Jou nal o Building Enginee ing
LONG-TERM ENVIRONMENTAL MONITORING FOR PREVENTIVE
CONSERVATION OF EXTERNAL HISTORICAL PLASTERWORKS
--Manusc ip D a --
Manusc ip Numbe : JBE-D-21-03287R2
A icle Type: Resea ch Pape
Sec ion/Ca ego y: Building ma e ials
Keywo ds: plas e wo k; empe a u e; ela i e humidi y; a chi ec u al he i age; p ese a ion
Co esponding Au ho : VICENTE FLORES-ALÉS, Ph.D.
Uni e sidad de Se illa
SEVILLE, SPAIN
Fi s Au ho : Ma a To es-González, PhD aspi an
O de o Au ho s: Ma a To es-González, PhD aspi an
Ca los Rubio-Bellido, Ph. D.
Da id Bien enido-Hue as, Ph. D.
Juan Manuel Alducin-Ochoa
VICENTE FLORES-ALÉS, Ph.D.
Abs ac : In his s udy, an analysis o he en i onmen al condi ions in a Csa clima ic zone o he
conse a ion o plas e wo k o he Real Alcáza o Se ille is ca ied ou . The
measu emen s ob ained om in-si u moni o ing a e compa ed wi h he measu emen s
p o ided by AEMET (S a e Me eo ological Agency in Spain) du ing he e e ence yea
and he s udy is comple ed by es ima ing u u e en i onmen al condi ions using wo
al e na i e app oaches: a mo phing p ocess om he EPW o he clima ic zone and he
applica ion o M5P da a mining algo i hms.
An op imal empe a u e ange is es ablished o he conse a ion o he plas e wo k
ha p e en s hei dehyd a ion o he eezing o wa e pa icles con ained. The
ans o ma ion o gypsum in o bassani e, he isks associa ed wi h exposu es o high
ela i e humidi y and he consequences o he sligh hyg oscopici y o he ma e ial and
he en i onmen al condi ions ha mus be de eloped o a o he g ow h o mold on
he su ace o he c acking o polych omies ha embellish hese plas e deco a ions on
nume ous occasions a e analyzed. The esul s ob ained allow us o es ablish
p e en i e conse a ion measu es no only on he plas e wo k bu also on he Real
Alcáza o Se ille and ha a chi ec u al he i age loca ed in he sub opical d y-summe
clima e.
Sugges ed Re iewe s: Da id Sanz, Ph. D.
Lec u e , Poly echnic Uni e si y o Mad id: Uni e sidad Poli ecnica de Mad id
[email p o ec ed]
He is an expe esea ch abou he i age and adi ional ma e ials. He is specialized in
he s udy o
gypsum.
Ana Sil a, Ph. D.
Resea ch, Uni e si y o Lisbon Highe Technical Ins i u e: Uni e sidade de Lisboa
Ins i u o Supe io Tecnico
[email p o ec ed]
D a. Sil a is a esea che wi h an ex ensi e expe ience in building se ice li e
p edic ion and du abili y.
Opposed Re iewe s:
Response o Re iewe s: Dea edi o :
We ha e ecei ed he second e ision o ou manusc ip JBE-D-21-03287R1 "LONG-
TERM ENVIRONMENTAL MONITORING FOR PREVENTIVE CONSERVATION OF
EXTERNAL HISTORICAL PLASTERWORKS".
Powe ed by Edi o ial Manage ® and P oduXion Manage ® om A ies Sys ems Co po a ion
#Re iewe 1 now conside i in condi ions o accep ance, bu he e iew o
#Re iewe 2 is he same ha we ecei ed p e iously- i means he same commen s
and he same pd a ached o mino commen s ha we conside ed in he second
e sion o ou manusc ip . As we indica ed in he p e ious e ision, all he sugges ions
ha e been aken in o conside a ion. The ex was ew i ing o imp o e he
comp ehension and highligh ed in yellow colou
Powe ed by Edi o ial Manage ® and P oduXion Manage ® om A ies Sys ems Co po a ion
Jou nal o Building Enginee ing
Dea Edi o :
Le me in oduce mysel , my name is Vicen e Flo es-Alés and I am a Ma e ial Science
Lec u e a he Uni e si y o Se illa (Spain). I submi he manusc ip en i led “LONG-
TERM MONITORING FOR PREVENTIVE CONSERVATION OF HISTORICAL
PLASTERWORKS” o you conside a ion in o de o be published in Jou nal o
Building Enginee ing.
.
The aim o his pape is e alua ing he in luence o en i onmen al condi ions on he
conse a ion o his o ical plas e wo k. The chosen case s udy is he Real Alcáza o
Se ille and was based on (i) he annual moni o ing o he en i onmen al pa ame e s
Tempe a u e and Rela i e Humidi y, (ii) he es ima ion o hou ly alues in u u e
scena ios using a i icial in elligence and (iii) he use o he h eshold alues o he
conse a ion o plas e wo k de ined in he li e a u e.
Look o wa d o you a ou able conside a ion.
Mos since ely,
Vicen e Flo es Alés
Au ho s: Ma a To es-González, Ca los Rubio-Bellido, Da id Bien enido-Hue as, Juan
Manuel Alducin-Ochoa, Vicen e Flo es-Alés(*)
[*Co esponding au ho ]
Mailing add ess: Vicen e Flo es Alés, E.T.S.I.E., A da/ Reina Me cedes, nº4A, 41012-Se illa
ESPAÑA. E-mail add ess: lo [email p o ec ed] Phone numbe : 34 954556656 Fax: 34 954556691
Co e Le e
REVIEWER #1:
Thank you o you commen s.
REVIEWER #2:
As we indica ed in he p e ious e ision, all he sugges ions ha e been aken in o conside a ion.
The ex was ew i ing o imp o e he comp ehension and highligh ed in yellow colou
Response o Re iewe s (R2)
· The in luence o empe a u e and ela i e humidi y in he p ese a ion o
plas e wo k.
· Compa a i e analysis be ween di e en app oaches.
· Clima e change will a ec he main enance o ma e ials.
· Fu u e es ima ions allow o es ablish mi iga ion ac ions on a chi ec u al
he i age.
Highligh s
LONG-TERM ENVIRONMENTAL MONITORING FOR PREVENTIVE CONSERVATION OF
EXTERNAL HISTORICAL PLASTERWORKS
Ma a To es-González a,b *, Ca los Rubio-Bellido a, Da id Bien enido-Hue as a, JM Alducin-
Ochoa a, V. Flo es-Alés a,
a Depa men o A chi ec u al Cons uc ion II, Uni e si y o Se ille
b Depa men o Ci il Enginee ing, A chi ec u e and Geo esou ces, Uni e si y o Lisbon
* Co esponding au ho : m o [email protected] , A Reina Me cedes 4, Se ille, 41012
KEYWORDS: plas e wo k; empe a u e; ela i e humidi y; a chi ec u al he i age; p ese a ion;
ABSTRACT:
In his s udy, an analysis o he en i onmen al condi ions in a Csa clima ic zone o he
conse a ion o plas e wo k o he Real Alcáza o Se ille is ca ied ou . The measu emen s
ob ained om in-si u moni o ing a e compa ed wi h he measu emen s p o ided by AEMET
(S a e Me eo ological Agency in Spain) du ing he e e ence yea and he s udy is comple ed
by es ima ing u u e en i onmen al condi ions using wo al e na i e app oaches: a mo phing
p ocess om he EPW o he clima ic zone and he applica ion o M5P da a mining algo i hms.
An op imal empe a u e ange is es ablished o he conse a ion o he plas e wo k ha
p e en s hei dehyd a ion o he eezing o wa e pa icles con ained. The ans o ma ion o
gypsum in o bassani e, he isks associa ed wi h exposu es o high ela i e humidi y and he
consequences o he sligh hyg oscopici y o he ma e ial and he en i onmen al condi ions
ha mus be de eloped o a o he g ow h o mold on he su ace o he c acking o
polych omies ha embellish hese plas e deco a ions on nume ous occasions a e analyzed.
The esul s ob ained allow us o es ablish p e en i e conse a ion measu es no only on he
plas e wo k bu also on he Real Alcáza o Se ille and ha a chi ec u al he i age loca ed in he
sub opical d y-summe clima e.
Manusc ip File [Fo Re ision, Please upload clean e sion o
Re ised manusc ip ]
Click he e o iew linked Re e ences
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
1. INTRODUCTION
1.1 The mo-hyg ome ic condi ions and he conse a ion o he a chi ec u al he i age.
The p e en i e conse a ion o he i age is c ucial so ha u u e gene a ions can know he
iden i y ai s o hei cul u e [1]. The e o e, he moni o ing and analysis o speci ic
en i onmen al condi ions [2] and hei impac on a chi ec u al he i age, as well as u u e
ulne abili ies due o global wa ming [3], a e cu en ly a challenge.
The e a e nume ous s udies ha analyze bo h he impac o en i onmen al condi ions on
a chi ec u al he i age [4,5] and he condi ioning o in e io spaces o a o he p ese a ion o
objec s (e.g., objec s ha belong o exhibi ions) [6,7]. In bo h cases, he e a e na ional and
in e na ional guidelines and s anda ds ha es ablish he in e io ai condi ioning condi ions o
hese buildings o such pu poses, such as UNI 10829 S anda d [8], UNE EN 15757 [9], ASHRAE
STANDARD 55-2020[10] o he B i ish PAS 198 [11]. Likewise, EN 15759-1 [12] also conside s
he possibili y o using hea ing sys ems o achie e he mal com o and o he op imal
conse a ion o he i age elemen s.
In addi ion o in e na ional s anda ds, nume ous s udies ha e ca ied ou he mo-hyg ome ic
moni o ing o a chi ec u al he i age in ecen decades, esul ing in a e y use ul ool o assess
he s a e o conse a ion o he building and o know he causes o he deg ada ion in o de o
de e mine a u u e es o a ion in e en ion o o es ablish p e en i e conse a ion c i e ia,
uling ou he use o des uc i e echniques. Thus, long- e m s udies o up o 20 yea s du a ion
[13,14] and s udies in which moni o ing is limi ed o one yea [15] ha e been published. In his
ega d, i is common o e alua e a leas annual moni o ing pe iods ha a e complemen ed by
he mal models on which di e en es ima es o en i onmen al condi ions a e es ablished.
En i onmen al moni o ing and simula ion ools make i possible o es ablish es ima es o
di e en ac o s, such as he modi ica ion o ope a ional pa e ns, he mal com o [16-18], o
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
ene gy conse a ion measu es [19]. Howe e , he impac o clima e change, he basis o
nume ous in es iga ions ela ed o ci ies [20,21] has been spa sely discussed in e e ence o
p e en i e conse a ion o he i age and long- e m ad e se aspec s, which could become e y
signi ican [22,23].
1.2 Rela i e humidi y and empe a u e in luence in he p ese a ion o his o ical
plas e wo k.
The e a e s udies abou how ambien condi ions could a ec cons uc i e ma e ials [24–28] bu
a e insu icien wi h espec o gypsum o plas e wo k. In his sense, i is wo h highligh ing he
impo ance o ollow-up campaigns o iden i y possible isks o he conse a ion o he
plas e wo k, p e en i e e sible damage and ad ance conse a ion s a egies. Howe e , o
accoun o all he possible damage mechanisms, a comp ehensi e me hodology should be
used o e alua e some isk indexes based on ambien empe a u e (T) and ela i e humidi y
(RH) measu emen s.
Gypsum, due o i s po ous mic os uc u e o med by c ys alline g oups o hyd a ed calcium
sul a e, can inco po a e wa e molecules inside, in he o m o wa e apo o e en in he o m
o liquid wa e , when en i onmen al condi ions a o capilla y condensa ion. Fu he mo e,
gypsum has he abili y o elimina e his wa e when en i onmen al condi ions change, educing
he con en o wa e apo in con ac wi h he ma e ial [29].
