Air Quality and Active Transportation Modes: A Spatiotemporal Concurrence Analysis in Guadalajara, Mexico

sus ainabili y A icle Ai Quali y and Ac i e T anspo a ion Modes: A Spa io empo al Concu ence Analysis in Guadalaja a, Mexico Gab iela Ochoa-Co a ubias 1,* , Ca los González-Figue edo 2, Hugo DeAlba-Ma ínez 2 and Alejand o L. G indlay 3   Ci a ion: Ochoa-Co a ubias, G.; González-Figue edo, C.; DeAlba-Ma ínez, H.; G indlay, A.L. Ai Quali y and Ac i e T anspo a ion Modes: A Spa io empo al Concu ence Analysis in Guadalaja a, Mexico. Sus ainabili y 2021,13, 13904. h ps://doi.o g/10.3390/su132413904 Academic Edi o s: O iol Ma que and Monika Wik o ia Maciejewska Recei ed: 29 Oc obe 2021 Accep ed: 3 Decembe 2021 Published: 16 Decembe 2021 Publishe ’s No e: MDPI s ays neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a il- ia ions. Copy igh : © 2021 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion (CC BY) license (h ps:// c ea i ecommons.o g/licenses/by/ 4.0/). 1Depa men o Habi a and U ban De elopmen , ITESO, The Jesui Uni e si y o Guadalaja a, Tlaquepaque 45604, Mexico 2Depa men o Technological and Indus ial P ocesses, ITESO, The Jesui Uni e si y o Guadalaja a, Tlaquepaque 45604, Mexico; [email p o ec ed] (C.G.-F.); [email p o ec ed] (H.D.-M.) 3 Depa men o U ban and Regional Planning, Uni e si y o G anada, 18071 G anada, Spain; g indlay@ug .es *Co espondence: [email p o ec ed] Abs ac : The p o ec ion o pedes ians, cyclis s, and public anspo a ion passenge s om en i on- men al pollu ion is a global conce n. This s udy ills he gap in he exis ing knowledge o empo al exposu e o ai pollu ion in La in Ame ican me opolises. The pape p oposes a me hodology ad- d essing he ela ionship be ween wo objec s o s udy, i.e., he use s o ac i e modes o anspo and ai quali y. This new me hodology assesses he spa io empo al concu ence o bo h objec s wi h s a is ical analysis o la ge open-access da abases, o p omo e heal hy and sus ainable u ban mobili y. The applica ion o he empi ical me hodology es ima ed he numbe o use s o ac i e anspo a ion modes exposed o poo ai quali y episodes in he Guadalaja a me opoli an a ea (Mexico) in 2019. The s udy conside ed wo pollu an s, ozone (O 3 ) and pa icula e ma e (PM 10 ), and wo ac i e modes, cycling and bus apid ansi (BRT). Spa io empo al analyses we e ca ied ou wi h geog aphic in o ma ion sys ems, as well as wi h nume ic compu ing pla o ms. Fi s , big da a we e used o coun he numbe o use s o each mode wi hin he a ea o in luence o he ai quali y moni o ing s a ions. Second, he numbe o ai pollu ion episodes was ob ained using he ai quali y index p oposed by he En i onmen al P o ec ion Agency (USA) on an hou ly basis. Thi d, he spa io empo al concu ence be ween ai quali y episodes and ac i e mode use s was calcula ed. In pa icula , he ai quali y moni o ing da a om he Jalisco A mosphe ic Moni o ing Sys em we e compa ed o use s o he public bicycle sha e sys em, known as MiBici, and o a bus apid ansi line, known as Mi Mac o Calzada. The esul s showed ha he numbe o cyclis s and BRT passenge s exposed o poo ai quali y episodes was conside able in absolu e e ms, ha is, 208,660 use s, while i was ma ginal when compa ed o he o al numbe o use s exposed o be e ai quali y ca ego ies in he s udy a ea, who ep esen ed only 10%. To apply he esul s a he me opoli an scale, he spa ial dis ibu ion o he ai quali y moni o ing sys em should be imp o ed, as well as he a ailabili y o da a on pedes ians and con en ional bus passenge s. Keywo ds: pedes ian; bicycle; ac i e anspo a ion modes; ai quali y; isk; heal h; mass anspo sys em; bus apid ansi ; GIS; big da a 1. In oduc ion Ai quali y is a big conce n in ci ies. Since 2016, 90% o wo ldwide ci y dwelle s li ing in low- and middle-income coun ies ha e b ea hed ai ha did no mee he sa e y s anda ds se by he Wo ld Heal h O ganiza ion (WHO), causing 4.2 million p ema u e dea hs due o ai pollu ion [1,2]. Ai pollu ion is he p esence in he ai o ma e (solid pa icles and gases) o o ms o ene gy ha cause isk, damage, o se ious annoyance o people and o he li ing o ganisms. Acco ding o he WHO [ 2 ], PM, ca bon monoxide (CO), ozone (O 3 ), ni ogen dioxide (NO 2 ), Sus ainabili y 2021,13, 13904. h ps://doi.o g/10.3390/su132413904 h ps://www.mdpi.com/jou nal/sus ainabili y Sus ainabili y 2021,13, 13904 2 o 15 and sul u dioxide (SO 2 ) a e he pollu an s