key: cord-0229956-v53v4553
authors: Beigi, Pedram; Haque, Mohaiminul; Rajabi, Mohammad Sadra; Hamdar, Samer
title: Bike Share's Impact on COVID-19 Transmission and Bike Share's Responses to COVID-19: A case study of Washington DC
date: 2022-05-10
journal: nan
DOI: nan
sha: 7b6001f59e1595210d6fcff47b798eb5961f7757
doc_id: 229956
cord_uid: v53v4553

Due to the wide-ranging travel restrictions and lockdowns applied to limit the diffusion of the SARS-CoV2 virus, the coronavirus disease of 2019 (COVID-19) pandemic has had an immediate and significant effect on human mobility at the global, national, and local levels. At the local level, bike-sharing played a significant role in urban transport during the pandemic since riders could travel outdoors with reduced infection risk. However, based on different data resources, this non-motorized mode of transportation was still negatively affected by the pandemic (i.e., relative reduction in ridership). This study has two objectives: 1) to investigate the impact of the COVID-19 pandemic on the numbers and duration of trips conducted through a bike-sharing system - the Capital Bikeshare in Washington, DC, USA; and 2) to explore whether land use and household income in the nation's capital influence the spatial variation of ridership during the pandemic. Towards realizing these objectives, this research looks at the relationship between bike sharing and COVID-19 transmission as a two-directional relationship rather than a one- directional causal relationship. Accordingly, this study models i) the impact of COVID-19 infection numbers and rates on the use of the Capital Bikeshare system and ii) the risk of COVID-19 transmission among individual bike-sharing users. In other words, we examine i) the cyclist's behavior as a function of the COVID-19 transmission evolution in an urban environment and ii) the possible relationship between the bike share usage and the COVID-19 transmission through adopting a probabilistic contagion model. The findings show the risk of using a bike-sharing system during the pandemic and whether bike sharing remains a healthier alternative mode of transportation in terms of infection risk.

COVID-19 was firstly reported in Wuhan, Hubei, China, on March 11, and the World Health Organization (WHO) investigated and declared COVID-19 as a pandemic on March 11. By the end of October 2021, WHO had received reports of about 247,400,000 cases and 5,000,000 fatalities caused by COVID-19, most of which occurred in America and Europe (WHO 2022) . Many nations employed various lockdowns to postpone the pandemic's peak and smooth the curve as early as the first months of the infection (Razavi, et al. 2022 ) (Javadinasr, et al. 2021) . Several studies have demonstrated that unrestricted mobility would have hastened the development of COVID-19 (Zargari, et al. 2022 ) (Shakerian, et al. 2022) , and those travel limitations, in general, tended to slow the disease's global growth (Anzai, et al. 2020 ) (Oztig and Askin 2020) (Chinazzi, et al. 2020 ) (Linka, et al. 2020 ).

Corresponding travel restrictions and various lockdowns had a significant impact on the entire transportation system globally (Huang, et al. 2020 ) (Shang, et al. 2021 ) (Mogaji 2020) . (Tian, et al. 2021 ) offered a different perspective on urban traffic and air pollution in sample Canadian cities affected by the outbreak. (Sun, Wandelt and Zhang 2020) used a network science technique to conduct a complete empirical investigation of the impact of the COVID-19 epidemic on air travel from a complex system standpoint. In another study, (Aloi, et al. 2020 ) examined the influence of the quarantine imposed in Spain on March 15, 2020, on urban mobility in the northern city of Santander. They discovered a 76 percent drop in total mobility and a 93 percent drop in public transportation utilization in Santander, Spain. (Gajendran 2020) employed descriptive research approaches to analyze travel scenarios under the normal situation, pre-lockdown, and until the COVID-19 epidemic ends with understanding better the influence of coronavirus on Indian people's travel patterns. Moreover, in another study in the United States by (Doucette, et al. 2021) , the effect of the lockdown on daily vehicle miles traveled (VMT) and MVCs in Connecticut has been investigated. The results declare that in the post-stay-at-home timeframe of 2020, the mean daily vehicle miles traveled (VMT) declined by 43 percent (Doucette, et al. 2021 ).