Va ia ions in T and RH can induce phase ansi ions. In gene al, gypsum (CaSO4·2H2O)
dehyd a ion occu s in d y en i onmen s [30], which is he eason why empe a u e below 40°C
is desi able o he co ec p ese a ion o plas e wo k. Howe e , he e may be si ua ions o
high empe a u es combined wi h high ela i e humidi y in which he e a e no al e a ions in he
mine al phases o he gypsum [31]. In his ega d, Winkle & Wilhelm (1970) indica ed he
ela ionship ha mus exis be ween ambien empe a u e and ela i e humidi y o ans o m
gypsum (CaSO4·2H2O) in o bassani e (CaSO4·0.5H2O) [31], In his case, he ma e ial is likely o
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
lose s abili y i exposu e o hese un a o able en i onmen al condi ions is p olonged o e ime
[32] (Fig. 1-A, 1-B) .
Fig. 1 (A) T-HR a io o gypsum-bassani e ans o ma ion a 1 a mosphe e p essu e [31] and (B) Rela ionship
be ween he ela i e humidi y and he mois u e con en o he gypsum [33].
Howe e , his ans o ma ion is pa o a e e sible p ocess in plas e wo k, which means ha a
ehyd a ion would allow he eco e y o he ini ial p ope ies as long as he dehyd a ion had
no been p olonged in ime un il i emained cons an , due o a change in clima ic condi ions,
causing he mechanical weakening o he plas e wo k [32].
Addi ionally, gypsum has low hyg oscopici y [33,34] and medium-low solubili y [35] ha do no
ep esen a di ec isk in he p ese a ion o plas e wo k [36]. Howe e , i is impo an o
con ol he en i onmen al condi ions because he condensa ion o mois u e on he su ace o
he plas e s a o s he al e a ion o he polych ome binde s and hei possible de achmen [33].
The adso p ion hyg oscopici y o gypsum wi h 80% RH is only 4 g/kg (0.4%) [33], eaching a
high adso p ion hyg oscopici y only by capilla y sa u a ion (99% RH) [34]. Likewise, humidi y
a o s he nes ing o bio ic o ganisms and he al e a ion de i ed om hei ac i i y [33].
The e o e, an RH close o o g ea e han 90% weakens he s uc u e o he plas e wo k and
a o s he p oli e a ion o mic oo ganisms and biological agen s, as well as he ixa ion o
en i onmen al pollu an s.[37–39] (Fig. 01-B).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
2.2. Measu emen p ocedu e
To de e mine ambien empe a u e and ela i e humidi y condi ions, a Lasca
he mohyg ome e , model EL-USB 2 LCD wi h wo eco ding channels, wi hin he ange -35ºC
o 80ºC and 0% RH o 100% RH, wi h a esolu ion o ± 0.5ºC and ± 3% RH was used.
In his case, he measu emen s ha e been ca ied ou o e a pe iod o one yea , a
unin e up ed one-hou in e als om Ma ch 2020 o Feb ua y 2021 wi h a he mo-hyg ome e
placed in he sou heas galle y o he Cou ya d o he Maidens (Fig. 6).
The ou doo measu emen s we e ob ained h ough he AEMET (S a e Me eo ological Agency
in Spain). The me eo ological s a ion used was he one loca ed in Se ille. This me eo ological
s a ion is equipped wi h a VAISALA HMP45D p obe wi h a measu ing ange om 40 o 60 ºC
and 0.8 o 100% RH, and an accu acy o ± 0.2°C and ± 2% RH. The measu emen da a ob ained
a e also om one-hou in e als.
To de e mine he in e al o a ia ion o RH by mon hs, he a e age o he minimum and
maximum daily humidi ies has been calcula ed and he mon hly a i hme ic means o hese
alues ha e been ob ained. These da a ha e been s udied ollowing he indica ions o he UNE-
EN 15757: 2011 s anda d[9].
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Fig. 6 Loca ion o he moni o ing equipmen on he g ound loo o he Ped o I Palace.
2.3. Ob aining u u e clima e da a
Acco ding o he da a ob ained du ing he annual moni o ing, wo app oaches ha e been
ca ied ou o ob ain u u e ime se ies: (i) es ima ion o u u e ou doo clima e da a h ough
he EPW mo phing p ocess o he clima e zone o he case o s udy, and (ii) he es ima ion o
he ime se ies o he in e io poin h ough he applica ion o da a mining algo i hms.
2.3.1. Mo phing om he EPW o he clima e zone o he case s udy
The i s o hem was based on he clima ic iles p o ided by he Spanish Technical Building
Code (CTE)[53] o he clima e zone o Se ille, Csa (Medi e anean) acco ding o Köppen –
Geige classi ica ion [54]. In he i s place, T and RH da a esul ing om he moni o ing ha e
been en e ed in he Se ille clima e a chi e o be able o ca y ou he e alua ion o u u e
scena ios. Secondly, T and RH da a measu ed ou doo s ha e been compa ed wi h hose
exis ing in he CTE o see i i is easible o es ablish conside a ions, no only in he case o s udy
bu also in plas e wo k in simila "sub opical d y-summe " clima es. The calib a ion c i e ia o
he Fede al Ene gy Managemen P og am (FEMP)[55], ASHRAE GUIDELINE S anda d 14-2002
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[56], [57] and he In e na ional Pe o mance Measu emen and Ve i ica ion P o ocol (IPMVP)
[58] ha e been ollowed. This alida ion p ocess makes i possible o e i y i he esul s o he
es ima ions o he clima e zone o Se ille a e ela ed o he eal alues moni o ed wi h a
ma gin o e o . Fo his, he ollowing o mulas a e used:
𝑀𝐵𝐸 =∑(𝑚𝑖−𝑠𝑖)
𝑛
𝑖=1
𝑛 (1)
𝑀𝐵𝐸 =∑(𝑚𝑖−𝑠𝑖)
𝑛
𝑖=1
𝑛 (2)
𝑁𝑀𝐵𝐸 =1
𝑚
∙∑(𝑚𝑖−𝑠𝑖)
𝑛
𝑖=1
𝑛−𝑝 (3)
In o de o scale he esul s o he Mean Bias E o (MBE) (Equa ion 1), No malized Mean Bias
E o (NMBE) is used by di iding MBE by he mean o measu ed alues (𝑚), gi ing he global
di e ence be ween he eal alues and he p edic ed ones (Equa ion 2); In hese equa ions mi
is he measu ed alue, si is he simula ed one and n he numbe o measu ed da a poin s. The
Coe icien o Va ia ion o he Roo Mean Squa e E o CV (RMSE) measu es he a iabili y o
he e o s be ween measu ed and simula ed alues (Equa ion 3).
The limi alues associa ed wi h hese pa ame e s a y depending on whe he hey a e hou ly
o mon hly. Conside ing ha he en i onmen al a iables a e hou ly, he limi alues a e
be ween -10% and 10% in NMBE and less han 30% in CV (RMSE) [56,57].
To es ima e u u e clima e scena ios in he clima e zone, he CCWo ldWea he Gen ool om
he UK Me O ice Hadley Cen e Coupled Model 3 HadCM3 has been used. Th ough a mo phing
p ocess, his ool gene a es wea he iles adap ed o clima e change om any loca ion in he
wo ld and c ea es iles compa ible wi h mos building pe o mance simula ion p og ams [59].
The mo phing o he clima e iles has been ca ied ou o scena io A2 o g eenhouse gas
emissions acco ding o he IPCC [60], esul ing in clima e scena io iles o yea s 2050 and 2080.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
2.3.2. Es ima ion o u u e ime se ies o he case s udy
The da a mining app oach o his s udy was pe o med using he M5P algo i hm. This algo i hm
allows es ima ing he ime se ies o he yea s 2050 and 2080 in he measu ed in e io poin o
he case s udy. The M5P algo i hm (also known as M5') is an e olu ion o he Classi ica ion and
eg ession ee (CART) [61,62]. Unlike CART, M5P combines decision ees wi h mul i a ia e
eg ession; a decision ee is buil ollowing he same in e ed ee s uc u e o e ed by CART,
bu a mul iple linea eg ession (MLR) model is i ed on each shee (Equa ion 4). The e o e, he
algo i hm wo ks by de eloping an MLR model in each sub egion (Fig. 7). In he p ocess o
de eloping he M5P algo i hm ee, ins ead o maximizing he in o ma ion gain, he in e nal
a ia ion o he subse s o he class alues o each b anch is minimized. Once he model is buil ,
p uning educes o e i ing. The ad an ages o he models gene a ed by his algo i hm a e ha
hey e icien ly handle la ge amoun s o nume ical a iables and a e obus in he absence o
alues in he ins ances o he analyzed da ase [63,64]. Thus, i s use o he cha ac e iza ion o
di e en aspec s o buildings has inc eased in ecen yea s [65,66].
𝑌
𝑀𝐿𝑅 = 𝛽0+∑ (𝛽𝑖𝑥𝑖)
𝑣
𝑖=1 + 𝜀 (4)
Whe e 𝛽0 is he independen e m, 𝛽𝑖 a e he eg ession coe icien s, 𝑥𝑖 a e he p edic o
a iables, and 𝜀 is he e o .
Fig. 7. Scheme o an M5P model.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
The app oach designed had a con igu a ion o 6 inpu a iables co esponding o he hou ly
alues o empe a u e, ela i e humidi y, dew poin empe a u e, ho izon al in a ed adia ion
in ensi y, global ho izon al adia ion, and di ec ion no mal sola i adiance, di use ho izon al
sola i adiance o he ime se ies o he ou e poin o Se ille. The ou pu a iables co espond
o he ela i e humidi y and he empe a u e a he in e io poin o he case s udy. Fo each
ou pu a iable a di e en M5P was designed.
The aining o each M5P model was ca ied ou using he da a om he cu en scena io. Once
he models we e ained, ime se ies o he yea s 2050 and 2080 we e es ima ed. I is wo h
no ing ha he M5P models we e alida ed using he coe icien o de e mina ion and he MBE
index as s a is ical pa ame e s.
3. RESULTS AND DISCUSSION
3.1 The mo-hyg ome ic condi ions o he Palace o Ped o I
The esul s ob ained om he moni o ing ca ied ou be ween Ma ch 2020 and Feb ua y 2021
in he sou heas galle y o he Cou ya d o he Maidens a e shown in Fig. 8 and able 1:
Fig. 8 En i onmen al condi ions in he sou heas co ne o he Cou ya d o he Maidens om Ma ch 2020 o
Feb ua y 2021.
The measu emen o he T and RH pa ame e s is ele an mainly o he s udy o he
luc ua ions and he mal jumps ha may occu du ing he day-nigh hou s o du ing he
di e en seasons o he yea . In his ega d, i is obse ed ha he mos p onounced he mal
jumps be ween day and nigh hou s do no exceed 5ºC (July and Augus ), a ac ha -by i sel -
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
does no ep esen any isk in e ms o ma e ial expansion. Finally, a high RH in win e is
eco ded and mus be con olled wi h adequa e en ila ion, al hough i is wi hin he accep able
ange o he co ec p ese a ion o plas e wo k.
Table 1. Summa y o en i onmen al pa ame e s om 01/03 /2020 o 28 / 02/2021.
