Available via license: CC BY
Content may be subject to copyright.
Sustainability 2020, 12, 3870; doi:10.3390/su12093870 www.mdpi.com/journal/sustainability
Article
Effects of the COVID-19 Lockdown on Urban
Mobility: Empirical Evidence from the City of
Santander (Spain)
Alfredo Aloi
1,2
, Borja Alonso
3,
*, Juan Benavente
3
, Rubén Cordera
4
, Eneko Echániz
4
,
Felipe González
5
, Claudio Ladisa
1,6
, Raquel Lezama-Romanelli
1
, Álvaro López-Parra
1
,
Vittorio Mazzei
1,2
, Lucía Perrucci
1,6
, Darío Prieto-Quintana
1
, Andrés Rodríguez
3
and
Roberto Sañudo
4
1
E.T.S. de Ingenieros de Caminos, Canales y Puertos, Universidad de Cantabria, 39005 Santander, Spain;
aal104@alumnos.unican.es (A.A.); cla682@alumnos.unican.es (C.L.); rlr896@alumnos.unican.es (R.L.-R.);
alp00@alumnos.unican.es (Á.L.-P.); vma252@alumnos.unican.es (V.M.); lpe759@alumnos.unican.es (L.P.);
dpq761@alumnos.unican.es (D.P.-Q.)
2
Dipartimento di Ingegneria Civile, Università Della Calabria, 87036 Rende, Italy
3
Transport System Research Group (GIST), Universidad de Cantabria, 39005 Santander, Spain;
juan.benavente@unican.es (J.B.); andres.rodriguez@unican.es (A.R.)
4
Research Group on Sustainable Mobility and Railways Engineering (SUM
+
Lab), Universidad de Cantabria,
39005 Santander, Spain; corderar@unican.es (R.C.); eneko.echaniz@unican.es (E.E.);
roberto.sanudo@unican.es (R.S.)
5
Faculty of Engineering and Sciences, Department of Industrial Engineering, Universidad Diego Portales,
8320000 Santiago, Chile; felipe.gonzalezr@udp.cl
6
Dipartimento di Ingegneria Civile, Ambientale, del Territorio, Edile e di Chimica, Politecnico di Bari,
70126 Bari, Italy
* Correspondence: alonsobo@unican.es; Tel.: +34-9422-02-294
Received: 14 April 2020; Accepted: 5 May 2020; Published: 9 May 2020
Abstract: This article analyses the impact that the confinement measures or quarantine imposed in
Spain on 15 March 2020 had on urban mobility in the northern city of Santander. Data have been
collected from traffic counters, public transport ITS, and recordings from traffic control cameras and
environmental sensors to make comparisons between journey flows and times before and during
the confinement. This data has been used to re-estimate Origin-Destination trip matrices to obtain
an initial diagnostic of how daily mobility has been reduced and how the modal distribution and
journey purposes have changed. The impact on externalities such as NO
2
emissions and traffic
accidents have also been quantified. The analysis revealed an overall mobility fall of 76%, being less
important in the case of the private car. Public transport users dropped by up to 93%, NO
2
emissions
were reduced by up to 60%, and traffic accidents were reduced by up to 67% in relative terms.
Keywords: COVID-19; coronavirus; mobility; traffic; confinement; quarantine; outbreak
1. Introduction
At the end of 2019, when health officials in Wuhan (China) reported a group of pneumonia cases
caused by an unknown agent, later identified as SARS-CoV-2 in mid-January, the rest of the world
saw it as a local problem occurring a long way away. The World Health Organization (WHO) named
this new illness COVID-19, and it quickly spread throughout the population, with particularly
serious impacts on Italy and Spain during its first phase of expansion outside China to the rest of the
world. On 11 March, the WHO declared this illness to be a “global pandemic”, and at the date of this
Sustainability 2020, 12, 3870 2 of 19
article, there are currently over 1,800,000 confirmed cases and many thousands of dead. Some
publications have already analyzed and modelled the expansion of this virus [1–3], and recent studies
have assessed the utility of Big Data tools to support the analysis of available data to make predictions
about this and other illnesses [4,5].
Regarding the repercussions of the virus on the transport sector and mobility, most research has
concentrated on the effects global mobility has had on China and its influence on how the virus has
spread in that country [6–10]. Regional, interregional, and municipal aggregated analysis has been
found using mobile phone tracking [11] or mobile applications [12]. All these studies provide
irrefutable evidence that provincial and interprovincial mobility has reduced analogously to how it
was observed in China in preceding studies. Moreover, Ivanov [13] reported using supply chain risk
disruption under different outbreak scenarios to assess the impacts on mobility.
However, the repercussions the pandemic has had on internal mobility within towns and cities,
above all in Europe, have not been reported on in any detailed way until now. Partial reports, news,
press notes, etc., have been found where some reference has been made to the fall in mobility due to
social distancing measures and the reduction in journeys being made. A report produced by INRIX
[14] in the city of Seattle (USA) using data from mid-March found a reduction in commuting journeys
of up to 60% and improvements in journey times of 26% with a 13% reduction in vehicle/km. A global
analysis of mobility has recently been published by Google [15] revealing a worldwide drop in
journeys in practically every country.
All the analyzed published data coincides in that mobility has dropped around the world as the
spread of the virus has increased and crossed frontiers. The success of the lockdown policy
implemented in Wuhan led the most affected countries to apply similar measures which restricted
mobility [16,17]. The analysis carried out by TomTom published at TomTom Traffic Index [18]
quantifies the congestion levels in many cities around the world. They found that current congestion
levels are below 10% due to the COVID-19 effect, where typical congestion levels in these cities are
normally around 50–70%. Public transport systems are the most prejudiced in this decrease, with
many users refusing to use them to avoid social contact and reduce risk of contagion. Where
limitations have been imposed on travel, the drop in the number of journeys being made has always
been greater percentage-wise for public transport than for private traffic. For example, Wuhan
(China) or Delhi (India) have registered reductions of 80–90% in the number of users [19]. On 23
March, the city of Bogotá (Colombia) performed a simulation of journey limitation measures, and the
TransMilenio system registered a fall in the number of users of 87% [20]. There are also some
guidelines and recommendations based reports published by official entities such as the International
Association of Public Transport (UITP) [21] or the Transportation Research Board (TRB) [22] to
provide best practices on planning and operating public transport systems under health emergency
conditions. In fact, reports are already available highlighting the economic consequences that these
measures could have on service providers due to the enormous switch in supply and demand and
the possible reductions in bus drivers to address the current adjustment in supply [23,24]. Figure 1
shows how volumes of public transport users have evolved in different cities around the world, as
reported by the Moovit platform [25]. The figure shows how large cities have experienced strong
declines as the virus has extended and more restrictions have been imposed.