wi h he s onges e idence o being public heal h conce ns. The En i onmen al P o ec ion Agency (EPA) p oposed he uni e sally ecognized ai quali y index (AQI) [ 3 ] o assess ai pollu ion. The EPA de ined six ca ego ies acco ding o pollu an concen a ions. The ca ego ies we e labeled co esponding o he le el o conce n, ha is, 1, 2, 3, 4, 5, o 6 o good, mode a e, unheal hy o sensi i e g oups, unheal hy, e y unheal hy, and haza dous conce n, espec i ely. Speci ically, O 3 and PM pollu ion a g ound le el poses a high isk o heal h [ 4 , 5 ] and ha e mul iscale empo al pa e ns. The concen a ions may a y acco ding o he hou o he day and he season, e.g., O 3 concen a ions a e highe du ing he day ime han a nigh , and pa icula e ma e , such as PM 10 , is mo e concen a ed in he win e han in he summe . P ecipi a ions also a ec PM concen a ions [6]. Exposu e o excessi e O 3 can cause b ea hing p oblems, igge as hma, and educe lung unc ion, leading o lung disease. Addi ionally, b ea hing ai wi h high concen a ions o PM is conside ed he leading en i onmen al cause o dea h and disease. Bo h long- e m and sho - e m exposu e o PM a e associa ed wi h mo bidi y and mo ali y om ca dio ascula and espi a o y diseases [2]. Di e en modes o anspo a e associa ed wi h speci ic e ec s o bad ai condi ions on heal h [ 7 , 8 ], o ins ance, he p i a e ca is ecognized as he mos pollu ing mode o anspo [ 9 – 11 ]. Recen ly, u ban policies in la ge me opolises ha e p omo ed ac i e ans- po a ion modes h ough in es men s in in as uc u e, in pa icula , o imp o emen s o pedes ians, cyclis s, and use s o mass anspo sys ems. Howe e , ac i e modes, such as walking, cycling, and accessing public anspo s a ions o bus s ops, in ol e a physical e o [ 12 , 13 ]. Thus, he ac i i y in ol ed o hese modes equi es mo e inhala ion and, consequen ly, use s a e mo e sensi i e o ai quali y [ 14 , 15 ]. The in eg a ion o public and nonmo o ized anspo ne wo ks in ci ies, as well as he educ ion o he use o p i a e ca s, will con ibu e o minimizing he heal h isks o pedes ians, cyclis s, and mass public anspo use s and p omo e sus ainable ci ies [16]. This s udy p oposes a no el me hodology o es ima e he exposu e o use s o ac i e modes o ai pollu ion. I also seeks o gi e in o ma ion o hese use s on heal hie h eshold hou s o walking o cycling. This new me hodology in ol es add essing he ela ionship be ween wo objec s o s udy, i.e., he use s o ac i e modes and ai quali y, assessing he spa io empo al concu ence o bo h objec s wi h he s a is ical analysis o la ge open access da abases, o suppo heal hy and sus ainable u ban mobili y. This s udy ca ied ou an empi ical analysis in he Guadalaja a me opoli an a ea (GMA), Mexico. This me opolis is one o he mos pollu ed in he coun y and in La in Ame ica [ 17 ]. Mo eo e , he ci y is suscep ible o pollu ion impac s, since nonmo o ized and public anspo jou neys p edomina e in he ci y’s modal spli by 40% and 37%, espec i ely. This sugges s ha a leas wo- hi ds o he popula ion, he ac i e mode use s, may be ulne able o poo ai quali y condi ions. This ac shows he impo ance o c oss- empo al and spa ial s udies o iden i y he pol- lu ion impac s on ci y inhabi an s. This is also ecognized in exis ing li e a u e [6,7,18–24]. Despi e s udies in Eu ope [ 7 , 25 – 27 ], he e is a lack o analysis o la ge da abases o empo al exposu e o ai pollu ion conce ning ac i e mode use s in La in Ame ican me opolises. A g ea numbe o use s we e expec ed o be exposed due o he his o ical poo ai quali y episodes in he ci y and he high numbe o ac i e mode use s. Ne e heless, he esul s showed ha his numbe was smalle han ini ially assumed. E en hough he ela i e igu es we e low, he absolu e numbe o use s exposed should p omp he adop ion o public policies o p o ec hem, such as he educ ion o au omobile dependency and he p omo ion o ac i e anspo a ion modes in he ci y, bene i ing he en i e popula ion. This a icle is s uc u ed as ollows. The s udy a ea is de ined in Sec ion 2, wi h a ocus on he ai quali y moni o ing sys em’s a eas o in luence, he bicycle sha e sys em, and he bus apid ansi (BRT) sys em. The ma e ials and me hods a e p esen ed in Sec ion 3, Sus ainabili y 2021,13, 13904 3 o 15 ollowed by he esul s in Sec ion 4. The discussion and conclusions a e included in Sec ions 4and 5, espec i ely. 2. S udy A ea The s udy a ea was de ined by he o e lap be