Sustainability 12 03870 v2 PDF

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 NO2 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%, NO2 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 article,
Sustainability 2020, 12, 3870; doi:10.3390/su12093870 www.mdpi.com/journal/sustainability

Sustainability 2020, 12, 3870 2 of 18
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.
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.
Sustainability 2020, 12, x FOR PEER REVIEW 3 of 19
22 apr. 2020
r. r. r. r. r. r. r.
ap ap ap ap ap ap ap
Figure1.1.Evolution
Figure Evolution of
of public
public transport
transport users
users since
since 15
15 January
January (source:
(source:www.moovit.com).
www.moovit.com).
InThis
the drop
case ofinSpain,
usagethehassimultaneous
been causedclosure
by both the reticence
of centers of the and
of education users themselves
restrictions onand the
mobility
recommendations of health authorities to avoid using public transport as much
were introduced on Sunday, 15 March 2020 and extended 15 days later, limiting all travel to only those as possible,
recommending
journeys considered thetouse of individual
be essential. means
This date of transport
is clearly such
reflected in as 2,
Figure bicycles. In spite
representing of such
the evolution
recommendations, research has also been found highlighting the risk of contagion
of public transport use for various Spanish cities. A strong decline in the number of users began associated withon
the use of shared bicycle systems [26].
the 15th and stabilized approximately one week later. The drop in user numbers is similar for all the
cities, These reductions
at around 80–90%.inAn travel haveofhad
analysis thesimilar
Googlerepercussions
report [15] foron the once
Spain mainagain
externalities normally
found falls of 94%
associated with motorized transport modes, mainly journey times, accidents, and pollution.
in journeys made for reasons of leisure or shopping, 77% in journeys for food or trips to pharmacies, Data has
been published in different media revealing reductions of 35% in accident rates in
89% in mobility around stations and transport nodes, and 68% towards work places. These reductions Istanbul (Turkey),
where
have thefound
been only action taken duringfor
to be homogenous theall
period of study
the Spanish was toincluding
regions, close the Cantabria,
schools [27]. Reductions
where in
Santander
emissions
(the haveofalso
case study thisbeen reported
work) from different countries around the world, especially significant
is located.
in the cases of NOx, CO2, and PM2.5 [28–30]. In China, for example, the measures to minimize the
The reported data proves that overall mobility in many countries of the world has fallen as
spread of SARS-CoV-2 have resulted in reductions of 25% in CO2 emissions, and NO2 levels were
a consequence of the sudden stop to human activities due to lockdown (quarantine) policies. This has led
36% lower than in the same period in 2019. In Italy, significant reductions in NO2 concentrations have
to the extreme breakdown of mobility patterns [33,34] and halted the current trend towards sustainable
been found to be mainly due to the reduction in the use of diesel vehicles for transport. In France,
transport in cities. In spite of increasing motorized traffic as the economic crisis was being overcome,
minimum NOx levels have been reached due to the restrictions adopted to fight COVID-19 in
especially on regional networks [35,36], the use of personal transport systems (both shared and private)
economic activities and transport. During the spread of the COVID-19 pandemic in New York, traffic
such as bicycles or scooters was exponentially growing in urban areas [37].
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
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.
22 apr. 2020
r. r. r. r. r. r. r.
ap ap ap ap ap ap ap
Figure2.2.Evolution
Figure Evolution of
of public
public transport
transport users
users in
in Spain
Spainsince
since15
15January
January(source:
(source:www.moovit.com).
www.moovit.com).
2. CaseTheStudy Analysis
reported data proves that overall mobility in many countries of the world has fallen as a
consequence of the
The particular casesudden
beingstop to human
studied in this activities duecity
article is the to lockdown (quarantine)
of Santander, located onpolicies.
the NorthThis has
Spanish
led to the extreme breakdown of mobility patterns [33,34] and halted the current trend towards
coast. Santander is the capital of the autonomous region of Cantabria. With nearly 180,000 inhabitants
sustainable transport in cities. In spite of increasing motorized traffic as the economic crisis was being
according to the National Institute of Statistics, trade and services provide employment for more than
overcome, especially on regional networks [35,36], the use of personal transport systems (both shared
70 percent of its active population.
and private) such as bicycles or scooters was exponentially growing in urban areas [37].
According to the 2013 mobility survey updated with traffic transit and pedestrian counts in 2018,
The authors, therefore, decided to report on a more detailed and integrated analysis of the
the aggregated picture of mobility before (including commuting) the quarantine can be seen in Figure 3,
mobility changes that have occurred within a city. This study covers various modes of transport and
where 42% of urban mobility is done by walking, 48% by private motorized transport (driving or
their effect on other externalities, and this research will provide data which can be used to make more
passenger), and 8% by bus. The remaining 2% is done by bike/scooter. It should be highlighted that
thorough detailed comparisons with other case studies.
public transport is more often used for study and health related trips. The daily trips profile was
estimated usingAnalysis
2. Case Study this data.
Figure 3 also shows three peak periods in a working day as many people return home for lunch
The particular case being studied in this article is the city of Santander, located on the North
due to the Spanish working day. These peak hours and patterns match for all modes throughout the
Spanish coast. Santander is the capital of the autonomous region of Cantabria. With nearly 180,000
day, with the exception of the afternoon peak hour, which is slightly longer for the private vehicle mode.