In addition to the ephemeral alterations in urban mobility that occurred during lockdowns (Dindar, Ourang and Ghadikola 2022) (Erfani, et al. 2021) , COVID-19 prompted permanent changes in transportation policy and practice (Budd and Ison 2020) (Campisi, et al. 2020 ) (Musselwhite, Avineri and Susilo 2020) . Many megacities, as well as medium-sized cities throughout the world, have redistributed public space in favor of cycling and walking, imposed automobile traffic limits, and even provided financial assistance to inhabitants for the purchase of bicycles. According to several experts (De Vos 2020) (Batty 2020) (Megahed and Ghoneim 2020) , the pandemic will result in a long-term avoidance of public transportation, an increased bicycling and walking, and a reduced total trips due to greater teleworking. These projections are backed up by research that has already been published. According to a survey conducted in the Netherlands, individuals are now more favorable toward vehicles and more negative against public transportation (de Haas, Faber and Hamersma 2020) . According to the same survey, a high number of people who worked from home during the epidemic anticipate doing so more frequently in the future (de Haas, Faber and Hamersma 2020). Another research based on data from Budapest, Hungary, found an 80 percent decline in public transportation demand, but only a 23 percent fall in biking and a 2% reduction in bike-sharing (Bucsky 2020) . In Beijing, China, the decline in bike-sharing utilization was even more pronounced, with 40 percent fewer rentals made compared to the same time in 2019 (Chai, et al. 2020 ).

Bike-sharing systems have previously been shown to be effective in the face of unexpected events such as public transportation strikes (Saberi, et al. 2018) . During the COVID-19 crisis, bike-sharing was likewise proven to be more resilient than a subway system in New York City, with a less-significant ridership drop. Although research on the influence of COVID-19 on passengers' views toward private automobiles, public transportation, bicycles, and walking has already been conducted, little research has been done to investigate the impact of covid-19 on bike-sharing mobility systems. In particular, the pandemic's influence on bike-sharing in the Washington DC metropolitan area is of significant interest, and the exact correlations between COVID-19 infection rate and bike-sharing ridership are still unclear.

As a result, the present study will examine the impact of the COVID-19 pandemic and household property on bike-sharing ridership, as well as the impact of the bike-sharing system on COVID-19 spread and whether it should continue to operate in the event of a pandemic.

The methodology for the analysis is based on i) a numerical comparison with ordinary least square (OLS) regressions to assess the relation of COVID-19 Cases and Bike Sharing ridership, and ii) adapting contagion models generally used for buses and light rail into a contagion model for shared bike services. We began by employing a time series analysis to examine the bike-sharing system's weekly average ridership with OLS regression to explore their relationship with the number of COVID-19 cases. We utilize spatial analysis to look at the influence of household income, unemployment rate, and density of D.C. tracts on bike ridership in that tract to see how people deal with bicycles during the pandemic.

COVID-19's influence on the bike-sharing system was assessed using a statistical method. In this investigation, ordinary least square (OLS) regressions were used. Ordinary Least Square (OLS) regression is a method for analyzing the relationship between a dependent variable and one or more independent variables by minimizing the sum of squared residuals (the difference between the observed and predicted values) of the dependent variable, which is shaped like a straight line (linear relationship) (Field 2013) . The aim of the regression modeling is to determine the significance and effect of COVID-19 on the behavior of the bike-sharing system rather than to produce a predictive application because there are so many exogenous elements at play. Figure 2 represents this variation and compares it with the reported number of daily new COVID-19 cases and its daily ridership. To investigate the impact of the Covid-19 pandemic on the D.C. bikeshare, the after-pandemic ridership data (i.e., Mar 2020 to Sep 2021) is compared to the ridership of 2019. The statistical model of multivariate linear regression is as follows:

Where is the dependent variable, " are the independent variables, ! is the model constant, % is the model coefficients, is the number of independent variables, and ε is the random error. Given the amount of skewness in Table 1 , the variable of the COVID-19 case does not have a normal distribution. To solve this problem, we use the natural logarithm of variables. Therefore, the final model is as follows where is bike ridership, and is COVID-19 Cases.

The District of Columbia has 179 census tracts, 450 block groups, and 6,507 census blocks. The average annual income, unemployment rate, and density of D.C.'s tracks were used for spatial analysis to observe how these characteristics of zones can impact the behavior of people using bikes during the pandemic. The Figure 3 illustrates that changes in bike share usage are influenced not just by COVID-19 but also by other factors such as the built environment. The use of the bicycle is affected by a combination of factors. We look at the influence of income, unemployment rate, and density in each tract as independent variables on the dependent variable, bike ridership. We compare each station's pre-and-post-COVID19 bike riding rates to get a more accurate comparison.

