Seismic capacity and vulnerability assessment considering ageing effects. Case study: 2 Three local Portuguese RC buildings

Depósi o de in es igación de la Uni e sidad de Se illa h ps://idus.us.es/ Es a es la e sión acep ada del a ículo publicado en: This is a accep ed manusc ip o a pape published in: Bulle in o Ea hquake Enginee ing (2020): 20/02/22024 DOI: h ps://dx.doi.o g/10.1007/s10518-020-00955-4 Copy igh : © Sp inge Na u e B.V. 2020 El acceso a la e sión publicada del a ículo puede eque i la susc ipción de la e is a. Access o he published e sion may equi e subsc ip ion. “This e sion o he a icle has been accep ed o publica ion, a e pee e iew (when applicable) and is subjec o Sp inge Na u e’s AM e ms o use, bu is no he Ve sion o Reco d and does no e lec pos -accep ance imp o emen s, o any co ec ions. The Ve sion o Reco d is a ailable online a : [h ps://dx.doi.o g/10.1007/s10518-020-00955-4]” Bulle in o Ea hquake Enginee ing Seismic capaci y and ulne abili y assessmen conside ing ageing e ec s.Case s udy: Th ee local Po uguese RC buildings --Manusc ip D a -- Manusc ip Numbe : Full Ti le: Seismic capaci y and ulne abili y assessmen conside ing ageing e ec s.Case s udy: Th ee local Po uguese RC buildings A icle Type: S.I. : Recen Ad ances in Seismic F agili y and Vulne abili y Keywo ds: RC buildings; Ageing e ec s; Pe o mance-based seismic assessmen ; Non-linea s a ic analyses; F agili y Cu es Co esponding Au ho : Ri a Ben o, Ph.D CERIS, Ins i u o Supe io Técnico, Uni e sidade de Lisboa, Po ugal Lisbon, PORTUGAL Co esponding Au ho Seconda y In o ma ion: Co esponding Au ho 's Ins i u ion: CERIS, Ins i u o Supe io Técnico, Uni e sidade de Lisboa, Po ugal Co esponding Au ho 's Seconda y Ins i u ion: Fi s Au ho : Ri a Cou o Fi s Au ho Seconda y In o ma ion: O de o Au ho s: Ri a Cou o Ma ia Vic o ia Requena-Ga cía-C uz Ri a Ben o, Ph.D An onio Mo ales-Es eban O de o Au ho s Seconda y In o ma ion: Funding In o ma ion: Abs ac : A high pe cen age o ein o ced conc e e (RC) buildings in Po ugal we e designed and buil be o e he in oduc ion o mode n seismic codes. This esea ch aims o assess he seismic capaci y and ulne abili y o RC buildings in he ci y o Lisbon. Fo ha pu pose, nonlinea s a ic p ocedu es ha e been used and agili y cu es ha e been de eloped. These buildings a e eaching he end o hei nominal li e. The e o e, ageing e ec s ha e been aken in o accoun , as well as he p esence o smoo h eba . To do so, a sensi i i y analysis has been pe o med by conside ing he chlo ide- induced co osion o he ein o cemen s eel eba and he deg ada ion o he conc e e co e . To illus a e he e ec s o ageing and he p ocedu e adop ed o he seismic agili y assessmen o old RC s uc u es, h ee RC buildings wi h mason y in ills ha e been selec ed as case s udies. They we e all buil be ween 1960 and 1980, and hey a e ep esen a i e o he cu en building s ock in Lisbon. The seismic capaci y o he buildings has been de e mined by means o nonlinea s a ic analyses o h ee- dimensional nume ical models. The N2 me hod and i s ex ended e sion ha e been conside ed o de e mine he a ge displacemen . The seismic sa e y o he buildings has been es ima ed in e ms o he demand/capaci y a io (DCR) o each e ical s uc u al elemen (columns and walls) acco ding o he bending and he shea ailu es. Then, a se o agili y cu es has been de eloped o all he buildings’ RC columns and walls o ep esen he p obabili y o eaching o exceeding he signi ican damage s a e. Resul s ha e shown ha he conc e e s eng h deg ada ion has had mo e in luence han educ ion o he eba diame e in he seismic capaci y. When conside ing s eel co osion, i has been demons a ed ha he co osion a e has educed he capaci y mo e han he ime o exposu e. I can be concluded ha ageing a ec s he seismic beha iou o RC s uc u es, inc easing he ulne abili y o hese buildings. Powe ed by Edi o ial Manage ® and P oduXion Manage ® om A ies Sys ems Co po a ion Sugges ed Re iewe s: F ancisco Ma ínez-Ál a ez, D Associa e P o esso , Pablo de Ola ide Uni e si y [email p o ec ed] Based in his expe ise. He is an expe in seismic enginee ing, ime se ies and da a mining. Daniel Cela ec Facul y o Ci il and Geode ic Enginee ing, Uni e si y o Ljubljana, Slo enia [email p o ec ed] He is an expe in seismic enginee ing assessmen , modelling unce ain ies and ein o ced conc e e buildings beha io . He has published se e al pape s on he seismic pe o mance o hese buildings as well as he simula ion o he ageing e ec s which ha e been conside ably ci ed. Powe ed by Edi o ial Manage ® and P oduXion Manage ® om A ies Sys ems Co po a ion 1 Seismic capaci y and ulne abili y assessmen conside ing ageing e ec s. Case s udy: 1 Th ee local Po uguese RC buildings 2 R. Cou o1, M.V. Requena-Ga cía-C uz2, R. Ben o1(*), A. Mo ales-Es eban2 3 4 1CERIS, Ins i u o Supe io Técnico, Uni e sidade de Lisboa, Po ugal, Add ess: A . Ro isco Pais 1049-001 Lisbon, Po ugal, 5 [email p o ec ed] ; [email p o ec ed] 6 2 Depa men o Building S uc u es and Geo echnical Enginee ing, Uni e si y o Se ille, Spain. Add ess: A . Reina Me cedes, 2, 41012, Se ille, Spain, 7 m equena[email p o ec ed], [email p o ec ed] 8 (*) Co esponding