Mon hs
RELATIVE HUMIDITY (%)
TEMPERATURE (° C)
Min. Values
egis e ed
Max. alues
egis e ed
Δ
(max-min)
Min. Values
egis e ed
Max. alues
egis e ed
Δ
(max-min)
A e age
SD
A e age
SD
A e age
SD
A e age
SD
Feb 20
55.3
11,001
73.8
9,454
18.5
14
1,319
18.1
1,577
4.1
Ap. 20
60.6
6,672
77.7
4.79
17.1
15.7
1,534
19.2
1,558
3.5
May 20
43.3
10,818
64.1
9,947
20.8
21.3
2,337
26.1
2,832
4.8
Jun 20
41.4
5,608
62
7.87
20.6
22.3
2,434
27.3
2,314
5
Jul. 20
38
5,877
56
7.43
18
27.6
1,448
31.8
1,328
4.2
Aug. 20
38
8,637
56.5
7,966
18.5
26.3
1,675
30.5
1,581
4.2
Sep . 20
44.1
9,493
58.3
8,287
14.2
24
1.9
27.4
2,105
3.4
Oc . 20
49.3
12,985
63.1
11,674
13.8
18.1
1,996
21.3
2,136
3.2
No . 20
70
8,459
78.6
7,134
8.6
15.6
1,964
17.9
1,965
2.3
Dec. 20
74.2
10,822
84.4
8,133
10.2
11.4
2,366
13.7
1,935
2.3
Jan. 21
75.2
9,545
86.1
8,141
10.9
8.6
3,493
11.2
3,093
2.6
Feb. 21
77.4
5,562
83.1
5,129
5.7
12.9
1,127
15.1
0.854
2.2
Fig. 9 G aphical ep esen a ion o ambien condi ion (T and RH) om 01/03 /2020 o 28 / 02/2021.
F om a de ailed analysis o he measu emen s ob ained (Table 1 and Fig 9), he isk due o high
RH is concen a ed om No embe o Feb ua y, bo h inclusi e; being Janua y he mos
un a o able mon h since 27.02% o he measu emen s eco ded du ing ha mon h exceed he
limi o 90% RH. Likewise, he isk o dehyd a ion o he gypsum and ans o ma ion in o
bassani e (CaSO4·1/2H2O) occu s om May o Oc obe , bo h inclusi e; being July and Augus
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
he mos c i ical mon hs o such pu poses. I is wo h o highligh ha a 56.71% o he
measu emen s eco ded in he yea emain below he limi es ablished in he ma ix (Fig. 10).
The da a ob ained co espond o a o al o 8,760 measu emen s, o which only 11.76% a e
ou side he limi s es ablished in he ma ix ha ela es he mo-hyg ome ic condi ions wi h he
mine alogical phases o calcium sul a e, depending on hei deg ee o hyd a ion.
Fig. 10 Dispe sion g aph o he measu emen s om 01/03 /2020 o 28 / 02/2021 a he sou heas co ne o he
Cou ya d o he Maidens.
Rega ding he ype o isk de ec ed, i can be seen in Fig. 10 how i is mainly dis inguished
be ween isk due o in e nal s uc u al ans o ma ions (associa ed wi h he summe mon hs)
and isk due o high RH (associa ed wi h he win e mon hs). The 7.95% o he hou ly eco ds
co espond o measu emen s ha a e below he cu e es ablished by Winkle & Wilhelm [31] ,
while he isk due o a high RH co esponds o 3.65% and he isk due o a low T co esponds o
only 10 obse a ions h oughou he yea , so he cu en clima ic condi ions in he Cou ya d o
he Maidens a e a o able in his sense.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Fig. 11 Mon hly ep esen a ion o he hou ly measu emen s ob ained in he Cou ya d o he Maidens om Ma ch
2020 o Feb ua y 2021.
Howe e , he esul s ob ained om moni o ing he clima e o he ci y o Se ille show di e en
ends in conse a ion isks (Fig. 11). As i can be seen, al hough he mos p ominen isks a e
once again ha o in e nal ans o ma ions in summe and high RH in win e , he e is a g ea e
numbe o obse a ions ha emain wi hin hese isks. Thus, he obse a ions a isk o in e nal
ans o ma ions co espond o alues be ween 348 and 360 in July and Augus , while in he
Cou ya d o he Maidens he alues we e 229 and 168, espec i ely. Likewise, i was also
de ec ed how he isk due o high RH ex ends o a g ea e numbe o mon hs han in he case
o he Cou ya d o he Maidens and wi h a signi ican ly highe numbe o obse a ions.
Fig. 12 Mon hly ep esen a ion o he hou ly measu emen s ob ained in Se ille om Ma ch 2020 o Feb ua y 2021.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
The e o e, signi ican a ia ions a e de ec ed in he esul s ob ained wi h he wo app oaches.
While he app oach o he analysis o he Cou ya d o he Maidens poin (wi h a clea
mic oclima e) p esen s less se e e condi ions o he conse a ion o plas e wo k, ou side he
condi ions would be di e en , p esen ing a g ea e isk due o he clima e in hese elemen s.
Likewise, he a iabili y de ec ed be ween he wo app oaches shows he g ea a ia ion ha
he analysis app oaches o he in luence o clima e can p esen on he conse a ion o
plas e wo k. Thus, he use o a join app oach o he ou doo clima e and mic oclima es in his
ype o case s udy allows us o show a b oade pe spec i e o he isk o conse a ion o
plas e wo k.
3.2 Calib a ion o ou doo measu emen s. The clima e ile o Se ille.
Despi e he possible isks de ec ed in he cu en scena io, clima e change can gene a e a
change in he conse a ion condi ions o plas e wo k. Fo ha , he u u e ime se ies o he
wo p e iously alida ed app oaches used we e es ima ed in his s udy. In he case o he
ou doo clima e app oach, T and RH pa ame e s measu ed du ing he ou doo moni o ing a e
compa ed wi h T and RH da a ex ac ed om he CTE o he clima e zone o Se ille Csa
(Medi e anean). As he en i onmen al a iables analyzed a e hou ly, he limi alues a e
be ween -10% and 10% in NMBE and less han 30% in CV (RMSE). Acco ding o Table 2, i is
de ec ed ha he CV (RMSE) alues a e wi hin he es ablished e o limi s. Howe e , i is
de ec ed ha T is sligh ly highe (0.88%) han he limi s in he NMBE and in ma e s ela ed o
RH he alue exceeds he limi in he NMBE by 6.40% (Table 2). F om hese esul s we can
es ablish he ollowing conside a ions:
In he i s place, conside ing ha a measu emen is being compa ed unde speci ic
en i onmen al condi ions wi h a ypical me eo ological yea acco ding o he c i e ia
es ablished in he CTE and based on he a ailable me eo ological s a ions, he le el o
app oxima ion is good, especially i we conside he CV (RMSE).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Table 2. Calib a ion o measu ed da a in RAS compa ed o CTE. Limi alues acco ding o ASHRAE Guideline14.
Va iable
NMBE [%]
CV (RMSE) [%]
Ob ained alues
Limi alues
Ob ained alues
Limi alues
Tempe a u e (T)
11.03
± 10
25.34
30
Rela i e humidi y (RH)
15.52
± 10
27.08
30
The e o e, o make p edic ions o clima e scena ios on he case s udy, i is necessa y o use
he ac ual measu ed alues o T and RH. Howe e , in o de o es ablish conside a ions on
simila Csa d y-summe sub opical o Medi e anean clima es, we can conside he
app oxima ion o he eal measu emen s as accep able, ecommending he e i ica ion o he
speci ic mic oclima ic condi ions using p ocedu es simila o he one used.
In he case o he in e io poin app oach, u u e es ima es we e ob ained by M5P algo i hm.
Table 3 shows he alues associa ed wi h each M5P model, while Fig. 13 shows he cloud o
poin s be ween he ac ual and es ima ed alues o he aining and alida ion phase. As i can
be seen, he coe icien o de e mina ion anged be ween 74% and 87%, while he mean e o
anged be ween 2.6ºC and 8% RH in each a iable. The es ima es we e mo e limi ed in he RH
a iable in a simila way o ha de ec ed in o he s udies o es ima ing u u e ime se ies o RH
[14]. In any case, he esul s can be conside ed alid o ha e a knowledge o he u u e ime
se ies expec ed in he add essed case s udy.
Table 3. Pe o mance ob ained by he M5P models in he aining and alida ion phase.
App oach
Ou pu a iable
De e mina ion coe icien (R2)
MBE
M5P 1
Tempe a u e (T)
86.53%
2.6
M5P 2
Rela i e Humidi y (RH)
74.93%
8.0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[9] A.E. de N. UNE, UNE-EN 15757: Conse a ion o Cul u al P ope y - Speci ica ions o
empe a u e and ela i e humidi y o limi clima e-induced mechanical damage in
o ganic hyg oscopic ma e ials, (2011).
[10] Ansi/Ash ae, ANSI/ASHRAE 55:2004 The mal En i onmen al Condi ions o Human
Occupancy, Ash ae. (2004) 30.
[11] PAS 198:2012 - Speci ica ion o managing en i onmen al condi ions o cul u al
collec ions – BSI B i ish S anda ds, (n.d.).
h p://shop.bsig oup.com/P oduc De ail/?pid=000000000030219669.
[12] UNE-EN-15759-1, Conse a ion o cul u al p ope y. Indoo clima e. Pa 1: Guidelines
o hea ing chu ches, chapels and o he places o wo ship, (2012).
[13] C. Bonacina, P. Baggio, F. Cappelle i, P. Romagnoni, A.G. S e an, The Sc o egni
Chapel: The esul s o o e 20 yea s o indoo clima e moni o ing, Ene gy Build. 95
(2015) 144–152. h ps://doi.o g/10.1016/j.enbuild.2014.12.018.
[14] D. Bien enido-Hue as, M. León-Muñoz, J.J. Ma ín-del-Río, C. Rubio-Bellido, Analysis
o clima e change impac on he p ese a ion o he i age elemen s in his o ic buildings
wi h a de icien indoo mic oclima e in wa m egions, Build. En i on. 200 (2021) 107959.
h ps://doi.o g/10.1016/j.builden .2021.107959.
[15] G. Zonno, R. Aguila , R. Bo oschek, P.B. Lou enço, Analysis o he long and sho - e m
e ec s o empe a u e and humidi y on he s uc u al p ope ies o adobe buildings
using con inuous moni o ing, Eng. S uc . 196 (2019) 109299.
h ps://doi.o g/10.1016/j.engs uc .2019.109299.
[16] R.P. K ame , M.P.E. Maas, M.H.J. Ma ens, A.W.M. an Schijndel, H.L. Schellen, Ene gy
conse a ion in museums using di e en se poin s a egies: A case s udy o a s a e-o -
he-a museum using building simula ions, Appl. Ene gy. 158 (2015) 446–458.
h ps://doi.o g/10.1016/j.apene gy.2015.08.044.
[17] M. Napp, T. Kalamees, Ene gy use and indoo clima e o conse a ion hea ing,
dehumidi ica ion and adap i e en ila ion o he clima e con ol o a mediae al chu ch
in a cold clima e, Ene gy Build. 108 (2015) 61–71.
h ps://doi.o g/10.1016/j.enbuild.2015.08.013.
[18] C. Co na o, V.A. Puggioni, R.M. S ollo, Dynamic simula ion and on-si e measu emen s
o ene gy e o i o complex his o ic buildings: Villa Mond agone case s udy, J. Build.
Eng. 6 (2016) 17–28. h ps://doi.o g/10.1016/j.jobe.2016.02.001.
[19] T. Ca dinale, G. Rospi, N. Ca dinale, The in luence o indoo mic oclima e on he mal
com o and conse a ion o a wo ks: The case s udy o he Ca hed al o Ma e a (Sou h
I aly), in: Ene gy P ocedia, Else ie L d, 2014: pp. 425–432.
h ps://doi.o g/10.1016/j.egyp o.2014.10.398.
[20] A. Re i, D. Sa e hwai e, F. A agon-Du and, J. Co ee-Mo lo , R.. R. Kiunsi, M. Pelling,
D. Robe s, W. Solecki, S.P. Gajja , A. S e dlik, Towa ds ans o ma i e adap a ion in
ci ies: he IPCC’s Fi h Assessmen , En i on. U ban. 26 (2014) 11–28.
h ps://doi.o g/10.1177/0956247814523539.
[21] A.J. P ie o, K. Ve iche , A. Sil a, J. de B i o, On he impac s o clima e change on he
unc ional de e io a ion o he i age buildings in Sou h Chile, Build. En i on. 183 (2020)
107138. h ps://doi.o g/10.1016/j.builden .2020.107138.