ween he ac i e mode s a ions whe e da a we e a ailable and he a ea o in luence o he ai quali y moni o ing s a ions (AQSs) o he Jalisco A mosphe ic Moni o ing Sys em (SIMAJ, by i s ini ials in Spanish) in he GMA. 2.1. The Guadalaja a Me opoli an A ea: Geog aphical and Mobili y Con ex The GMA is he capi al o he Jalisco p o ince. I is an u ban a ea o 5.2 million inhabi an s in wes e n Mexico [ 28 ], comp ising nine municipali ies: Guadalaja a, Zapopan, San Ped o Tlaquepaque, Tonalá, El Sal o, Tlajomulco de Zúñiga, Juanaca lán, Ix lahuacán de los Memb illos, and Zapo lanejo. As shown in Figu e 1, he ci y is su ounded by moun- ains o he no hwes , wes , and sou heas . The p e ailing wind ci cula es wes -sou hwes , combined wi h a “chimney e ec ” caused by no h-no heas winds [ 29 ], inc easing he pollu an concen a ions in he sou h-sou heas sec o o he ci y. Sus ainabili y 2021, 13, x FOR PEER REVIEW 3 o 15 A g ea numbe o use s we e expec ed o be exposed due o he his o ical poo ai quali y episodes in he ci y and he high numbe o ac i e mode use s. Ne e heless, he esul s showed ha his numbe was smalle han ini ially assumed. E en hough he el- a i e igu es we e low, he absolu e numbe o use s exposed should p omp he adop ion o public policies o p o ec hem, such as he educ ion o au omobile dependency and he p omo ion o ac i e anspo a ion modes in he ci y, bene i ing he en i e popula ion. This a icle is s uc u ed as ollows. The s udy a ea is de ined in Sec ion 2, wi h a ocus on he ai quali y moni o ing sys em’s a eas o in luence, he bicycle sha e sys em, and he bus apid ansi (BRT) sys em. The ma e ials and me hods a e p esen ed in Sec- ion 3, ollowed by he esul s in Sec ion 4. The discussion and conclusions a e included in Sec ions 4 and 5, espec i ely. 2. S udy A ea The s udy a ea was de ined by he o e lap be ween he ac i e mode s a ions whe e da a we e a ailable and he a ea o in luence o he ai quali y moni o ing s a ions (AQSs) o he Jalisco A mosphe ic Moni o ing Sys em (SIMAJ, by i s ini ials in Spanish) in he GMA. 2.1. The Guadalaja a Me opoli an A ea: Geog aphical and Mobili y Con ex The GMA is he capi al o he Jalisco p o ince. I is an u ban a ea o 5.2 million in- habi an s in wes e n Mexico [28], comp ising nine municipali ies: Guadalaja a, Zapopan, San Ped o Tlaquepaque, Tonalá, El Sal o, Tlajomulco de Zúñiga, Juanaca lán, Ix lahuacán de los Memb illos, and Zapo lanejo. As shown in Figu e 1, he ci y is su ounded by moun ains o he no hwes , wes , and sou heas . The p e ailing wind ci cula es wes - sou hwes , combined wi h a “chimney e ec ” caused by no h-no heas winds [29], in- c easing he pollu an concen a ions in he sou h-sou heas sec o o he ci y. Figu e 1. Ai quali y moni o ing s a ions and ac i e modes s a ions in he Guadalaja a me opoli an a ea, Mexico. The Figu e shows ai quali y moni o ing s a ions (AQSs) o he Jalisco A mosphe ic Figu e 1. Ai quali y moni o ing s a ions and ac i e modes s a ions in he Guadalaja a me opoli an a ea, Mexico. The Figu e shows ai quali y moni o ing s a ions (AQSs) o he Jalisco A mosphe ic Moni o ing Sys em (SIMAJ, om i s ini ials in Spanish) and hei a ea o in luence (AI), he s a ions o he bicycle sha e sys em (MiBici), con en ional bus s ops, and s a ions o he U ban Elec ic T ain Sys em (SITEUR, om i s ini ials in Spanish), i.e., ligh ail ansi (LRT), bus apid ansi (BRT), and high-quali y buses (HQB). Sou ce: au ho s based on [30–35]. The me opoli an a ea has a semiwa m clima e, wi h some hea island episodes. The e a e empe a u e in e sions in he ci y on 78% o he days o he yea , mainly be ween No embe and June. This phenomenon a o s he o ma ion o O 3 and makes i di icul o ola ile compounds such as PM 10 o dispe se. Howe e , he si ua ion imp o es in he ainy Sus ainabili y 2021,13, 13904 4 o 15 season (June–Augus ) as he pollu an s, mainly he PM 10 , suspended in he a mosphe e p ecipi a e [29]. Ai quali y is a g owing conce n in he GMA. Quan i ying and moni o ing ai pollu ion is he i s s ep o educe exposu e o ai pollu ion [ 2 ]. The local go e nmen moni o s he ai quali y h ough he SIMAJ [ 36 ] and o e s eal- ime ai pollu ion measu emen s. Figu e 1shows he en ixed AQSs. The AQSs measu e he concen a ion o i e pollu an s (CO, PM 10 , O 3 , SO 2 , NO 2 ) as well as me eo ological a iables (ambien empe a u e, ela i e humidi y, and wind speed and di ec ion). The da a a e used