Au ho – Ri a Ben o, i a.ben o@ ecnico.ulisboa.p , Tel: +351 218 418 205 9 10 Abs ac (150-250 wo ds) 11 A high pe cen age o ein o ced conc e e (RC) buildings in Po ugal we e designed and buil be o e he in oduc ion o mode n 12 seismic codes. This esea ch aims o assess he seismic capaci y and ulne abili y o RC buildings in he ci y o Lisbon. Fo ha 13 pu pose, nonlinea s a ic p ocedu es ha e been used and agili y cu es ha e been de eloped. These buildings a e eaching he 14 end o hei nominal li e. The e o e, ageing e ec s ha e been aken in o accoun , as well as he p esence o smoo h eba . To do 15 so, a sensi i i y analysis has been pe o med by conside ing he chlo ide-induced co osion o he ein o cemen s eel eba and he 16 deg ada ion o he conc e e co e . To illus a e he e ec s o ageing and he p ocedu e adop ed o he seismic agili y assessmen 17 o old RC s uc u es, h ee RC buildings wi h mason y in ills ha e been selec ed as case s udies. They we e all buil be ween 1960 18 and 1980, and hey a e ep esen a i e o he cu en building s ock in Lisbon. The seismic capaci y o he buildings has been 19 de e mined by means o nonlinea s a ic analyses o h ee-dimensional nume ical models. The N2 me hod and i s ex ended e sion 20 ha e been conside ed o de e mine he a ge displacemen . The seismic sa e y o he buildings has been es ima ed in e ms o he 21 demand/capaci y a io (DCR) o each e ical s uc u al elemen (columns and walls) acco ding o he bending and he shea 22 ailu es. Then, a se o agili y cu es has been de eloped o all he buildings’ RC columns and walls o ep esen he p obabili y 23 o eaching o exceeding he signi ican damage s a e. Resul s ha e shown ha he conc e e s eng h deg ada ion has had mo e 24 in luence han educ ion o he eba diame e in he seismic capaci y. When conside ing s eel co osion, i has been demons a ed 25 ha he co osion a e has educed he capaci y mo e han he ime o exposu e. I can be concluded ha ageing a ec s he seismic 26 beha iou o RC s uc u es, inc easing he ulne abili y o hese buildings. 27 28 Keywo ds (max 6): 29 RC buildings; Ageing e ec s; Pe o mance-based seismic assessmen ; Non-linea s a ic analyses; F agili y Cu es. 30 31 32 1. In oduc ion 33 In ecen yea s, ein o ced conc e e (RC) buildings wi h low seismic capaci ies ha e been seismically damaged in 34 ea hquake-p one coun ies. Such is he case o he 1999 Kocaeli Ea hquake, he 1999 Chi-chi Ea hquake, he 2011 35 Eas Japan Ea hquake, he 2011 Ch is chu ch Ea hquake and he 2017 Puebla Ea hquake (Ea hquake Enginee ing 36 Resea ch Ins i u e 2020). Mo eo e , p e ious s udies s a ed ha buildings cons uc ed p io o seismic codes a e likely 37 o be damaged due o an ea hquake (Kam e al. 2011). This is because hey ha e only been g a i y load designed and 38 buil wi h an inadequa e la e al load esis ance. In ac , hey ha e been widely analysed and p o ed o be o limi ed 39 duc ili y capaci y, i.e. seismically ulne able (A e a e al. 2019). Hence, his accoun s o he need o p opose eliable 40 seismic isk assessmen and ea hquake loss models o such s uc u es. 41 42 A majo issue o adequa e seismic isk assessmen and ea hquake loss modelling is he p ope cha ac e isa ion o he 43 expec ed beha iou o hese s uc u es. In o de o ackle his issue, models conside ing he pe o mance o di e en 44 building classes (e.g. ulne abili y unc ions, agili y unc ions, e c.) ha e been employed. Building class de ini ions 45 ha e ollowed no uni o m ule. Fo ins ance, HAZUS-MH used be ween 190 and 5,300 classes (Na ional Ins i u e o 46 Building Sciences and Fede al Eme gency Managemen Agency (NIBS and FEMA) 2003), depending on how hey 47 a e coun ed. The Global Ea hquake Model (GEM) p oposed an inno a i e gene ic axonomy (Scaw ho n e al. 2013) 48 ha could classi y buildings in unlimi ed ways. Some au ho s e en de eloped models o he agili y o ulne abili y 49 o di e en classes based on indi idual buildings (Maio and Tsionis 2016). In gene al, hese buildings we e chosen 50 due o hei ep esen a i eness o a egion o in e es . In addi ion, in o de o conside he a ie y o di e en ea u es 51 o he classes, a p obabilis ic model o he building class was also es ablished. 52 53 The de e io a ion e ec o he s uc u es du ing hei li e ime has been commonly neglec ed in seismic ulne abili y 54 assessmen s udies (Rosse o and Elnashai 2005; Tsionis e al. 2011; Ca uso e al. 2019a). Howe e , a conside able 55 numbe o old RC s uc u es su e om he ageing and he deg ada ion o ma e ials, which may ad e sely a ec hei 56 s uc u al pe o mance. Among di e en de e io a ion mechanisms, co osion in ein o cemen is one o he majo 57 p oblems in he du abili y o exis ing RC s uc u es. P e ious s udies showed ha co osion can a ec he conc e e 58 co e o he ein o cemen (Al-Ha hy e al. 2011), he mechanical beha iou o he ein o cemen (Du e al. 2005) 59 Manusc ip Click he e o access/download;Manusc ip ;_BEE_AgingE ec s_S uc u e_202 Click he e o iew linked Re e ences 2 and he bond be ween he conc e e and he ein o cemen (Lundg en 2007). These co osion e ec s we e s udied in 60 old RC buildings subjec ed o seismic ac ion (Cela ec e al. 2011; Yalcine e al. 2015). None heless, acco ding o he 61 au