inhabitants according to the National Institute of Statistics, trade and services provide employment
The new urban mobility patterns resulting from the quarantine are calculated from different
for more than 70 percent of its active population.
data obtained by the authors thanks to an agreement with the city hall, which provides real time
According to the 2013 mobility survey updated with traffic transit and pedestrian counts in 2018,
information to a data processing laboratory located at the University of Cantabria. A total of 480
the aggregated picture of mobility before (including commuting) the quarantine can be seen in Figure
electromagnetic traffic counters reported values for intensity and occupation every minute. Public
3, where 42% of urban mobility is done by walking, 48% by private motorized transport (driving or
transport GPS positioning data were analyzed to provide speed and number of stops from the onboard
passenger), and 8% by bus. The remaining 2% is done by bike/scooter. It should be highlighted that
Intelligent Transport System (ITS), ticketing data provided the number of users at any given moment,
public transport is more often used for study and health related trips. The daily trips profile was
atestimated
what stop, line,this
using and vehicle. Finally, 45 traffic control cameras were analyzed, and using image
data.
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.
Figure3.3.Basic
Figure Basicmobility
mobility(including
(including commuting)
commuting) data
data before
beforethe
thequarantine
quarantine(source:
(source:Mobility
Mobilitysurvey,
survey,
2013.Universidad
2013. UniversidaddedeCantabria).
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
Figure
of aFigure 4. 4.Images
residential Images
zone, taken fromtraffic
showing
taken from traffic
the control
queue cameras.
ofcameras.
control In
In aa residential
people waiting zone
to enter
residential (left)
(left)and
andthe
a supermarket,
zone themain
and
mainstreet
on ofof
the
street right
is thethe
themaincity center (right).
street(right).
city center 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
The analysis started with general
Finally,
Department data wasfrom
also fixed sensors
analyzed along withmotorized
installed by thefrom
data traffic,
Regional
mobile where a strong
Government’s decline was
Environmental
sensors installed on offound
Department
the roofs local
wasthroughout
also analyzedthe city. As awith
along reference,
data before
from confinement,
mobile sensors the main entrance
installed on the to theofcity
roofs registered
local buses. about
buses.
42,000
The vehicles
analysison a working
started day. After
with general confinement
motorized traffic,was
wheredeclared, thedecline
a strong traffic reduced
was found by throughout
64%, and
during week 3, the reduction rose to 78%. In order to obtain an overall result, a Network
the city. As a reference, before confinement, the main entrance to the city registered about 42,000 vehicles Macroscopic
Fundamental Diagram—NMFD [38] was estimated for all the Mondays in the month of March. Days
on a working day. After confinement was declared, the traffic reduced by 64%, and during week 3,
2 and 7 of March can be considered as normal, days 16 and 23 enter the first period of quarantine,
the reduction rose to 78%. In order to obtain an overall result, a Network Macroscopic Fundamental
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
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
Diagram—NMFD [38] was
[38] was estimated forestimated for all the
all the Mondays inMondays
the month in of
theMarch.
month of March.
Days 2 and Days
7 of
2 and 7 of March can be considered as normal, days 16 and 23 enter the first
March can be considered as normal, days 16 and 23 enter the first period of quarantine, being the period of quarantine,
being
first andthe first and
second secondMonday
working workingofMonday of the quarantine
the quarantine (days 2 and (days
9), 2respectively.
and 9), respectively.
The 30thThe 30th
of March
of March (Monday) enters the second period of quarantine (16th day of
(Monday) enters the second period of quarantine (16th day of quarantine), where the restrictions quarantine), where the
restrictions on mobility are even more severe (and still are, at the date of writing
on mobility are even more severe (and still are, at the date of writing this article), trips being this article), trips
being limited to essential activities only and the purchase of basic necessities. The data was
limited to essential activities only and the purchase of basic necessities. The data was aggregated into
aggregated into periods of 5 min. Figure 5 shows how the normal days before quarantine show
periods of 5 min. Figure 5 shows how the normal days before quarantine show practically identical
practically identical behavior, with average maximum intensities of around 600 veh/h and always
behavior, with average maximum intensities of around 600 veh/h and always within the stable range,
within the stable range, never reaching capacity. Flow and occupancy drastically reduce during the
never reaching capacity. Flow and occupancy drastically reduce during the quarantine, with an even
quarantine, with an even greater reduction (over 65%) on the 30th of March. The figure on the right
greater reduction (over 65%) on the 30th of March. The figure on the right shows the same behavior
shows the same behavior throughout the day, clearly showing that during confinement, morning
throughout the day, clearly showing that during confinement, morning mobility is greater than during
mobility is greater than during the afternoon, where afternoon rush hour has all but disappeared,
the afternoon, where afternoon rush hour has all but disappeared, and morning rush hour smoothed
and morning rush hour smoothed its peak after the restrictions were hardened as fewer people were
its peak after the restrictions were hardened as fewer people were allowed to travel for reasons of work.
allowed to travel for reasons of work. An increase in the average flow during the night period
An increase1:00
(between in the
a.m.average
and 5:00flow
a.m.)during the
has also night
been period
noted, but(between
no further1:00
dataa.m. and 5:00 at
are available a.m.) has also
this time to
been noted, but
explain this fact. no further data are available at this time to explain this fact.
Network Macroscopic Fundamental Diagram

