New insights into the climate of northern Iberia during the Younger Dryas and Holocene: The Mendukilo multi-speleothem record

New insigh s in o he clima e o no he n Ibe ia du ing he Younge
D yas and Holocene: The Mendukilo mul i-speleo hem eco d
J.L. Be nal-Wo mull
a
,
b
,
*
, A. Mo eno
a
, M. Ba olom
e
a
, M. A iolabengoa
c
,
C. P
e ez-Mejías
d
, E. I ia e
e
,C.Os

aca
b
,C.Sp
€
o l
, H. S oll
g
, I. Cacho
h
, R.L. Edwa ds
i
,
H. Cheng
d
,
j
a
Py enean Ins i u e o Ecology - CSIC, 50059, Za agoza, Spain
b
Depa men o Ea h Sciences, Uni e si y o Za agoza, C/ Ped o Ce buna 12, 50009, Za agoza, Spain
c
Depa men o Geology, Uni e si y o he Basque Coun y, Leioa, Spain
d
Ins i u e o Global En i onmen al Change, Xi'an Jiao ong Uni e si y, 710054, Xi'an, China
e
Labo a o y o Human E olu ion-IsoTOPIK S able Iso ope Labo a o y, Depa men o His o y, Geog aphy &Communica ion, Edificio de IþDþi, Uni e sidad
de Bu gos, Pl. Misael Ba~
nuelos s/n, 09001, Bu gos, Spain
Ins i u e o Geology, Uni e si y o Innsb uck, Inn ain 52, 6020, Innsb uck, Aus ia
g
Depa men o Ea h Sciences, ETH Zü ich, 8092, Zü ich, Swi ze land
h
GRC Geoci
encies Ma ines, Dep . Din
amica de la Te a i de l'Oce
a, Facul a de Ci
encies de la Te a, Uni e si a de Ba celona, 28080, Ba celona, Spain
i
Uni e si y o Minneso a, Minneapolis, MN, 55455, USA
j
Ins i u e o Ea h En i onmen , Chinese Academy o Sciences, Xi'an, China
a icle in o
A icle his o y:
Recei ed 7 No embe 2022
Recei ed in e ised o m
25 Janua y 2023
Accep ed 10 Feb ua y 2023
A ailable online xxx
Handling Edi o : Mi a Ma hews
Keywo ds:
Holocene
Younge d yas
Ibe ian Peninsula
Speleo hem
No h A lan ic
S able iso opes
Ab up changes
abs ac
Recen hyd oclima e s udies on he Ibe ian Peninsula ha e shown a complex egional pa e n in iming
and in ensi y o clima e change spanning he Younge D yas and he Holocene. These changes a e due o
mul i ace ed in e ac ions be ween clima e a iabili y ha cha ac e izes he A lan ic Ocean egion and
hyd oclima ic p ocesses associa ed wi h he Medi e anean clima e, hus making i di ficul o econ-
s uc cen ennial- and millennial-scale a iabili y in ain all. In his s udy we p esen a composi e and
con inuous iso opic eco d (
d
13
C and
d
18
O) consis ing o ou s alagmi es om Mendukilo ca e (MEN
composi e) in he wes e n Py enees co e ing he Younge D yas and he en i e Holocene. This eco d
e eals millennial-scale shi s in ca bon iso opes in esponse o changes in he hyd oclima e in he
no he n pa o Ibe ia. The MEN oxygen iso opes show li le a ia ion on millenial ime scales bu e eal
cen ennial changes ha co ela e wi h No h A lan ic e en s (e.g., he 8.2 ky BP cooling e en ). We
obse e a delay in he onse o humid condi ions in he ea ly Holocene and a subsequen end owa ds
d ie and colde condi ions be ween 6.0 and 2.5 ky BP. This new, high- esolu ion and eplica ed spe-
leo hem eco d deno es he complex connec ion ha exis s be ween he No h A lan ic and Wes e n
Eu ope du ing las millennia and he s ong egional he e ogenei y o he hyd oclima e o Ibe ia du ing
his ime.
©2023 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY license
(h p://c ea i ecommons.o g/licenses/by/4.0/).
1. In oduc ion
The ansi ion om he cold and d y Younge D yas o he ea ly
Holocene was one o he mos apid wa ming e en s on a global
scale. This wa ming was associa ed wi h a esump ion o he
A lan ic Me idional O e u ning Ci cula ion (AMOC) (McManus
e al., 2004). In he No h A lan ic, he onse o he Holocene was
punc ua ed by sho cold pe iods due o a iable mel wa e inpu .
A la i udes below 45

N, including he Medi e anean egion,
deglacial wa ming was influenced by o bi ally d i en changes
(Renssen e al., 2009) ha modula ed Holocene hyd oclima e
a iabili y. The gene al e olu ion om we e o d ie clima ic
condi ions be ween he ea ly and la e Holocene h oughou
sou he n Eu ope (Robe s e al., 2019) is consis en wi h he long-
e m change in insola ion, and hence i is influenced by season-
ali y (Wanne e al., 2008). In ac , he Holocene Medi e anean
clima e has been adi ionally di ided in o h ee in e als
*Co esponding au ho . Py enean Ins i u e o Ecology - CSIC, 50059, Za agoza,
Spain.
E-mail add ess: [email p o ec ed] (J.L. Be nal-Wo mull).
Con en s lis s a ailable a ScienceDi ec
Qua e na y Science Re iews
jou nal homepage: www.else ie .com/loca e/quasci e
h ps://doi.o g/10.1016/j.quasci e .2023.108006
0277-3791/©2023 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY license (h p://c ea i ecommons.o g/licenses/by/4.0/).
Qua e na y Science Re iews 305 (2023) 108006
acco ding o wa e a ailabili y: (1) a fi s pa cha ac e ized by
humid condi ions (11.7e7.0 ky BP), (2) he mid-Holocene wi h
inc eased hyd oclima ic a iabili y bu in gene al highe empe a-
u es (7.0e5.5 ky BP), and (3) a dec ease in humidi y since ~5.5 ky
BP associa ed wi h he onse o he cu en Medi e anean- ype
clima e (Jalu e al., 2009;Magny e al., 2011). In he case o he
Ibe ian Peninsula which is influenced by wo con as ing clima e
egimes (A lan ic and Medi e anean), he hyd oclima e e olu ion
du ing he deglacial wa ming and he Holocene was cha ac e ized
by a s ong egional he e ogenei y (e.g., Ca i
on e al., 2010). Thus,
he obse ed changes in con inen al eco ds (lake sedimen s and
speleo hems) (e.g., Gonz
alez-Samp
e iz e al., 2017;Mo ell
on e al.,
2018;Mo eno e al., 2017) in Ibe ia a e no consis en in space o
ime wi h he simplified pic u e ou lined o he Medi e anean
egion leading o appa en con adic ions in empe a u e and
p ecipi a ion econs uc ions ha need o be unde s ood on a
egional scale. Di e en eco ds in he egion ha e highligh ed ha
he some imes con adic o y pa e n o clima e p oxy esul s may
be due o seasonali y changes in ain all (Baldini e al., 2019;
Mo ell
on e al., 2009;Mo eno e al., 2017;Walczak e al., 2015),
po en ially due shi s in he posi ion and s eng h o he Azo es
High.