ho s’ knowledge, he e a e ew s udies on he in luence o ageing e ec s on he seismic pe o mance o exis ing 62 RC buildings. In ac , he e is a dis inc lack o s udies ha ocus on sensi i ely analysing he in luence o hese 63 di e en pa ame e s ( he co osion a e o he ein o cemen , he ime o co osion ini ia ion, he spalling o he 64 conc e e co e and he a ec ed s uc u al elemen s) while conside ing he agili y le el. 65 66 The scope o his s udy is o in es iga e he seismic capaci y and ulne abili y o RC buildings in he ci y o Lisbon 67 h ough non-linea s a ic analyses and he de elopmen o agili y cu es. These buildings a e eaching he end o 68 hei nominal li e. The e o e, ageing e ec s ha e been aken in o accoun . To do so, a sensi i i y analysis has been 69 pe o med by conside ing he chlo ide-induced co osion o he ein o cemen s eel eba and he deg ada ion o he 70 conc e e co e . To illus a e he e ec s o ageing and he p ocedu e adop ed o he seismic agili y and ulne abili y 71 assessmen o old RC s uc u es, h ee RC buildings wi h mason y in ills ha e been selec ed as case s udies. They 72 we e all buil be ween 1960 and 1980, and hey a e ep esen a i e o he cu en building s ock in Lisbon. 73 74 75 2. Case S udies o Building S uc u es 76 In Po ugal, acco ding o he 2011 Census Su ey (Ins i u o Nacional de Es a ís ica (INE) 2012), RC buildings 77 ep esen abou 50% o he o al building s ock in he coun y. Mo eo e , 49% o hese buildings we e cons uc ed 78 wi hou adequa e seismic p o isions and based on ea ly seismic codes (p e–1983). Acco ding o Sil a e al. (2014), in 79 2011, RC buildings in Po ugal hos ed 60% o he na ional popula ion, i.e. 6,342,000 people. Fu he mo e, he seismic 80 haza d in Po ugal has been widely p o ed o be o conside able impo ance (Campos Cos a e al. 2008). Recen 81 s udies (Ama o-Mellado e al. 2017a; Ama o-Mellado e al. 2017b) ha e shown ha he maximum eco ded magni ude 82 in he a ea is 6.6 and he aul ’s maximum magni ude is 6.5, which shows an ou s anding seismic isk. 83 84 In Po ugal, he i s design codes ha p oposed some p o isions ega ding seismic ac ion we e he RSCCS (Diá io 85 do Go e no 1958) and he RSEP (Diá io do Go e no 1961). In 1983, he design code RSA (Diá io do Go e no 1983) 86 was in oduced o inally es ablish adequa e and es ic i e seismic pe o mance equi emen s. In his s udy, he 87 classi ica ion p oposed by Sil a e al. (2014) has been ollowed; buildings cons uc ed be o e 1958 ha e been 88 classi ied as p e-code (PC), buildings cons uc ed be ween 1958 and 1983 ha e been ca ego ised as mid-code (MC), 89 and inally, he ones buil a e 1983 ha e been ca ego ised as pos -code (C). 90 91 In he case o Lisbon, acco ding o he 2011 Census Su ey, he e a e 52,496 buildings and 42% o hem a e RC 92 buildings, o which 71% a e PC. In his wo k, a su ey in he a ea o “Al alade” (Fig. 1(a)) has been ca ied ou o 93 cha ac e ise he exis ing buildings, ocusing on he numbe , da e o cons uc ion and hei RC s uc u al con igu a ion. 94 This a ea is limi ed by he “A enida do B asil”, he “A enida Almi an e Gago Cou inho”, he “A enida do Campo 95 G ande” and he ain line (Fig. 1(b)). 96 97 98 (a) (b) 99 Fig. 1. Plan o Lisbon (a) indica ing Al alade’s a ea and (b) limi s. 100 101 This assessmen included 2,249 buildings, o which 28% ha e an RC s uc u e and 71% ha e a mixed mason y- 102 conc e e s uc u e. The la e buildings a e e y common in Lisbon (and Po ugal in gene al) and hey we e cons uc ed 103 3 be ween 1930 and 1960. Thei seismic pe o mance was b oadly analysed in di e en s udies (Milose ic e al. 2018, 104 2019). The o he 1% ep esen s he Gaiolei o buildings (Simões e al. 2019a, b), ga ages and some unknown buildings 105 (due o he lack o da a). The assessmen has been ca ied ou by analysing and ga he ing he da a om he bluep in s 106 a ailable a he “A qui o Municipal do Lisboa” in A cGIS (En i onmen al Sys ems Resea ch Ins i u e (ESRI) 2011). 107 108 Rega ding he RC s uc u es, he s uc u al con igu a ion o 88% o hem has been iden i ied: 46% a e amed and 109 42% a e wall- ame. Fig. 2(a) shows he dis ibu ion o he RC s uc u es iden i ied in he a ea o Al alade acco ding 110 o hei s uc u al con igu a ion: amed in blue, wall- ame in ed and unknown in yellow. Mos o he RC buildings 111 we e buil be o e 1980 (97% o all o hem), while 70% o hese we e cons uc ed be ween 1950 and 1970 (when RC 112 s uc u es s a ed o be buil ) (Fig. 2(b)). 113 114 115 (a) (b) 116 Fig. 2. Dis ibu ion o RC buildings in Al alade ega ding (a) cons uc ion da e in 3D iew and (b) iden i ica ion o 117 hei s uc u al con igu a ion in g aph o ma . 118 119 The numbe o s o eys has been iden i ied as one o he main pa ame e s o classi y he buildings in ypologies. Fig. 120 3(a) shows he dis ibu ion o he RC buildings acco ding o hei s uc u al con igu a ion and he numbe o s o eys 121 in Al alade. The majo i y o he amed s uc u es (51%) a e medium-heigh buildings o h ee o i e s o eys. In he 122 case o he wall- ame s uc u es, mos o hem (49%) a e all buildings o mo e han nine s o eys (Fig. 3(b)). A high 123 pe cen age o all he RC buildings p esen s i egula i ies in ele a ion: 57% and 67% o he amed and wall- ame 124 buildings, espec i ely. The wall- ame buildings (62%) a e also cha ac e ised by he p esence o so s o eys. This 125 has been p o ed o be one o he ypical ulne abili ies ha inc ease seismic damage in buildings (Ruiz-Pinilla e al. 126 2016). Also, he c oss-sec ion o he columns in 30% o he buildings o bo h s uc u al con igu a ions dec eases in 127 heigh . 