Sustainability 2020, 12, 3870 3 of 19
Figure 1. Evolution of public transport users since 15 January (source: www.moovit.com).
This drop in usage has been caused by both the reticence of the users themselves and the
recommendations of health authorities to avoid using public transport as much as possible,
recommending the use of individual means of transport such as bicycles. In spite of such
recommendations, research has also been found highlighting the risk of contagion associated with
the use of shared bicycle systems [26].
These reductions in travel have had similar repercussions on the main externalities normally
associated with motorized transport modes, mainly journey times, accidents, and pollution. Data has
been published in different media revealing reductions of 35% in accident rates in Istanbul (Turkey),
where the only action taken during the period of study was to close the schools [27]. Reductions in
emissions have also been reported from different countries around the world, especially significant
in the cases of NOx, CO2, and PM2.5 [28–30]. In China, for example, the measures to minimize the
spread of SARS-CoV-2 have resulted in reductions of 25% in CO2 emissions, and NO2 levels were
36% lower than in the same period in 2019. In Italy, significant reductions in NO2 concentrations have
been found to be mainly due to the reduction in the use of diesel vehicles for transport. In France,
minimum NOx levels have been reached due to the restrictions adopted to fight COVID-19 in
economic activities and transport. During the spread of the COVID-19 pandemic in New York, traffic
levels were estimated to be down 35% compared with a year ago. CO2 levels dropped 5–10%, and
significant decreases in the emissions of CO and methane were also detected [31]. Wang et al. [32]
applied the Community Multi-scale Air Quality model considering the constraints caused by the
situation with COVID-19 in China. In particular, a traffic reduction of 80% was considered in the most
restrictive case. After analyzing the model results, they concluded that anthropogenic emissions
decreased, especially in transport and industry, with a reduction in PM2.5 concentrations. However,
the study concluded that the reduction was not enough to avoid the occurrence of severe air pollution
events in most of the areas being studied.
In the case of Spain, the simultaneous closure of centers of education and restrictions on mobility
were introduced on Sunday, 15 March 2020 and extended 15 days later, limiting all travel to only
those journeys considered to be essential. This date is clearly reflected in Figure 2, representing the
evolution of public transport use for various Spanish cities. A strong decline in the number of users
began on the 15th and stabilized approximately one week later. The drop in user numbers is similar
for all the cities, at around 80–90%. An analysis of the Google report [15] for Spain once again found
falls of 94% in journeys made for reasons of leisure or shopping, 77% in journeys for food or trips to
pharmacies, 89% in mobility around stations and transport nodes, and 68% towards work places.
These reductions have been found to be homogenous for all the Spanish regions, including Cantabria,
where Santander (the case study of this work) is located.
22 apr. 2020
apr.
apr.
apr.
apr.
apr.
apr.
apr.
Sustainability 2020, 12, 3870 4 of 19
Figure 2. Evolution of public transport users in Spain since 15 January (source: www.moovit.com).
The reported data proves that overall mobility in many countries of the world has fallen as a
consequence of the sudden stop to human activities due to lockdown (quarantine) policies. This has
led to the extreme breakdown of mobility patterns [33,34] and halted the current trend towards
sustainable transport in cities. In spite of increasing motorized traffic as the economic crisis was being
overcome, especially on regional networks [35,36], the use of personal transport systems (both shared
and private) such as bicycles or scooters was exponentially growing in urban areas [37].
The authors, therefore, decided to report on a more detailed and integrated analysis of the
mobility changes that have occurred within a city. This study covers various modes of transport and
their effect on other externalities, and this research will provide data which can be used to make more
thorough detailed comparisons with other case studies.
2. Case Study Analysis
The particular case being studied in this article is the city of Santander, located on the North
Spanish coast. Santander is the capital of the autonomous region of Cantabria. With nearly 180,000
inhabitants according to the National Institute of Statistics, trade and services provide employment
for more than 70 percent of its active population.
According to the 2013 mobility survey updated with traffic transit and pedestrian counts in 2018,
the aggregated picture of mobility before (including commuting) the quarantine can be seen in Figure
3, where 42% of urban mobility is done by walking, 48% by private motorized transport (driving or
passenger), and 8% by bus. The remaining 2% is done by bike/scooter. It should be highlighted that
public transport is more often used for study and health related trips. The daily trips profile was
estimated using this data.
22 apr. 2020
apr.
apr.
apr.
apr.
apr.
apr.
apr.
Sustainability 2020, 12, 3870 5 of 19
Figure 3. Basic mobility (including commuting) data before the quarantine (source: Mobility survey,
2013. Universidad de Cantabria).
Figure 3 also shows three peak periods in a working day as many people return home for lunch
due to the Spanish working day. These peak hours and patterns match for all modes throughout the
day, with the exception of the afternoon peak hour, which is slightly longer for the private vehicle mode.
The new urban mobility patterns resulting from the quarantine are calculated from different
data obtained by the authors thanks to an agreement with the city hall, which provides real time
information to a data processing laboratory located at the University of Cantabria. A total of 480
electromagnetic traffic counters reported values for intensity and occupation every minute. Public
transport GPS positioning data were analyzed to provide speed and number of stops from the onboard
Intelligent Transport System (ITS), ticketing data provided the number of users at any given moment,
at what stop, line, and vehicle. Finally, 45 traffic control cameras were analyzed, and using image
processing techniques the authors were able to discretize pedestrian flows. Thanks to the recording
system, we were able to process and compare the images before and during the quarantine.