No he n Ibe ia shows la ge en i onmen al changes along a W-
E ansec con olled by he al i ude and p oximi y o he ocean,
esul ing in a la ge empe a u e and humidi y a iabili y du ing
he Holocene (Finn
ee al.,2019). Fo example, he Holocene
ege a ion and clima e succession econs uc ed om lakes and
pea bogs in NW Ibe ia and he Eu osibe ian egion (excluding
eas e n Ibe ia) e eals he adi ional Holocene egional ipa i e
pa e n (Allen e al.,1996;L
opez-Me ino e al., 2012;Mo eno e al.,
2011): Lake eco ds om he Py enees show a p og ession om a
dominan ly A lan ic pa e n in he wes e n sec o (Gonz
alez-
Samp
e iz e al., 2006) o a s ong Medi e anean imp in in he
cen al egion (P
e ez-Sanz e al., 2013). Consis en ly, lake eco ds
p esen in he cen al-sou he n egion o he p e-Py enees show a
delay in he onse o we condi ions (9.5 ky BP) and a dec ease in
humidi y a e 5e4.5 ky BP (Mo ell
on e al., 2009), a pa e n ha
is suppo ed by e es ial p oxies ob ained om wes e n Medi-
e anean ma ine eco ds (Fle che e al., 2013;F igola e al.,
2007). Se e al Holocene speleo hem eco ds ha e also been ob-
ained om ca es in no he n Spain, bu some a e discon inuous
(e.g., Se a do Cou el; Railsback e al., 2011) o do no co e he
en i e in e glacial (Ma ín-Chi ele e al., 2011;Rossi e al., 2018).
In gene al, o bi al o cing clea ly exe ed a fi s -o de con ol on
speleo hem g ow h, as shown by an inc ease in he ela i e
abundance o speleo hems (S oll e al., 2013). Holocene speleo-
hem eco ds also e ealed he impo an ole o he A lan ic
Ocean in shaping cen ennial-scale oscilla ions (Domínguez-Villa
e al., 2017;Smi h e al., 2016), while a compila ion o speleo-
hem eco ds om no he n Spain highligh ed he ole o ain all
seasonali y (Baldini e al., 2019).
He e, we p esen ou new s alagmi es collec ed om Mendu-
kilo ca e in no he n Ibe ia, which p o ide a well- eplica ed, high-
esolu ion and con inuous eco d o no he n Ibe ian clima e om
he beginning o he Younge D yas o p esen -day. S able iso opes,
combined wi h ace elemen s and suppo ed by a de ailed moni-
o ing o he ca e, allow econs uc ing empe a u e and amoun o
ain all du ing las 12,700 yea s, hus add essing he ole o
di e en mechanisms in causing cen ennial-scale oscilla ions, such
as he 8.2 ky e en . The new da a is discussed in bo h si e-specific
and egional con ex s, oge he wi h o he published eco ds om
he A lan ic ma gin o Eu ope, hus o e ing new insigh s in o he
Ibe ian clima e e olu ion on millennial o cen ennial ime scales
since he Younge D yas.
2. Si e desc ip ion
Mendukilo ca e is loca ed a 750 m a.s.l. and 40 km om he
Can ab ian coas (Fig. 1A and B) in a si e o communi y impo ance
wi hin he lands o he As i z council (La aun alley, Na a a), on
he slopes o he eas e n pa o he Sie a de A ala (42

58
0
25
00
N,
1

53
0
45
00
W; Fig. 1B). The ca e de eloped wi hin ee limes ones o
Lowe C e aceous (U gonian, Albian-Ap ian) age a he eas e n
bounda y o he Basque-Can ab ian basin. The clima e o his e-
gion is empe a e wi h cool summe s (Wes coas ma i ime clima e
ollowing he K€
oppen-Geige classifica ion). Rain, og and d izzle
a e abundan , making his a ea one o he ainies loca ions in Spain
(1365 mm o annual p ecipi a ion). Rain all is concen a ed in all,
win e and sp ing, and empe a u es a e mild, wi hou eaching
ex emes alues, eflec ing he p oximi y o he A lan ic Ocean
(Supp. Ma e ial Fig. S1). The su oundings o he ca e a e made up
o small ka s ic moun ain pla eaus and dolines, wi h meadows in
he highes pa s and o es s o beech, oak and abundan moss
co e ing he slopes. Soils a e well de eloped, ich in clay, and
usually less han 50 cm in hickness.
The ca e shows a descending mo phology and consis s o se en
main ooms (En ance hall, In e media e galle y, Laminosin galle y,
Los Lagos galle y, D agon hall, Gue e o and Caballo galle y's)
(Fig. 1C and D). The o al passage leng h is 869 m and he e ical
ex ension is 59 m (see also Supp. Ma e ial).
3. Me hods
3.1. Rain all moni o ing
In o de o explo e he ela ionship be ween he iso opic
composi ion o ain all (
d
18
O
and
d
D
) and he clima e pa ame e s
ou side he ca e, ainwa e was sampled in he Uni e si y o he
Basque Coun y (Bilbao), an a ea wi h a mainly A lan ic sou ce o
ain all and 95 km away om Mendukilo ca e (Fig.1). The sampling
was ca ied ou be ween July 2018 and Feb ua y 2021 and 144 ain
e en s we e sampled. These samples we e analyzed o hei s able
iso opic composi ion using ca i y ing-down spec oscopy (PIC-
ARRO L2130-i) a he Py enean Ins i u e o Ecology (Za agoza). The
esul s a e epo ed in pe mil wi h espec o Vienna S anda d
Mean Ocean Wa e (V-SMOW) and he ep oducibili y o he
measu emen s is ypically 0.1
‰
o
d
18
O and 0.5
‰
e1
‰
o
d
D. The
iso opic esul s we e compa ed o he ai empe a u e measu ed a
he Bilbao ai po me eo ological s a ion on he day o he
espec i e ain e en (Supp. Ma e ial Fig. S2).
3.2. Ca e moni o ing: sampling and analyses
Mendukilo ca e was moni o ed on a seasonal basis o em-
pe a u e, humidi y, pCO
2
, d ipwa e composi ion and a med ca -
bona es om 2018 o 2021. Ca e-ai empe a u e and ela i e
humidi y we e eco ded using HOBO®U23 p o 2 da a logge s in
h ee di e en galle ies (Laminosin, D agon hall and Gue e o;
Fig. 1D). Addi ionally, pCO
2
was measu ed using a pSense Po able
CO
2
me e (model AZ-0001) a he same sampling poin s. P ecipi-
a ion and empe a u e da a ob ained om he Alda z me eo o-
logical s a ion, loca ed 5 km no heas o Mendukilo ca e (Fig. 1B),
we e used o compa ison.
D ipwa e samples we e ob ained om ou di e en d ip si es
(MEN-A, MEN-B, MEN-D and MEN-E; Fig.1D) and, addi ionally, ou
plu iome e s (RAIN-O-MATIC-HOBO coupled wi h a HOBO da a
logge UA-003-64) we e used o moni o d ip a e (Fig. 1D). All
d ipwa e samples we e analyzed o oxygen (
d
18
O
d ip
) and
hyd ogen (
d
D
d ip
) iso ope composi ion ia ca i y ing-down
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
2
spec oscopy (PICARRO L2130-i) a he Py enean Ins i u e o Ecol-
ogy (Za agoza). Resul s a e epo ed in pe mil wi h espec o
Vienna S anda d Mean Ocean Wa e (V-SMOW) and he ep o-
ducibili y o he measu emen s is ypically 0.1
‰
o
d
18
O and
0.5
‰
e1
‰
o
d
D(Fig. 2).