128 129 130 (a) (b) 131 Fig. 3. Dis ibu ion o RC buildings in Al alade ega ding (a) he numbe o s o eys in 3D iew and (b) iden i ica ion 132 o hei s uc u al con igu a ion in g aph o ma . 133 134 Th ee case s udies o buildings ha e been selec ed as he mos ep esen a i e o hei s uc u al con igu a ions in he 135 a ea o Al alade (Fig. 4): one amed (Model C) and wo wall- ame buildings (Models A and B). They a e MC 136 4 buildings cons uc ed wi h ypical design de ails o p e-70s RC buildings in Lisbon as well as wi h smoo h eba . The 137 analysis o he bluep in s has e ealed ha ein o cemen is no adequa e in he join s since: i) he spacing and he 138 a io o he s i ups is no enough in columns and beams, leading o shea ailu es and plas ic hinge gene a ion; ii) he 139 longi udinal a io is insu icien in he bo om pa o he beams; iii) he longi udinal ein o cemen was designed o 140 esis code-based momen s; and i ) he sec ion o he columns dec eases in heigh which educes he s i ness o he 141 sys em. A ypical cons uc i e de ail o he beam-column o wall-beam join s is shown in Fig. 5. In Table 1, he 142 a ia ion o he ein o cemen a ios o he s uc u al elemen s o each model is lis ed. Mo eo e , he h ee buildings 143 only ha e in ills in he uppe s o eys, which leads o he gene a ion o so -s o ey mechanisms in he g ound loo 144 (Models A and B) and in he i s loo (Model C). 145 146 147 Fig. 4. S uc u al schemes and plans o he h ee models. 148 149 150 Fig. 5. Typical cons uc i e de ail o he beam-column o wall-beam join s. 151 152 Table 1. Geome ical p ope ies and minimum and maximum ein o cemen o he s uc u al elemen s o each model. 153 (T and B e e o op and bo om beam ein o cemen s, espec i ely). 154 155 Cha ac e is ic Columns Beams Walls Models A B C A B C A B C Dimensions (cm) 12.7x11.3- 28.6x27.2 23x25- 25x87 20x31- 30x40 10x31- 40x45 13x31- 25x68 10x40- 20x60 300x15- 400x25 250x25 - 5 C oss-sec ion (cm2) 143.5-777.9 575-2175 558-1240 310-1800 403-1700 440-1380 4500- 1000 6250 - Longi udinal eba (cm2) 4.04-61.68 0.63-7.29 0.31-38.79 T: 3.30-12.26 T:2.26-29.76 T: 0.63-16.46 3.41 15.78 - B:4.28-16.04 B: 3.93-33.58 B: 1.42-11.10 - T ans e sal eba (cm2) 3.18 0.18-0.71 1.58 3.2-10.0 1.48 2.11-2.47 2.45 3.15 - Spacing o s i ups (cm) 20 5-19 15 20 20-50 15-18 25 15 156 Fo each building, he mass o he s uc u es has been di ided in o dead and li e loads. The dead loads ha e included 157 he sel -weigh o he RC elemen s, mason y in ills and coa ings, i.e. 8 kN/m2. The li e loads ha e been de ined 158 acco ding o Pa -1 o Eu ocode 8 (EC8-1) (Eu opean Union 2004). The o al mass and heigh o he buildings a e 159 shown in Table 2, as well as he undamen al pe iod o ib a ion o each o he models o he ini ial si ua ion (wi hou 160 co osion, and he ein called “IS”). 161 162 Table 2. Geome ical cha ac e is ics o he buildings: mass, heigh and undamen al pe iod o ib a ion o he IS. 163 To al mass ( on) To al heigh (m) Fundamen al pe iod o ib a ion (IS) (s) X Y Model A 3677 24.6 0.90 0.87 Model B 720 15.4 0.31 0.41 Model C 1507 17.02 0.51 0.54 164 Model A was buil in he 1960s and i is an eigh -s o ey RC wall- ame, wi h a o al heigh o 24.6 m. All s o eys a e 165 3.0 m high, excep o he g ound loo , which is 3.6 m high. The s uc u e has h ee main ames in he X di ec ion, 166 wi h all he columns o ien ed in his di ec ion. In he Y di ec ion, he s uc u e has RC walls on he s ai s and a li 167 co e. The e o e, he ho izon al loads in X di ec ion a e esis ed by he ames, while in he Y di ec ion hey a e mainly 168 esis ed by he walls. The g ound loo has a so s o ey con igu a ion. 169 170 Model B was buil in he 1950s and i is a i e-s o ey RC wall- ame, wi h a o al heigh o 15.5 m. All s o eys ha e 171 he same heigh o 3.0 m; excep o he g ound loo , which is 3.5 m high. I has h ee and i e ames in he X and 172 Y di ec ion, espec i ely. The s ai s and he li a e included in he cen e o he building, su ounded by a shea wall 173 in he X di ec ion. Simila o Model A, he g ound loo has a so s o ey con igu a ion. 174 175 Model C was buil in he 1960s and i is a six-s o ey amed RC building, wi h a o al heigh o 17.0 m. The heigh o 176 all loo s is i egula . E en be ween he g ound and i s loo , a signi ican i egula i y in heigh can be ound, which 177 leads o he gene a ion o sho columns. In he plan, he s uc u al con igu a ion is conside ably symme ical in he X 178 di ec ion bu signi ican ly i egula in he Y di ec ion. Mo eo e , he columns a e all o ien ed wi h hei s ong axis in 179 he X di ec ion. I also includes a s ai case in he cen e o he building, as well as in e nal pa i ions o conside able 180 hickness and a so s o ey mechanism in he g ound loo . 