as he inpu o he announcemen o en i onmen al ale s in he ci y. Figu e 1shows he he e ogenous spa ial pa e n o he AQSs ha pa ially assess he ambien pollu ion in he me opolis. Juanaca lán, Ix lahuacán de los Memb illos, Tlajomulco de Zúñiga, and Zapo lanejo a e mos ly ou side he SIMAJ’s a ea o in luence, while El Sal o, Tonalá, Tlaquepaque, and Zapopan a e only pa ly inside his a ea. Only he Guadalaja a municipali y has accep able co e age by he SIMAJ. Acco ding o he Na ional Ins i u e o Ecology and Clima e Change (Mexico) [ 10 ], in 2019, he numbe o days on which he maximum ecommended h esholds we e exceeded, in a leas one pa ame e , o any pollu an in he GMA was 267 days. This is 6% mo e han he numbe egis e ed in 2018 and ep esen s wo hi ds o he yea wi h excessi e le els o ai pollu ion. The igu es in Table 1 e eal ha PM 10 and O 3 a e he pollu an s ha mos equen ly exceed he s anda ds. Gi en hese igu es, his s udy ocused on PM10 and O3as he pollu an s o be analyzed. Table 1. Ambien pollu an s in he Guadalaja a me opoli an a ea in 2019. Sou ce: au ho s, based on [10]. Pollu an Assessmen Uni Value S anda d [NOM-Mexico] Numbe o Days abo e S anda d O38 h ppm 0.146 0.070 39.50% PM10 24 h µg/m3261 75 41.10% CO 8 h ppm 3.6 11 N/A NO21 h ppm 0.203 0.210 N/A SO28 h ppm 0.013 0.2 N/A N/A: Nonapplicable; ppm = pa s pe million. P i a e ehicles a e ecognized as he main sou ce o ambien ai pollu ion in ci ies [ 37 ], and he GMA is no excep ion [ 38 , 39 ]. Be ween 1980 and 2010, he low-densi y u ban de elopmen pa adigm in he ci y inc eased indi idual mo o ized ips by mo e han 942% [17]. Mass anspo sys ems a e ecognized as a solu ion o sus ainable anspo de- elopmen [ 26 ] and a ca alys o be e ai quali y, despi e he nega i e impac s du ing hei cons uc ion [ 27 ]. Ne e heless, hei con ibu ion o he GMA modal sha e is e y limi ed. Acco ding o he Jalisco S a e Go e nmen [ 17 ], 37% o he ips in he GMA a e made on oo , 3% by bicycle, 37% by public anspo a ion, and 23% by p i a e ehicles. The mass anspo sys ems ba ely con ibu e, wi h only 3% o he modal sha e. I is ecognized ha commu ing on oo and by bicycle can signi ican ly educe u ban pollu an emissions [ 40 , 41 ]. Acco ding o he Na ional Ins i u e o S a is ics and Geog aphy (INEGI, om i s ini ials in Spanish) [ 28 ], 25% o dwellings in he GMA ha e a bicycle as a po en ial means o anspo a ion. This igu e con as s wi h he con ibu ion o cycling o he modal spli in he ci y [42]. His o ically, ca s ha e been a o ed in he GMA wi h la ge in as uc u e in es men s. Recen ly, he go e nmen , mo i a ed by communi y associa ions and academic ins i u ions, has p omo ed a sus ainable mobili y pa adigm, pa icula ly h ough ac i e modes, e.g., imp o emen s o he ne wo k o bike lanes, he expansion o he bicycle sha e sys em (known as MiBici), as well as he enhancemen o he U ban Elec ic T ain Sys em (SITEUR, om i s ini ials in Spanish). MiBici is he public bicycle sha e sys em in he ci y, es ablished in 2014 and ecognized as one o he bes bicycle sha e sys ems in he coun y in 2019 [ 43 ]. The sys em se es mo e han 86,000 cyclis s h ough 3200 bicycles and 300 s a ions [ 44 ]. Balde as e al. [ 16 ] Sus ainabili y 2021,13, 13904 5 o 15 s a e ha one o he main easons use s choose MiBici is he low cos compa ed wi h o he anspo a ion modes. The Cicloci ies Ranking 2019 [ 43 ] s a es ha he local go e nmen s o Guadalaja a, Zapopan, and Tlaquepaque ha e enhanced he cycling in as uc u e by deploying public specialis s in mobili y and by gi ing i p io i y in planning bodies. Figu e 1shows ha he spa ial dis ibu ion o MiBici is loca ed mainly in Guadalaja a, o a ce ain ex en in Zapopan, and e y li le in Tlaquepaque. This ne wo k is nonexis en in he es o he GMA municipali ies. Cycling ep esen s a ound 2% o he o al modal spli in he GMA [ 42 ]. The peak hou s o use o he MiBici sys em a e 8:00 a.m. and 6:00 p.m. Acco ding o Jalisco Cómo Vamos [ 45 ], MiBici egis e ed an a e age o mo e han 2.8 million annual cycling ips in he pe iod 2014–2020. The same sou ce epo s ha he kilome e s o bicycle lanes inc eased i e- old be ween 2015 and 2020, o aling 182 km. The highes numbe o ips since he ope a ion began was in 2019, wi h 4,660,470 ips aken by mo e han 27 housand use s, wi h an a e age a el ime o 12:32 min [44]. The he e ogeneous spa ial dis ibu ion o MiBici shows a wes –eas pa e n limi ed o he co e o he ci y (Figu e 1). The SITEUR has ope a ed since 1989. The public ca ie se es mo e han 98.5 million passenge s a yea h ough 121 s a ions, i.e., 3.23% o he o al modal spli in he ci y, mainly wi h ligh ail ain and BRT [ 17 ]. This s udy coun ed he BRT Line 1 passenge s, due o he da a a ailabili y when his esea ch was conduc ed. The BRT Line 1, known as Mi Mac o Calzada, is 16.6 km long wi h 27 s a ions, and has been in ope a ion since 2009. I accoun s o 30.3% o SITEUR ips. The main line plus he 15 eeding lines se ed 44.2 million passenge s in 2020, [ 46 ] wi h a i e-minu e equency (mo ning peak hou ) [ 47 ]. As shown in Figu e 1, he BRT is no h–sou h o ien ed and only se es he Guadalaja a municipali y up o he Tlaquepaque bounda y. 2.2. The Jalisco A mosphe ic Moni o ing Sys em’s A ea o In luence The s udy a ea was de e mined by he SIMAJ’s co e age and he da a a ailabili y on an hou ly basis. The spa ial dis ibu ion o he AQSs does no co e he GMA’s en i e u ban oo p in , as shown in Figu e 1. Mo eo e , acco ding o he SIMAJ [ 48 ], he AQSs ha e a wo-kilome e a ea o in luence, o which eal- ime hou ly measu emen s o PM 10 concen a ions in mic og ams pe cubic me e and O 3 in pa s pe million a e eco ded; hese concen a ions a e used o calcula e hei ela ed AQIs. The spa ial concu ence be ween he AQSs’ a eas o in luence and he ac i e modes o anspo s ops/s a ions was iden i ied h ough a geog aphic in o ma ion sys em (A cGis P o, ESRI ® , Redlands, CA, USA). The calcula ions included 10 AQSs, 287 bike sha e s a ions, 9073 con en ional bus s ops, 77 BRT s a ions, and 48 ligh ail ansi s a ions [30,35]. Ai quali y moni o ing da a a ec ing mos o he ac i e mode use s we e no a ailable. Figu e 2highligh s he na ow co e age o he AQSs wi h espec o he ac i e mode s ops/s a ions. Thi y-se en pe cen o he MiBici s a ions, 28% o he con en ional bus s ops, 44% o he BRT s a ions, 22% o he high-quali y bus s ops, and 38% o he ligh ail ain s a ions we e ou side he a ea o in luence o he AQSs in he ci y. Only 107 o he MiBici s a ions we e included in he a ea o in luence o ou AQSs: A emajac, Cen o, Tlaquepaque, and Valla a. Simila ly, only wel e BRT s a ions we e conside ed o his s udy, as hey ell inside o he AQSs’ a eas o in luence. Speci ically, he BRT s a ions loca ed wi hin he a ea o in luence o he Cen o AQS we e Ciencias de la Salud (10), Juan Ál a ez (11), Alameda (12), San Juan de Dios (13), Bicen ena io (14), La Paz (15), and Niños Hé oes (16). The s a ions in he a ea o in luence o he Mi a alle AQS we e López de Legazpi (23), Clemen e O ozco (24), A es Plás icas (25), Escul u as (26), and F ay Angélico (27). Sus ainabili y 2021,13, 13904 6 o 15 Sus ainabili y 2021, 13, x FOR PEER REVIEW 6 o 15 s udy, as hey ell inside o he AQSs’ a eas o in luence. Speci ically, he BRT s a ions loca ed wi hin he a ea o in luence o he Cen o AQS we e Ciencias de la Salud (10), Juan Á l a ez (11), Alameda (12), San Juan de Dios (13), Bicen ena io (14), La Paz (15), and Ni- ños Hé oes (16). The s a ions in he a ea o in luence o he Mi a alle AQS we e López de Legazpi (23), Clemen e O ozco (24), A es Plás icas (25), Escul u as (26), and F ay An- gélico (27). Un o una ely, no da a o he con en ional and high-quali y buses o he ligh ail ains we e a ailable on an hou ly basis. Thus, due o he SIMAJ’s spa ial pa e n, oge he wi h he da a a ailabili y, i e AQSs, 107 MiBici s a ions, and 12 BRT s a ions we e con- side ed o he compu a ions. Figu e 2. S udy a ea. The igu e shows he a ea o in luence (AI) o he ai quali y moni o ing s a ions (AQSs) managed by he Jalisco A mosphe ic Moni o ing Sys em (SIMAJ, om i s ini ials in Spanish). The map highligh s he ac i e mode s a ions conside ed in he s udy, i.e., 107 s a ions o he MiBici bicycle sha e sys em and 12 bus apid ansi (BRT) s a ions. Sou ce: au ho s based on [30–34]. 