700
Average Flow (veh/h)
600
500
400
300
200
100
0
0 10 20 30 40 50
Average Occupancy (%)
A similar analysis can be performed by disaggregating the city into corridors [39]. Figure 6
Monday 02/03/20 Monday 09/03/20 Monday 16/03/20 Monday 23/03/20
shows the
Monday corridor
30/03/20 with the
Monday 06/04/20 most
Monday 20/04/20traffic (Marqués de la Hermida street) which is also the main
entrance route to Santander from the East (direction Bilbao).

Figure
The
Figure 5. 5.
resultingNetwork
diagrams
Network Macroscopic
(FigureFundamental
Macroscopic 6 left) follow
Fundamental Diagram (NMFD)
the same
Diagram andand
pattern
(NMFD) average
of flowas
behavior
average daily profile
thedaily
flow of
case explained
profile
above, Santander.
although with higher intensity and occupancy values. The same occurs with the traffic profiles
of Santander.
(Figure 6 right), where the morning peak is quite striking as it is a Monday and the main access point
from A the
similar analysis can
surrounding be performed
catchment area. Thisby disaggregating
peak continues the city into
to appear corridors
during [39]. Figure
quarantine, but is6much
shows
the corridor with the most traffic (Marqués de la Hermida street) which is also
less striking with lower traffic intensity. As before, once morning rush hour has passed, the amount the main entrance route
toofSantander from the East (direction Bilbao).
traffic drops and afternoon rush hour disappears completely.
Figure 6. 6.
Figure NMFD (left)
NMFD and
(left) average
and flow
average daily
flow profile
daily (right)
profile of the
(right) main
of the corridor
main of Santander.
corridor of Santander.
Theresulting
The drop in mobility
diagramswas much6 steeper
(Figure for the
left) follow thecity’s
samepublic transport
pattern system.
of behavior Thecase
as the dramatic fall
explained
can bealthough
above, seen in Figure 7 (left),
with higher with anand
intensity average of overvalues.
occupancy 90%, the morning
The peak disappearing,
same occurs with the trafficand the
profiles
mid-day peak being slightly maintained. As happened with general traffic, a 35% decrease
(Figure 6 right), where the morning peak is quite striking as it is a Monday and the main access point in users
appears
from in the afternoon
the surrounding period compared
catchment area. This with
peakthe numbers
continues toregistered in thequarantine,
appear during morning. If but
the is
data is
much
analyzed
less strikingby buslower
with line (Figure
traffic 7intensity.
right), lines 1 and 2 once
As before, carrymorning
the mostrush
passenger
hour hasloads, and their
passed, demandof
the amount
dropped
traffic dropsbyand
more than an rush
afternoon averagehourofdisappears
85%. The lines servicing the university campus (among other
completely.
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 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
Sustainability in12,
2020, Figure7 (left), with an average of over 90%, the morning peak disappearing, and7the
3870 of 18
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
The drop
analyzed in mobility
by bus was7much
line (Figure right),steeper for the
lines 1 and city’s
2 carry thepublic
most transport
passengersystem. Thetheir
loads, and dramatic
demand fall
can be seen in (Figure 7 left), with an average of over 90%, the morning peak disappearing,
dropped by more than an average of 85%. The lines servicing the university campus (among other and the
mid-day peak being
zones) suffered slightlyreduction
an average maintained. As and
of 92% happened with general
those serving the citytraffic, a 35%
periphery decrease
reduced by in users
88%.
appearsThe much lower demand, together with the reduction in general traffic and a lack of congestion,is
in the afternoon period compared with the numbers registered in the morning. If the data
analyzed
has meantby the
busjourney
line (Figure
times7of right), lineshave
the lines 1 and 2 carry
been the reduced.
notably most passenger loads, and
As an example, their8demand
Figure shows
dropped by more than an average of 85%. The lines servicing the university campus
the journey times by time of day for two lines of high and low demand (lines 1 and 6c2 respectively). (among other
zones) suffered an average reduction of 92%
Both cases registered 30% reductions in cycle times. and those serving the city periphery reduced by 88%.
Figure7.7.Public
Figure Publictransport
transporttrips
trips throughout
throughout a working day
day (left)
(left) and
anddaily
dailytrips
tripsper
perline
line(right).
(right).
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
Sustainability 30%
2020, 12, x FOR reductions
PEER REVIEW in cycle times. 8 of 19
Line 1 Travel times and demand profiles Line 6c2 Travel times and demand profiles
0:57:36 800 0:57:36 800
0:50:24 700 0:50:24 700
0:43:12 600 0:43:12 600
Travel time
Travel time
0:36:00 500 0:36:00 500