[22] V. Rajčić, A. Skende , D. Damjano ić, An inno a i e me hodology o assessing he
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
clima e change impac on cul u al he i age, In . J. A chi . He i . 12 (2018) 21–35.
h ps://doi.o g/10.1080/15583058.2017.1354094.
[23] A. Haugen, C. Be olin, G. Leijonhu ud, T. Ols ad, T. B os öm, A me hodology o
long- e m moni o ing o clima e change impac s on his o ic buildings, Geosci. 8 (2018).
h ps://doi.o g/10.3390/geosciences8100370.
[24] S. Ramí ez, M. Za zo, A. Pe les, F.J. Ga cía-Diego, A me hodology o disc iminan ime
se ies analysis applied o mic oclima e moni o ing o esco pain ings, Senso s
(Swi ze land). 21 (2021) 1–29. h ps://doi.o g/10.3390/s21020436.
[25] M. Gómez He as, La Tempe a u a en los Ma e iales del Pa imonio, (2012) 87–95.
h p://digi al.csic.es/handle/10261/46794.
[26] Z. Pa lík, J. Foř , R. Če ný, An in si u moni o ing sys em o he s udy o en i onmen al
in luences on du abili y and he des uc i e p ocess o building ma e ials and s uc u es,
WIT T ans. Modelling Simul. 55 (2013) 287–296. h ps://doi.o g/10.2495/CMEM130231.
[27] S. Michalski, Tempe a u a Inco ec a, Can. Conse . Ins i u e. ICCROM. (2009).
[28] S. Michalski, Humedad ela i a Inco ec a, Can. Conse . Ins i u e. ICCROM. (2009).
[29] L. Villanue a Domínguez, A. Ga cía San os, Manual del yeso, CIE In e siones Edi o iales,
2001.
[30] D. F eye , W. Voig , C ys alliza ion and Phase S abili y o CaSO4 and CaSO 4 - Based
Sal s, Mona she e Fu Chemie. 134 (2003) 693–719. h ps://doi.o g/10.1007/s00706-
003-0590-3.
[31] E.M. Winkle , E.J. Wilhelm, Sal bu s by hyd a ion p essu es in a chi ec u al s one in
u ban a mosphe e, Geol. Soc. Am. Bull. 81 (1970) 567–572.
h ps://doi.o g/h ps://doi.o g/10.1130/0016-7606(1970)81[567:SBBHPI]2.0.CO;2.
[32] L. Ri e bach, P. Becke , Tempe a u e and humidi y dependen o ma ion o
CaSO4·xH2O (x = 0...2) phases, Glob. Plane . Change. 187 (2020).
h ps://doi.o g/10.1016/j.gloplacha.2020.103132.
[33] E. Goossens, Mois u e ans e p ope ies o coa ed gypsum, Technische Uni e si ei
Eindho en, Facul ei Bouwkunde, 2003. h ps://doi.o g/10.6100/IR571306.
[34] C.A.C. Mesqui a, Re es imien os Con inuos In e io es de Va ias Capas con
Ca ac e ís icas de Ba e a de Vapo e Hig oscopicidad, Doc o al disse a ion,
Poly echnic Uni e si y o Mad id, 2012.
[35] R. Rubio Domene, Yese ías de la Alhamb a. His o ia, écnica y conse ación, Pa ona o
de la Alhamb a y Gene ali e. Uni e si y o G anada, G anada, 2010.
[36] M. To es-González, F.J. Alejand e, V. Flo es-alés, A.I. Cale o-cas illo, F.J. Blasco-lópez,
Analysis o he s a e o conse a ion o his o ical plas e wo k h ough isual inspec ion
and non-des uc i e es s . The case o he uppe ieze o he Toledanos Room ( The
Royal Alcáza o Se ille , Spain ), J. Build. Eng. 40 (2021) 1–14.
h ps://doi.o g/h ps://doi.o g/10.1016/j.jobe.2021.102314.
[37] R. Campos de Al ea , The main enance and he p e en i e p ese a ion measu es o he
cul u al goods in he Royal Alcáza o Se ille, Apun . Del Alcáza Se illa. 18 (2018) 71–
87.
[38] E. Co ea Gómez, R. Rubio Domene, El yeso. Las deco aciones de yeso en época naza í,
in: P. de la A. y Gene ali e (Ed.), Man. Buenas P ác icas. Res au ación Made a, Yeso y
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Ce ámica, Pa ona o, Conseje ía de Educación, Cul u a y Depo e de la Jun a de
Andalucía, G anada, 2014: pp. 43–52.
[39] J.M. Cab e a Ga ido, La in luencia de los con aminan es en el Pa imonio a ís ico
Nacional, Econ. Ind. 107 (1972) 51–60.
[40] H. Vii anen, T. Ojanen, Imp o ed Model o P edic Mold G ow h in Building Ma e ials,
in: The m. Pe o m. Ex e . En el. Whole Build. X–P oceedings CD, 2007: pp. 2–7.
[41] G.A. Tsongas, F. Rio oan, Minimum condi ions o isible mold g ow h, ASHRAE J. 58
(2016) 32–43.
[42] D. Sanz A auz, Análisis del yeso empleado en e es imien os ex e io es median e
écnicas geológicas, Doc o al disse a ion, Uni e sidad Poli écnica de Mad id, 2009.
h p://oa.upm.es/1711/1/DAVID_SANZ_ARAUZ.pd .
[43] A.I. Cale o Cas illo, A. Ga cía Bueno, O. López C uz, V.J. Medina Fló ez, La polic omía
o iginal de las yese ías del Pa io de las Doncellas del Real Alcáza de Se illa. Ma e iales
cons i u i os y écnicas de ejecución, A queol. y Te i . Medie . 24 (2017) 255–290.
h ps://doi.o g/10.17561/ay m. 24i0.9.
[44] C. Sabbioni, P. B imblecombe, M. Cassa , The A las o Clima e Change Impac on
Eu opean Cul u al He i age: Scien i ic Analysis and Managemen S a egies, An hem
P ess, London, UK, 2010.
[45] A. Almag o Go bea, El Alcáza de Se illa Un palacio musulmán pa a un ey c is iano, in:
C is. y Musulmanes En La Península Ibé ica La Gue . La F on . y La Con i encia. XI
Cong . Es ud. Medie ., 2007: pp. 331–365.
[46] F.J. Blasco López, Yese ías medie ales de adición islámica del Real Alcáza de Se illa:
Re isión His o iog á ica, Me odología pa a la ca ac e ización, e aluación de su
du abilidad y elabo ación de un in en a io, Doc o al disse a ion, Uni e si y o Se ille,
2011.
[47] A.I. Cale o Cas illo, Ma e iales, écnicas y p ocedimien os en la deco ación
a qui ec ónica. Aplicaciones a la conse ación y es au ación de las yese ías del Pa io de
las Doncellas. Real Alcáza de Se illa., Doc o al disse a ion, Uni e sidad de G anada,
2016. h p://hdl.handle.ne /10481/43864.
[48] A. Pleguezuelo, Tile-wo k in he mudéja palace in he Royal Alcáza o Se ille. A
P elimina y Visual Analysis., Apun . Del Alcáza Se illa. 16 (2015) 219–230.
[49] C. En íquez Díaz, J.R. Baeza Ál a ez, A p ojec o he es o a ion o he ilings o he
g ound loo o he mudéja palace, Apun . Del Alcáza Se illa. 19 (2019) 65–77.
[50] C. Cañas Palop, Las a madu as de cubie as mudéja es del palacio de Ped o I, del Alcáza
de Se illa: análisis in eg al y p opues as pa a la es au ación, Doc o al disse a ion,
Uni e si y o Se ille, 2006. h ps://dialne .uni ioja.es/se le / esis?codigo=23333.
[51] S. Fe nández Aguile a, Po a en ane os mudéja es en el Real Alcáza de Se illa, A chi o
Hi, Dipu ación P o incial de Se illa, Se illa, 2012.
[52] F.M. Tubino, Es udios sob e el a e en España. La a qui ec u a hispano- isigoda y á abe
española. El Alcáza de Se illa. Una iglesia mozá abe, 1886.
[53] The Go e nmen o Spain, Royal Dec ee 314/2006. App o ing he Spanish Technical
Building Code, Mad id, Spain, 2013.
[54] M. Ko ek, J. G iese , C. Beck, B. Rudol , F. Rubel, Wo ld map o he Köppen-Geige
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
clima e classi ica ion upda ed, Me eo ol. Zei sch i . 15 (2006) 259–263.
h ps://doi.o g/10.1127/0941-2948/2006/0130.
[55] L. Webs e , J. B ad o d, Measu emen and Ve i ica ion o Fede al Ene gy (M&V
Guidelines), 2008.
[56] ANSI/ASHRAE, ASHRAE Guideline 14-2002 Measu emen o Ene gy and Demand
Sa ings, Ash ae. 8400 (2002) 170.
[57] G.R. Ruiz, C.F. Bande a, Valida ion o calib a ed ene gy models: Common e o s,
Ene gies. 10 (2017). h ps://doi.o g/10.3390/en10101587.
[58] E iciency Valua ion O ganiza ion, In e na ional Pe o mance Measu emen &
Ve i ica ion P o ocol, Handb. Financ. Ene gy P oj. I (2016) 122.
[59] M.F. Jen sch, A.S. Bahaj, P.A.B. James, CCWo ldWea he Gen, Clima e change wo ld
wea he ile gene a o , Ve sion 1.8, Sus ain. Ene gy Res. G . (2013).
[60] I.P. on C. Change, Summa y o Policymake s, in: In e go e nmen al Panel on Clima e
Change (Ed.), Clim. Chang. 2013 - Phys. Sci. Basis, Camb idge Uni e si y P ess,
Camb idge, 2014: pp. 1–30. h ps://doi.o g/10.1017/CBO9781107415324.004.
[61] J.R. Quinlan, o he s, Lea ning wi h con inuous classes, in: 5 h Aus . J . Con . A i . In ell.,
1992: pp. 343–348.
[62] Y. Wang, I.H. Wi en, Induc ion o model ees o p edic ing con inuous classes, in: Eu .
Con . Mach. Lea n., P ague: Uni e si y o Economics, Facul y o In o ma ics and
S a is ics, 1997.
[63] A. Behnood, V. Behnood, M. Modi i Gha eh e an, K.E. Alyamac, P edic ion o he
comp essi e s eng h o no mal and high-pe o mance conc e es using M5P model ee
algo i hm, Cons . Build. Ma e . 142 (2017) 199–207.
h ps://doi.o g/10.1016/j.conbuildma .2017.03.061.
[64] L. Lin, Q. Wang, A.W. Sadek, A combined M5P ee and haza d-based du a ion model
o p edic ing u ban eeway a ic acciden du a ions, Accid. Anal. P e . 91 (2016) 114–
126. h ps://doi.o g/10.1016/j.aap.2016.03.001.
[65] F. A sa ian, A. Sabe , A. Pou zangba , A.G. Olabi, M.A. Khanmohammadi, Analysis o
ecycled agg ega es e ec on ene gy conse a ion using M5″ model ee algo i hm,
Ene gy. 156 (2018) 264–277. h ps://doi.o g/10.1016/j.ene gy.2018.05.099.
[66] C.F. Je ey Kuo, C.H. Lin, M.H. Lee, Analyze he ene gy consump ion cha ac e is ics
and a ec ing ac o s o Taiwan’s con enience s o es-using he big da a mining
app oach, Ene gy Build. 168 (2018) 120–136.
h ps://doi.o g/10.1016/j.enbuild.2018.03.021.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Figu e 1 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 01.jpg
Figu e 2 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 02.jpg
Figu e 3 Click he e o access/download;Figu e (Do no upload pic u e,
.s g o ma . Upload only PDF, eps o i o ma );Figu e 03.jpg
Figu e 4 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 04.jpg
Figu e 5 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 05.jpg
Figu e 6 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 06.jpg
Figu e 13 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 13.jpg
Figu e 14 Click he e o access/download;Figu e (Do no upload pic u e, .s g o ma . Upload
only PDF, eps o i o ma );Figu e 14.jpg
Table 1. Summa y o en i onmen al pa ame e s om 01/03 /2020 o 28 / 02/2021.