181 182 183 3. Building S uc u al Modelling 184 The h ee buildings ha e been analysed wi h he OpenSees so wa e (McKenna e al. 2000). The columns, beams and 185 shea walls ha e been modelled wi h o ce-based ib e elemen s. The beha iou o he co e and co e conc e e ib es 186 has been simula ed employing he model p oposed by Popo ics (Conc e e04 ma e ial in OpenSees) (Popo ics 1973). 187 The ein o cemen s eel ib es ha e been modelled by using he uniaxial Giu e-Menego o-Pin o model (S eel02 188 ma e ial in OpenSees) (Filippou e al. 1983). Table 3E o ! Re e ence sou ce no ound. p o ides he mean alues 189 o he p ope ies o he ma e ials adop ed in his s udy. I should be men ioned ha he conc e e ensile s eng h has 190 been simula ed as 10% o he comp essi e s eng h. 191 192 Table 3. S uc u al ma e ial p ope ies. 193 Conc e e c (MPa) ɛc (‰) ɛcu (‰) Ec (GPa) Co e 28 2 200 30 Co e 28 2 4 30 S eel y (MPa) u (MPa) ɛsu (‰) Es (GPa) 6 Ribbed eba 370 36 240 210 Smoo h eba 222 216 168 126 Mason y Gw (GPa) α τc (MPa) Ew (GPa) In ills 1240 0.05 280 4092 c -comp essi e maximum s eng h; ɛc - s ain a maximum s eng h; ɛcu - ul ima e s ain; Ec - modulus o elas ici y o conc e e; y - yielding s eng h; 194 u - ul ima e s eng h; ɛsu - ul ima e s eng h; and Es - modulus o elas ici y o s eel; Gw - elas ic shea modulus; α - pos -capping deg ading b anch 195 coe icien ; τc - shea c acking s ess; Ew - modulus o elas ici y o he mason y. 196 197 The e ec s o he smoo h eba ha e been aken in o accoun by modi ying he s eel cons i u i e law as p oposed by 198 Ca uso e al. (2019c). In ills can wo sen he seismic beha iou o buildings i hey a e no egula ly dis ibu ed. In he 199 cases unde s udy, hey a e sligh ly he e ogeneously dis ibu ed in plan and ele a ion. The e o e, he in ill panels ha e 200 been conside ed and modelled by assuming he wo-diagonal uss app oach es ablished in Cela ec e al. (2012). Also, 201 he conc e e slabs p esen signi ican s i ness in all he buildings as shown in Fig. 5. The e o e, hei e ec s ha e 202 been simula ed by connec ing he RC beams by a igid diaph agm a each loo le el. Then, he masses ha e been 203 applied a he cen e o each loo . 204 205 206 4. Damage Limi S a es 207 Seismic demand has been de ined acco ding o he EC8-1 elas ic esponse spec um o soil ype A and he p o isions 208 es ablished in he Po uguese EC8-3 Annex (Ins i u o Po uguês da Qualidade 2017). As es ablished in his documen , 209 he e u n pe iods o conside in he assessmen o exis ing buildings a e: 73, 308 and 975 yea s o he damage 210 limi a ion (DL), he signi ican damage (SD) and he nea -collapse (NC) limi s a es, espec i ely. Acco ding o each 211 o he limi s a es, he e e ence g ound accele a ion (agR) is hen mul iplied by a coe icien . Only he esponse 212 spec um o he Type 1 seismic ac ion has been aken in o accoun , he e o e, he PGA de ined o Lisbon (Zone 1.3) 213 is 0.15g (Ins i u o Po uguês da Qualidade 2017). 214 215 Seismic sa e y o he buildings has been es ima ed in e ms o he demand/capaci y a io (DCR) o each e ical 216 s uc u al elemen (columns and walls). The p ocedu es es ablished in Pa 3 o Eu ocode 8 (EC8-3) (Eu opean Union 217 2005) ha e been ollowed o de e mine he limi s a es. The SD s a e has been used o he seismic assessmen , as 218 equi ed in he EC8-3 Po uguese Annex o esiden ial buildings. I has been assumed ha he mos c i ical e ical 219 elemen , conside ing a duc ile o agile ailu e, con ols he s a e o he s uc u e. To do so, he bending and shea 220 capaci ies o he SD limi s a e ha e been ob ained h ough Equa ion A.1 and A.12 om he EC8-3. Then, as 221 es ablished in he documen , he ul ima e cho d o a ion (θu) has been mul iplied by 0.75 and 0.58 o de e mine he 222 ailu e in columns and walls, espec i ely. Mo eo e , he educ ion coe icien s p o ided by he EC8-3 ha e been used 223 o ake in o accoun he smoo h eba . 224 225 Se e al me hods we e p oposed o ake in o accoun he deg ada ion o he capaci y o buildings o e ime, i.e. he 226 ageing e ec s (Be o e al. 2009). They ha e been conside ed by means o : i) he educ ion o he longi udinal and he 227 ans e sal eba sec ion; ii) he educ ion o he ul ima e eba de o ma ion; and iii) he deg ada ion o he conc e e 228 co e . In his s udy, he educ ion o he ul ima e eba de o ma ion has been neglec ed due o he negligible alues 229 ob ained in expe imen al es s (Cela ec e al. 2011). The e o e, only wo app oaches ha e been ollowed: he simpli ied 230 model o ake in o accoun he co osion o he longi udinal and ans e se ein o cemen in he ex e io (i.e. exposed) 231 elemen s; and he model o simula e he spalling and deg ada ion o he conc e e co e . I has only been assumed ha 232 he co osion in luences he diame e o he s eel eba , e en hough i can also a ec he bond s ess be ween he 233 conc e e and he s eel. 234 235 236 5. Sensi i i y Analysis 237 The capaci y o each o he models conside ed in he s udy has been ob ained by means o non-linea s a ic analyses 238 using he OpenSEEs so wa e and handling hei ou pu s in Ma lab (The Ma hWo ks Inc. 2018). Fi s , a modal analysis 239 has been ca ied ou o ob ain he pe iods o ib a ion. Then, a displacemen -con ol in eg a o has been used o scale 240 he o ces o each a displacemen . The analyses ha e been ob ained o he ±X and ±Y di ec ions, and wo load 241 pa e ns ha e been conside ed: modal, p