Trips
Trips
0:28:48 400 0:28:48 400
0:21:36 300 0:21:36 300
0:14:24 200 0:14:24 200
0:07:12 100 0:07:12 100
0:00:00 0 0:00:00 0
00
00
10 0
11 0
12 0
13 0
14 0
15 0
16 0
17 0
18 0
19 0
20 0
21 0
0
00
10 0
11 0
12 0
13 0
14 0
15 0
16 0
17 0
18 0
19 0
0
0
:0
:0
:0
:0
:0
:0
:0
:0
:0
:0
:0
:0
0
:0
:0
:0
:0
:0
:0
:0
:0
:0
:0
7:
8:
9:
8:
9:
Travel time before Travel time after Trips before Trips after Travel time before Travel time after Trips before Trips after
Figure8.8.Cycle
Figure Cycle times
times throughout
throughout the
the day
day of
of two
twopublic
publictransport
transportlines.
lines.
Figure99shows
Figure shows the
the evolution
evolution of of the
the daily
dailydemand
demandand andthethetotal running
total running length
lengthperper
dayday
as aas
a performance
performance indicator
indicator ofofthe
theservice
serviceprovided.
provided.As Ascan
canbebeseen,
seen,the
thedemand
demandstarted
starteddropping
droppingslightly
slightly
atatthe
theend
endofofthe
theweek
weekbefore
beforequarantine.
quarantine. However,
However, during
during the
the first
first working
workingday dayofofthe
thelockdown
lockdown
period,the
period, the number
number of users
of users fell fell to levels
to levels that were
that were even lower
even lower than reported
than those those reported on a Saturday.
on a normal normal
Saturday. These results for the public transport system are in line with those already
These results for the public transport system are in line with those already reported in Section 1 and, reported in
assection
can be1 deduced,
and, as canthey
be deduced,
represent they represent
a strong blowa strong
to theblow to the budgetary
budgetary balance ofbalance of the since
the system, system,the
since the operational cost is now more disproportionate to the new scenario
operational cost is now more disproportionate to the new scenario of reduced demand. The public of reduced demand. The
public transport
transport operatoroperator
kept thekept
usualtheworking
usual working day services
day services during during one week,
one week, but thebutservice
the service
had had
to be
to be adapted to readjust the line intervals to this new
adapted to readjust the line intervals to this new demand (Figure 9).demand (Figure 9).
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.
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
Sustainability 2020, 12,operator
3870 kept the usual working day services during one week, but the service8 of
had18
to be adapted to readjust the line intervals to this new demand (Figure 9).
Figure 9.
Figure 9. Evolution
Evolution of
of public
public transport
transport demand
demand and
and total
total running
running length
length serviced per day.
serviced per day.
A decision support tool1.isHeadway

Table needed tocomparison of the
optimize the publicoftransport
design the newsystem.
services. Applying the social
optimum results of the frequency optimization model for this
Headways (min)same network [40,41] provides a rapid
Line
response for the operator to be able to readapt the service, as well as taking into account the new
Before Optimun (Quarantine) Current
cycle times recorded. Table 1 shows a precise comparison between the situation prior to confinement
with the calculated1 optimal intervals
20 30
and the current situation of the service33during confinement,
2 20 30 33
where a great similarity
3 can be seen.20 35 42
4 15 35 33
5c1 Table 1. Headway
12 comparison of the 24
public transport system.20
5c2 12 24 20
6c1 30 Headways 60 (min) 60
Line
6c2 Before
30 Optimun (Quarantine)
60 Current
60
7c1 1 20
20 60
30 3355
7c2 20 60
2 20 30 3355
11 30 60 60
12 3 20
30 35
40 4245
13 4 15
30 35
60 3355
145c1 12
22 24
60 2055
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 € 2663.00 € -€
−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).
Figure10.
Figure 10.Traffic
Trafficand
andpedestrian
pedestrian flow
flow comparison
comparison before
before (left)
(left) and
and during
during(right)
(right)quarantine.
quarantine.
Alldifferences
The the updated information
are was introduced
revealing about into
the resulting the city
drop mobilityTo
in mobility. model, the the
estimate O-Ddaily
matrices were
trip profile
re-estimated, and the data was expanded to include night time. This process provides
for pedestrians and bikes, the authors correlated the pedestrian and bikes flows obtained from 9 the new overall
modal distribution
different for thewith
camera recordings entire
theday and for
existing separateand
pedestrian periods
bikesduring the day as from
flow assignments well theas the trip
mobility
model that the University of Cantabria had calibrated for the city and the profile obtained fromcar
profiles shown in Figure 11. As can be seen, the new modal distribution changes significantly, as the
journeys
latest change from 48%
Origin-Destination to 77%
(O-D) (it isupdated
survey, importantwith tomanual
clarify that theup
counts number of total
to 2018. car journeys
The same is
correlation
much lower during the quarantine), and public transport from 8% to 2%. Pedestrian
has been applied using the new flows obtained from the images taken during the quarantine period, journeys also
show
once a significant
again dropthe
re-estimating in their share from
trip matrices and42% to 19%.
finding the It wastravel
daily estimated that overall mobility has
profile.
fallen by 76%, with variations throughout the day, reaching falls of up to 85% at certain times. Car
The resulting reduction coincided with that observed for general traffic and public transport:
travel fell by 68% (in some periods by up to 85%) and bus travel fell by 93%. These results are
The morning and evening peaks disappeared drastically with the appearance of a mid-day peak period
consistent with the data provided by Moovit [23] and the Spanish Ministry of Mobility [11]. Figure 3
between 11:00 and 13:00 when many people leave their homes to do basic shopping. Mobility during
shows that the proportion of trips using bicycles or scooters (labeled as “others”) was very low before
the afternoon period falls by approximately 55% compared with the morning, where for a normal day
the quarantine. As happened with other modes of transport, overall mobility using these means of
the drop is slightly lower than was found for motorized vehicles, not reaching 70%.
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
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.
Figure11.
Figure 11.Mode
Modeshare,
share,trip
tripreason,
reason, and
and total
total trips
trips daily
daily profile
profilebefore
beforeand
andduring
duringthe
thequarantine.
quarantine.
Thelockdown
The lockdownmeasures
measurescan canalso
also cause
cause changes
changes to
to proximity
proximitymobility
mobilityduring
duringthe theSARS-CoV-2
SARS-CoV-2
pandemic.This
pandemic. Thiscan
canbebeseen
seenininchanges
changesininthe
thelength
lengthofofthe
thetrips
tripsbeing
beingmade
madebybycitizens
citizens(see
(seeFigure
Figure12)
as12) as reported
reported by thebySpanish
the Spanish Ministry
Ministry of Transport
of Transport basedbased on mobile
on mobile phone
phone data
data [11].
[11]. The
The lockdown
lockdown has
has caused the shortest trips (0.5–2 km) to clearly gain weight over long-distance trips
caused the shortest trips (0.5–2 km) to clearly gain weight over long-distance trips (more than 5 km). (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.
Trip length distribution before and during the quarantine