3. Da a Sou ces and Me hods The WHO s a es ha using la ge yea ly da abases allows ou comes o be e ined, aking in o accoun seasonal a ia ions [8]. The analysis assessed da a om he selec ed s a ions in 2019. This yea was chosen as he e e ence yea o p e en mis ep esen a ion due o changes in a el beha io s caused by he COVID-19 pandemic in 2020 and 2021. An hou ly ime scale was selec ed due o he O3 and PM10 empo al beha io , and da a Figu e 2. S udy a ea. The igu e shows he a ea o in luence (AI) o he ai quali y moni o ing s a ions (AQSs) managed by he Jalisco A mosphe ic Moni o ing Sys em (SIMAJ, om i s ini ials in Spanish). The map highligh s he ac i e mode s a ions conside ed in he s udy, i.e., 107 s a ions o he MiBici bicycle sha e sys em and 12 bus apid ansi (BRT) s a ions. Sou ce: au ho s based on [30–34]. Un o una ely, no da a o he con en ional and high-quali y buses o he ligh ail ains we e a ailable on an hou ly basis. Thus, due o he SIMAJ’s spa ial pa e n, o- ge he wi h he da a a ailabili y, i e AQSs, 107 MiBici s a ions, and 12 BRT s a ions we e conside ed o he compu a ions. 3. Da a Sou ces and Me hods The WHO s a es ha using la ge yea ly da abases allows ou comes o be e ined, aking in o accoun seasonal a ia ions [ 8 ]. The analysis assessed da a om he selec ed s a ions in 2019. This yea was chosen as he e e ence yea o p e en mis ep esen a ion due o changes in a el beha io s caused by he COVID-19 pandemic in 2020 and 2021. An hou ly ime scale was selec ed due o he O 3 and PM 10 empo al beha io , and da a a ailabili y. This s udy bene i ed om open da a a ailabili y o all AQSs [ 48 ], pa icula ly o O3and PM10. Nonmo o ized modes and public anspo accoun o 77% o ips in he ci y [ 17 ]. Disappoin ingly, no da a on pedes ian ips we e a ailable, no ips made by p i a ely owned bicycles. The MiBici open da a sys em [ 49 ] was accessed o ob ain he o al ip egis e s o o om any o i s ancho ing s a ions o 2019. The SITEUR epo ed he numbe Sus ainabili y 2021,13, 13904 7 o 15 o passenge s en e ing he BRT s a ions hou ly [ 50 ]. No da a o ligh ail ains we e a ailable a he equi ed accu acy. As a consequence o da a a ailabili y, he eina e , he e m “ac i e mode use s” e e s o ips ia MiBici and pedes ians accessing BRT s a ions in he a eas o in luence o AQSs. Table 2shows da a sou ces and eco ding equency o each kind o s a ion compu ed in his s udy, i.e., AQS, MiBici, and BRT. Table 2. Sou ces o da a on ai quali y moni o ing and ac i e anspo a ion modes in he Guadalaja a me opoli an a ea in 2019. S a ion Da a Tempo al Basis * Numbe o Reco ds Sou ce Ai quali y O3and PM10 concen a ion Hou 87,600 Minis y o he En i onmen and Te i o ial De elopmen (Semade , Jalisco, Mexico) SIMAJ [ 36 , 48 ] Ac i e anspo a ion mode Bicycle in–ou ancho s Minu e 2,388,884 Go e nmen o he S a e o Jalisco, Mexico: MiBici [ 49 ] Passenge s coming in o he BRT s a ion Hou 105,120 U ban Elec ic T ain Sys em, Guadalaja a [50] * Fo he yea 2019. A s a is ical spa io empo al app oach was used. The yea 2019 was chosen as he mos ecen yea wi hou po en ial changes o a el beha io s due o he COVID-19 pandemic [7]. Figu e 3shows he concep ual model used o iden i y he numbe o ac i e mode use s exposed o poo ai quali y episodes in he e e ence yea . Fi s , he ac i e mode da abase was o ganized o iden i y he numbe o ac i e mode use s in he AQSs’ a eas o in luence, on an hou ly basis. Then, he numbe o episodes pe hou in each AQS a ea was assessed using he ai quali y index [ 3 ]. Finally, he concu ence be ween pollu an s and use s o ac i e anspo a ion modes was assessed. Sus ainabili y 2021, 13, x FOR PEER REVIEW 8 o 16 Figu e 3. Concep ual model. Ac i e modes analyzed we e bicycle sha e sys em and bus apid ansi (BRT). Pollu an s compu ed we e ozone (O3) and pa icula e ma e (PM10). Ai quali y index (AQI) e e s o he En i onmen al P o ec ion Agency index [3]. The wo d “use s” e e s o bo h bicycle ips and BRT passenge s. 3.1. Ac i e Mode Da a Managemen Da a om he 107 MiBici s a ions we e used o calcula e he numbe o ips pe hou linked o each AQS. The da a we e managed using Py hon p og amming language and Pandas, among o he lib a ies. The ollowing assump ions we e made o compu e bicycle ips and c ea e a da abase ha would be compa ible wi h AQS da a. Fi s , one hou was added o he empo al da a. Second, bicycle ips a eling a leas one minu e inside he AQS’s a ea o in luence we e quan i ied as one ip in he co esponding hou . Thi d, he cyclis s we e exposed o ai quali y episodes bo h when s a ing and ending he ip a he s a ion, and hus coun ed wice. Howe e , i a cyclis s a ed and inished he ip a s a- ions linked o he same AQS, i was coun ed as only one ip. The SITEUR also p o ided hou ly da a on people accessing he 12 selec ed BRT s a- ions. The e we e no da a o passenge s lea ing he s a ions; hese esul s we e unde es- ima ed, since BRT use s we e only coun ed as passenge s accessing he BRT and no as pedes