An example of the data used is shown in Figure 4. On the left, the figure shows a moment taken
of a residential zone, showing the queue of people waiting to enter a supermarket, and on the right
is the main street in the center of the city with a clear absence of vehicles and pedestrians.
Finally, data from fixed sensors installed by the Regional Government’s Environmental
Department was also analyzed along with data from mobile sensors installed on the roofs of local buses.
Sustainability 2020, 12, 3870 6 of 19
Figure 4. Images taken from traffic control cameras. In a residential zone (left) and the main street of
the city center (right).
The analysis started with general motorized traffic, where a strong decline was found
throughout the city. As a reference, before confinement, the main entrance to the city registered about
42,000 vehicles on a working day. After confinement was declared, the traffic reduced by 64%, and
during week 3, the reduction rose to 78%. In order to obtain an overall result, a Network Macroscopic
Fundamental Diagram—NMFD [38] was estimated for all the Mondays in the month of March. Days
2 and 7 of March can be considered as normal, days 16 and 23 enter the first period of quarantine,
being the first and second working Monday of the quarantine (days 2 and 9), respectively. The 30th
of March (Monday) enters the second period of quarantine (16th day of quarantine), where the
restrictions on mobility are even more severe (and still are, at the date of writing this article), trips
being limited to essential activities only and the purchase of basic necessities. The data was
aggregated into periods of 5 min. Figure 5 shows how the normal days before quarantine show
practically identical behavior, with average maximum intensities of around 600 veh/h and always
within the stable range, never reaching capacity. Flow and occupancy drastically reduce during the
quarantine, with an even greater reduction (over 65%) on the 30th of March. The figure on the right
shows the same behavior throughout the day, clearly showing that during confinement, morning
mobility is greater than during the afternoon, where afternoon rush hour has all but disappeared,
and morning rush hour smoothed its peak after the restrictions were hardened as fewer people were
allowed to travel for reasons of work. An increase in the average flow during the night period
(between 1:00 a.m. and 5:00 a.m.) has also been noted, but no further data are available at this time to
explain this fact.
Figure 5. Network Macroscopic Fundamental Diagram (NMFD) and average flow daily profile of
Santander.
A similar analysis can be performed by disaggregating the city into corridors [39]. Figure 6
shows the corridor with the most traffic (Marqués de la Hermida street) which is also the main
entrance route to Santander from the East (direction Bilbao).
0
100
200
300
400
500
600
700
0 1020304050
Average Flow (veh/h)
Average Occupancy (%)
Network Macroscopic Fundamental Diagram
Monday 02/03/20 Monday 09/03/20 Monday 16/03/20 Monday 23/03/20
Monday 30/03/20 Monday 06/04/20 Monday 20/04/20
Sustainability 2020, 12, 3870 7 of 19
The resulting diagrams (Figure 6 left) follow the same pattern of behavior as the case explained
above, although with higher intensity and occupancy values. The same occurs with the traffic profiles
(Figure 6 right), where the morning peak is quite striking as it is a Monday and the main access point
from the surrounding catchment area. This peak continues to appear during quarantine, but is much
less striking with lower traffic intensity. As before, once morning rush hour has passed, the amount
of traffic drops and afternoon rush hour disappears completely.
Figure 6. NMFD (left) and average flow daily profile (right) of the main corridor of Santander.
The drop in mobility was much steeper for the city’s public transport system. The dramatic fall
can be seen in Figure 7 (left), with an average of over 90%, the morning peak disappearing, and the
mid-day peak being slightly maintained. As happened with general traffic, a 35% decrease in users
appears in the afternoon period compared with the numbers registered in the morning. If the data is
analyzed by bus line (Figure 7 right), lines 1 and 2 carry the most passenger loads, and their demand
dropped by more than an average of 85%. The lines servicing the university campus (among other
zones) suffered an average reduction of 92% and those serving the city periphery reduced by 88%.
The much lower demand, together with the reduction in general traffic and a lack of congestion,
has meant the journey times of the lines have been notably reduced. As an example, Figure 8 shows
the journey times by time of day for two lines of high and low demand (lines 1 and 6c2 respectively).
Both cases registered 30% reductions in cycle times.
Figure 7. Public transport trips throughout a working day (left) and daily trips per line (right).
0
100
200
300
400
500
600
700
800
0:00:00
0:07:12
0:14:24
0:21:36
0:28:48
0:36:00
0:43:12
0:50:24
0:57:36
8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
Trip s
Trav el ti me
Line 6c2 Travel times and demand profiles
Travel time b efore Travel time after Trips before Trips after
0
100
200
300
400
500
600
700
800
0:00:00
0:07:12
0:14:24
0:21:36
0:28:48
0:36:00
0:43:12
0:50:24
0:57:36
7:00
8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
Trip s
Trav el ti me
Line 1 Travel times and demand profiles
Travel time before Trav el time after Trips before Trips after
Sustainability 2020, 12, 3870 8 of 19
Figure 8. Cycle times throughout the day of two public transport lines.
Figure 9 shows the evolution of the daily demand and the total running length per day as a
performance indicator of the service provided. As can be seen, the demand started dropping slightly
at the end of the week before quarantine. However, during the first working day of the lockdown
period, the number of users fell to levels that were even lower than those reported on a normal
Saturday. These results for the public transport system are in line with those already reported in
section 1 and, as can be deduced, they represent a strong blow to the budgetary balance of the system,
since the operational cost is now more disproportionate to the new scenario of reduced demand. The
public transport operator kept the usual working day services during one week, but the service had
to be adapted to readjust the line intervals to this new demand (Figure 9).
Figure 9. Evolution of public transport demand and total running length serviced per day.
A decision support tool is needed to optimize the design of the new services. Applying the social
optimum results of the frequency optimization model for this same network [40,41] provides a rapid
response for the operator to be able to readapt the service, as well as taking into account the new
cycle times recorded. Table 1 shows a precise comparison between the situation prior to confinement
with the calculated optimal intervals and the current situation of the service during confinement,
where a great similarity can be seen.
Table 1. Headway comparison of the public transport system.