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
Sustainability by3870
2020, 12, the Spanish Ministry of Transport based on mobile phone data [11]. The lockdown 11 of 18
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
Figure
for work 12 (mid-long
also showsdistance
how themainly)
weight decreases
of shorter and
tripsmost
increases at weekends,
of these because
trips are made for travelling for
buying basic
work (mid-long
products. distance mainly) decreases and most of these trips are made for buying basic products.
Trip length distribution before and during the quarantine

70
60
50
40
%
30
20
10
0
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
0.5-2 Km 2-5 Km 5-10 Km
Figure 12.12.
Figure Trip distribution
Trip length length distribution before
before and during and during
the quarantine (source:the quarantine (source:
https://www.mitma.gob.es).
https://www.mitma.gob.es).
As explained beforehand, this strong decline in motorized mobility has resulted in a reduction in
As explained beforehand, this strong decline in motorized mobility has resulted in a reduction
externalities such as emissions. In terms of emissions, the analysis has been focused on NO2 , as it is the
in externalities such as emissions. In terms of emissions, the analysis has been focused on NO2, as it
substance most directly related to urban traffic, as opposed to other pollutants coming from alternative
is the substance most directly related to urban traffic, as opposed to other pollutants coming from
sources such as PM and PM2.5 [42–44]. The analysis of the NO2 values from the static Department of
alternative sources 10such as PM10 and PM2.5 [42–44]. The analysis of the NO2 values from the static
the Environment stations located in the city center and in an industrial zone located in a nearby town
Department of the Environment stations located in the city center and in an industrial zone located
shows how the normal peaks that used to be registered disappeared completely at the beginning of the
in a nearby town shows how the normal peaks that used to be registered disappeared completely at
quarantine, with reductions of over 50–60% compared with the average values measured in the period
the beginning of the quarantine, with reductions of over 50–60% compared with the average values
2010–2019 (Figure 13).
measured in the period 2010–2019 (Figure 13).
Evolution of NO2 emissions (static devices)

35
NO2 emissions (ug/m3)
30
25
20
15
10
5
0
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
Industrial zone City center 2010-2019 values
Figure
Figure 13.13.
NO2 NO 2 measurements
measurements onenvironmental
on the static the staticstations
environmental stations (source:
(source: https://www.airecantabria.
https://www.airecantabria.com/index.php).
com/index.php).
The
Thevalues
valuesprovided
providedby bythe
thestatic
staticstations
stationscan
canbe
becomplemented
complementedby bydata
datafrom
fromthe
themobile
mobilesensors
sensors
installed
installedon
onthe
thecity
citybuses.
buses.AllAll
thethe
different NONO
different 2 measurements werewere
2 measurements usedused
in order to have
in order more more
to have data
available (Figure
data available 14) and
(Figure alland
14) the all
data from
the datathe month
from the of March
month ofhas beenhas
March aggregated. This has resulted
been aggregated. This has
in an overall reduction in NO2 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 13. NO2 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 NO2 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
resulted in anreduction
in an overall overall reduction in NO2 emissions
in NO2 emissions throughoutthroughout the especially
the city, but city, but especially in theCentral,
in the North, North,
Central,
and Westand Westwhere
zones, zones,a where a large number
large number of work of workcolleges,
places, places, colleges, and university
and university buildings
buildings are
are located.
located. The absence
The absence of data of datathe
from from the northern
northern zone iszone is because
because the buses
the buses equipped
equipped with with sensors
sensors werewerenot
not running
running in that
in that zonezone during
during the analyzed
the analyzed time time period.
period. The results
The results obtained
obtained are consistent
are consistent with
with those
those
foundfound in an independent
in an independent reportreport
mademade by a Spanish
by a Spanish ecologist
ecologist organization
organization [42]. [42].
Figure 14. NO22 measurements

