions:(1) The spatiotemporal patterns of air pollution in CA were influenced by the COVID-19 mitigation lockdown and reopening policies.(2) The lockdown policy generally reduced the concentration of air pollutants in CA; the reopening increased the emissions of air pollution back to a normal trend, as compared to previous years.(3) The concentration of CO has a sharper decline than that of NO 2 and PM 2.5 during the pandemic.(4) NO 2 emissions decreased over locations of major power plants and increased over populous residential areas, especially those serving as transportation hubs at the intersections of national highways.

Lung Association, 2020. State of the Air. Available at:http://www.stateoftheair.org/assets/SOTA-2020.pdf (2020) Accessed date: June 21 2020.Bates, J., Cargo volumes soar at Ontario International Airport. https://airport-world.com/cargo-