Line Headways (min)
Before Optimun (Quarantine) Current
1 20 30 33
2 20 30 33
3 20 35 42
4 15 35 33
5c1 12 24 20
5c2 12 24 20
6c1 30 60 60
6c2 30 60 60
7c1 20 60 55
7c2 20 60 55
11 30 60 60
12 30 40 45
13 30 60 55
14 22 60 55
Sustainability 2020, 12, 3870 9 of 19
These operational adjustments, assuming an operating cost of 3.5 euros/km (it is assumed that
there are no fleet or personnel savings, since the situation is provisional), show significant savings of
close to 50%, which allow the economic impact of the lower demand to be reduced, while maintaining
the social optimum in the system, as shown in Table 2.
Table 2. Estimated operational costs of the public transport system.
Before Optimun (Quarantine) Current *
Estimated operational costs (€/day) 37,053.67 € 19,449.45 € 19,260.85 €
−47.51% −48.02%
Estimated incomes (€/day) 24
,
392.50 € 2
,
663.00 € - €
−89.08%
Estimated revenue (€/day) −12,661.17 € −16,786.45 € −19,260.85 €
32.58% 52.13%
* The city council decided to avoid charging public transport fares during the quarantine period.
Because of the type of restrictions imposed, fewer journeys were made using public transport
for reasons of work in favor of using the private car. The same has happened with pedestrian mobility
during the morning rush hour, which has been drastically reduced. The only mobility now occurring
on foot has been to walk to local food shops and to buy other essential products. Due to the
impossibility of performing specific counts in person, the authors were able to access and process the
images from traffic control cameras, many of which are located in the streets with the highest
pedestrian flows in the city. The same approach has been followed to quantify the trips made by
bikes/scooters. In this sense, it is important to notice that the public bike system was closed during
the quarantine, and there are no private operators of any sharing mobility.
The results obtained revealed a similar reduction in pedestrian flows, although to a lesser degree
than was found with the public transport service. Figure 10 shows two examples of the comparison
between the same location at the same time of day between vehicle and pedestrian flows before (left)
and after (right) the restrictions were imposed. The upper example compares an access tunnel to a
central intermodal area, and the lower example compares one of the main pedestrian locations in the
city (Numancia square).
The differences are revealing about the resulting drop in mobility. To estimate the daily trip
profile for pedestrians and bikes, the authors correlated the pedestrian and bikes flows obtained from
9 different camera recordings with the existing pedestrian and bikes flow assignments from the
mobility model that the University of Cantabria had calibrated for the city and the profile obtained
from the latest Origin-Destination (O-D) survey, updated with manual counts up to 2018. The same
correlation has been applied using the new flows obtained from the images taken during the
quarantine period, once again re-estimating the trip matrices and finding the daily travel profile.
The resulting reduction coincided with that observed for general traffic and public transport:
The morning and evening peaks disappeared drastically with the appearance of a mid-day peak
period between 11:00 and 13:00 when many people leave their homes to do basic shopping. Mobility
during the afternoon period falls by approximately 55% compared with the morning, where for a
normal day the drop is slightly lower than was found for motorized vehicles, not reaching 70%.
Sustainability 2020, 12, 3870 10 of 19
Figure 10. Traffic and pedestrian flow comparison before (left) and during (right) quarantine.
All the updated information was introduced into the city mobility model, the O-D matrices were
re-estimated, and the data was expanded to include night time. This process provides the new overall
modal distribution for the entire day and for separate periods during the day as well as the trip
profiles shown in Figure 11. As can be seen, the new modal distribution changes significantly, as car
journeys change from 48% to 77% (it is important to clarify that the number of total car journeys is
much lower during the quarantine), and public transport from 8% to 2%. Pedestrian journeys also
show a significant drop in their share from 42% to 19%. It was estimated that overall mobility has
fallen by 76%, with variations throughout the day, reaching falls of up to 85% at certain times. Car
travel fell by 68% (in some periods by up to 85%) and bus travel fell by 93%. These results are
consistent with the data provided by Moovit [23] and the Spanish Ministry of Mobility [11]. Figure 3
shows that the proportion of trips using bicycles or scooters (labeled as “others”) was very low before
the quarantine. As happened with other modes of transport, overall mobility using these means of
transport also decreased. However, the resulting proportional reduction was lower than the one
observed for public transport or walking. This means that most people continued cycling (or using
scooters) during the quarantine, which is consistent with the results found by some bike sharing
companies [17]. Furthermore, if the origins and destinations of the estimated matrices are now
combined with the information provided by Google in their regional mobility report [15], the new
reasons for journey during the quarantine period can then be inferred. Logically, the journeys made
for reasons of education disappear; those made for work are clearly in the majority, moving from
35% before quarantine to 74% (over the whole day). The remaining journeys are explained by the
requirement to obtain basic products and the purchase of other available products.
Sustainability 2020, 12, 3870 11 of 19
Figure 11. Mode share, trip reason, and total trips daily profile before and during the quarantine.
The lockdown measures can also cause changes to proximity mobility during the SARS-CoV-2
pandemic. This can be seen in changes in the length of the trips being made by citizens (see Figure 12)
as reported by the Spanish Ministry of Transport based on mobile phone data [11]. The lockdown has
caused the shortest trips (0.5–2 km) to clearly gain weight over long-distance trips (more than 5 km).
Figure 12 also shows how the weight of shorter trips increases at weekends, because travelling for work
(mid-long distance mainly) decreases and most of these trips are made for buying basic products.
Figure 12. Trip length distribution before and during the quarantine (source:
https://www.mitma.gob.es).
As explained beforehand, this strong decline in motorized mobility has resulted in a reduction
in externalities such as emissions. In terms of emissions, the analysis has been focused on NO
2
, as it
0
10
20
30
40
50
60
70
03/03/20
05/03/20
07/03/20
09/03/20
11/03/20
13/03/20
15/03/20
17/03/20
19/03/20
21/03/20
23/03/20
25/03/20
27/03/20
29/03/20
31/03/20
02/04/20
04/04/20
06/04/20
08/04/20
%
Trip length dis tribution before a nd during the quarantine
0.5-2 Km 2-5 Km 5-10 Km
Sustainability 2020, 12, 3870 12 of 19
is the substance most directly related to urban traffic, as opposed to other pollutants coming from
alternative sources such as PM
10
and PM
2.5
[42–44]. The analysis of the NO
2
values from the static
Department of the Environment stations located in the city center and in an industrial zone located
in a nearby town shows how the normal peaks that used to be registered disappeared completely at
the beginning of the quarantine, with reductions of over 50–60% compared with the average values
measured in the period 2010–2019 (Figure 13).