Mon hs
RELATIVE HUMIDITY (%)
TEMPERATURE (° C)
Min. Values
egis e ed
Max. alues
egis e ed
Δ
(max-
min)
Min. Values
egis e ed
Max. alues
egis e ed
Δ
(max-
min)
A e ag
e
SD
A e ag
e
SD
A e ag
e
SD
A e ag
e
SD
Feb 20
55.3
11,001
73.8
9,454
18.5
14
1,319
18.1
1,577
4.1
Ap. 20
60.6
6,672
77.7
4.79
17.1
15.7
1,534
19.2
1,558
3.5
May 20
43.3
10,818
64.1
9,947
20.8
21.3
2,337
26.1
2,832
4.8
Jun 20
41.4
5,608
62
7.87
20.6
22.3
2,434
27.3
2,314
5
Jul. 20
38
5,877
56
7.43
18
27.6
1,448
31.8
1,328
4.2
Aug. 20
38
8,637
56.5
7,966
18.5
26.3
1,675
30.5
1,581
4.2
Sep . 20
44.1
9,493
58.3
8,287
14.2
24
1.9
27.4
2,105
3.4
Oc . 20
49.3
12,985
63.1
11,674
13.8
18.1
1,996
21.3
2,136
3.2
No . 20
70
8,459
78.6
7,134
8.6
15.6
1,964
17.9
1,965
2.3
Dec. 20
74.2
10,822
84.4
8,133
10.2
11.4
2,366
13.7
1,935
2.3
Jan. 21
75.2
9,545
86.1
8,141
10.9
8.6
3,493
11.2
3,093
2.6
Feb. 21
77.4
5,562
83.1
5,129
5.7
12.9
1,127
15.1
0.854
2.2
Table 1 Click he e o access/download;Table;Table 1.docx
Table 2. Calib a ion o measu ed da a in RAS compa ed o CTE. Limi alues acco ding o ASHRAE
Guideline14.
Va iable
NMBE [%]
CV (RMSE) [%]
Ob ained
alues
Limi
alues
Ob ained
alues
Limi
alues
Tempe a u e (T)
11.03
± 10
25.34
30
Rela i e humidi y
(RH)
15.52
± 10
27.08
30
Table 2 Click he e o access/download;Table;Table 2.docx
Table 3. Pe o mance ob ained by he M5P models in he aining and alida ion phase.
App oach
Ou pu a iable
De e mina ion coe icien (R2)
MBE
M5P 1
Tempe a u e (T)
86.53%
2.6
M5P 2
Rela i e Humidi y (RH)
74.93%
8.0
Table 3 Click he e o access/download;Table;Table 3.docx
Decla a ion o in e es s
x The au ho s decla e ha hey ha e no known compe ing inancial in e es s o pe sonal ela ionships
ha could ha e appea ed o in luence he wo k epo ed in his pape .
☐The au ho s decla e he ollowing inancial in e es s/pe sonal ela ionships which may be conside ed
as po en ial compe ing in e es s:
Jou nal o Building Enginee ing
Dea Edi o :
Manusc ip i le: “LONG-TERM MONITORING FOR PREVENTIVE CONSERVATION OF HISTORICAL
PLASTERWORKS”
As co esponding au ho , I ce i y ha he co-au ho s o his manusc ip ha e NO a ilia ions wi h
o in ol emen in any o ganiza ion o en i y wi h any inancial in e es (such as hono a ia;
educa ional g an s; pa icipa ion in speake s’ bu eaus; membe ship, employmen , consul ancies,
s ock owne ship, o o he equi y in e es ; and expe es imony o pa en -licensing
a angemen s), o non- inancial in e es (such as pe sonal o p o essional ela ionships,
a ilia ions, knowledge o belie s) in he subjec ma e o ma e ials discussed in his manusc ip .
Vicen e Flo es-Alés
Uni e sidad de Se illa
As co esponding au ho , I ce i y ha he co-au ho s o his manusc ip ha e NO a ilia ions wi h
o in ol emen in any o ganiza ion o en i y wi h any inancial in e es (such as hono a ia;
educa ional g an s; pa icipa ion in speake s’ bu eaus; membe ship, employmen , consul ancies,
s ock owne ship, o o he equi y in e es ; and expe es imony o pa en -licensing a angemen s),
o non- inancial in e es (such as pe sonal o p o essional ela ionships, a ilia ions, knowledge o
belie s) in he subjec ma e o ma e ials discussed in his manusc ip .
Con lic o In e es
AUTHOR STATEMENT
M. To es-González: In es iga ion, da a discussion, W i ing, Re iew & Edi ing
C.Rubio-Bellido: Concep ualiza ion, In es iga ion, da a discussion, Valida ion
D. Bien enido-Hue as: Concep ualiza ion, In es iga ion, So wa e, da a discussion
J.M. Alducin-Ochoa:, In es iga ion, S a is ical analysis, Fo mal analysis
V. Flo es-Alés: Concep ualiza ion, Me hodology, In es iga ion, Valida ion
Au ho S a emen
LONG-TERM ENVIRONMENTAL MONITORING FOR PREVENTIVE CONSERVATION OF
EXTERNAL HISTORICAL PLASTERWORKS
Ma a To es-González a,b *, Ca los Rubio-Bellido a, Da id Bien enido-Hue as a, JM Alducin-
Ochoa a, V. Flo es-Alés a,
a Depa men o A chi ec u al Cons uc ion II, Uni e si y o Se ille
b Depa men o Ci il Enginee ing, A chi ec u e and Geo esou ces, Uni e si y o Lisbon
* Co esponding au ho : m o [email protected] , A Reina Me cedes 4, Se ille, 41012
KEYWORDS: plas e wo k; empe a u e; ela i e humidi y; a chi ec u al he i age; p ese a ion;
ABSTRACT:
In his s udy, an analysis o he en i onmen al condi ions in a Csa clima ic zone o he
conse a ion o plas e wo k o he Real Alcáza o Se ille is ca ied ou . The measu emen s
ob ained om in-si u moni o ing a e compa ed wi h he measu emen s p o ided by AEMET
(S a e Me eo ological Agency in Spain) du ing he e e ence yea and he s udy is comple ed
by es ima ing u u e en i onmen al condi ions using wo al e na i e app oaches: a mo phing
p ocess om he EPW o he clima ic zone and he applica ion o M5P da a mining algo i hms.
An op imal empe a u e ange is es ablished o he conse a ion o he plas e wo k ha
p e en s hei dehyd a ion o he eezing o wa e pa icles con ained. The ans o ma ion o
gypsum in o bassani e, he isks associa ed wi h exposu es o high ela i e humidi y and he
consequences o he sligh hyg oscopici y o he ma e ial and he en i onmen al condi ions
ha mus be de eloped o a o he g ow h o mold on he su ace o he c acking o
polych omies ha embellish hese plas e deco a ions on nume ous occasions a e analyzed.
The esul s ob ained allow us o es ablish p e en i e conse a ion measu es no only on he
plas e wo k bu also on he Real Alcáza o Se ille and ha a chi ec u al he i age loca ed in he
sub opical d y-summe clima e.
Re ised Manusc ip wi h Changes Ma ked Click he e o iew linked Re e ences
1. INTRODUCTION
1.1 The mo-hyg ome ic condi ions and he conse a ion o he a chi ec u al he i age.
The p e en i e conse a ion o he i age is c ucial so ha u u e gene a ions can know he
iden i y ai s o hei cul u e [1]. The e o e, he moni o ing and analysis o speci ic
en i onmen al condi ions [2] and hei impac on a chi ec u al he i age, as well as u u e
ulne abili ies due o global wa ming [3], a e cu en ly a challenge.
The e a e nume ous s udies ha analyze bo h he impac o en i onmen al condi ions on
a chi ec u al he i age [4,5] and he condi ioning o in e io spaces o a o he p ese a ion o
objec s (e.g., objec s ha belong o exhibi ions) [6,7]. In bo h cases, he e a e na ional and
in e na ional guidelines and s anda ds ha es ablish he in e io ai condi ioning condi ions o
hese buildings o such pu poses, such as UNI 10829 S anda d [8], UNE EN 15757 [9], ASHRAE
STANDARD 55-2020[10] o he B i ish PAS 198 [11]. Likewise, EN 15759-1 [12] also conside s
he possibili y o using hea ing sys ems o achie e he mal com o and o he op imal
conse a ion o he i age elemen s.
In addi ion o in e na ional s anda ds, nume ous s udies ha e ca ied ou he mo-hyg ome ic
moni o ing o a chi ec u al he i age in ecen decades, esul ing in a e y use ul ool o assess
he s a e o conse a ion o he building and o know he causes o he deg ada ion in o de o
de e mine a u u e es o a ion in e en ion o o es ablish p e en i e conse a ion c i e ia,
uling ou he use o des uc i e echniques. Thus, long- e m s udies o up o 20 yea s du a ion
[13,14] and s udies in which moni o ing is limi ed o one yea [15] ha e been published. In his
ega d, i is common o e alua e a leas annual moni o ing pe iods ha a e complemen ed by
he mal models on which di e en es ima es o en i onmen al condi ions a e es ablished.
En i onmen al moni o ing and simula ion ools make i possible o es ablish es ima es o
di e en ac o s, such as he modi ica ion o ope a ional pa e ns, he mal com o [16-18], o
ene gy conse a ion measu es [19]. Howe e , he impac o clima e change, he basis o
nume ous in es iga ions ela ed o ci ies [20,21] has been spa sely discussed in e e ence o
p e en i e conse a ion o he i age and long- e m ad e se aspec s, which could become e y
signi ican [22,23].
1.2 Rela i e humidi y and empe a u e in luence in he p ese a ion o his o ical
plas e wo k.
The e a e s udies abou how ambien condi ions could a ec cons uc i e ma e ials [24–28] bu
a e insu icien wi h espec o gypsum o plas e wo k. In his sense, i is wo h highligh ing he
impo ance o ollow-up campaigns o iden i y possible isks o he conse a ion o he
plas e wo k, p e en i e e sible damage and ad ance conse a ion s a egies. Howe e , o
accoun o all he possible damage mechanisms, a comp ehensi e me hodology should be
used o e alua e some isk indexes based on ambien empe a u e (T) and ela i e humidi y
(RH) measu emen s.
Gypsum, due o i s po ous mic os uc u e o med by c ys alline g oups o hyd a ed calcium
sul a e, can inco po a e wa e molecules inside, in he o m o wa e apo o e en in he o m
o liquid wa e , when en i onmen al condi ions a o capilla y condensa ion. Fu he mo e,
gypsum has he abili y o elimina e his wa e when en i onmen al condi ions change, educing
he con en o wa e apo in con ac wi h he ma e ial [29].
Va ia ions in T and RH can induce phase ansi ions. In gene al, gypsum (CaSO4·2H2O)
dehyd a ion occu s in d y en i onmen s [30], which is he eason why empe a u e below 40°C
is desi able o he co ec p ese a ion o plas e wo k. Howe e , he e may be si ua ions o
high empe a u es combined wi h high ela i e humidi y in which he e a e no al e a ions in he
mine al phases o he gypsum [31]. In his ega d, Winkle & Wilhelm (1970) indica ed he
ela ionship ha mus exis be ween ambien empe a u e and ela i e humidi y o ans o m
gypsum (CaSO4·2H2O) in o bassani e (CaSO4·0.5H2O) [31], In his case, he ma e ial is likely o
The con ol and egula ion o en i onmen al pa ame e s wi hin a e e ence in e al a ound
a e age alues o ypical seasonal cycles o he his o ical RAS clima e will educe he isk o
physical-mechanical damage and will also allow o es ablish mi iga ion ac ions.