Fa med ca bona e p ecipi a ed on glass pla es (a d ip si es
MEN-A, MEN-D and MEN-E and addi ionally a MEN-F and MEN-G
si es wi hou d ip moni o ing) was analyzed o oxygen and ca bon
iso opes (
d
18
O
a med
and
d
13
C
a med
, epo ed as
‰
wi h espec o
he Vienna Pee Dee Belemni e (VPDB) s anda d). The samples we e
eco e ed seasonally. The MEN-E, MEN-F and MEN-G moni o ing
si es coincide wi h he loca ions o he s alagmi es examined in his
s udy (MEN-3, MEN-4 and MEN-5 espec i ely; MEN-2 was no
benea h an ac i e d ip si e). The fi s ba ch o iso opic analyses (7
samples) was analyzed a he Uni e si y o Innsb uck (Aus ia)
using a The moFishe Del a V Plus linked o a GasBench II, ollowing
he me hodology desc ibed in Sp€
o l (2011). The long- e m ep o-
ducibili y (1 sigma) o
d
18
O is 0.08
‰
and 0.06
‰
o
d
13
C(Sp€
o l,
Fig. 1. - A) Loca ion o Mendukilo ca e ( ed squa e) and o he eco ds o he Ibe ian Peninsula ci ed in his wo k (yellow squa es ¼ca e si es; g een s a s ¼ma ine eco ds; pu ple
hexagon ¼lake eco d; blue ci cle ¼ice ca e): [1] La Ga ma Ca e; [2] Cue a de Asiul; [3] Kai e Ca e; [4] El Pindal Ca e; [6] El Soplao Ca e; [7] Co a de A coia; [8] El Re ugio Ca e; [9]
Ejul e Ca e; [10] Molinos ca e; [11] ODP976; [12] ALB-2; [13] Basa de la Mo a lake; [14] A294 Ice Ca e. B) Regional se ing wi h he loca ion o he ca e ( ed s a ), me eo ological
s a ion (yellow squa e; whe e he empe a u e and ain all da abases o his wo k we e ob ained) and nea by majo ci ies (black hexagons). C) Digi al ele a ion model and plan
iew o he ca e (pink). D) Geomo phological map o Mendukilo ca e showing loca ions o d ipwa e and speleo hem samples.
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
3
Fig. 2. Moni o ing esul s om Mendukilo ca e om No embe 2018 o May 2021. A) Daily and mon hly p ecipi a ion ou side he ca e (Alda z me eo ological s a ion). B) D ip a e
o he di e en sampling poin s inside he ca e. C) Daily su ace a e age empe a u e ou side he ca e. D) Ca e empe a u e and E) CO
2
concen a ion a he Laminosin (ligh blue
line and do s), D agon (g ey line and do s) and Gue e o (blue line and do s) galle ies. F) D ip wa e s able iso ope composi ion o he di e en d ip si es. G)
d
18
O and H)
d
13
C alues
o a med calci e o he di e en moni o ing si es (see Fig. S3 in he Suppl. o mo e in o ma ion).
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
4
2011). The es o he samples we e analyzed a he Iso TOPIK
Labo a o y (Uni e si y o Bu gos; 29 samples) and a he Uni e si y
o Ba celona (10 samples) ollowing simila p ocedu es and iden-
ical equipmen (Fig. 2).
3.3. Mendukilo s alagmi es: pe og aphy, ace elemen s and s able
iso ope analyses
Fou s alagmi es we e collec ed om he Gue e o galle y
(Fig. 1D), a deep galle y a ound 200 m om he ca e en ance.
Excep o MEN-2 hese s alagmi es we e loca ed unde nea h
ac i e d ips ed by di e en ac u e ne wo ks and d ipping con-
di ions (sec ion 4.1). The s alagmi es we e cu pa allel o hei
g ow h axis, and he cen al segmen o he slab o each speleo hem
was sampled o UeTh da ing, s able iso opes, and majo and ace
elemen s. In MEN-3 and MEN-4, he opposi e slab was used o
ob ain hin sec ions o a pe og aphic s udy (Fig. 3A). MEN-4 is
mac oscopically e y simila o MEN-2 and MEN-5; hus hese wo
las s alagmi es we e no pe og aphically analyzed.
A o al o 1451 samples o
d
18
O and
d
13
C analysis we e ob ained
along he cen al axis a 1 mm inc emen s om s alagmi es MEN-2,
MEN-3, MEN-4 and MEN-5. Iso opic analyses we e pe o med a
he Uni e si y o Innsb uck, using a The moFishe Del a V Plus
linked o a GasBench II, ollowing he me hodology desc ibed in
Sp€
o l (2011). The long- e m ep oducibili y (1 sigma) o he
d
18
O
analyses is 0.08
‰
and 0.06
‰
o
d
13
C(Sp€
o l, 2011). All alues a e
epo ed as pe mil wi h espec o he Vienna Pee Dee Belemni e
s anda d (VPDB).
In addi ion o he iso ope samples, 267 ca bona e samples we e
analyzed o ace elemen s. Samples we e aken a 5 mm in e als
and measu ed o Mg/Ca and S /Ca a ios a ETH Zu ich (Agilen
QQQ 8800) using a s anda diza ion app oach simila o ha e-
po ed in S oll e al. (2022).
3.4. UeTh da ing and age model de elopmen : in eg a ion in o a
composi e eco d
A o al o 41 powde samples we e d illed o u anium-se ies
da ing om dis inc g ow h laye s along he cen al g ow h axis
o each speleo hem using a handheld d ill and a ungs en ca bide
d ill bi . A e chemical sepa a ion, U and Th iso ope measu emen s
we e pe o med using a MC-ICP-MS (The mo-Finnigan Nep une
Plus) a he Uni e si y o Xi'an and he Uni e si y o Minneso a
(USA), ollowing a p e iously desc ibed me hodology (Cheng e al.,
2013). To calcula e co ec ed ages an ini ial
230
Th/
232
Th a omic a io
o 4.4 ±2.2 10
6
was used.
Fig. 3. A) Thin sec ions showing he main ab ics o s alagmi es MEN 4 (columna ab ic [C]) and MEN 3 ( ansi ion be ween columna , columna mic oc ys alline [Cm] and
dend i ic [d] ab ics). B) Da ing esul s (black do s wi h e o ba s) and age models o he ou s alagmi es (g een line) wi h hei co esponding e o limi s ( ed lines) as ob ained
using S alAge. Changes in g ow h a e (blue line) a e also indica ed.
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
5

The indi idual age model o each s alagmi e has been ob ained
using he so wa e S alAge (Scholz and Ho mann, 2011)(Fig. 3B).
La e , he indi idual iso ope p ofiles o he ou s alagmi es we e
in eg a ed in o a single one (MEN composi e eco d) using he
so wa e Iscam (Fohlmeis e , 2012). This p og am looks o he
highes co ela ion be ween wo o mo e da ed p oxy signals
wi hin age unce ain ies using a linea in e pola ion be ween
adjacen UeTh da es. Two composi e eco ds ha e been con-
s uc ed by applying he
d
13
C and hen he
d
18
O da a om he
di e en s alagmi es o he Mendukilo ca e. The iso opic alues
be ween he di e en s alagmi es in he o e lapping in e als
co ela e well wi h each o he in each o he composi es (
d
13
C and
d
18
O), bu finally i has been decided o use he
d
13
C composi e (and
he esul ing age model) h oughou his wo k, since i p esen s a
g ea e a iabili y in i s iso opic alues o e he las 13 ky BP.