Figure 13. NO
2
measurements on the static environmental stations (source:
https://www.airecantabria.com/index.php).
The values provided by the static stations can be complemented by data from the mobile sensors
installed on the city buses. All the different NO
2
measurements were used in order to have more data
available (Figure 14) and all the data from the month of March has been aggregated. This has resulted
in an overall reduction in NO
2
emissions throughout the city, but especially in the North, Central,
and West zones, where a large number of work places, colleges, and university buildings are located.
The absence of data from the northern zone is because the buses equipped with sensors were not
running in that zone during the analyzed time period. The results obtained are consistent with those
found in an independent report made by a Spanish ecologist organization [42].
Figure 14. NO
2
measurements before (left) and during (right) quarantine with mobile devices.
Finally, accident data reported by the local city police was analyzed. Data from approximately
one and a half months before quarantine was compared with data produced during quarantine: From
the start of February to the 14th of March, and from the 15th of March to the 23rd of April. The number
of accidents in absolute terms has fallen from the start of quarantine (Figure 15) from a total of 105 to
0
5
10
15
20
25
30
35
1/3/20
3/3/20
5/3/20
7/3/20
9/3/20
11/3/20
13/3/20
15/3/20
17/3/20
19/3/20
21/3/20
23/3/20
25/3/20
27/3/20
29/3/20
31/3/20
NO2 emissions (ug/m3)
Evolution of NO2 emissions (static devices)
Indu strial zo ne City center 2010-2019 values
Sustainability 2020, 12, 3870 13 of 19
17 (−83.8%), from 17.5 accidents per week to 2.83. However, if this data is weighted to consider the
average daily traffic intensity registered in the city, the fall is around 67%.
Figure 15. Number of traffic accidents per week (source: Santander local police).
3. What Is Next?
Evidently, urban mobility, and Santander is no exception, has been put on hold during the
quarantine period. The great unknown being faced by many cities is how the demand for transport
and mobility will evolve during the following stages in the process (the recovery period).
Many cities are planning measures to coordinate the slow increase in mobility during the
recovery period with the social distancing required to avoid the probability of new contagion and the
return of the illness. It is a worry that the move towards sustainable mobility has been halted and we
have returned to the ratios of motorized transport found over a decade ago. Considering the
experience of China, the trend is to favor trips being made on foot or using individual means of
sustainable transport such as bicycles or scooters [17]. To support this change, investment has been
made in providing the infrastructure required for these means of transport by removing lanes that
were previously used for traffic and reducing city center speed limits to 30 km/h [45–47]. The
measures that have simultaneously been taken to maintain social distancing between people means
that pedestrian areas have to be redesigned along with crossings (Figure 16) and traffic lights,
removing buttons [48] and lengthening the crossing time phase for pedestrians, which partly
compensates for the reduced capacity caused by a smaller crossing section and thereby avoids the
accumulation of pedestrians waiting to cross. These measures will inevitably result in a reduction in
the capacity of the infrastructure provided for motorized transport.
Sustainability 2020, 12, 3870 14 of 19
Figure 16. Pedestrian crossing proposed in Santander (source: AC Proyectos: www.acproyectos.es).
Nevertheless, for all the effort that is used introducing these measures, their success will
inevitably depend on the behavior of public transport. The effect will be clearer in those cities with
large public transport systems (metros, BRTs, etc.) where the transport capacity will be reduced (and
is already being seen in cities undergoing a period of recuperation) due to social distancing measures
between passengers inside the vehicles [49] and the increased frequencies required to reduce waiting
times and passenger accumulation at stops and stations. All these factors inevitably mean more
vehicles are required, and more drivers are required for non-automatic systems. Both these cases
imply significant increases in the operating costs of systems that will be carrying a reduced demand
compared to the situation as it was before quarantine. Data is already available from some cities in
China where the passenger demand for public transport is slowly recovering, reaching 30% by the
4th week into the recovery period [50]. Faced with this situation, many administrations and operators
are working with different recovery hypotheses and, although market prediction models are
available [51–53], the intrinsic characteristics of the nature of this crisis and the abnormal situation it
generates may affect their efficiency.
As an example and in the specific case of Santander, where the normal bus load at the rush hour
is 82% in the city center corridor, different scenarios can be analyzed for both the growth in demand
and the system operation. Assuming similar rates of growth to the Chinese case [50] and that social
distancing rules are maintained, the resulting reduction in vehicle capacity will be around 70% of the
nominal capacity, possibly rising to 50% with any potential relaxing of distancing rules by the
authorities when the situation warrants it. Figure 17 shows the evolution of the required operating
costs to satisfy the predicted demand whilst maintaining “safe” occupancy rates. Assuming a
recovery period lasting 2 months, the operator will have to double their normal operating costs for
over 3 weeks and at the same time carry a much reduced demand. With all these factors, Table 3
provides details for total income and costs during the 2 months of the recovery period for the case
study. In the case of a 30% occupancy rate, the operator would multiply by 2.5 their losses compared
to a normal period, while a 50% occupancy rate would result in a 52% increase in losses. Both cases
will also have to factor in the feasibility of having to increase the required resources (fleet and drivers)
to provide the service.
These results affect the economic viability of the entire public transport system. It would be
unsustainable during only this brief period of time, and the service needs to be provided during the
entire quarantine period, as shown in Table 3, where the overall losses would be 148% higher. The
implication is that public administrations will need to increase subsidies paid to guarantee the
economic viability of the transport system. As well as all the above, assuming that it will not be viable
KEEP
DISTANCE
KEEP
DISTANCE
KEEP
DISTANCE
Sustainability 2020, 12, 3870 15 of 19
to duplicate the current fleet size, the focus of any action should not only be on the supply, but also
on the demand. Times for entering work places and educational establishments can be scaled to
minimize the rush hour factor that results in high vehicle occupancy on public transport systems.