In overall, during the COVID-19, public bicycle utilization was 28% lower than before. If the ratio of COVID-19 to pre-COVID-19 public bicycle usage is nearly the same for each station, the shift in utilization is due to the epidemic and has nothing to do with other factors. Suppose this ratio varies substantially from station to station. In that case, it indicates that the change is influenced not just by the COVID-19 but also by the station's and surrounding area's features, such as the built environment and demography. Figure 4 depicts the spatial analysis of this ratio with respect to income, density, and unemployment rate.

COVID-19 can be transmitted directly via human-to-human contact/proximity due to droplet formation and movement. The droplets are forming at times of breathing, talking, sneezing, and coughing varies in both size and content. The size of the droplets significantly depends on various environmental factors, such as gravity, the direction and strength of local airflows, temperature, and relative humidity. Droplets >5 μm in diameter fall rapidly to the ground under gravity and therefore are transmitted only over a limited distance (e.g., ≤1 m). In contrast, if the nuclei of the droplets are ≤5 μm in diameter, they can remain suspended in the air for significant periods of time, allowing them to be transmitted over distances >1 m. The term droplet is often taken to refer to droplets >5 μm in diameter (Chartier and Pessoa-Silva 2009) .

Therefore, in this study, we'll suppose the droplets have a diameter of roughly 5 μm and can spread throughout a 1.5-meter-diameter circular region. Droplets of this size can stay in the air for 8-12 minutes, according to (Stadnytskyi, et al. 2020) . As a result, we'll use airtime of 8 minutes, 10 minutes, and 12 minutes for the droplets while retaining the spread diameter at 1.5 meters. Furthermore, we will assume that the probability of a person becoming infected as a result of being in close proximity to someone who is infected (P(A/B) -which is dependent on crowding conditions and adherence to social distancing at the bike-share station) is equal to 1 if the relative distance between them is less than or equal to 1 m, and 0 if the relative distance is greater than 1.5m (five feet). The majority of D.C.'s bike-share stations have 15 bike docks with an average of 17 docks, each measuring 3ft*6ft. Only 10 cyclists should be allowed at any given time to maintain the 1m social separation. For a 1.5m separation barrier, only 6 cyclists should be permitted. The possibility wearing masks (and the quality of the mask) or using cleaners/hand sanitizers are not taken into account in this contagion probability analysis.

Based on the above, the ratio between the number of infected riders at stations and the total number of riders in a given track segment or station defines the probability that a person is standing next to someone infected (P(B)), which is related to the number of people infected in the area. For the scenario where one, two, or three infected riders are present at a station, an illustration of this probability for different riders' numbers is presented below. Given such reasoning, if it is assumed that the riders are exposed to the infected person either at the start or the end location (i.e., origin and destination -O.D.), the probability that each person's susceptibility can be calculated as follows:

Here, the subscript start represents the starting station, and the subscript end represents the ending station of a given user's trip. The virus can also be transmitted via the human-to-surface-to-human path from one person to another. Droplets created by the infected person when talking, sneezing, and other activities will ultimately fall on nearby surfaces. Anyone who contacts that surface and subsequently touches his mouth, nose, or eye might become infected. This is a critical transmission mode in shared bikes since everyone who rides them will be in close contact with the bike's handles. The coronavirus can be stable for up to four hours on copper, up to 24 hours on cardboard, and up to two to three days on plastic and stainless steel, Viruses were applied to copper, cardboard, stainless steel, and plastic for 7 days at a temperature of 21 to 23°C and relative humidity of 40%. Their study shows that the virus lasted 15 hours on copper, 24 hours on cardboard, 76 hours on stainless steel, and 76 hours on plastic based on tissue-culture infectious dose (TCID50) per liter of collecting material. For this study, we will use a stable period of 1, 2, and 3 days. In other words, we will assume a person using the same bike as the infected user within the specific time period will get infected.