Finally, he age model ha p oduced he highes cohe ence
(applying a smoo hing o 100 yea s) was selec ed and subsequen ly
he da a om o he p oxies o his wo k (
d
18
O and ace elemen s)
ha e been assigned a ch onology based on his new
d
13
C composi e
age model.
4. Resul s
4.1. Moni o ing
The a e age annual ain all in he a ea (1921e2021) is 1365 mm
(Supp. Ma e ial Fig. S1). Rain all amoun du ing he moni o ed
yea s shows an impo an in e annual a iabili y, wi h a dec easing
annual amoun om 2019 (1931 mm) o 2020 (1371 mm) (Fig. 2A).
D ip da a (No embe 2018 o May 2021) e eal ha he MEN-D and
MEN-E d ip si es we e con inuously d ipping. They show a lowe
d ip- a e a iabili y be ween di e en seasons compa ed o he
MEN-A and MEN-B d ip si es, which end o almos cease in sum-
me (Fig. 2B). Episodes o highes d ip a es pe day eco ded by he
MEN-A and MEN-B logge s co ela e wi h pe iods o in ense daily
(>50 L/m
2
) and mon hly ain all (win e and all in 2019, Fig. 2A).
D ip si es MEN-A, MEN-B and MEN-D a e ed by low discha ge
seasonal d ips wi h a medium o high a iabili y, wi h a maximum
discha ge a e o 2.49 ml min
1
, 0.15 ml min
1
and 0.17 ml min
1
and a coe ficien o a ia ion o 134, 90 and 59, espec i ely
( ollowing Sma and F iede ich (1987), Supp. Ma e ial Fig. S3). On
he o he hand, MEN-E is ed by seepage flow (low a iabili y, co-
e ficien o a ia ion o 21) showing he lowes maximum discha ge
a e o 0.06 ml min
1
. In he case o d ips wi hou d ip a e moni-
o ing (MEN-F and MEN-G) hey ha e been ca ego ized as con in-
uous d ips (no seasonal) in he di e en moni o ing campaigns
inside he ca e and wi h an appa en simila beha io be ween
MEN-E and MEN-D d ips. The classifica ion o d ips based on
discha ge a e and coe ficien o a ia ion has been shown o be
use ul o disce n i s influence on he iso opic composi ion o he
d ipwa e and o he a med calci e (P
e ez-Mejías e al., 2018).
The ai empe a u e inside Mendukilo ca e a ies be ween 8.4
and 9.2

C du ing he yea , i.e. 3.4

C lowe han he mean annual
su ace ai empe a u e in he moni o ed yea s (11.8e12.6

C,
Alda z me eo ological s a ion) showing a delay o abou 10e14 days
in ela ion o a ia ions in su ace empe a u e (Fig. 2C and D). Only
he empe a u e senso ins alled in he Gue e o oom, a om he
en ance, eco ded a cons an empe a u e (9.1e9.2

C). pCO
2
alues ollow he seasonal empe a u e pa e n inside he ca e,
being highe in summe and lowe in win e .
Fo he s udied pe iod,
d
18
O
d ips
and
d
D
d ips
alues o he ou
d ip si es ange om 8.7 o 7.3
‰
and 53.8 o 42.0
‰
,
espec i ely. The mean
d
18
O
d ips
alues o hese d ip si es a e
be ween 7.6
‰
(MEN-D and MEN-E) and 8.0
‰
(MEN-A and
MEN-B) and, in gene al, win e alues a e lowe han summe
alues (Fig. 2F). Conside ing he a e age alues o all d ip si es
du ing summe (7.6
‰
) and win e (8.1
‰
), he a e age seasonal
a iabili y is 0.5
‰
. Adding o his a iabili y, a sligh end owa ds
less nega i e alues is obse ed be ween 2018 and 2021 associa ed
wi h a dec easing amoun o ain all as desc ibed abo e. The
excep ion is he MEN-E d ip si e which ed he MEN-3 s alagmi e
and eco ds a s able d ip a e ha does no espond o seasons o
ain e en s and shows a low iso opic a iabili y (Fig. 2). The sea-
sonali y eco ded in he d ipwa e s (low a iabili y) alues was also
obse ed in ain all (high a iabili y; Fig. S2), al hough i was much
mo e educed in he fi s one due o homogeniza ion once wa e
en e s in o he epika s .
Ca bona e p ecipi a ed on glass slides all yea ound and he
d
18
O
a med
and
d
13
C
a med
alues ange om 6.8
‰
o 4.6
‰
and
om 12.0
‰
o 8.3
‰
, espec i ely (fi e di e en sampling
poin s, Fig. 2G and H). Excep o d ips MEN-E, MEN-F and MEN-G,
a med calci e esul s show a endency owa ds less nega i e alues
(s a is ically significan o
d
18
O
a med
) du ing he second hal o
2020 and he fi s pa o 2021 (mean
d
18
O
a med
5.3
‰
) compa ed
o he pe iod 2018e2020 (mean
d
18
O
a med
6.3
‰
;Fig. 2G).
Compa ing
d
18
O
d ip
wi h
d
18
O
a med
o each d ip si e shows ha
calci e p ecipi a ed close o iso opic equilib ium. The amoun o
ain all is hus likely influencing
d
18
O
a med
on an in e annual scale
bu he seasonal impac is less p onounced. This seems o be ue
o
d
13
C
a med
as well, al hough he MEN-D d ip si e shows seasonal
a iabili y in he 2020e2021 ime in e al whe e he highes
d
13
C
a med
alues coincide wi h low pCO
2
, hus sugges ing enhanced
degassing and possible p io calci e p ecipi a ion (PCP). An in e -
es ing ea u e is ha
d
13
C
a med
shows less nega i e alues o he
seepage d ip and nea by d ips o he same ca e galle y (MEN-F and
MEN-G) in compa ison o as e d ips (Fig. 2H).
4.2. Pe og aphy, ch onology and s able iso opes o he Mendukilo
s alagmi es
S alagmi es MEN-2 (31.0 cm), MEN-4 (25.6 cm) and MEN-5
(61.0 cm) consis o coa sely c ys alline calci e and a e mac o-
scopically homogeneous wi hou any signal o ec ys alliza ion.
They a e made o columna ab ic, lack g ow h hia i and do no
show mac oscopically isible laminae (Fig. 3A). S alagmi e MEN-3
(28.0 cm) is an excep ion showing a mo e po ous, columna
mic oc ys alline ab ic passing in o a dend i ic ype a he base ( he
fi s 4.0 cm) as well as close o he op (be ween 18.5 and 28.0 cm
om he base - Fig. 3A). Acco ding o F isia (2015), his ansi ion
sugges s a change om a ela i ely slow and cons an discha ge o
mo e a iable d ip a es.