Only by taking measures addressing both supply and demand can the viability of public transport
systems be guaranteed during the period of social distancing restrictions.
Figure 17. Evolution of public transport costs during the recovery period.
Table 3. Estimated incomes and costs during the recovery period.
30% Occupation Rate 50% Occupation Rate Normal
Incomes 706,873.25 € 706,873.25 € 1,492,277.60 €
Operating costs 3,751,221.19 € 2,493,789.37 € 2,697,507.37 €
Balance 3
,
044
,
347.95 € 1
,
786
,
916.12 € 1
,
205
,
229.77 €
4. Conclusions and Open Questions
This article has used available data to provide a preliminary report about how the imposition of
quarantine throughout Spain due to the SARS-CoV-2 virus has affected internal mobility in the city
of Santander. The analyzed data shows an overall fall of 76%, although this is less important in the
case of the private car. Public transport use has fallen the most with 93% fewer users. Mobility during
the morning and midday has dropped less than in the afternoon, when the fall is much more drastic
with the disappearance of afternoon peak traffic periods. The effect of confinement has logically
modified the purposes behind people’s journeys, work being by far the most important purpose. The
declining mobility has also produced a reduction in the emission of pollutants. The data collected
about NO
2
emissions has allowed the authors to quantify this reduction as practically 60%, and traffic
accidents have been reduced by up to 67% in relative terms. The implications on the public transport
system have also been analyzed, highlighting the importance and the need for policies which not
only maintain the economic exploitation itself, but also distribute the demand to avoid rush hours.
These effects should make us think on how mobility in cities will be after this crisis. There is a
real risk of a decline in the sustainability of mobility in urban areas. The main questions to be
answered will be, first of all, when, how, and to what extent the demand levels for public transport
systems will recover (if they ever do). The willingness of the user to take over not only collective but
also shared transport systems will be another important issue to be assessed. Further questions to be
investigated are how this will affect users’ perception of different transport services, and what new
strategies both public and private sector operators will need to follow to make public transport
systems attractive again. Variables such as vehicle cleanliness and hygiene, as well as vehicle
occupancy, are likely to increase their prominence in measurements of perceived quality.
Sustainability 2020, 12, 3870 16 of 19
Finally, it may be interesting to monitor changes in travel production habits: Will teleworking
increase? Will travel patterns change for leisure, shopping, etc.? How much of this decreased mobility
was really necessary? Research that allows a follow-up and monitoring of these new issues, as well
as a before and after comparison, will surely be an important contribution to the state of the art for a
better understanding of this kind of event.
Author Contributions: Conceptualization, B.A.; methodology, B.A., R.C., E.E., and A.R.; software, J.B. and A.R.;
validation, B.A., R.C., and F.G.; formal analysis, B.A., A.A., C.L., R.L.R., Á.L.P., V.M., L.P., D.P.-Q., and R.S.; data
curation, J.B. and A.R.; writing—original draft preparation, B.A., A.A., C.L., R.L.-R., Á.L.-P., V.M., L.P., D.P.-Q.;
writing—review and editing, B.A., R.C., E.E., F.G., A.R., and R.S.; supervision, B.A., R.C., E.E., and A.R. All
authors have read and agreed to the published version of the manuscript.
Funding: The infrastructure of the Traffic and Dynamic Modeling Laboratory of the University of Cantabria has
been partially funded by FEDER funds (Ref. No: UCAN10-4E-549). The image recognition methodology applied
in this research has been developed thanks to financing from the Spanish Ministerio de Economía, Industria y
Competitividad from the project referenced TRA2017-85853-C2-1-R.
Acknowledgments: The authors would like to thank Santander City Council for providing the necessary data
for this article. Part of the analysis presented in this work has been carried out by a group of civil engineering
students (all of them listed as co-authors of this article) at the University of Cantabria (Spain) as a practical case
study in the subject of transport engineering, which is currently being taught online due to the restrictions
imposed on 15 March in Spain.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Corman, V.M.; Landt, O.; Kaiser, M.; Molenkamp, R.; Meijer, A.; Chu, D.K.; Mulders, D.G. Detection of
2019 novel coronavirus (2019-nCoV) by real-time RT-PCR. Euro Surveill. 2020, 25, 3; doi:10.2807/1560-
7917.ES.2020.25.3.2000045.
2. Li, L.; Yang, Z.; Dang, Z.; Meng, C.; Huang, J.; Meng, H.; Shao, Y. Propagation analysis and prediction of
the COVID-19. Infect. Dis. Model. 2020, 5, 282–292; doi:10.1016/j.idm.2020.03.002.
3. Mizumoto, K.; Chowell, G. Transmission potential of the novel coronavirus (COVID-19) onboard the
diamond Princess Cruises Ship. Infect. Dis. Model. 2020, 5, 264–270; doi:10.1016/j.idm.2020.02.003.
4. Allam, Z.; Jones, D.S. On the coronavirus (COVID-19) outbreak and the smart city network: Universal data
sharing standards coupled with artificial intelligence (AI) to benefit urban health monitoring and
management. In Healthcare; Multidisciplinary Digital Publishing Institute: Basel, Switzerland, 2020;
Volume 8; doi:10.3390/healthcare8010046.
5. Zhou, C.; Su, F.; Pei, T.; Zhang, A.; Du, Y.; Luo, B.; Song, C. COVID-19: Challenges to GIS with Big Data.
Geogr. Sustain. 2020, 1, 77–87; doi:10.1016/j.geosus.2020.03.005.
6. Anzai, A.; Kobayashi, T.; Linton, N.M.; Kinoshita, R.; Hayashi, K.; Suzuki, A.; Nishiura, H. Assessing the
Impact of Reduced Travel on Exportation Dynamics of Novel Coronavirus Infection (COVID-19). J. Clin.
Med. 2020, 9, 601; doi:10.3390/jcm9020601.
7. Chinazzi, M.; Davis, J.T.; Ajelli, M.; Gioannini, C.; Litvinova, M.; Merler, S.; Viboud, C. The effect of travel
restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak. Science 2020, 368, 395–400;
doi:10.1126/science.aba9757.