opo ional o he undamen al mode o ib a ion; and uni o m, p opo ional 242 o he masses o he loo s. As in Requena-Ga cía-C uz e al. (2019), only he esul s co esponding o he modal 243 pa e n ha e been aken in o accoun , since hey ha e been mo e es ic i e. Despi e he sligh asymme y in he Y 244 di ec ion o all models, no signi ican di e ences ha e been ob ained be ween he posi i e and he nega i e 245 13 con igu a ion: he e a e ou walls in he Y di ec ion and only wo in he X di ec ion. In he case o Model B, his 442 in luence has been mo e p obable due o he walls’ co osion capaci y and he modal pa e n adop ed o he 443 dis ibu ion o la e al o ces. In he X di ec ion, he modal pa e n has been mo e es ic i e o he in e media e 444 s o eys, which canno be co oded. In he Y di ec ion, he modal pa e n is mo e demanding o he walls on he g ound 445 loo , which a e hose ha can be a ec ed by co osion. Acco ding o Fig. 14. Compa ison be ween Simula ion A and 446 Simula ion D o Models A and B (only models including walls) in he (a) X and (b) Y di ec ions.Fig. 14, Model B 447 does no comply wi h he EC8-1 p o isions in any o he simula ions. Fu he mo e, conside ing he di e en co osion 448 a es, he di e ence ob ained is abou 1% in he X di ec ion o bo h models and in he Y di ec ion 4% and 3% o 449 Model A and Model B, espec i ely. 450 451 452 (a) (b) 453 Fig. 13. No malised pusho e cu es o Simula ion D in he (a) X and (b) Y di ec ions. 454 455 456 (a) 457 458 (b) 459 Fig. 14. Compa ison be ween Simula ion A and Simula ion D o Models A and B (only models including walls) in 460 he (a) X and (b) Y di ec ions. 461 462 Fig. 15(a) and (b) show he dis ibu ion o damage o he ini ial si ua ion in he e ical elemen s in he X and Y 463 di ec ions, espec i ely. Di e en le els ha e been de e mined acco ding o he DCR o he lexu al and shea 464 ailu es. I has been assumed ha he signi ican damage limi s a e is a ained when he s o ey shea (Vdemand) o he 465 s o ey cho d o a ion (θdemand) is equal o o g ea e han 1 (Vdemand/VR ≥ 1 and θdemand/θum,SD≥1, espec i ely). Fo he 466 lexu al damage, wo mo e le els ha e been de e mined o isualise he elemen s ha almos eached he DCR o he 467 14 SD limi s a e (θdemand/θy>0.9) and he elemen s ha al eady exceeded he cho d o yielding (θdemand/θy>1). In he case 468 o Model A, mos o he walls will be damaged due o lexu al ailu e. In bo h di ec ions, he walls in he in e media e 469 loo s will be conside ably damaged. E en in he Y di ec ion, mos o hem will collapse due o lexu al ailu e. In 470 addi ion, all he columns will be close o yielding. In he case o Model B, in he X di ec ion, he in e media e walls 471 will collapse due o lexu al ailu e while he g ound loo wall will collapse due o shea ailu e. In he Y di ec ion, 472 only he g ound loo elemen s will be damaged, and some will e en collapse due o bo h lexu al and shea ailu es. 473 Mos o he damaged elemen s a e in he g ound loo due o he so s o ey mechanism. In he case o Model C, in 474 he X di ec ion, he columns o he so s o ey loo will each and exceed he yielding. In he Y di ec ion, some o he 475 columns will be damaged due o lexu al yielding and shea ailu e. I should be men ioned ha he ageing e ec s do 476 no modi y he elemen ’s ype o collapse. Mo eo e , i has been e i ied ha , o e e y single simula ion, he ypes 477 o ailu e and he mo e ulne able elemen s ha e no changed compa ed o he ini ial si ua ion. 478 479 480 (a) 481 482 (b) 483 Fig. 15. Damage in e ical elemen s o he ini ial si ua ion in he X (a) and Y (b) di ec ions. 484 485 Finally, he agili y cu es o he mos demanding simula ion and he wo s di ec ion ha e been ob ained o each 486 model. Fig. 16 shows he agili y cu es in he Y di ec ion o Models A (b), B (c) and C (d). Fo Models A and B, 487 he agili y cu es ha e been ob ained o Simula ion D; while o Model C, Simula ion A has been conside ed. In 488 Fig. 16(a), e e y cu e has been plo ed o compa e he esul s. Fo he ollowing igu es, each model’s agili y cu e 489 has been p esen ed wi h an app op ia e scale. The cu es wi h he ageing e ec ha e 20% mo e p obabili y o 490 exceedance o Model A and Model B, and only 0.7% o Model C. These cu es lead o he same conclusions as 491 p e iously men ioned: i) he ulne abili y o Model B is conside able; ii) Model A has p esen ed he bes seismic 492 beha iou o all he cases; and iii) o Model C, he in luence o co osion has been ba ely obse ed. 493 15 Ne e heless, i is wo h highligh ing ha i highe alues o dispe sion had been adop ed o he s uc u es whe e 494 co osion was conside ed, wi h highe amoun s o b ds as he ime in e al Δ and he le els o he co osion a e 495 inc ease, he seismic pe o mance o he RC buildings could become much wo se. 496 497 498 (a) (b) 499 500 501 (c) (d) 502 Fig. 16. F agili y cu es o he mos demanding simula ion and o he wo s di ec ion. 503 504 505 8. Conclusions 506 This pape has ocused on he assessmen o he seismic ulne abili y o RC buildings in he ci y o Lisbon h ough 507 he de ini ion o analy ical damage agili y cu es. The ageing e ec s ha e been aken in o accoun , as well as he 508 p esence o smoo h eba . To do so, a sensi i i y analysis has been pe o med by conside ing he chlo ide-induced 509 co osion o he eba and he deg ada ion o he conc e e co e . To illus a e he e ec s o ageing and he p ocedu e 510 adop ed in his s udy, h ee RC buildings wi h mason y in ills ha e been selec ed as case s udies. They we e all buil 511 be ween 1960 and 1980, and hey a e ep esen a i e o he cu en building s ock o Lisbon. 