The study area is Washington DC, with 179 census tracts. Washington DC's Capital Bikeshare, in August 2008, became the first city in North America to launch a bike-sharing system. Capital Bikeshare (CaBi) is metro D.C.'s bike-share system, with more than 4,300 bikes available at 654 stations across seven jurisdictions. The variables used in this study include Bike ridership (BR), COVID-19 daily cases (CC), average annual household income per capita (AI), unemployment rate (UR), and density (DE) of D.C.'s tracks. ACS Economic Characteristics and Population of D.C. Census Tract is publicly available and also Capital Bike-Share Ridership data are available and include the start and end time of each trip and stations used (start and end) and including the type of user divided between annual members and casual users. Ridership data of CaBi was retrieved from January 2019 to September 2021. The number of COVID-19 cases is available in the Centers for Disease Control and Prevention (CDC) database. The statistical summary of data is shown in Table 1 .

Some trips lasted a very short (or long) time, and those lasting more than four hours or less than one minute were excluded from the current study to increase accuracy. Short trips can be caused by various circumstances, such as when a user finds the bike is in bad condition and returns to exchange it for a better one or when the bike is promptly locked before a user can take it out of the dock. Since then, it's true that the majority of these one-minute-or-less trips began and finished at the same stations. Similarly, we discovered several extremely lengthy trips. These long trips might be the result of the dock system failing to lock the bike upon return, which the users were unaware of this situation, sometimes because the station is offline, or the bike is lost or stolen after it was checked out. 

Capital Bikeshare's trips duration was aggregated monthly to determine if and how this increase in the trip's duration is related to the coronavirus pandemic by comparing the monthly trip duration before and during the pandemic. The data reveals a continued growth in the average trip duration, from a 13-minute daily average at the beginning of Jan 2019 to a 19-minute average by the end of September 2021, translating into a 44% increase. From Jan 2019 to Feb 2020, the average monthly trip duration is 15.5 minutes, while after the pandemic from March 2020 to September 2021, the average monthly trip duration is 21 minutes. This could mean that people prefer short trips on foot. Only those who have always used a bicycle for daily commuting continue to use it. Since the average percentage of casual users before the pandemic is only 11%, and after a pandemic is 43%. The average duration of trips for casual users is 35 min before and 30 min after the pandemic. This average for members is 13 min before and 14.5 after the pandemic. It is also important to note that tourists have made many trips, which have been drastically reduced due to COVID-19 restrictions. 3 / 1 / 2 0 2 0 4 / 1 / 2 0 2 0 5 / 1 / 2 0 2 0 6 / 1 / 2 0 2 0 7 / 1 / 2 0 2 0 8 / 1 / 2 0 2 0 9 / 1 / 2 0 2 0 1 0 / 1 / 2 … 1 1 / 1 / 2 … 1 2 / 1 / 2 … 1 / 1 / 2 0 2 1 2 / 1 / 2 0 2 1 3 / 1 / 2 0 2 1 4 / 1 / 2 0 2 1 5 / 1 / 2 0 2 1 6 / 1 / 2 0 2 1 7 / 1 / 2 0 2 1 8 / 1 / 2 0 2 1 9 / 1 / Considering ridership, as shown in Figure 4 , the ridership fluctuated around 6300 to 11900 before the COVID-19. After the COVID-19, it was followed by a quick decline from nearly 3700 to 8500. The potential daily average bike ridership of 2020 is 10560 trips (CapitalBikeshare 2022) while 5834, meaning a 45% decrease in daily ridership.

Regression model results are shown in Table 2 . Model 1) the natural logarithm of three days average ridership of Capital Bikeshare trips as the dependent variable and the average weekly number of new COVID-19 cases as the independent variable, Model 2) the ratio of average daily ridership of Capital Bikeshare stations trips in tracks before and after COVID-19 as the dependent variable and the census data of tracts as the independent variable. Ln (Bike Ridership)

In the estimated model, the coefficient is negative, which means a decrease in the Bike Ridership with increasing COVID-19 Cases. These results seem to be in line with reality. In the fitted model, ! indicates the elasticity of the Bike ridership relative to COVID-19 Cases. As the COVID-19 Cases increases, people reduce unnecessary trips, and society will feel more afraid of the pandemic and getting infected, which means that the bike-share ridership will decrease. Figure 3 demonstrates that during the COVID-19, the ridership of most stations decreased while that of certain stations increased, with the ratio varying substantially. It demonstrates that before COVID-19, the stations with the lowest ridership ratio were focused around downtown D.C. The office employment density is a significant factor; it is usually negatively correlated with the ridership. It's most likely because the COVID-19 stay-at-home directive reduced the number of individuals working in offices, reducing the use of public bicycles in locations with a large concentration of office workers. As a result, the detrimental impact is readily apparent in the city center. While it is reasonable to assume that the highest strata of society use personal vehicles during the pandemic and that bicycle riding declines in these areas while the opposite is true in low-income communities, regression analysis shows that there is no significant relationship between wealth, population density, unemployment rate and bicycle ridership before and after the pandemic.