8. Fan, C.; Cai, T.; Gai, Z.; Wu, Y. The relationship between the migrant population’s migration network and
the risk of COVID-19 transmission in China—Empirical analysis and prediction in prefecture-level cities.
Int. J. Environ. Res. Public Health 2020, 17, 2630; doi:10.3390/ijerph17082630.
9. Kraemer, M.U.; Yang, C.H.; Gutierrez, B.; Wu, C.H.; Klein, B.; Pigott, D.M.; Brownstein, J.S. The effect of
human mobility and control measures on the COVID-19 epidemic in China. Science 2020, 368, 493–497;
doi:10.1126/science.abb4218.
10. Tian, H.; Liu, Y.; Li, Y.; Wu, C.H.; Chen, B.; Kraemer, M.U.; Wang, B. The impact of transmission control
measures during the first 1 50 days of the COVID-19 epidemic in China. MedRxiv 2020;
doi:10.1101/2020.01.30.20019844.
11. Ministerio de Transporte, Movilidad y Agenda Urbana. Gobierno de España. Análisis de la movilidad en
España durante el Estado de Alarma. Available online: https://www.mitma.gob.es/ministerio/covid-
19/evolucion-movilidad-big-data (accessed on 27April 2020).
Sustainability 2020, 12, 3870 17 of 19
12. MapBox. Where and When Local Travel Decreased from COVID-19 around the World. Available online:
https://blog.mapbox.com/movement-changes-around-the-world-from-covid-19-
cc79db7e04c7?gi=c2399c995c69 (accessed on 27April 2020).
13. Ivanov, D. Predicting the impacts of epidemic outbreaks on global supply chains: A simulation-based
analysis on the coronavirus outbreak (COVID-19/SARS-CoV-2) case. Transp. Res. Part E 2020, 136, 101922;
doi:10.1016/j.tre.2020.101922.
14. INRIX. Understanding the Impact of COVID-19 on Commuting and Retail: An Analysis of the U.S.
Epicenter 2020. Available online: https://inrix.com/campaigns/coronavirus-impact-study/ (accessed on 27
April 2020).
15. Google. COVID-19 Community Mobility Reports. 2020. Available online:
https://www.google.com/covid19/mobility/(accessed on 27April 2020).
16. Pirouz, B.; Shaffiee, S.; Piro, P. Investigating a serious challenge in the sustainable development process:
Analysis of confirmed cases of COVID-19 (New Type of Coronavirus) through a binary classification using
artificial intelligence andregression analysis. Sustainability 2020, 12, 2427; doi:10.3390/su12062427.
17. La, V.P.; Pham, T.H.; Ho, M.T.; Nguyen, M.H.; P Nguyen, K.L.; Vuong, T.T.; Vuong, Q.H. Policy response,
social media and science journalism for the sustainability of the public health system amid the COVID-19
outbreak: The Vietnam lessons. Sustainability 2020, 12, 2931; doi:10.3390/su12072931.
18. TomTom. TomTom Traffic Index 2020. Available online: https://www.tomtom.com/en_gb/traffic-
index/(accessed on 27 April 2020).
19. TUMI. 2020. Available online: https://www.transformative-mobility.org/news/the-covid-19-outbreak-and-
implications-to-public-transport-some-observations(accessed on 27 April 2020).
20. Voz de América. Bogotá, Desolada en el Primer día de Simulacro de Aislamiento por el COVID-19.
Available online: https://www.voanoticias.com/a/bogota-desolada-primer-dia-simulacro-de-aislamiento-
covid-19/5338003.html (accessed on 27 April 2020).
21. UITP. Management of COVID-19 Guidelines for Public Transport Operators. 2020. Available online:
https://www.uitp.org/sites/default/files/cck-focus-papers-files/Corona%20Virus_EN.pdf (accessed on 27
April 2020).
22. National Academies of Sciences, Engineering and Medicine. A Guide for Public Transportation Pandemic
Planning and Response; The National Academies Press: Washington, DC, 2014; doi:10.17226/22414.
23. APTA. Passenger Transport. 2020, 78, 7. https://www.nxtbook.com/ygsreprints/APTA/PT_04-06-
20_allpgstoYGS_Update2/index.php#/1 (accessed on 27 April 2020)
24. González, D.; Ochoa, M.A. La Salud del Transporte Público Español. Think Tank Movilidad. 2020.
Available online: https://www.fundacioncorell.es/la-salud-del-transporte-publico-espanol/ (accessed on 27
April 2020).
25. Moovit. Impact of Coronavirus (COVID-19) on Public Transit Usage. 2020. Available online:
https://moovitapp.com/insights/en/Moovit_Insights_Public_Transit_Index-countries (accessed on 27 April
2020).
26. Kucharski, R.; Cats, O. When Sharing is not Always Caring: On the Spreading Processes in Ride-Sharing
Networks. Available online: https://www.linkedin.com/pulse/when-sharing-always-caring-spreading-
processes-rafa%25C5%2582-kucharski/?trackingId=S28punKTRECg2lbMhrc70g%3D%3D (accessed on 27
April 2020).
27. Anadolu Agency. Traffic Accidents Reduced in Istanbul Amid COVID-19. Available online:
https://www.aa.com.tr/en/latest-on-coronavirus-outbreak/traffic-accidents-reduced-in-istanbul-amid-
covid-19/1782717 (accessed on 27 April 2020).
28. Isalfan, R.J. The dramatic impact of Coronavirus outbreak on air quality: Has it saved as much as it has
killed so far? Glob. J. Environ. Sci. Manag. 2020, 6, 275-288; doi:10.22034/gjesm.2020.03.01.
29. Anjum, N.A. Good in the worst: COVID-19 restrictions and eas e in global air pollution. Preprints 2020, 2020,
040069; doi:10.20944/preprints202004.0069.v1.
30. EPA. Air Data: Air Quality Data Collected at Outdoor Monitors Across the U.S. Available online:
https://www.epa.gov/outdoor-air-quality-data (accessed on 27 April 2020).
31. BBC. McGrath M (2020) Coronavirus: Air Pollution and CO2 Fall Rapidly as Virus Spreads; BBC: London, UK,
2020; Available online: https://www.bbc.com/news/science-environment-51944780 (accessed on 27 April
2020).