Figure 14. measurements before
before (left) and during (right) quarantine with mobile devices.
Finally, accident data reported by the

Finally,
Sustainability accident
2020, data
12, x FOR reported
PEER REVIEW by thelocal city
local police
city policewaswas analyzed. Data
analyzed. from
Data approximately
from one
approximately
13 of 19
and a half
one and months
a half before
months quarantine
before waswas
quarantine compared
compared withwith
datadata
produced during
produced quarantine:
during From
quarantine: the
From
start of February to the 14th of March, and from the 15th of March to the 23rd of April.
the start of February to the 14th of March, and from the 15th of March to the 23rd of April. The number The number of
accidents
of accidentsin absolute
in absoluteterms has
terms fallen
has from
fallen from the
thestart
startofofquarantine
quarantine(Figure
(Figure15)
15)from
fromaatotal
totalof
of105
105 toto
17 (−83.8%), from 17.5 accidents per week to 2.83. However, if this data is weighted
17 (−83.8%), from 17.5 accidents per week to 2.83. However, if this data is weighted to consider the to consider the
average
average daily
daily traffic
trafficintensity
intensityregistered
registeredininthe
thecity,
city,the
thefall
fallisisaround
around67%.67%.
Figure15.
Figure 15. Number
Number of
of traffic
traffic accidents
accidents per
per week
week (source:
(source: Santander
Santander local
local police).
police).
3.
3. What
What IsIs Next?
Next?
Evidently,
Evidently, urbanurban mobility,
mobility, and
and Santander
Santander isis nono exception,
exception, hashas been
been put
put on
on hold
hold during
during the
the
quarantine
quarantine period. The great unknown being faced by many cities is how the demand for transport
period. The great unknown being faced by many cities is how the demand for transport
and
and mobility
mobility willwill evolve
evolveduring
duringthethefollowing
followingstages
stagesininthe
theprocess
process(the
(therecovery
recoveryperiod).
period).
Many cities are planning measures to coordinate the slow increase
Many cities are planning measures to coordinate the slow increase in mobility in mobility during the recovery
during the
period with the social distancing required to avoid the probability of new contagion
recovery period with the social distancing required to avoid the probability of new contagion and the and the return
of the illness.
return It is a worry
of the illness. that thethat
It is a worry move
the towards sustainable
move towards mobility
sustainable has been
mobility hashalted and we
been halted have
and we
returned to the ratios of motorized transport found over a decade ago. Considering
have returned to the ratios of motorized transport found over a decade ago. Considering the the experience of
China, the trend is to favor trips being made on foot or using individual means
experience of China, the trend is to favor trips being made on foot or using individual means ofof sustainable transport
such as bicycles
sustainable or scooters
transport such as[17]. To support
bicycles this change,
or scooters [17]. Toinvestment
support this haschange,
been made in providing
investment the
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
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.
KEEP
DISTANCE
KEEP KEEP
DISTANCE DISTANCE
Figure 16.
Figure 16. Pedestrian
Pedestrian crossing
crossing proposed in Santander
proposed in Santander (source:
(source: AC
AC Proyectos:
Proyectos: www.acproyectos.es).
www.acproyectos.es).
Nevertheless,
Nevertheless,for forallall
thetheeffort that that
effort is usedis introducing these measures,
used introducing their success
these measures, theirwill inevitably
success will
depend
inevitably on depend
the behavior
on the ofbehavior
public transport.
of publicThe effect will
transport. Thebeeffect
clearerwillinbe
those cities
clearer inwith
thoselarge
citiespublic
with
transport
large public systems (metros,
transport systems BRTs, etc.) where
(metros, BRTs, the etc.)transport
where thecapacity
transport will be reduced
capacity will be (and is already
reduced (and
being seenbeing
is already in cities
seenundergoing a period ofa recuperation)
in cities undergoing due to social
period of recuperation) duedistancing measuresmeasures
to social distancing between
passengers inside theinside
between passengers vehicles [49] and[49]
the vehicles the and