UeTh da ing e ealed ha hese ou s alagmi es g ew con in-
uously o e di e en in e als o he Holocene and Younge D yas
(YD). Taking oge he , hey co e con inuously he en i e Holocene
and he YD (i.e., since 12.8 ky BP) wi h a good o e lap (Fig. 3):
MEN-2 g ew be ween 12.8 and 6.3 ky BP, MEN-3 be ween 6.0 and
0 ky BP, MEN-4 co e s he las 3.0 ky , and MEN-5 spans he las
8.8 ky . The da ing o he base and op pa s o s alagmi e MEN-3
was challenging gi en he change in pe og aphy (see abo e;
Supp. Ma e ial Table S1). Measu ed U concen a ions in Mendukilo
s alagmi es ange be ween 73 and 350 ppb, and he measu ed
230
Th/
232
Th ac i i y a io a ies be ween 20 and 9500. Each o he
MEN s alagmi e age models ha e ai ly low unce ain y o he
co ec ed ages anging be ween 0.005 and 0.250 ky BP ( he
a e age e o o he MEN-2, MEN-4 and MEN-5 age models a e
0.055, 0.041 and 0.040 ky BP espec i ely). In he case o MEN-3
he e o s a e la ge fluc ua ing be ween 0.060 and 0.800 ky BP
(a e age e o o 0.212 ky BP). A e age g ow h a es o MEN-3
(145
m
m/y ) and MEN-4 (224
m
m/y ) a e highe han o MEN-2
(68
m
m/y ) and MEN-5 (77
m
m/y ), indica ing highe speleo hem
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
6
g ow h a es du ing he la e Holocene in his ca e compa ed o he
YD and ea ly Holocene (Fig. 3).
The
d
18
O and
d
13
C alues o he ou s alagmi es ange om 6.3
o 4.3
‰
and om 10.4 o 4.4
‰
, espec i ely, wi h simila
ampli udes in he o e lapping in e als (Suppl. Ma e ial Fig. S4 and
Fig. S5). This eplica ion suppo s he use o hese iso opic p ofiles
as eliable p oxy eco ds. The only excep ion is he uppe pa o
s alagmi e MEN-3 which di e s ma kedly om MEN-4 and MEN-5
in i s s able iso ope alues (Suppl. Ma e ial Fig. S4 and Fig. S5).
Gi en he ac ha his pa o MEN-3 also shows a di e en ype o
calci e ab ic we excluded his pa .
4.3. Composi e eco d om Mendukilo ca e
The MEN composi e gene a ed wi h Iscam o bo h
d
13
C and
d
18
O
iso opic eco ds co e s he las 12.7 ky BP wi h an a e age g ow h
a e o 130
m
m/y . The
d
13
C composi e eco d shows alues
be ween 9.5 and 4.5
‰
(MEN
d
13C mean
¼6.6
‰
; MEN
d
13C
STD
¼0.8
‰
;Fig. 4). In he case o
d
18
O he alues a e cha ac e ized
by a lowe ampli ude and a iabili y, anging om 6.3 o 4.3
‰
(MEN
d
18
O mean
¼5.4
‰
; MEN
d
18
O STD
¼0.3
‰
;Fig. 4). Fo he com-
posi e we included he s alagmi es om he younges o he oldes
one o maximize he co ela ion coe ficien . The
d
13
C co ela ion
be ween MEN-3 and MEN-4 is high ( ¼0.97) bu hey only in e -
sec a a gap o abou 0.5 ky ; when adding MEN-5 he co ela ion
dec eases ( ¼0.37) eflec ing bad in e ela ion in he o e lapping
in e als o s alagmi es MEN-5, MEN-4 and MEN-3 du ing he las 3
ky (Suppl. Ma e ial Fig. S4 and Fig. S5). Thus, paleoclima ic in-
e p e a ions o he las 3 ky should be iewed wi h cau ion.
Finally, when combining his composi e wi h he MEN-2 eco d, he
final co ela ion is e y high ( ¼0.91). The MEN composi e da a
con ains esul s om ma hema ical in e pola ion, he e o e, ab up
a ia ions in he MEN composi e should be ea ed and in e p e ed
wi h cau ion and always aking in o accoun he iso opic (Fig. S4
and Fig. S5) and ace elemen (Fig. S6 and Fig. S7) eco ds o
each s alagmi e sepa a ely.
The composi e eco d p oduced by Iscam shows ha he YD
s a s wi h opposi e ends when compa ing bo h iso opic eco ds
(Fig. 4). The
d
13
C alues inc ease ( om 6.7 o 5.6
‰
) and hose o
d
18
O dec ease ( om 4.6 o 5.2
‰
) a he onse o he YD. The
alues emain s able du ing he YD, and a i s end he
d
13
C alues
g adually dec ease ( om 5.6 o 7.8
‰
) while he
d
18
O alues
inc ease ab up ly ( om 5.2
‰
in he YD o 4.3
‰
a 11.6 ky BP)
eaching he highes alues o he en i e eco d a he YD-Holocene
ansi ion (Fig. 4).
MEN composi e
d
13
C da a show a significan end owa ds mo e
nega i e alues du ing he fi s pa o he G eenlandian,
om 6.0
‰
a 11.7 ky BP o 8.1
‰
a 10 ky BP (Fig. 4). A e -
wa ds, he alues emain s able, only in e up ed by a sho in e al
o highe alues be ween 9.3 and 9.1 ky BP. The highes
d
18
O alues
o he composi e a e p esen in he ea ly G eenlandian and show a
end owa ds lowe alues (as he
d
13
C eco d) bu las ing un il he
end o he G eenlandian ( om 5.0
‰
a 11.68 ky BP o 5.7
‰
a
8.23 ky BP). Du ing he No hg ippian (8.2e4.2 ky BP) he
d
13
C
eco d shows a p og essi e end owa ds highe alues:
om 8.0
‰
a 7.9 ky BP o 4.8
‰
a 4.1 ky BP (Fig. 4). Du ing he
fi s pa o he No hg ippian (8.2e6.5 ky BP) he
d
18
O alues a e
e y nega i e and inc ease in he second pa (mean 5.2
‰
be ween
6.5 and 4.2 ky BP). The lowes
d
18
O alues o he MEN composi e
eco d a e p esen by ex eme alues o 6.3
‰
a 7.0 and 8.1 ky BP.
The
d
18
O a iabili y du ing he Meghalayan (4.2 ky BP o p esen )
is small and lacks a end.
d
13
C alues show he highes alues om
Fig. 4. Mendukilo iso ope composi es o
d
13
C (uppe panel; no e in e ed y-axis) and
d
18
O (lowe panel) o he las 13 ky ob ained using Iscam combining ou s alagmi es. The
Holocene di ision ollows Walke e al. (2019).
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
7
4.2 o 2.5 ky BP, bu by compa ing he
d
13
C alues o he composi e
wi h hose o s alagmi es MEN-3 and MEN-5 i is possible o de ec
ha he e is a dis o ion in he o me du ing he ime in e al o
4.2 and 3.5 ky BP. This o e -exagge a ion o he
d
13
C da a in he
composi e is clea ly an a i ac o he ma hema ical algo i hm, so
ha in e p e a ions a millennial and cen ennial scales du ing ha
ime in e al mus ake in o conside a ion he
d
13
C alues shown by
s alagmi es MEN-3 and MEN-5 sepa a ely om he composi e. The
MEN
d
13
C composi e shows a apid shi owa ds lowe iso ope
alues a 2.5 ky BP (Fig. 4).
Mg/Ca anges be ween 0.40 and 1.60 mmol/mol (Suppl. Ma e ial
Fig. S6) and shows g ea e a ia ions in ends among he ou
s alagmi es. S /Ca measu ed in he ou s alagmi es show simila
empo al pa e ns and a e spanning he same ange o alues
(0.12e0.06 mmol/mol) (Suppl. Ma e ial Fig. S7). S /Ca alues a e
low du ing he YD and inc ease du ing he G eenlandian, eaching
high alues a ound 10 ky BP. This is ollowed by a dec easing end
un il 3.5 ky BP and an inc easing end un il mode n condi ions.