Sustainability 2020, 12, 3870 18 of 19
32. Wang, P.; Chen, K.; Zhu, S.; Zhang, H. Severe air pollution events not avoided by reduced anthropogenic
activities during COVID-19 outbreak. Resour. Conserv. Recycl. 2020, 158, 104814;
doi:10.1016/j.resconrec.2020.104814.
33. González, M.; Hidalgo, C.; Barabási, A. Understanding individual human mobility patterns. Nature2008.
453, 779–782; doi:10.1038/nature06958.
34. Kaltenbrunner, A.; Meza, R.; Grivolla, J.; Codina, J.; Branchs, R. Urban cycles and mobility patterns:
Exploring and predicting trends in a bicycle-based public transport system. Pervasive Mob. Comput. 2010, 6,
455-466; doi:10.1016/j.pmcj.2010.07.002.
35. Ministerio de Transportes, Movilidad y Agenda Urbana. Evolución del Tráfico Desde el Año 2000.
Gobierno de España. Available online: https://www.mitma.gob.es/carreteras/trafico-velocidades-y-
accidentes-mapa-estimacion-y-evolucion/evolucion-del-trafico-desde-el-año-2000/evolucion-del-trafico-
desde-el-año-2000-rce-y-todas-las-redes (accessed on 27 April 2020).
36. Macioszek, E. Changes in values of traffic volume—Case study based on general traffic measurements in
Opolskie Voivodeship (Poland). In Directions of Development of Transport Networks and Traffic Engineering.
Lecture Notes in Networks and Systems 51; Springer International Publishing: Cham, Switzerland, 2019;
p. 66–76.
37. Hardt, C.; Bogenberger, K. Usage of e-scooters in urban environments. Transp. Res. Procedia 2019, 37, 155–
162; doi:10.1016/j.trpro.2018.12.178.
38. Geroliminis, N.; Daganzo, C.F. Existence of Urban-scale macroscopic fundamental diagrams: Some
experimental findings. Transp. Res. Part B 2008, 42, 9, 759–770; doi:10.1016/j.trb.2008.02.002.
39. Alonso, B.; Ibeas, A.; Musolino, G.; Rindone, C.; Vitetta, A. Effects of traffic control regulation on Network
Macroscopic Fundamental Diagram: A statistical analysis of real data. Transp. Res. Part A 2019, 126, 136–
151; doi:10.1016/j.tra.2019.05.012.
40. Dell’Olio, L.; Ibeas, A.; Ruisanchez, F. Optimizing bus-size and headway in transit networks. Transportation
2012, 39, 449–464; doi:10.1007/s11116-011-9332-2.
41. Ibeas, A.; Alonso, B.; dell’Olio, L.; Moura, J.L. Bus size and headways optimization model considering
elastic demand. J. Transp. Eng. 2013, 140, 04013021; doi:10.1061/(ASCE)TE.1943-5436.0000641.
42. Ecologistas en Acción. Efectos de la Crisis de la COVID-19 en la Calidad del Aire Urbano en España. 2020.
Available online: https://www.ecologistasenaccion.org/wp-content/uploads/2020/04/informe-calidad-aire-
covid-19.pdf (accessed on 27 April 2020).
43. Kurtenbach, R.; Kleffmann, J.; Niedojadlo, A.; Wiesen, P. Primary NO2 emissions and their impact on air
quality in traffic environments in Germany. Environ. Sci. Eur. 2012, 24, 21.
44. Beevers, S.D.; Westmoreland, E.; de Jong, M.C.; Williams, M.L.; Carslaw, D.C. Trends in NOx and NO2
emissions from road traffic in Great Britain. Atmos. Environ. 2012, 54, 107–116;
doi:10.1016/j.atmosenv.2012.02.028.
45. “Consider Bicycles” during Covid-19 Outbreak, Says World Health Organisation. Available online:
https://cyclingindustry.news/consider-bicycles-during-covid-19-outbreak-says-world-health-
organisation/ (accessed on 27 April 2020).
46. Barcelona Ampliará Aceras y Carriles Bici Para Minimizar Los Contagios. Available online:
https://www.elperiodico.com/es/barcelona/20200425/barcelona-ampliara-aceras-y-carriles-bici-para-
minimizar-los-contagios-covid-7940149 (accessed on 27 April 2020).
47. Milan Announces Ambitious Scheme to Reduce Car Use after lockdown. Available online:
https://www.theguardian.com/world/2020/apr/21/milan-seeks-to-prevent-post-crisis-return-of-traffic-
pollution (accessed on 27 April 2020).
48. Sydney Traffic Lights Automated to Combat COVID-19. Available online:
https://www.governmentnews.com.au/sydney-traffic-lights-automated-to-combat-covid-19/ (accessed on
28 April 2020).
49. COVID-19: Additional Safe Distancing Measures Implemented on Public Transport. Available online:
https://Www.Channelnewsasia.Com/News/Singapore/Lta-Safe-Distancing-Covid-19-Public-Transport-
12623902 (accessed on 27 April 2020).
50. UITP Webinar: ¿Cómo Será el Día Después y Cómo Prepararnos Desde el Sector del Transporte Público?
Available online: https://www.uitp.org/events/webinar-¿cómo-será-el-d%C3%ADa-después-y-cómo-
prepararnos-desde-el-sector-del-transporte-público (accessedon 28 April 2020).
Sustainability 2020, 12, 3870 19 of 19
51. Kostrzewski, M. Sensitivity analysis of selected parameters in the order picking process simulation model,
with randomly generated orders. Entropy 2020, 22, 1021; doi:10.3390/e22040423.
52. Czwajda, L.; Kosacka-Olejnik, M.; Kudelska, I.; Kostrzewski, M.; Sethanan, K.; Pitakaso, R. Application of
prediction markets phenomenon as decision support instrument in vehicle recycling sector. LogForum 2019,
15, 265–278; doi:10.17270/J.LOG.2019.329.
53. Snowberg, E.; Wolfers, J.; Zitzewitz, E. Prediction markets for economic forecasting. In Handbook of Economic
Forecasting; NBER: Cambridge, MA, USA, 2012; p. 18222.
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