On he g ound loo o he Palace, he in e io a eas a e no physically closed by ca pen y o
locksmi hs and a e pe manen ly open; he e a e la ge access a ches ha allow ansi be ween
di e en ooms, as well as be ween hese and he ou doo spaces.
In his way, wo a eas acco ding o exposu e o en i onmen al condi ions could be de ined. An
a ea ha includes hose plas e wo ks ha ecei e di ec incidence o sola adia ion has been
conside ed (yellow zone) (Fig. 5-A), and ano he zone wi h hose plas e wo ks no exposed o
di ec sunligh and wi h a highe ela i e humidi y ( ed zone), being he mos ep esen a i e
and un a o able case in e ms o he en i onmen al pa ame e s (Fig. 5-B).
Fig. 5 A) Sola pa h on 03/20/2020 in he geog aphical coo dina es co esponding o he Palace o Ped o I
(Re ie ed om h ps://www.sunea h ools.com/) B) Two di e en zones on he g ound loo o he Ped o I Palace
acco ding o sunligh .
Addi ionally, i has been possible o e i y ha he humidi y and empe a u e alues measu ed
in he in e io ooms o he Palace o Ped o I [36] a e e y simila o hose ob ained in he
sou heas galle y o he Cou ya d o he Maidens (ex e io ) because his a ea does no ecei e
he di ec incidence o sola adia ion (Fig. 5-A).
2.2. Measu emen p ocedu e
To de e mine ambien empe a u e and ela i e humidi y condi ions, a Lasca
he mohyg ome e , model EL-USB 2 LCD wi h wo eco ding channels, wi hin he ange -35ºC
o 80ºC and 0% RH o 100% RH, wi h a esolu ion o ± 0.5ºC and ± 3% RH was used.
In his case, he measu emen s ha e been ca ied ou o e a pe iod o one yea , a
unin e up ed one-hou in e als om Ma ch 2020 o Feb ua y 2021 wi h a he mo-hyg ome e
placed in he sou heas galle y o he Cou ya d o he Maidens (Fig. 6).
The ou doo measu emen s we e ob ained h ough he AEMET (S a e Me eo ological Agency
in Spain). The me eo ological s a ion used was he one loca ed in Se ille. This me eo ological
s a ion is equipped wi h a VAISALA HMP45D p obe wi h a measu ing ange om 40 o 60 ºC
and 0.8 o 100% RH, and an accu acy o ± 0.2°C and ± 2% RH. The measu emen da a ob ained
a e also om one-hou in e als.
To de e mine he in e al o a ia ion o RH by mon hs, he a e age o he minimum and
maximum daily humidi ies has been calcula ed and he mon hly a i hme ic means o hese
alues ha e been ob ained. These da a ha e been s udied ollowing he indica ions o he UNE-
EN 15757: 2011 s anda d[9].
Fig. 6 Loca ion o he moni o ing equipmen on he g ound loo o he Ped o I Palace.
2.3. Ob aining u u e clima e da a
Acco ding o he da a ob ained du ing he annual moni o ing, wo app oaches ha e been
ca ied ou o ob ain u u e ime se ies: (i) es ima ion o u u e ou doo clima e da a h ough
he EPW mo phing p ocess o he clima e zone o he case o s udy, and (ii) he es ima ion o
he ime se ies o he in e io poin h ough he applica ion o da a mining algo i hms.
2.3.1. Mo phing om he EPW o he clima e zone o he case s udy
The i s o hem was based on he clima ic iles p o ided by he Spanish Technical Building
Code (CTE)[53] o he clima e zone o Se ille, Csa (Medi e anean) acco ding o Köppen –
Geige classi ica ion [54]. In he i s place, T and RH da a esul ing om he moni o ing ha e
been en e ed in he Se ille clima e a chi e o be able o ca y ou he e alua ion o u u e
scena ios. Secondly, T and RH da a measu ed ou doo s ha e been compa ed wi h hose
exis ing in he CTE o see i i is easible o es ablish conside a ions, no only in he case o s udy
bu also in plas e wo k in simila "sub opical d y-summe " clima es. The calib a ion c i e ia o
he Fede al Ene gy Managemen P og am (FEMP)[55], ASHRAE GUIDELINE S anda d 14-2002
[56], [57] and he In e na ional Pe o mance Measu emen and Ve i ica ion P o ocol (IPMVP)
[58] ha e been ollowed. This alida ion p ocess makes i possible o e i y i he esul s o he
es ima ions o he clima e zone o Se ille a e ela ed o he eal alues moni o ed wi h a
ma gin o e o . Fo his, he ollowing o mulas a e used:
𝑀𝐵𝐸 =∑(𝑚𝑖−𝑠𝑖)
𝑛
𝑖=1
𝑛 (1)
𝑀𝐵𝐸 =∑(𝑚𝑖−𝑠𝑖)
𝑛
𝑖=1
𝑛 (2)
𝑁𝑀𝐵𝐸 =1
𝑚
∙∑(𝑚𝑖−𝑠𝑖)
𝑛
𝑖=1
𝑛−𝑝 (3)
In o de o scale he esul s o he Mean Bias E o (MBE) (Equa ion 1), No malized Mean Bias
E o (NMBE) is used by di iding MBE by he mean o measu ed alues (𝑚), gi ing he global
di e ence be ween he eal alues and he p edic ed ones (Equa ion 2); In hese equa ions mi
is he measu ed alue, si is he simula ed one and n he numbe o measu ed da a poin s. The
Coe icien o Va ia ion o he Roo Mean Squa e E o CV (RMSE) measu es he a iabili y o
he e o s be ween measu ed and simula ed alues (Equa ion 3).
The limi alues associa ed wi h hese pa ame e s a y depending on whe he hey a e hou ly
o mon hly. Conside ing ha he en i onmen al a iables a e hou ly, he limi alues a e
be ween -10% and 10% in NMBE and less han 30% in CV (RMSE) [56,57].
To es ima e u u e clima e scena ios in he clima e zone, he CCWo ldWea he Gen ool om
he UK Me O ice Hadley Cen e Coupled Model 3 HadCM3 has been used. Th ough a mo phing
p ocess, his ool gene a es wea he iles adap ed o clima e change om any loca ion in he
wo ld and c ea es iles compa ible wi h mos building pe o mance simula ion p og ams [59].
The mo phing o he clima e iles has been ca ied ou o scena io A2 o g eenhouse gas
emissions acco ding o he IPCC [60], esul ing in clima e scena io iles o yea s 2050 and 2080.
2.3.2. Es ima ion o u u e ime se ies o he case s udy
The da a mining app oach o his s udy was pe o med using he M5P algo i hm. This algo i hm
allows es ima ing he ime se ies o he yea s 2050 and 2080 in he measu ed in e io poin o
he case s udy. The M5P algo i hm (also known as M5') is an e olu ion o he Classi ica ion and
eg ession ee (CART) [61,62]. Unlike CART, M5P combines decision ees wi h mul i a ia e
eg ession; a decision ee is buil ollowing he same in e ed ee s uc u e o e ed by CART,
bu a mul iple linea eg ession (MLR) model is i ed on each shee (Equa ion 4). The e o e, he
algo i hm wo ks by de eloping an MLR model in each sub egion (Fig. 7). In he p ocess o
de eloping he M5P algo i hm ee, ins ead o maximizing he in o ma ion gain, he in e nal
a ia ion o he subse s o he class alues o each b anch is minimized. Once he model is buil ,
p uning educes o e i ing. The ad an ages o he models gene a ed by his algo i hm a e ha
hey e icien ly handle la ge amoun s o nume ical a iables and a e obus in he absence o
alues in he ins ances o he analyzed da ase [63,64]. Thus, i s use o he cha ac e iza ion o
di e en aspec s o buildings has inc eased in ecen yea s [65,66].
𝑌
𝑀𝐿𝑅 = 𝛽0+∑ (𝛽𝑖𝑥𝑖)
𝑣
𝑖=1 + 𝜀 (4)
Whe e 𝛽0 is he independen e m, 𝛽𝑖 a e he eg ession coe icien s, 𝑥𝑖 a e he p edic o
a iables, and 𝜀 is he e o .
Fig. 7. Scheme o an M5P model.
The app oach designed had a con igu a ion o 6 inpu a iables co esponding o he hou ly
alues o empe a u e, ela i e humidi y, dew poin empe a u e, ho izon al in a ed adia ion
in ensi y, global ho izon al adia ion, and di ec ion no mal sola i adiance, di use ho izon al
sola i adiance o he ime se ies o he ou e poin o Se ille. The ou pu a iables co espond
o he ela i e humidi y and he empe a u e a he in e io poin o he case s udy. Fo each
ou pu a iable a di e en M5P was designed.
The aining o each M5P model was ca ied ou using he da a om he cu en scena io. Once
he models we e ained, ime se ies o he yea s 2050 and 2080 we e es ima ed. I is wo h
no ing ha he M5P models we e alida ed using he coe icien o de e mina ion and he MBE
index as s a is ical pa ame e s.
3. RESULTS AND DISCUSSION
3.1 The mo-hyg ome ic condi ions o he Palace o Ped o I
The esul s ob ained om he moni o ing ca ied ou be ween Ma ch 2020 and Feb ua y 2021
in he sou heas galle y o he Cou ya d o he Maidens a e shown in Fig. 8 and able 1:
Fig. 8 En i onmen al condi ions in he sou heas co ne o he Cou ya d o he Maidens om Ma ch 2020 o
Feb ua y 2021.
The measu emen o he T and RH pa ame e s is ele an mainly o he s udy o he
luc ua ions and he mal jumps ha may occu du ing he day-nigh hou s o du ing he
di e en seasons o he yea . In his ega d, i is obse ed ha he mos p onounced he mal
jumps be ween day and nigh hou s do no exceed 5ºC (July and Augus ), a ac ha -by i sel -
does no ep esen any isk in e ms o ma e ial expansion. Finally, a high RH in win e is
eco ded and mus be con olled wi h adequa e en ila ion, al hough i is wi hin he accep able
ange o he co ec p ese a ion o plas e wo k.
Table 1. Summa y o en i onmen al pa ame e s om 01/03 /2020 o 28 / 02/2021.
Mon hs
RELATIVE HUMIDITY (%)
TEMPERATURE (° C)
Min. Values
egis e ed
Max. alues
egis e ed
Δ
(max-min)
Min. Values
egis e ed
Max. alues
egis e ed
Δ
(max-min)
A e age
SD
A e age
SD
A e age
SD
A e age
SD
Feb 20
55.3
11,001
73.8
9,454
18.5
14
1,319
18.1
1,577
4.1
Ap. 20
60.6
6,672
77.7
4.79
17.1
15.7
1,534
19.2
1,558
3.5
May 20
43.3
10,818
64.1
9,947
20.8
21.3
2,337
26.1
2,832
4.8
Jun 20
41.4
5,608
62
7.87
20.6
22.3
2,434
27.3
2,314
5
Jul. 20
38
5,877
56
7.43
18
27.6
1,448
31.8
1,328
4.2
Aug. 20
38
8,637
56.5
7,966
18.5
26.3
1,675
30.5
1,581
4.2
Sep . 20
44.1
9,493
58.3
8,287
14.2
24
1.9
27.4
2,105
3.4
Oc . 20
49.3
12,985
63.1
11,674
13.8
18.1
1,996
21.3
2,136
3.2
No . 20
70
8,459
78.6
7,134
8.6
15.6
1,964
17.9
1,965
2.3
Dec. 20
74.2
10,822
84.4
8,133
10.2
11.4
2,366
13.7
1,935
2.3
Jan. 21
75.2
9,545
86.1
8,141
10.9
8.6
3,493
11.2
3,093
2.6
Feb. 21
77.4
5,562
83.1
5,129
5.7
12.9
1,127
15.1
0.854
2.2
Fig. 9 G aphical ep esen a ion o ambien condi ion (T and RH) om 01/03 /2020 o 28 / 02/2021.