S /Ca is an ico ela ed wi h
d
13
C a he millennial ime scale (co -
ela ion coe ficien 0.37, n ¼1288, p ≪0.01) (Suppl. Ma e ial
Fig. S8). The Ba/Ca a io o he MEN composi e shows s ong sim-
ila i ies wi h S /Ca, bu co ela ion among he ou s alagmi es is
less good, so we only use S /Ca o compa e wi h o he p oxies. In
pa icula , he ends du ing he las 3000 yea s a e no well
eplica ed among he ou speleo hems and hus no conside ed o
disce ning he clima e signal. We also assess he po en ial e ec o
in-ca e p ocesses such as PCP on he
d
13
C eco d looking o he
mos obus p oxy ends associa ed o mo e cons an Mg/Ca a ios
o he MEN-5 s alagmi e (Suppl. Ma e ial Fig. S9).
5. Discussion
5.1. Con ols on MEN geochemical p oxy da a
T ace elemen s in speleo hems a e use ul paleoclima e in-
dica o s in ca es whe e seasonal and long- e m changes in wa e
balance esul in la ge and sys ema ic change in Mg/Ca, S /Ca and
Ba/Ca in d ipwa e associa ed wi h PCP du ing he d y season
(Fai child and T eble, 2009). In Mendukilo s alagmi es, Mg/Ca and
S /Ca show a con as ing pa e n (Suppl. Ma e ial Fig. S6 and
Fig. S7), indica ing ha PCP canno be he dominan con ol on bo h
a ios. One explana ion could be a significan non-bed ock sou ce
o Mg o S which a ied empo ally. In ca es e y close o he coas ,
such as Pindal Ca e wi h galle ies 200 m om he sea cli , ma ine
ae osols can be a significan sou ce o Mg in he Holocene (Mo eno
e al., 2010). Howe e , in he ca e o Mendukilo, 50 km om he
A lan ic coas and a 750 m ele a ion, su ounded by moun ain
anges o highe al i ude, a dominan ma ine ae osol con ibu ion
o Mg is no expec ed, and empo al changes in Mg/Ca a e unlikely
o eflec changes in ma ine ae osol Mg deli e y. An al e na i e
explana ion o he di e ences in Mg and S could be decoupled
empo al changes in he pa i ioning coe ficien o ei he S o Mg.
Many s udies ha e documen ed a ia ions in he S pa i ioning
coe ficien d i en by calci e g ow h a e (Lo ens, 1981;Nielsen
e al., 2013;Tan e al., 2014;Tang e al., 2008;Teso ie o and
Pankow, 1996), and his is a common explana ion o he decou-
pling o Mg/Ca and S /Ca in s alagmi es (S oll e al., 2012;Wa ken
e al., 2018). We obse e ha S /Ca in Mendukilo s alagmi es
posi i ely co ela es wi h g ow h a e (co ela ion
coe ficien ¼0.55, n ¼1288, p ≪0.01) deduced om he UeTh age
models (Suppl. Ma e ial Fig. S10), sugges ing ha g ow h a e
con ibu ed o he S /Ca a ia ions. Consequen ly, we in e ha S /
Ca may be d i en by he g ow h a e e ec supe imposed on PCP,
whe eas Mg/Ca may eflec empo al a ia ions in PCP. In ac , S /
Mg o s alagmi es allows isola ing he co ela ion be ween S /Ca
and g ow h a e (Suppl. Ma e ial Fig. S8). Since s alagmi es ed by
di e en d ip ou es may ha e di e en sensi i i ies o PCP, Mg/Ca
o coe al s alagmi es may no co a y as is he case in his wo k. On
he con a y, he exis ence o common d i e s o d ipwa e sa u-
a ion by soil pCO
2
may lead o mo e ep oducible a ia ions in
g ow h a e and S /Ca.
Speleo hem
d
13
C is egula ed by bo h soil/ ege a ion p ocesses
as well as in-ca e p ocesses. The
d
13
C o DIC in d ipwa e is ini ially
se by he
d
13
C o soil CO
2
, which is mo e nega i e when he soil
pCO
2
is high unde wa m and mois condi ions which a o high
he e o ophic and au o ophic espi a ion. The
d
13
C o DIC in
d ipwa e may be modified sligh ly du ing ca bona e dissolu ion,
bu he low dead ca bon pe cen age in mos ecen s alagmi es,
including hose om No he n Ibe ia, sugges s ha dissolu ion
commonly occu s in open sys ems and he impac on he
d
13
Co
DIC is limi ed (Lechlei ne e al., 2021). In he ca e o ka s ca i ies,
he coupled p ocesses o degassing and PCP can gene a e a signi -
ican posi i e shi in
d
13
C o DIC be o e he wa e eaches he
s alagmi e. The mode n
d
13
C
a med
o he high d ip loca ion (11
o 12
‰
) is in he ange o
d
13
C expec ed o CO
2
in soils wi h a
dominan C3 ege a ion in a wa m and humid clima e (Lechlei ne
e al., 2021), po en ially a ec ed by only limi ed PCP. The mode n
d
13
C
a med
o he seepage loca ion (MEN-E) and o he nea by d ips
o he same galle y (MEN-F and G) a e less nega i e (9 o10
‰
)
and hese d ips may ha e expe ienced some degassing and PCP
and/o a g ea e ex en o p ecipi a ion om he a i ing d ip (in
si u PCP), and/o a mo e closed dissolu ion egime. Seasonal
changes in
d
13
C
a med
a highe and in e media e d ip a e si es may
eflec seasonal a ia ions in soil pCO
2
o in he significance o PCP.
D ip a e moni o ing and a med calci e da a sugges ha he MEN
composi e eco d is no seasonally biased bu a he eflec s
con inuous, yea - ound deposi ion. Con inuous calci e deposi ion is
also suppo ed by he columna ab ic and he absence o isible
laminae as well as he lack o dissolu ion ea u es (i.e., he d ip
wa e was always supe sa u a ed wi h espec o calci e).
Long- e m changes in
d
13
C alues a e a he un ela ed o he
Mg/Ca pa e n in he MEN s alagmi es. Fo example, in he
d
13
C
MEN5 eco d, he main long- e m ends exhibi simila Mg/Ca
a ios (low Mg/Ca a iabili y be ween he di e en
d
13
C la ge
anomalies; Suppl. Ma e ial Fig. S9). The e o e, he empo al ends
in
d
13
C a e mos likely d i en una ec ed by PCP, al hough he
seepage d ips may be cha ac e ized by some PCP. We sugges ha
he long- e m ends in
d
13
C a e d i en by clima ic p ocesses con-
olling soil pCO
2
, i.e. soil empe a u e and soil mois u e. Some
s udies sugges ha he empe a u e componen domina es o e
soil pCO
2
, such du ing MIS 3 and 4 in Villa s ca e in SW F ance
(Gen y e al., 2006) and du ing he las deglacia ion in El Pindal ca e
in NW Spain (Mo eno e al., 2010). Compa ison o he Mendukilo
composi e
d
13
C eco d wi h paleo empe a u e econs uc ions om
Py enean lakes (e.g., Ta a s e al., 2018) and SST eco ds (Ca al
a
e al., 2019;Ma a e al., 2014) suppo s a empe a u e influence
a millennial scale (Fig. 5). The e o e, highe (lowe )
d
13
C alues a e
a guably linked o colde (wa me ) and/o d ye (we e ) in e als
wi h educed (enhanced) soil espi a ion and ege a ion p oduc-
i i y. We no e ha he in e se co ela ion be ween
d
13
C and S /Ca
in he MEN s alagmi es (Suppl. Ma e ial Fig. S10), while opposi e o
he end expec ed om PCP con ol, is consis en wi h he ex-
pec ed ela ionship o low
d
13
C in wa m and humid pe iods o high
soil CO
2
which also lead o highe d ipwa e o e sa u a ion and
highe g ow h a es p omo ing highe S /Ca.