However, the figure 3 shows bike-share reduced more significantly in commercial and office areas which are in downtown of D.C. Table 3 shows the average, 75 percentile, and maximum number of users during the peak hours at 8, 10, and 12-minute intervals. Which indicates human-human transmission would be extremely unlikely. Two strangers at a station simultaneously are extremely unlikely to come into close proximity. And only 19 percent of time intervals contain two or more that people in a station. However, we consider the actual average, 50% capacity and 100% capacity to investigate the different scenarios. In Table 4 , the number of infected persons using the contagion model is shown in the case of 1, 2, or 3 infected persons using a single bike-share station during a day while assuming that the aerosols stay in the air for a duration ranging between 8 and 12 minutes.

On the other hand, human-surface-human may be possible. The numbers of trips per bike per 1, 2, and 3 days are calculated from the data for the human-to-surface-to-human transmission. The number of persons infected via surface to human transmission is shown in Table 5 . In (Krishnakumari 2020), the authors used a similar contagion model to study the transmission in D.C. Metro. The authors found that 3 infected persons can infect 20000 people during a 3-day period. According to our study, during a 3-day period, three infected persons can infect 27 people while using the bike-share stations. In other words, comparing the D.C. Metro service to the bike-share service, biking remains a healthier transportation alternative in terms of infection risk.

Despite this finding, the use of bike-sharing systems may need to be regulated as the full capacity operation may still lead to a significant rate of infection and transmission. To see the impact of bike-share on virus transmission we assume that the network of bike-share is connected, and the infected people are coming back to use the service every day and contribute to the transmission of the virus. We can make such an assumption because the users with membership are more likely to be using the services for their daily commute. According to (Lauer, et al. 2020 ) (CDC n.d.) the average incubation period seems to be around 5 days and 95% of the infected people recover after ten days. So, we will assume a 5-day incubation period and ten days recovery period for the modeling. In other words, each person will transmit the virus between days 5 to 10 after getting infected. With such assumptions Figure 5 shows the impact of the bike-share services on the transmission of the virus with a different operating capacity of the bike stations when one infected person uses the service on day one. It takes 12 days to reach exponential contagion at full station operating capacity, 15 days for 75% station operating capacity. However, with a 50% operating capacity, the contagion doesn't reach exponential contagion within 27 days. 

In this paper, the impact of Covid-19 on the D.C. bike-share system as well as the bike-share usage impact on disease transmission is explored D.C.'s Capital Bikeshare program lost around 10,000 daily users at the beginning of the Pandemic in March of 2020. Later on, the ridership numbers increased, but the Capital Bikeshare services are still below their daily averages from earlier years by an estimated 9,000 trips per day. The data reveals continued growth in the average trip duration from 13 minutes to 19 minutes. This could mean that people prefer short trips on foot instead of cycling as before the pandemic. Moreover, the stations with the lowest ratio of ridership after to before the pandemic were focused around downtown D.C. The office employment density is a significant factor in bike ridership. While it is reasonable to assume that the upper class of the society use personal vehicles during the COVID-19 and that bicycle riding declines in these areas while the opposite is true in low-income communities, regression analysis shows that there is no significant relationship between wealth, population density and bicycle ridership before and after the pandemic. However, it has been shown that land use plays an essential role in bike-sharing ridership, which it reduced more significantly in commercial and office areas in downtown D.C.

With such usage, to analyze the impact of bike-sharing on the dynamics of the Covid-19 reported case, the transmission process is classified into two types: i) Human-to-human transmission and ii) Human-tosurface-to-human transmission. A probabilistic contagion model is used to represent the first form of transmission. Transmission from one person to another would be infrequent. It's exceedingly improbable that two strangers at a station will come into close proximity at the same time. On the other hand, the second type is more likely to happen, and it will be anticipated using the average number of trips per bike throughout the study interval and the corresponding duration during which the virus remains stable on a given surface.