ians a e hei ansi jou ney. 3.2. Alloca ion o Episodes by AQI Ca ego y AQI was assessed o alloca e a ca ego y o O3 and PM10 pollu ion in he i e AQSs s udied. The AQI was designed by he EPA [51], and calcula ions we e pe o med using he MATLAB p og amming language. Fi s , mo ing a e ages o pollu an concen a ions we e calcula ed o 8 and 24 h ime windows, o O3 and PM10, espec i ely. The AQI was calcula ed by subs i u ing he mo - ing a e ages di ec ly in Equa ion (1). Nex , he AQI was classi ied acco ding o he es ab- lished ca ego ies desc ibed in he in oduc ion (Sec ion 1). Finally, he numbe o episodes in AQI ca ego ies 3, 4, and 5 we e linked o AQSs o each hou o he yea ; hese h ee ca ego ies co espond o poo ai quali y episodes. Figu e 3. Concep ual model. Ac i e modes analyzed we e bicycle sha e sys em and bus apid ansi (BRT). Pollu an s compu ed we e ozone (O 3 ) and pa icula e ma e (PM 10 ). Ai quali y index (AQI) e e s o he En i onmen al P o ec ion Agency index [3]. The wo d “use s” e e s o bo h bicycle ips and BRT passenge s. Sus ainabili y 2021,13, 13904 8 o 15 3.1. Ac i e Mode Da a Managemen Da a om he 107 MiBici s a ions we e used o calcula e he numbe o ips pe hou linked o each AQS. The da a we e managed using Py hon p og amming language and Pandas, among o he lib a ies. The ollowing assump ions we e made o compu e bicycle ips and c ea e a da abase ha would be compa ible wi h AQS da a. Fi s , one hou was added o he empo al da a. Second, bicycle ips a eling a leas one minu e inside he AQS’s a ea o in luence we e quan i ied as one ip in he co esponding hou . Thi d, he cyclis s we e exposed o ai quali y episodes bo h when s a ing and ending he ip a he s a ion, and hus coun ed wice. Howe e , i a cyclis s a ed and inished he ip a s a ions linked o he same AQS, i was coun ed as only one ip. The SITEUR also p o ided hou ly da a on people accessing he 12 selec ed BRT s a ions. The e we e no da a o passenge s lea ing he s a ions; hese esul s we e unde - es ima ed, since BRT use s we e only coun ed as passenge s accessing he BRT and no as pedes ians a e hei ansi jou ney. 3.2. Alloca ion o Episodes by AQI Ca ego y AQI was assessed o alloca e a ca ego y o O 3 and PM 10 pollu ion in he i e AQSs s udied. The AQI was designed by he EPA [ 51 ], and calcula ions we e pe o med using he MATLAB p og amming language. Fi s , mo ing a e ages o pollu an concen a ions we e calcula ed o 8 and 24 h ime windows, o O 3 and PM 10, espec i ely. The AQI was calcula ed by subs i u ing he mo ing a e ages di ec ly in Equa ion (1). Nex , he AQI was classi ied acco ding o he es ablished ca ego ies desc ibed in he in oduc ion (Sec ion 1). Finally, he numbe o episodes in AQI ca ego ies 3, 4, and 5 we e linked o AQSs o each hou o he yea ; hese h ee ca ego ies co espond o poo ai quali y episodes. AQI = [(AQIHi −AQILo)/(ConcHi −ConcLo)] ×(Conci−concLo) + (AQILo) (1) whe e Conci= Inpu concen a ions o a gi en pollu an ; ConcLo = The concen a ion b eakpoin ha is less han o equal o Conci; ConcHi = The concen a ion b eakpoin ha is g ea e han o equal o Conci; AQILo = The AQI alue/b eakpoin co esponding o ConcLo; AQIHi = The AQI alue/b eakpoin co esponding o ConcHi. 3.3. Concu ence Analysis The concu ence analysis implied s a i ied compu a ions o wo inpu s: i s , o he numbe o MiBici ips and BRT passenge s pe hou in he i e AQSs’ a eas o in luence; hen, o he numbe o poo ai quali y episodes alling in o ca ego ies 3, 4, o 5. I was assumed ha he en i e s udy a ea had he same compensa ion o calcula e exposu e o pollu an s. The concu ence analysis was compu ed wi h dynamic ables on a sp eadshee (Excel, MSO ice ® , Redmond, WA, USA), associa ing spa io empo al co espondence. As a esul , he numbe o ac i e mode use s exposed o ca ego y 3, 4, o 5 episodes was es ima ed, bo h o O3and PM10. 