F om a de ailed analysis o he measu emen s ob ained (Table 1 and Fig 9), he isk due o high
RH is concen a ed om No embe o Feb ua y, bo h inclusi e; being Janua y he mos
un a o able mon h since 27.02% o he measu emen s eco ded du ing ha mon h exceed he
limi o 90% RH. Likewise, he isk o dehyd a ion o he gypsum and ans o ma ion in o
bassani e (CaSO4·1/2H2O) occu s om May o Oc obe , bo h inclusi e; being July and Augus
he mos c i ical mon hs o such pu poses. I is wo h o highligh ha a 56.71% o he
measu emen s eco ded in he yea emain below he limi es ablished in he ma ix (Fig. 10).
The da a ob ained co espond o a o al o 8,760 measu emen s, o which only 11.76% a e
ou side he limi s es ablished in he ma ix ha ela es he mo-hyg ome ic condi ions wi h he
mine alogical phases o calcium sul a e, depending on hei deg ee o hyd a ion.
Fig. 10 Dispe sion g aph o he measu emen s om 01/03 /2020 o 28 / 02/2021 a he sou heas co ne o he
Cou ya d o he Maidens.
Rega ding he ype o isk de ec ed, i can be seen in Fig. 10 how i is mainly dis inguished
be ween isk due o in e nal s uc u al ans o ma ions (associa ed wi h he summe mon hs)
and isk due o high RH (associa ed wi h he win e mon hs). The 7.95% o he hou ly eco ds
co espond o measu emen s ha a e below he cu e es ablished by Winkle & Wilhelm [31] ,
while he isk due o a high RH co esponds o 3.65% and he isk due o a low T co esponds o
only 10 obse a ions h oughou he yea , so he cu en clima ic condi ions in he Cou ya d o
he Maidens a e a o able in his sense.
Fig. 11 Mon hly ep esen a ion o he hou ly measu emen s ob ained in he Cou ya d o he Maidens om Ma ch
2020 o Feb ua y 2021.
Howe e , he esul s ob ained om moni o ing he clima e o he ci y o Se ille show di e en
ends in conse a ion isks (Fig. 11). As i can be seen, al hough he mos p ominen isks a e
once again ha o in e nal ans o ma ions in summe and high RH in win e , he e is a g ea e
numbe o obse a ions ha emain wi hin hese isks. Thus, he obse a ions a isk o in e nal
ans o ma ions co espond o alues be ween 348 and 360 in July and Augus , while in he
Cou ya d o he Maidens he alues we e 229 and 168, espec i ely. Likewise, i was also
de ec ed how he isk due o high RH ex ends o a g ea e numbe o mon hs han in he case
o he Cou ya d o he Maidens and wi h a signi ican ly highe numbe o obse a ions.
Fig. 12 Mon hly ep esen a ion o he hou ly measu emen s ob ained in Se ille om Ma ch 2020 o Feb ua y 2021.
The e o e, signi ican a ia ions a e de ec ed in he esul s ob ained wi h he wo app oaches.
While he app oach o he analysis o he Cou ya d o he Maidens poin (wi h a clea
mic oclima e) p esen s less se e e condi ions o he conse a ion o plas e wo k, ou side he
condi ions would be di e en , p esen ing a g ea e isk due o he clima e in hese elemen s.
Likewise, he a iabili y de ec ed be ween he wo app oaches shows he g ea a ia ion ha
he analysis app oaches o he in luence o clima e can p esen on he conse a ion o
plas e wo k. Thus, he use o a join app oach o he ou doo clima e and mic oclima es in his
ype o case s udy allows us o show a b oade pe spec i e o he isk o conse a ion o
plas e wo k.
3.2 Calib a ion o ou doo measu emen s. The clima e ile o Se ille.
Despi e he possible isks de ec ed in he cu en scena io, clima e change can gene a e a
change in he conse a ion condi ions o plas e wo k. Fo ha , he u u e ime se ies o he
wo p e iously alida ed app oaches used we e es ima ed in his s udy. In he case o he
ou doo clima e app oach, T and RH pa ame e s measu ed du ing he ou doo moni o ing a e
compa ed wi h T and RH da a ex ac ed om he CTE o he clima e zone o Se ille Csa
(Medi e anean). As he en i onmen al a iables analyzed a e hou ly, he limi alues a e
be ween -10% and 10% in NMBE and less han 30% in CV (RMSE). Acco ding o Table 2, i is
de ec ed ha he CV (RMSE) alues a e wi hin he es ablished e o limi s. Howe e , i is
de ec ed ha T is sligh ly highe (0.88%) han he limi s in he NMBE and in ma e s ela ed o
RH he alue exceeds he limi in he NMBE by 6.40% (Table 2). F om hese esul s we can
es ablish he ollowing conside a ions:
In he i s place, conside ing ha a measu emen is being compa ed unde speci ic
en i onmen al condi ions wi h a ypical me eo ological yea acco ding o he c i e ia
es ablished in he CTE and based on he a ailable me eo ological s a ions, he le el o
app oxima ion is good, especially i we conside he CV (RMSE).
7. ACKNOWLEDGEMENTS
This esea ch has been ca ied ou hanks o he inancing o he p ojec "DEVELOPMENT AND
EVALUATION OF DURABILITY MODELS AND PREVENTIVE CONSERVATION OF
DECORATIVE ELEMENTS FROM THE HISTORICAL PLASTERS OF THE ROYAL ALCÁZAR DE
SEVILLA" (PGC2018-093470-B-I00) by he Minis y o Science, Inno a ion and Uni e si ies o
he Go e nmen o Spain and he inancing ecei ed om he VI PPIT-2021-I.3 om he
Uni e si y o Se ille.
The au ho s wish o hank he collabo a ion o he Boa d o T us ees o he Real Alcáza o
Se ille o he acili ies p o ided o achie e isual inspec ion and measu emen o
en i onmen al pa ame e s.
8. REFERENCES
[1] E. Lucchi, Re iew o p e en i e conse a ion in museum buildings, J. Cul . He i . 29
(2018) 180–193. h ps://doi.o g/10.1016/j.culhe .2017.09.003.
[2] D. Camu o, Mic oclima e o Cul u al He i age: Conse a ion, Res o a ion, and
Main enance o Indoo and Ou doo Monumen s, Else ie , 2019.
[3] I. Cook, R. Johns on, K. Selby, Clima e Change and Cul u al He i age: A Landscape
Vulne abili y F amewo k, J. Isl. Coas . A chaeol. (2019).
h ps://doi.o g/10.1080/15564894.2019.1605430.
[4] C. Fe ei a, J. Ba elas, A. Sil a, J. de B i o, I.S. Dias, I. Flo es-Colen, Impac o
en i onmen al exposu e condi ions on he main enance o acades’ claddings,
Buildings. 11 (2021). h ps://doi.o g/10.3390/buildings11040138.
[5] V. Cos anzo, K. Fabb i, E. Schi o, M. P e elli, L. Ma le a, Mic oclima e moni o ing and
conse a ion issues o a Ba oque chu ch in I aly: a isk assessmen analysis, Build. Res.
In . 0 (2021) 1–19. h ps://doi.o g/10.1080/09613218.2021.1899797.
[6] S.P. Co gna i, M. Filippi, Assessmen o he mo-hyg ome ic quali y in museums:
Me hod and in- ield applica ion o he “ Duccio di Buoninsegna” exhibi ion a San a
Ma ia della Scala (Siena, I aly), J. Cul . He i . 11 (2010) 345–349.
h ps://doi.o g/10.1016/j.culhe .2009.05.003.
[7] M. Za zo, A. Fe nández-Na ajas, F.J. Ga cía-Diego, Long- e m moni o ing o esco
pain ings in he ca hed al o Valencia (Spain) h ough humidi y and empe a u e senso s
in a ious loca ions o p e en i e conse a ion, Senso s. 11 (2011) 8685–8710.
h ps://doi.o g/10.3390/s110908685.
[8] N. I aliana, UNI 10829. Wo ks o a o his o ical impo ance. Ambien condi ions o he
conse a ion. Measu emen and analysis. [Beni di in e esse s o ico e a is ico.
Condizioni ambien ali di conse azione. Misu azione ed analisi.], (1999) 1–24.
[9] A.E. de N. UNE, UNE-EN 15757: Conse a ion o Cul u al P ope y - Speci ica ions o
empe a u e and ela i e humidi y o limi clima e-induced mechanical damage in
o ganic hyg oscopic ma e ials, (2011).
[10] Ansi/Ash ae, ANSI/ASHRAE 55:2004 The mal En i onmen al Condi ions o Human
Occupancy, Ash ae. (2004) 30.
[11] PAS 198:2012 - Speci ica ion o managing en i onmen al condi ions o cul u al
collec ions – BSI B i ish S anda ds, (n.d.).
h p://shop.bsig oup.com/P oduc De ail/?pid=000000000030219669.
[12] UNE-EN-15759-1, Conse a ion o cul u al p ope y. Indoo clima e. Pa 1: Guidelines
o hea ing chu ches, chapels and o he places o wo ship, (2012).
[13] C. Bonacina, P. Baggio, F. Cappelle i, P. Romagnoni, A.G. S e an, The Sc o egni
Chapel: The esul s o o e 20 yea s o indoo clima e moni o ing, Ene gy Build. 95
(2015) 144–152. h ps://doi.o g/10.1016/j.enbuild.2014.12.018.
[14] D. Bien enido-Hue as, M. León-Muñoz, J.J. Ma ín-del-Río, C. Rubio-Bellido, Analysis
o clima e change impac on he p ese a ion o he i age elemen s in his o ic buildings
wi h a de icien indoo mic oclima e in wa m egions, Build. En i on. 200 (2021) 107959.
h ps://doi.o g/10.1016/j.builden .2021.107959.
[15] G. Zonno, R. Aguila , R. Bo oschek, P.B. Lou enço, Analysis o he long and sho - e m
e ec s o empe a u e and humidi y on he s uc u al p ope ies o adobe buildings
using con inuous moni o ing, Eng. S uc . 196 (2019) 109299.
h ps://doi.o g/10.1016/j.engs uc .2019.109299.
[16] R.P. K ame , M.P.E. Maas, M.H.J. Ma ens, A.W.M. an Schijndel, H.L. Schellen, Ene gy
conse a ion in museums using di e en se poin s a egies: A case s udy o a s a e-o -
he-a museum using building simula ions, Appl. Ene gy. 158 (2015) 446–458.
h ps://doi.o g/10.1016/j.apene gy.2015.08.044.
[17] M. Napp, T. Kalamees, Ene gy use and indoo clima e o conse a ion hea ing,
dehumidi ica ion and adap i e en ila ion o he clima e con ol o a mediae al chu ch
in a cold clima e, Ene gy Build. 108 (2015) 61–71.
h ps://doi.o g/10.1016/j.enbuild.2015.08.013.
[18] C. Co na o, V.A. Puggioni, R.M. S ollo, Dynamic simula ion and on-si e measu emen s
o ene gy e o i o complex his o ic buildings: Villa Mond agone case s udy, J. Build.
Eng. 6 (2016) 17–28. h ps://doi.o g/10.1016/j.jobe.2016.02.001.
[19] T. Ca dinale, G. Rospi, N. Ca dinale, The in luence o indoo mic oclima e on he mal
com o and conse a ion o a wo ks: The case s udy o he Ca hed al o Ma e a (Sou h
I aly), in: Ene gy P ocedia, Else ie L d, 2014: pp. 425–432.
h ps://doi.o g/10.1016/j.egyp o.2014.10.398.
[20] A. Re i, D. Sa e hwai e, F. A agon-Du and, J. Co ee-Mo lo , R.. R. Kiunsi, M. Pelling,
D. Robe s, W. Solecki, S.P. Gajja , A. S e dlik, Towa ds ans o ma i e adap a ion in
ci ies: he IPCC’s Fi h Assessmen , En i on. U ban. 26 (2014) 11–28.
h ps://doi.o g/10.1177/0956247814523539.