The numerical analysis indicated that one infected person could infect a maximum of 5 persons, and three infected persons can infect a maximum of 28 persons during a 3-day interval (ignoring the possible infections beyond the bikers' community). The peak value of the 7-day moving average of infected persons is 194.

The findings show that bike-sharing has a minor impact on the Covid-19 infection rate, and that the government and decision-makers should consider it as a safe mode of transportation that should be maintained because it can encourage people to use it instead of the subway or bus. In comparison to other modes of public transit, bike-share remains a relatively healthier option.

The limitation is that there does not appear to be any real-world assessment of how bike-share users interact at stations. Also, for future studies, given the more specific data sets (i.e., the trip history dataset with a user identification number, socio-economic characteristics in a given location, datasets provided at this stage with routing information), the interaction of users at a station can be modeled much more realistic. Moreover, with the travel paths data complemented by pedestrian density information along some routes, we can predict the total impact of bike-share services at the origins, destinations, and in between the origindestination (O.D.) pairs.

Public transit ridership analysis during the COVID-19 pandemic

Effects of the COVID-19 lockdown on urban mobility: Empirical evidence from the city of Santander (Spain)

Assessing the impact of reduced travel on exportation dynamics of novel coronavirus infection (COVID-19)

The Coronavirus crisis: What will the post-pandemic city look like?

Modal share changes due to COVID-19: The case of Budapest

Responsible Transport: A post-COVID agenda for transport policy and practice

Cycling master plans in Italy: The I-BIM feasibility tool for cost and safety assessments

Capital Bikeshare System Data

Ending Isolation and Precautions for People with COVID-19: Interim Guidance

Analysis of Spatial-temporal Behavior Pattern of the Share Bike Usage during COVID-19 Pandemic in Beijing

Natural ventilation for infection control in health-care settings

The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak

How COVID-19 and the Dutch 'intelligent lockdown'change activities, work and travel behaviour: Evidence from longitudinal data in the Netherlands

The effect of COVID-19 and subsequent social distancing on travel behavior

Development of a Communication-Assisted Adaptive Overcurrent Protection Scheme in Smart Distribution Networks in Presence of Wind and Solar Generation

Initial impact of COVID-19's stay-athome order on motor vehicle traffic and crash patterns in Connecticut: an interrupted time series analysis

Heterogeneous or homogeneous? A modified decision-making approach in renewable energy investment projects

Discovering statistics using IBM SPSS statistics. sage

Impact of novel Coronavirus (COVID-19) pandemic on travel pattern: A case study of India

Understanding the impact of the COVID-19 pandemic on transportation-related behaviors with human mobility data

The Enduring Effects of COVID-19 on Travel Behavior in the United States: A Panel Study on Observed and Expected Changes in Telecommuting

Virus spreading in public transport networks: the alarming consequences of the business as usual scenario

The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application

Outbreak dynamics of COVID-19 in Europe and the effect of travel restrictions

Antivirus-built environment: Lessons learned from Covid-19 pandemic

Impact of COVID-19 on transportation in

Editorial JTH 16--The Coronavirus Disease COVID-19 and implications for transport and health

New coronavirus stable for hours on surfaces

Human mobility and coronavirus disease 2019 (COVID-19): a negative binomial regression analysis

An automatic system to monitor the physical distance and face mask wearing of construction workers in covid-19 pandemic

Understanding the impacts of a public transit disruption on bicycle sharing mobility patterns: A case of Tube strike in London

Hybrid Simulation-based Resource Planning and Constructability Analysis of RCC Pavement Projects

Impacts of COVID-19 pandemic on user behaviors and environmental benefits of bike sharing: A big-data analysis

The airborne lifetime of small speech droplets and their potential importance in SARS-CoV-2 transmission

How did COVID-19 impact air transportation? A first peek through the lens of complex networks

Assessing the impact of COVID-19 pandemic on urban transportation and air quality in Canada

Aerosol and surface stability of SARS-CoV-2 as compared with SARS-CoV-1

2022. WHO Coronavirus (COVID-19) Dashboard. Accessed 2022

Impact of mobility on COVID-19 spread--A time series analysis