4. Resul s 4.1. Ac i e Mode Da a Managemen As expec ed, he numbe o passenge s who walked o access he 12 mass anspo BRT s a ions was highe han he numbe o ips logged a he 104 MiBici s a ions, ha is, 20,028,250 and 2,388,884, espec i ely. Figu e 4displays he numbe o annual ac i e mode use s s udied on an hou ly basis. Figu e 4a shows ha mos o he MiBici ips we e in he a ea o in luence o he Cen o AQS. All s a ions egis e ed wo peak hou s, i.e., 9:00 and 19:00. A 9:00, he e we e 131,608, 17,005, and 41,602 ips in he a ea o in luence o AQS Cen o, Tlaquepaque, and Valla a, espec i ely. These igu es inc eased a 19:00 in Sus ainabili y 2021,13, 13904 9 o 15 he same a eas o in luence, ha is, 153,314, 21,062, and 58,179 ips, espec i ely This was concu en wi h he s anda d wo king hou s in he ci y. In addi ion, he ips linked o he Cen o s a ion also showed a peak hou a 14:00 (lunch ime), i.e., 129,274 ips. Sus ainabili y 2021, 13, x FOR PEER REVIEW 9 o 15 he same a eas o in luence, ha is, 153,314, 21,062, and 58,179 ips, espec i ely This was concu en wi h he s anda d wo king hou s in he ci y. In addi ion, he ips linked o he Cen o s a ion also showed a peak hou a 14:00 (lunch ime), i.e., 129,274 ips. The spa ial pa e n o he BRT passenge s was simila , al hough he s a ions ha con- ibu ed o he peak hou s we e clea ly dis inguished. Figu e 4b shows he h ee peak hou s a he BRT s a ions in he a ea o in luence o he Cen o and Mi a alle AQSs, ha is, be ween 8:00 and 9:00, be ween 14:00 and 15:00, and be ween 19:00 and 20:00. O he 20 million passenge s using hese s a ions, 60% we e in he a ea o in luence o he Cen o AQS and 40% in ha o he Mi a alle s a ion, i.e., 12,208,712 and 7,819,538, espec i ely. Mo eo e , i was shown ha 1,832,788 passenge s walked o each BRT s a ions in he icini y o he Mi a alle AQS be ween 8:00 and 9:00, and 2,152,791 passenge s accessed BRT s a ions a ound he Cen o AQS be ween 19:00 and 20:00. This can be explained by he ac ha he o me AQS se es a mos ly esiden ial a ea, while he la e se es a eas wi h a high densi y o wo kplaces, pa icula ly e ail ade and public se ices. (a) (b) (c) (d) Figu e 4. Ac i e mode use s and poo ozone (O3) o pa icula e ma e (PM10) episodes in he s udy a ea on an hou ly basis. The o de o magni ude be ween igu es mus be conside ed o p ope ly compa e hem. (a,b) e e o he annual numbe o MiBici ips and bus apid ansi (BRT) passenge s, espec i ely. (c,d) show he numbe o days wi h episodes wi hin he wo s ai quali y index (AQI) ca ego ies o O3 and PM10, espec i ely. 4.2. Alloca ion o Episodes by AQI Ca ego y The AQI calcula ions emphasized ca ego ies h ee, ou , and i e. Acco ding o he EPA, ca ego ies ou and i e s ongly a ec ed use s in ac i e modes. No consensus was ound ega ding he impac o ca ego y h ee episodes on ac i e mode use s. E en hough he En i onmen al P o ec ion Agency does no explici ly include cyclis s as a sensi i e g oup, some o ganiza ions p omo e he use o ace masks o cyclis s iding wi hin a eas in AQI ca ego y h ee. Thus, his ca ego y was compu ed o accoun o cyclis s’ exposu e o pollu an s. Figu e 4. Ac i e mode use s and poo ozone (O 3 ) o pa icula e ma e (PM 10 ) episodes in he s udy a ea on an hou ly basis. The o de o magni ude be ween igu es mus be conside ed o p ope ly compa e hem. ( a , b ) e e o he annual numbe o MiBici ips and bus apid ansi (BRT) passenge s, espec i ely. ( c , d ) show he numbe o days wi h episodes wi hin he wo s ai quali y index (AQI) ca ego ies o O3and PM10, espec i ely. The spa ial pa e n o he BRT passenge s was simila , al hough he s a ions ha con ibu ed o he peak hou s we e clea ly dis inguished. Figu e 4b shows he h ee peak hou s a he BRT s a ions in he a ea o in luence o he Cen o and Mi a alle AQSs, ha is, be ween 8:00 and 9:00, be ween 14:00 and 15:00, and be ween 19:00 and 20:00. O he 20 million passenge s using hese s a ions, 60% we e in he a ea o in luence o he Cen o AQS and 40% in ha o he Mi a alle s a ion, i.e., 12,208,712 and 7,819,538, espec i ely. Mo eo e , i was shown ha 1,832,788 passenge s walked o each BRT s a ions in he icini y o he Mi a alle AQS be ween 8:00 and 9:00, and 2,152,791 passenge s accessed BRT s a ions a ound he Cen o AQS be ween 19:00 and 20:00. This can be explained by he ac ha he o me AQS se es a mos ly esiden ial a ea, while he la e se es a eas wi h a high densi y o wo kplaces, pa icula ly e ail ade and public se ices. 4.2. Alloca ion o Episodes by AQI Ca ego y The AQI calcula ions emphasized ca ego ies h ee, ou , and i e. Acco ding o he EPA, ca ego ies ou and i e s ongly a ec ed use s in ac i e modes. No consensus was ound ega ding he impac o ca ego y h ee episodes on ac i e mode use s. E en hough he En i onmen al P o ec ion Agency does no explici ly include cyclis s as a sensi i e g oup, some o ganiza ions p omo e he use o ace masks o cyclis s iding wi hin a eas in AQI ca ego y h ee. Thus, his ca ego y was compu ed o accoun o cyclis s’ exposu e o pollu an s.