[21] A.J. P ie o, K. Ve iche , A. Sil a, J. de B i o, On he impac s o clima e change on he
unc ional de e io a ion o he i age buildings in Sou h Chile, Build. En i on. 183 (2020)
107138. h ps://doi.o g/10.1016/j.builden .2020.107138.
[22] V. Rajčić, A. Skende , D. Damjano ić, An inno a i e me hodology o assessing he
clima e change impac on cul u al he i age, In . J. A chi . He i . 12 (2018) 21–35.
h ps://doi.o g/10.1080/15583058.2017.1354094.
[23] A. Haugen, C. Be olin, G. Leijonhu ud, T. Ols ad, T. B os öm, A me hodology o
long- e m moni o ing o clima e change impac s on his o ic buildings, Geosci. 8 (2018).
h ps://doi.o g/10.3390/geosciences8100370.
[24] S. Ramí ez, M. Za zo, A. Pe les, F.J. Ga cía-Diego, A me hodology o disc iminan ime
se ies analysis applied o mic oclima e moni o ing o esco pain ings, Senso s
(Swi ze land). 21 (2021) 1–29. h ps://doi.o g/10.3390/s21020436.
[25] M. Gómez He as, La Tempe a u a en los Ma e iales del Pa imonio, (2012) 87–95.
h p://digi al.csic.es/handle/10261/46794.
[26] Z. Pa lík, J. Foř , R. Če ný, An in si u moni o ing sys em o he s udy o en i onmen al
in luences on du abili y and he des uc i e p ocess o building ma e ials and s uc u es,
WIT T ans. Modelling Simul. 55 (2013) 287–296. h ps://doi.o g/10.2495/CMEM130231.
[27] S. Michalski, Tempe a u a Inco ec a, Can. Conse . Ins i u e. ICCROM. (2009).
[28] S. Michalski, Humedad ela i a Inco ec a, Can. Conse . Ins i u e. ICCROM. (2009).
[29] L. Villanue a Domínguez, A. Ga cía San os, Manual del yeso, CIE In e siones Edi o iales,
2001.
[30] D. F eye , W. Voig , C ys alliza ion and Phase S abili y o CaSO4 and CaSO 4 - Based
Sal s, Mona she e Fu Chemie. 134 (2003) 693–719. h ps://doi.o g/10.1007/s00706-
003-0590-3.
[31] E.M. Winkle , E.J. Wilhelm, Sal bu s by hyd a ion p essu es in a chi ec u al s one in
u ban a mosphe e, Geol. Soc. Am. Bull. 81 (1970) 567–572.
h ps://doi.o g/h ps://doi.o g/10.1130/0016-7606(1970)81[567:SBBHPI]2.0.CO;2.
[32] L. Ri e bach, P. Becke , Tempe a u e and humidi y dependen o ma ion o
CaSO4·xH2O (x = 0...2) phases, Glob. Plane . Change. 187 (2020).
h ps://doi.o g/10.1016/j.gloplacha.2020.103132.
[33] E. Goossens, Mois u e ans e p ope ies o coa ed gypsum, Technische Uni e si ei
Eindho en, Facul ei Bouwkunde, 2003. h ps://doi.o g/10.6100/IR571306.
[34] C.A.C. Mesqui a, Re es imien os Con inuos In e io es de Va ias Capas con
Ca ac e ís icas de Ba e a de Vapo e Hig oscopicidad, Doc o al disse a ion,
Poly echnic Uni e si y o Mad id, 2012.
[35] R. Rubio Domene, Yese ías de la Alhamb a. His o ia, écnica y conse ación, Pa ona o
de la Alhamb a y Gene ali e. Uni e si y o G anada, G anada, 2010.
[36] M. To es-González, F.J. Alejand e, V. Flo es-alés, A.I. Cale o-cas illo, F.J. Blasco-lópez,
Analysis o he s a e o conse a ion o his o ical plas e wo k h ough isual inspec ion
and non-des uc i e es s . The case o he uppe ieze o he Toledanos Room ( The
Royal Alcáza o Se ille , Spain ), J. Build. Eng. 40 (2021) 1–14.
h ps://doi.o g/h ps://doi.o g/10.1016/j.jobe.2021.102314.
[37] R. Campos de Al ea , The main enance and he p e en i e p ese a ion measu es o he
cul u al goods in he Royal Alcáza o Se ille, Apun . Del Alcáza Se illa. 18 (2018) 71–
87.
[38] E. Co ea Gómez, R. Rubio Domene, El yeso. Las deco aciones de yeso en época naza í,
in: P. de la A. y Gene ali e (Ed.), Man. Buenas P ác icas. Res au ación Made a, Yeso y
Ce ámica, Pa ona o, Conseje ía de Educación, Cul u a y Depo e de la Jun a de
Andalucía, G anada, 2014: pp. 43–52.
[39] J.M. Cab e a Ga ido, La in luencia de los con aminan es en el Pa imonio a ís ico
Nacional, Econ. Ind. 107 (1972) 51–60.
[40] H. Vii anen, T. Ojanen, Imp o ed Model o P edic Mold G ow h in Building Ma e ials,
in: The m. Pe o m. Ex e . En el. Whole Build. X–P oceedings CD, 2007: pp. 2–7.
[41] G.A. Tsongas, F. Rio oan, Minimum condi ions o isible mold g ow h, ASHRAE J. 58
(2016) 32–43.
[42] D. Sanz A auz, Análisis del yeso empleado en e es imien os ex e io es median e
écnicas geológicas, Doc o al disse a ion, Uni e sidad Poli écnica de Mad id, 2009.
h p://oa.upm.es/1711/1/DAVID_SANZ_ARAUZ.pd .
[43] A.I. Cale o Cas illo, A. Ga cía Bueno, O. López C uz, V.J. Medina Fló ez, La polic omía
o iginal de las yese ías del Pa io de las Doncellas del Real Alcáza de Se illa. Ma e iales
cons i u i os y écnicas de ejecución, A queol. y Te i . Medie . 24 (2017) 255–290.
h ps://doi.o g/10.17561/ay m. 24i0.9.
[44] C. Sabbioni, P. B imblecombe, M. Cassa , The A las o Clima e Change Impac on
Eu opean Cul u al He i age: Scien i ic Analysis and Managemen S a egies, An hem
P ess, London, UK, 2010.
[45] A. Almag o Go bea, El Alcáza de Se illa Un palacio musulmán pa a un ey c is iano, in:
C is. y Musulmanes En La Península Ibé ica La Gue . La F on . y La Con i encia. XI
Cong . Es ud. Medie ., 2007: pp. 331–365.
[46] F.J. Blasco López, Yese ías medie ales de adición islámica del Real Alcáza de Se illa:
Re isión His o iog á ica, Me odología pa a la ca ac e ización, e aluación de su
du abilidad y elabo ación de un in en a io, Doc o al disse a ion, Uni e si y o Se ille,
2011.
[47] A.I. Cale o Cas illo, Ma e iales, écnicas y p ocedimien os en la deco ación
a qui ec ónica. Aplicaciones a la conse ación y es au ación de las yese ías del Pa io de
las Doncellas. Real Alcáza de Se illa., Doc o al disse a ion, Uni e sidad de G anada,
2016. h p://hdl.handle.ne /10481/43864.
[48] A. Pleguezuelo, Tile-wo k in he mudéja palace in he Royal Alcáza o Se ille. A
P elimina y Visual Analysis., Apun . Del Alcáza Se illa. 16 (2015) 219–230.
[49] C. En íquez Díaz, J.R. Baeza Ál a ez, A p ojec o he es o a ion o he ilings o he
g ound loo o he mudéja palace, Apun . Del Alcáza Se illa. 19 (2019) 65–77.
[50] C. Cañas Palop, Las a madu as de cubie as mudéja es del palacio de Ped o I, del Alcáza
de Se illa: análisis in eg al y p opues as pa a la es au ación, Doc o al disse a ion,
Uni e si y o Se ille, 2006. h ps://dialne .uni ioja.es/se le / esis?codigo=23333.
[51] S. Fe nández Aguile a, Po a en ane os mudéja es en el Real Alcáza de Se illa, A chi o
Hi, Dipu ación P o incial de Se illa, Se illa, 2012.
[52] F.M. Tubino, Es udios sob e el a e en España. La a qui ec u a hispano- isigoda y á abe
española. El Alcáza de Se illa. Una iglesia mozá abe, 1886.
[53] The Go e nmen o Spain, Royal Dec ee 314/2006. App o ing he Spanish Technical
Building Code, Mad id, Spain, 2013.
[54] M. Ko ek, J. G iese , C. Beck, B. Rudol , F. Rubel, Wo ld map o he Köppen-Geige
clima e classi ica ion upda ed, Me eo ol. Zei sch i . 15 (2006) 259–263.
h ps://doi.o g/10.1127/0941-2948/2006/0130.
[55] L. Webs e , J. B ad o d, Measu emen and Ve i ica ion o Fede al Ene gy (M&V
Guidelines), 2008.
[56] ANSI/ASHRAE, ASHRAE Guideline 14-2002 Measu emen o Ene gy and Demand
Sa ings, Ash ae. 8400 (2002) 170.
[57] G.R. Ruiz, C.F. Bande a, Valida ion o calib a ed ene gy models: Common e o s,
Ene gies. 10 (2017). h ps://doi.o g/10.3390/en10101587.
[58] E iciency Valua ion O ganiza ion, In e na ional Pe o mance Measu emen &
Ve i ica ion P o ocol, Handb. Financ. Ene gy P oj. I (2016) 122.
[59] M.F. Jen sch, A.S. Bahaj, P.A.B. James, CCWo ldWea he Gen, Clima e change wo ld
wea he ile gene a o , Ve sion 1.8, Sus ain. Ene gy Res. G . (2013).
[60] I.P. on C. Change, Summa y o Policymake s, in: In e go e nmen al Panel on Clima e
Change (Ed.), Clim. Chang. 2013 - Phys. Sci. Basis, Camb idge Uni e si y P ess,
Camb idge, 2014: pp. 1–30. h ps://doi.o g/10.1017/CBO9781107415324.004.
[61] J.R. Quinlan, o he s, Lea ning wi h con inuous classes, in: 5 h Aus . J . Con . A i . In ell.,
1992: pp. 343–348.
[62] Y. Wang, I.H. Wi en, Induc ion o model ees o p edic ing con inuous classes, in: Eu .
Con . Mach. Lea n., P ague: Uni e si y o Economics, Facul y o In o ma ics and
S a is ics, 1997.
[63] A. Behnood, V. Behnood, M. Modi i Gha eh e an, K.E. Alyamac, P edic ion o he
comp essi e s eng h o no mal and high-pe o mance conc e es using M5P model ee
algo i hm, Cons . Build. Ma e . 142 (2017) 199–207.
h ps://doi.o g/10.1016/j.conbuildma .2017.03.061.
[64] L. Lin, Q. Wang, A.W. Sadek, A combined M5P ee and haza d-based du a ion model
o p edic ing u ban eeway a ic acciden du a ions, Accid. Anal. P e . 91 (2016) 114–
126. h ps://doi.o g/10.1016/j.aap.2016.03.001.
[65] F. A sa ian, A. Sabe , A. Pou zangba , A.G. Olabi, M.A. Khanmohammadi, Analysis o
ecycled agg ega es e ec on ene gy conse a ion using M5″ model ee algo i hm,
Ene gy. 156 (2018) 264–277. h ps://doi.o g/10.1016/j.ene gy.2018.05.099.
[66] C.F. Je ey Kuo, C.H. Lin, M.H. Lee, Analyze he ene gy consump ion cha ac e is ics
and a ec ing ac o s o Taiwan’s con enience s o es-using he big da a mining
app oach, Ene gy Build. 168 (2018) 120–136.
h ps://doi.o g/10.1016/j.enbuild.2018.03.021.