512 513 Resul s ha e shown ha he numbe o walls and he o ien a ion o he columns a e decisi e in de e mining he 514 capaci y o RC buildings. In addi ion, his is ela ed o he p ema u e b i le shea collapse, as was clea ly iden i ied 515 a he beginning o he pusho e cu es in all cases. The e o e, a e o i ing scheme ocused on sol ing he shea 516 p oblems should be aken in o accoun o imp o e he seismic pe o mance o hese RC buildings. Besides, he 517 in luence o he in ills has been mo e c i ical in he case o he model wi hou walls, i.e. in he amed building. This 518 is because, o he o he models, he buildings’ beha iou basically depends on he shea walls. Conce ning he models’ 519 beha iou , Model B has p esen ed he wo s beha iou , ollowed by Model A and hen C. This is no only due o he 520 s uc u al con igu a ion o he buildings, bu also due o he o al numbe o elemen s exposed o he ex e io 521 en i onmen and he e o e a ec ed by ageing. 522 523 By conside ing he ageing e ec s, he seismic pe o mance o hese buildings can wo sen by up o 20%, as concluded 524 by he agili y analysis. Among he aspec s analysed, i has been obse ed ha conc e e s eng h deg ada ion has 525 signi ican ly educed he capaci y o all he models when compa ed o he educ ion o he eba diame e , i.e. s eel 526 16 co osion app oach. Mo eo e , in he hypo hesis o he s eel co osion app oach, co osion a e has had mo e in luence 527 han he ime o exposu e. In conclusion, ageing a ec s he seismic beha iou o RC s uc u es, inc easing he 528 ulne abili y o hese buildings. 529 530 531 Acknowledgmen s 532 The i s au ho would like o acknowledge he inancial suppo o Fundação pa a a Ciência e a Tecnologia (FCT, 533 Minis é io da Educação e Ciência, Po ugal) h ough he schola ship BL105/2019 by he FCT Resea ch P og am: 534 Mi Risk - F amewo k o seismic isk educ ion eso ing o cos -e ec i e e o i ing solu ions, POCI-01-01456- 535 Fede -031865. In addi ion, he schola ship p o ided by he VI-PPI o he Uni e si y o Se ille is acknowledged. 536 537 538 Re e ences 539 Al-Ha hy AS, S ewa MG, Mulla d J (2011) Conc e e co e c acking caused by s eel ein o cemen co osion. 540 63:655–667. h ps://doi.o g/10.1680/mac .2011.63.9.655 541 Ama o-Mellado JL, Mo ales-Es eban A, Asencio-Co és G, Ma ínez-Ál a ez F (2017a) Compa ing seismic 542 pa ame e s o di e en sou ce zone models in he Ibe ian Peninsula. Tec onophysics 717:449–472. 543 h ps://doi.o g/10.1016/J.TECTO.2017.08.032 544 Ama o-Mellado JL, Mo ales-Es eban A, Ma ínez-Ál a ez F (2017b) Mapping o seismic pa ame e s o he Ibe ian 545 Peninsula by means o a geog aphic in o ma ion sys em. Cen Eu J Ope Res. h ps://doi.o g/10.1007/s10100- 546 017-0506-7 547 A e a CA, Ca illo J, A chbold J, e al (2019) Response o Mid-Rise Rein o ced Conc e e F ame Buildings o he 548 2017 Puebla Ea hquake. Ea hq Spec a 35:1763–1793. h ps://doi.o g/10.1193/061218EQS144M 549 Be o L, Vi aliani R, Sae a A, Simioni P (2009) Seismic assessmen o exis ing RC s uc u es a ec ed by deg ada ion 550 phenomena. S uc Sa 31:284–297. h ps://doi.o g/10.1016/j.s usa e.2008.09.006 551 Campos Cos a A, Sousa ML, Ca alho A (2008) Seismic Zona ion o Po uguese Na ional Annex o Eu ocode 8. 8– 552 15 553 Ca uso C (2019) Seismic e o i op ions o an old RC wall- ame building in Lisbon: impac on loss es ima ion and 554 cos -bene i analysis. Is i u o Técnico Supe io , Uni e sidade de Lisboa, Po uga 555 Ca uso C, Ben o R, Cas o JM (2019a) A con ibu ion o he seismic pe o mance and loss assessmen o old RC wall- 556 ame buildings. Eng S uc 197:109369. h ps://doi.o g/10.1016/j.engs uc .2019.109369 557 Ca uso C, Ben o R, Cas o JM (2019b) Seismic isk assessmen o an old RC ame-wall building. In: Symposium 558 Guima ães 2019 Towa ds a Resilien Buil En i onmen –Risk and Asse Managem (IABSE2019). Guima ães, 559 Po ugal, pp 1480–1487 560 Ca uso C, Ben o R, Sousa R, Co eia AA (2019c) Modelling s ain pene a ion e ec s in RC walls wi h smoo h s eel 561 ba s. Mag Conc Res 71:894–906. h ps://doi.o g/10.1680/jmac .18.00052 562 Cela ec D, Ricci P, Dolšek M (2012) The sensi i i y o seismic esponse pa ame e s o he unce ain modelling 563 a iables o mason y-in illed ein o ced conc e e ames. Eng S uc 35:165–177. 564 h ps://doi.o g/10.1016/j.engs uc .2011.11.007 565 Cela ec D, Vam a sikos D, Dolšek M (2011) Simpli ied es ima ion o seismic isk o ein o ced conc e e buildings 566 wi h conside a ion o co osion o e ime. Bull Ea hq Eng 9:1137–1155. h ps://doi.o g/10.1007/s10518-010- 567 9241-3 568 Diá io do Go e no (1958) Regulamen o de Segu ança das Cons uções Con a os Sismos (RSCCS), Dec e o-Lei 569 n◦41658, Lisbon, Po ugal [In Po uguese] 570 Diá io do Go e no (1961) Regulamen o de Solici ações em Edi ício e Pon es (RSEP), Dec e o-Lei n. 44041, Lisbon, 571 Po ugal [In Po uguese]. 