Va ia ion in he d ipwa e and s alagmi e
d
18
O eco d may
eflec changes in
d
18
O o he su ace ocean in he mois u e sou ce
a ea, as well as changes in he a mosphe ic p ocesses which
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
8
Fig. 5. Pal eoen i onmen al eco ds om he Ibe ian Peninsula and Mo occo co e ing he YD and he Holocene. A) S acked eco d o no malized
d
13
C s alagmi e da a om Kai e
ca e; Cue a del Cob e; Cue a Mayo om no he n Ibe ia (Ma ín-Chi ele e al., 2011). B)
d
13
C eco d o he SIR-1 speleo hem om El Soplao ca e in no he n Ibe ia (Rossi e al.,
2018). C) Summe insola ion a 42.D)
d
13
C eco d o he BG speleo hem om wes e nmos Ibe ia (Bu aca Glo iosa Ca e, Po ugal) (Tha che e al., 2020). No e ha ampli ude o
d
13
C axis in his eco d (high a iabili y) is no he same compa ed o he es o Ibe ian s alagmi e eco ds. E) MEN
d
13
C composi e eco d ( his wo k). F) G. bulloides Mg/Ca SST (ALB-
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
9
Fohlmeis e , J., 2012. A s a is ical app oach o cons uc composi e clima e eco ds
o da ed a chi es. Qua . Geoch onol. 14, 48e56. h ps://doi.o g/10.1016/
j.quageo.2012.06.007.
F igola, J., Mo eno, A., Cacho, I., Canals, M., Sie o, F.J., Flo es, J.A., G imal , J.O.,
Hodell, D.A., Cu is, J.H., 2007. Holocene clima e a iabili y in he wes e n
Medi e anean egion om a deepwa e sedimen eco d: holocene clima e
a iabili y. Paleoceanog aphy 22 (2). h ps://doi.o g/10.1029/2006PA001307.
F isia, S., 2015. Mic os a ig aphic logging o calci e ab ics in speleo hems as ool
o palaeoclima e s udies. In . J. Speleol. 17.
Ga cía-Esc
a zaga, A., Gu i
e ez-Zugas i, I., Ma ín-A oyo, A.B., Fe nandes, R., Nú~
nez
de la Fuen e, S., Cuenca-Solana, D., I ia e, E., Sim~
oes, C., Ma ín-Chi ele , J.,
Gonz
alez-Mo ales, M.R., Robe s, P., 2022. Human o age esponse o ab up
clima e change a 8.2 ka on he A lan ic coas o Eu ope. Sci. Rep. 12 (1), 6481.
h ps://doi.o g/10.1038/s41598-022-10135-w.
Ga cía-Ruiz, J.M., Palacios, D., And 
es, N., L
opez-Mo eno, J.I., 2020. Neoglacia ion in
he Spanish Py enees: a mul ip oxy challenge. Medi e . Geosci. Re . 2 (1),
21e36. h ps://doi.o g/10.1007/s42990-020-00022-9.
Gella ly, A.F., G o e, J.M., Swi su , V.R., 1992. Mid-Holocene glacial ac i i y in he
Py enees. Holocene 2 (3), 266e270. h ps://doi.o g/10.1177/
095968369200200309.
Gen y, D., Blama , D., Ghaleb, B., Plagnes, V., Causse, Ch, Bakalowicz, M., Zoua i, K.,
Chki , N., Hells om, J., Waine , K., 2006. Timing and dynamics o he las
deglacia ion om Eu opean and No h A ican
d
13C s alagmi e p ofi-
lesdcompa ison wi h Chinese and Sou h Hemisphe e s alagmi es. Qua . Sci.
Re . 25 (17e18), 2118e2142. h ps://doi.o g/10.1016/j.quasci e .2006.01.030.
Gil-Rome a, G., Gonz
alez-Samp
e iz, P., Lashe as-

Al a ez, L., Se illa-Callejo, M.,
Mo eno, A., Vale o-Ga c
es, B., L
opez-Me ino, L., Ca i
on, J.S., P
e ez Sanz, A.,
A anba i, J., Ga cía-P ie o F once, E., 2014. Biomass-modula ed fi e dynamics
du ing he las glacialein e glacial ansi ion a he cen al Py enees (Spain).
Palaeogeog . Palaeoclima ol. Palaeoecol. 402, 113e124. h ps://doi.o g/10.1016/
j.palaeo.2014.03.015.
Gim
enez, R., Ba olom
e, M., G
azquez, F., Iglesias, M., Mo eno, A., 2021. Unde lying
clima e con ols in iple oxygen (16 O, 17 O, 18 O) and hyd ogen (1 H, 2 H)
iso opes composi ion o ain all (cen al Py enees). F on . Ea h Sci. 9, 633698.
h ps://doi.o g/10.3389/ ea .2021.633698.
Gonz
alez-Samp
e iz, P., A anba i, J., P
e ez-Sanz, A., Gil-Rome a, G., Mo eno, A.,
Leunda, M., Se illa-Callejo, M., Co ella, J.P., Mo ell
on, M., Oli a, B., Vale o-
Ga c
es, B., 2017. En i onmen al and clima e change in he sou he n Cen al
Py enees since he Las Glacial Maximum: a iew om he lake eco ds. Ca ena
149, 668e688. h ps://doi.o g/10.1016/j.ca ena.2016.07.041.
Gonz
alez-Samp
e iz, P., Vale o-Ga c
es, B.L., Mo eno, A., Jalu , G., Ga cía-Ruiz, J.M.,
Ma í-Bono, C., Delgado-Hue as, A., Na as, A., O o, T., Dedouba , J.J., 2006.
Clima e a iabili y in he Spanish Py enees du ing he las 30,000 y e ealed
by he El Po ale sequence. Qua . Res. 66 (1), 38e52. h ps://doi.o g/10.1016/
j.yq es.2006.02.004.
Ilyashuk, E.A., Koinig, K.A., Hei i, O., Ilyashuk, B.P., Psenne , R., 2011. Holocene
empe a u e a ia ions a a high-al i ude si e in he Eas e n Alps: a chi onomid
eco d om Schwa zsee ob S€
olden, Aus ia. Qua . Sci. Re . 30 (1e2), 176e191.
h ps://doi.o g/10.1016/j.quasci e .2010.10.008.
I ia e-Chiapusso, M.J., 2016. Re iewing he La eglacial-Holocene ansi ion in NW
Ibe ia: a palaeoecological app oach based on he compa ison be ween dis-
simila egions. Qua . In . 26.
Jalu , G., Dedouba , J.J., Fon ugne, M., O o, T., 2009. Holocene ci cum-
Medi e anean ege a ion changes: clima e o cing and human impac . Qua .
In . 200 (1e2), 4e18. h ps://doi.o g/10.1016/j.quain .2008.03.012.