increased frequencies
the increased required
frequencies to reduce
required waiting
to reduce times
waiting
and
timespassenger accumulation
and passenger at stops at
accumulation andstops
stations.
and All these factors
stations. All theseinevitably
factors mean more vehicles
inevitably mean more are
required,
vehicles areandrequired,
more drivers and aremore required
drivers forare
non-automatic
required forsystems. Both these
non-automatic cases imply
systems. significant
Both these cases
increases in the operating
imply significant increasescosts in the ofoperating
systems that costswill be carrying
of systems thatawill
reduced demand
be carrying compared
a reduced to the
demand
situation
comparedastoitthe wassituation
before quarantine.
as it was before Data quarantine.
is already available from some
Data is already cities in
available fromChina
some where
citiesthe
in
passenger
China where demand for publicdemand
the passenger transport forispublic
slowlytransport
recovering, reaching
is slowly 30% by the
recovering, 4th week
reaching 30%into by the
the
recovery
4th week period
into the[50]. Faced
recovery with[50].
period this Faced
situation,
withmany administrations
this situation, and operatorsand
many administrations areoperators
working
with different with
are working recovery hypotheses
different recovery and,hypotheses
although market prediction
and, although models
market are available
prediction models [51–53],
are
the intrinsic characteristics of the nature of this crisis and the abnormal situation
available [51–53], the intrinsic characteristics of the nature of this crisis and the abnormal situation it it generates may affect
their efficiency.
generates may affect their efficiency.
As
As anan example
exampleand andin inthe
thespecific
specificcasecaseofofSantander,
Santander,wherewherethe thenormal
normal busbusload
load at at
thethe
rushrushhour
houris
82% in the city center corridor, different scenarios can be analyzed for
is 82% in the city center corridor, different scenarios can be analyzed for both the growth in demand both the growth in demand
and the system
system operation.
operation. Assuming similar rates of growth to the Chinese case [50] and that social social
distancing rules are are maintained,
maintained,the theresulting
resultingreduction
reductioninin vehicle
vehicle capacity
capacity will
will be be around
around 70%70% of
of the
the nominal
nominal capacity,
capacity, possibly
possibly rising
rising to to
50% 50% withany
with anypotential
potentialrelaxing
relaxingofof distancing
distancing rules rules by the
authorities when the situation warrants it. Figure 17 shows the evolution of the required required operating
operating
costs to satisfy the predicted demand whilst maintaining “safe”
costs to satisfy the predicted demand whilst maintaining “safe” occupancy rates. Assumingoccupancy rates. Assuming a recoverya
period
recovery lasting
period 2 months,
lasting the operator
2 months, will
the have towill
operator doublehavetheir normaltheir
to double operating
normal costs for overcosts
operating 3 weeksfor
and
overat3 the sameand
weeks time atcarry a much
the same timereduced
carry demand.
a much reduced With all demand.
these factors,WithTable 3 provides
all these factors,details
Tablefor3
total income
provides andfor
details costs during
total income theand
2 months of the recovery
costs during period
the 2 months of for
thethe case study.
recovery period Infor
thethecase of
case
astudy.
30% occupancy
In the case rate, the operator
of a 30% occupancy would
rate, multiply
the operator by 2.5 theirmultiply
would losses compared
by 2.5 their to alosses
normal period,
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
economic viability of the transport system. As well as all the above, assuming that it will not be viable
to duplicate
Sustainability the12,
2020, current
3870 fleet size, the focus of any action should not only be on the supply, but 14 also
of 18
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.
while
Only bya 50% occupancy
taking measuresrateaddressing
would result in supply
both a 52% increase in losses.
and demand canBoth cases willofalso
the viability havetransport
public to factor
in the feasibility of having to increase the required resources (fleet
systems be guaranteed during the period of social distancing restrictions. and drivers) to provide the service.
Figure 17. Evolution