572 Diá io do Go e no (1983) Regulamen o de Segu ança e Acções pa a Es u u as de Edi ícios e Pon es (RSA), Dec e o- 573 Lein. 235/83, Lisbon, Po ugal [In Po uguese] 574 Dolšek M, Faj a P (2005) Simpli ied non-linea seismic analysis o in illed ein o ced conc e e ames. Ea hq Eng 575 S uc Dyn 34:49–66. h ps://doi.o g/10.1002/eqe.411 576 Du YG, Cla k LA, Chan AHC (2005) Residual capaci y o co oded ein o cing ba s. Mag Conc Res 57:135–147. 577 h ps://doi.o g/10.1680/mac .2005.57.3.135 578 Ea hquake Enginee ing Resea ch Ins i u e (2020) Lea ning om Ea hquakes Reconnaissance A chi e. 579 h ps://www.ee i.o g/p ojec s/lea ning- om-ea hquakes-l e/l e- econnaissance-a chi e/ 580 En i onmen al Sys ems Resea ch Ins i u e (ESRI) (2011) A cGIS Desk op: Release 10 581 Eu opean Union (2004) Eu ocode 8: Design o s uc u es o ea hquake esis ance. Pa 1: Gene al ules, seismic 582 17 ac ions and ules o buildings. Belgium 583 Eu opean Union (2005) Eu ocode-8: Design o s uc u es o ea hquake esis ance. Pa 3: Assessmen and e o i ing 584 o buildings. Belgium 585 Filippou FE, Popo EP, Be e o V V (1983) E ec s o bond de e io a ion on hys e e ic beha io o ein o ced conc e e 586 join s. Be keley, Cali o nia 587 Ins i u o Nacional de Es a ís ica (INE) (2012) “Censos 2011 Resul ados De ini i os - Po ugal” , I.P. Lisboa [In 588 Po uguese]. h p://censos.ine.p / 589 Ins i u o Po uguês da Qualidade (2017) Eu ocódigo 8: P oje o de es u u as pa a esis eência aos sismos. Pa e 3: 590 A aliaçao e eabili açao de edi icios 591 Kam WY, Pampanin S, Elwood K (2011) Seismic pe o mance o ein o ced conc e e buildings in he 22 Feb ua y 592 Ch is chu ch (Ly el on) ea hquake. Bull New Zeal Soc Ea hq Eng 44:239–278. 593 h ps://doi.o g/10.5459/bnzsee.44.4.239-278 594 Lundg en K (2007) E ec o co osion on he bond be ween s eel and conc e e: an o e iew. Mag Conc Res 59:447– 595 461. h ps://doi.o g/10.1680/mac .2007.59.6.447 596 Maio R, Tsionis G (2016) Seismic agili y cu es o he Eu opean building s ock: Re iew and e alua ion o 597 analy ical agili y cu es 598 McKenna F, Fen es GL, Sco MH (2000) OpenSees: Open sys em o ea hquake enginee ing simula ion 599 Milose ic J, Ca a i S, Ben o R (2018) Sensi i i y analysis o he seismic pe o mance o ancien mixed mason y-RC 600 buildings in Lisbon. In J Mason Res Inno 3:108–154. h ps://doi.o g/10.1504/IJMRI.2018.092459 601 Milose ic J, Ca a i S, Ben o R (2019) De ini ion o agili y cu es h ough nonlinea s a ic analyses: p ocedu e and 602 applica ion o a mixed mason y-RC building s ock. Bull Ea hq Eng. h ps://doi.o g/10.1007/s10518-019- 603 00694-1 604 Na ional Ins i u e o Building Sciences and Fede al Eme gency Managemen Agency (NIBS and FEMA) (2003) 605 Mul i-haza d Loss Es ima ion Me hodology, Ea hquake Model, HAZUS MH Technical Manual, Fede al 606 Eme gency Managemen Agency. Washing on, D. C. 607 Pan azopoulou SJ, Papoulia KD (2001) Modeling Co e -C acking due o Rein o cemen Co osion in RC S uc u es. 608 J Eng Mech 127:342–351. h ps://doi.o g/10.1061/(ASCE)0733-9399(2001)127:4(342) 609 Popo ics S (1973) A nume ical app oach o he comple e s ess-s ain cu e o conc e e. Cem Conc Res 3:583–599. 610 h ps://doi.o g/10.1016/0008-8846(73)90096-3 611 Requena-Ga cía-C uz MV, Mo ales-Es eban A, Du and-Ney a P, Es ê ão JMC (2019) An index-based me hod o 612 e alua ing seismic e o i ing echniques. Applica ion o a ein o ced conc e e p ima y school in Huel a. PLoS 613 One 14:e0215120. h ps://doi.o g/10.1371/jou nal.pone.0215120 614 Rosse o T, Elnashai A (2005) A new analy ical p ocedu e o he de i a ion o displacemen -based ulne abili y 615 cu es o popula ions o RC s uc u es. Eng S uc 27:397–409. h ps://doi.o g/10.1016/j.engs uc .2004.11.002 616 Ruiz-Pinilla JG, Adam JM, Pé ez-Cá cel R, e al (2016) Lea ning om RC building s uc u es damaged by he 617 ea hquake in Lo ca, Spain, in 2011. Eng Fail Anal 68:76–86. 618 h ps://doi.o g/10.1016/J.ENGFAILANAL.2016.05.013 619 Scaw ho n C, Cha leson A, Allen L, e al (2013) GEM global ea hquake model GEM Building Taxonomy Ve sion 620 2.0 exposu e modelling 621 Sil a V, C owley H, Va um H, e al (2014) In es iga ion o he cha ac e is ics o Po uguese egula momen - ame 622 RC buildings and de elopmen o a ulne abili y model. Bull Ea hq Eng 13:1455–1490. 623 h ps://doi.o g/10.1007/s10518-014-9669-y 624 Simões AG, Ben o R, Lagoma sino S, e al (2019a) Seismic assessmen o nine een h and wen ie h cen u ies URM 625 buildings in Lisbon: s uc u al ea u es and de i a ion o agili y cu es. Bull Ea hq Eng. 626 h ps://doi.o g/10.1007/s10518-019-00618-z 627 Simões AG, Ben o R, Lagoma sino S, e al (2019b) F agili y Func ions o Tall URM Buildings a ound Ea ly 20 h 628 Cen u y in Lisbon. Pa 1: Me hodology and Applica ion a Building Le el. In J A chi He i 1–24. 629 h ps://doi.o g/10.1080/15583058.2019.1618974 630 The Ma hWo ks Inc. (2018) MATLAB and S a is ics Toolbox Release 2018b 631 Tsionis G, Papailia A, Fa dis M (2011) Analy ical F agili y Func ions o Mason y Buildings and Building Agg ega es 632 – UPAT me hodology 633 Yalcine H, Sensoy S, E en O (2015) Seismic Pe o mance Assessmen o a Co oded 50-Yea -Old Rein o ced 634 Conc e e Building. J S uc Eng 141:05015001. h ps://doi.o g/10.1061/(ASCE)ST.1943-541X.0001263 635 636