Figure 17. Evolution of
of public
public transport
transport costs
costs during
during the
the recovery
recovery period.
period.
Table 3.
Table Estimated incomes
3. Estimated incomes and
and costs
costs during
during the
the recovery
recovery period.
period.
30%
30% OccupationRate
Occupation Rate 50%Occupation
50% Occupation Rate
Rate Normal
Normal
Incomes
Incomes 706,873.25€€
706,873.25 706,873.25
706,873.25 €€ 1,492,277.60 €€
1,492,277.60
Operating costs
Operating costs 3,751,221.19 €€
3,751,221.19 2,493,789.37 €€
2,493,789.37 2,697,507.37 €€
2,697,507.37
Balance 3,044,347.95 € 1,786,916.12 € 1,205,229.77 €
Balance 3,044,347.95 € 1,786,916.12 € 1,205,229.77 €
These results
4. Conclusions andaffect
Openthe economic viability of the entire public transport system. It would be
Questions
unsustainable during only this brief period of time, and the service needs to be provided during
This article has used available data to provide a preliminary report about how the imposition of
the entire quarantine period, as shown in Table 3, where the overall losses would be 148% higher.
quarantine throughout Spain due to the SARS-CoV-2 virus has affected internal mobility in the city
The implication is that public administrations will need to increase subsidies paid to guarantee the
of Santander. The analyzed data shows an overall fall of 76%, although this is less important in the
economic viability of the transport system. As well as all the above, assuming that it will not be viable
case of the private car. Public transport use has fallen the most with 93% fewer users. Mobility during
to duplicate the current fleet size, the focus of any action should not only be on the supply, but also
the morning and midday has dropped less than in the afternoon, when the fall is much more drastic
on the demand. Times for entering work places and educational establishments can be scaled to
with the disappearance of afternoon peak traffic periods. The effect of confinement has logically
minimize the rush hour factor that results in high vehicle occupancy on public transport systems.
modified the purposes behind people’s journeys, work being by far the most important purpose. The
Only by taking measures addressing both supply and demand can the viability of public transport
declining mobility has also produced a reduction in the emission of pollutants. The data collected
systems be guaranteed during the period of social distancing restrictions.
about NO2 emissions has allowed the authors to quantify this reduction as practically 60%, and traffic
accidents have been
4. Conclusions reduced
and Open by up to 67% in relative terms. The implications on the public transport
Questions
system have also been analyzed, highlighting the importance and the need for policies which not
This article has used available data to provide a preliminary report about how the imposition of
only maintain the economic exploitation itself, but also distribute the demand to avoid rush hours.
quarantine throughout Spain due to the SARS-CoV-2 virus has affected internal mobility in the city of
These effects should make us think on how mobility in cities will be after this crisis. There is a
Santander. The analyzed data shows an overall fall of 76%, although this is less important in the case
real risk of a decline in the sustainability of mobility in urban areas. The main questions to be
of the private car. Public transport use has fallen the most with 93% fewer users. Mobility during the
answered will be, first of all, when, how, and to what extent the demand levels for public transport
morning and midday has dropped less than in the afternoon, when the fall is much more drastic with
systems will recover (if they ever do). The willingness of the user to take over not only collective but
the disappearance of afternoon peak traffic periods. The effect of confinement has logically modified
also shared transport systems will be another important issue to be assessed. Further questions to be
the purposes behind people’s journeys, work being by far the most important purpose. The declining
investigated are how this will affect users’ perception of different transport services, and what new
mobility has also produced a reduction in the emission of pollutants. The data collected about NO
strategies both public and private sector operators will need to follow to make public transport2
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.
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. [CrossRef] [PubMed]
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. [CrossRef] [PubMed]
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. [CrossRef] [PubMed]
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. Healthcare 2020, 8, 46. [CrossRef]
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. [CrossRef]
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. [CrossRef]
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.
[CrossRef]
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. [CrossRef]
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.
[CrossRef]
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. [CrossRef]
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 27 April 2020).
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 27 April 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. [CrossRef]
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 27 April 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. [CrossRef]
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. [CrossRef]
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, USA, 2014. [CrossRef]
23. APTA. Passenger Transport. 2020, Volume 78, p. 7. Available online: 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. [CrossRef]
29. Anjum, N.A. Good in the worst: COVID-19 restrictions and ease in global air pollution. Preprints 2020,
2020, 040069. [CrossRef]
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).
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. [CrossRef]
33. González, M.; Hidalgo, C.; Barabási, A. Understanding individual human mobility patterns. Nature 2008,
453, 779–782. [CrossRef]
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.
[CrossRef]
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-a~no-2000/evolucion-del-trafico-desde-el-a~no-
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;
pp. 66–76.
37. Hardt, C.; Bogenberger, K. Usage of e-scooters in urban environments. Transp. Res. Procedia 2019, 37, 155–162.
[CrossRef]
38. Geroliminis, N.; Daganzo, C.F. Existence of Urban-scale macroscopic fundamental diagrams:
Some experimental findings. Transp. Res. Part B 2008, 42, 759–770. [CrossRef]
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.
[CrossRef]
40. Dell’Olio, L.; Ibeas, A.; Ruisanchez, F. Optimizing bus-size and headway in transit networks. Transportation
2012, 39, 449–464. [CrossRef]
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. [CrossRef]
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. [CrossRef]
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. [CrossRef]
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-\T1\textquestiondowncómo-será-el-d%C3%ADa-
después-y-cómo-prepararnos-desde-el-sector-del-transporte-público (accessed on 28 April 2020).
51. Kostrzewski, M. Sensitivity analysis of selected parameters in the order picking process simulation model,
with randomly generated orders. Entropy 2020, 22, 423. [CrossRef]
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. [CrossRef]
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/).

Sustainability 12 03870 v2 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sustainability 12 03870 v2 PDF

Uploaded by

Copyright:

Available Formats

sustainability

Keywords: COVID-19; coronavirus; mobility; traffic; confinement; quarantine; outbreak

Sustainability 2020, 12, 3870; doi:10.3390/su12093870 www.mdpi.com/journal/sustainability

Network Macroscopic Fundamental Diagram

entrance route to Santander from the East (direction Bilbao).

0:36:00 500 0:36:00 500

A decision support tool1.isHeadway

Table 2. Estimated operational costs of the public transport system.

Before Optimun (Quarantine) Current *

Trip length distribution before and during the quarantine

Trip length distribution before and during the quarantine

Evolution of NO2 emissions (static devices)

Industrial zone City center 2010-2019 values

Figure 14. NO22 measurements

Finally, accident data reported by the

Figure 17. Evolution

You might also like