Jamb ina-En íquez, M., Rico, M., Mo eno, A., Lei a, M., Be n
a dez, P., P ego, R.,
Recio, C., Vale o-Ga c
es, B.L., 2014. Timing o deglacia ion and pos glacial
en i onmen al dynamics in NW Ibe ia: he Sanab ia Lake eco d. Qua . Sci. Re .
94, 136e158. h ps://doi.o g/10.1016/j.quasci e .2014.04.018.
Jim
enez-Mo eno, G., Rod íguez-Ramí ez, A., P
e ez-Asensio, J.N., Ca i
on, J.S., L
opez-
S
aez, J.A., Villa ías-Robles, J.J., Celes ino-P
e ez, S., Ce illo-Cuenca, E., Le
on,

A.,
Con e as, C., 2015. Impac o la e-Holocene a idifica ion end, clima e a i-
abili y and geodynamic con ol on he en i onmen om a coas al a ea in SW
Spain. Holocene 25 (4), 607e617. h ps://doi.o g/10.1177/0959683614565955.
Kaniewski, D., Ma ine , N., Mo hange, C., Fai e, S., O o, T., Van Campo, E., 2016.
Sola pacing o s o m su ges, coas al flooding and ag icul u al losses in he
Cen al Medi e anean. Sci. Rep. 6 (1), 25197. h ps://doi.o g/10.1038/s ep25197.
Kilha n, H., Couchoud, I., D ysdale, R.N., Rossi, C., Hells om, J., A naud, F., Wong, H.,
2022. The 8.2 ka e en in no he n Spain: iming, s uc u e and clima ic impac
om a mul i-p oxy speleo hem eco d. Clim. Pas 18 (10), 2321e2344. h ps://
doi.o g/10.5194/cp-18-2321-2022.
Klei en, H., Kikki, F., Kissel, C., Laj, C., Ninnemann, U.S., Rich e , T.O., Co ijo, E., 2008.
Reduced No h A lan ic deep wa e coe al wi h he glacial lake Agassiz esh-
wa e ou bu s . Science 319 (5859), 60e64. h ps://doi.o g/10.1126/
science.1148924.
Kuma , R., 2011. Neoglacia ion. Encyclopedia O Snow, Ice And Glacie s. Sp inge , USA.
La ocque-Toble , I., Hei i, O., Weh li, M., 2010. La e Glacial and Holocene empe a-
u e changes a Egelsee, Swi ze land, econs uc ed using sub ossil chi ono-
mids. J. Paleolimnol. 43 (4), 649e666. h ps://doi.o g/10.1007/s10933-009-
9358-z.
Lechlei ne , F.A., Day, C.C., Kos , O., Wilhelm, M., Haghipou , N., Hende son, G.M.,
S oll, H.M., 2021. S alagmi e ca bon iso opes sugges deglacial inc ease in soil
espi a ion in wes e n Eu ope d i en by empe a u e change. Clim. Pas 17 (5),
1903e1918. h ps://doi.o g/10.5194/cp-17-1903-2021.
LeG ande, A.N., Schmid , G.A., 2008. Ensemble, wa e iso ope-enabled, coupled
gene al ci cula ion modeling insigh s in o he 8.2 ka e en : ensemble,
d
18
O
GCM o he 8.2 ka e en . Paleoceanog aphy 23 (3). h ps://doi.o g/10.1029/
2008PA001610 n/a-n/a.
Leunda, M., Gonz
alez-Samp
e iz, P., Gil-Rome a, G., A anba i, J., Mo eno, A., Oli a-
U cia, B., Se illa-Callejo, M., Vale o-Ga c
es, B., 2017. The La e-Glacial and Ho-
locene Ma bo 
e Lake sequence (2612 m a.s.l., Cen al Py enees, Spain): es ing
high al i ude si es sensi i i y o millennial scale ege a ion and clima e a i-
abili y. Global Plane . Change 157, 214e231. h ps://doi.o g/10.1016/
j.gloplacha.2017.08.008.
Leunda, M., Gonz
alez-Samp
e iz, P., Gil-Rome a, G., Ba olom
e, M., Belmon e-
Ribas,

A., G
omez-Ga cía, D., Kal en iede , P., Rubiales, J.M., Schw€
o e , C.,
Tinne , W., Mo ales-Molino, C., Sancho, C., 2019. Ice ca e e eals en i onmen al
o cing o long- e m Py enean ee line dynamics. J. Ecol. 107 (2), 814e828.
h ps://doi.o g/10.1111/1365-2745.13077.
L
opez-Me ino, L., Sil a S
anchez, N., Kaal, J., L
opez-S
aez, J.A., Ma ínez Co izas, A.,
2012. Pos -dis u bance ege a ion dynamics du ing he la e pleis ocene and he
Holocene: an example om NW Ibe ia. Global Plane . Change 92e93, 58e70.
h ps://doi.o g/10.1016/j.gloplacha.2012.04.003.
Lo ens, R.B., 1981. S , Cd, Mn and Co dis ibu ion coe ficien s in calci e as a unc ion
o calci e p ecipi a ion a e. Geochem. Cosmochim. Ac a 45 (4), 553e561.
h ps://doi.o g/10.1016/0016-7037(81)90188-5.
Magny, M., Combou ieu-Nebou , N., de Beaulieu, J.L., Bou -Roumazeilles, V.,
Colomba oli, D., Desp a , S., F ancke, A., Joannin, S., O u, E., Pey on, O.,
Re el, M., Sado i, L., Siani, G., Sic e, M.A., Sama in, S., Simonneau, A., Tinne , W.,
Vanni
e e, B., Wagne , B., e al., 2013. No hesou h palaeohyd ological con as s
in he cen al Medi e anean du ing he Holocene: en a i e syn hesis and
wo king hypo heses. Clim. Pas 9 (5), 2043e2071. h ps://doi.o g/10.5194/cp-9-
2043-2013.
Magny, M., Vanni
e e, B., Calo, C., Mille , L., Le oux, A., Pey on, O., Zanche a, G., La
Man ia, T., Tinne , W., 2011. Holocene hyd ological changes in sou h-wes e n
Medi e anean as eco ded by lake-le el fluc ua ions a Lago P eola, a coas al
lake in sou he n Sicily, I aly. Qua . Sci. Re . 30 (19e20), 2459e2475. h ps://
doi.o g/10.1016/j.quasci e .2011.05.018.
Ma co , S.A., Shakun, J.D., Cla k, P.U., Mix, A.C., 2013. A econs uc ion o egional
and global empe a u e o he pas 11,300 yea s. Science 339 (6124),
1198e1201. h ps://doi.o g/10.1126/science.1228026.
Ma ín-Chi ele , J., Mu~
noz-Ga cía, M.B., Edwa ds, R.L., Tu e o, M.J., O ega, A.I.,
2011. Land su ace empe a u e changes in No he n Ibe ia since 4000y BP,
based on
d
13C o speleo hems. Global Plane . Change 77 (1e2), 1e12. h ps://
doi.o g/10.1016/j.gloplacha.2011.02.002.
Ma a , B., G imal , J.O., Lopez-Ma inez, C., Cacho, I., Sie o, F.J., Flo es, J.A.,
Zahn, R., Canals, M., Cu is, J.H., Hodell, D.A., 2004. Ab up Tempe a u e Changes
in he Wes e n Medi e anean o e he Pas 250,000 Yea s. Science 306 (5702),
1762e1765. h ps://doi.o g/10.1126/science.1101706.
Ma a , B., Jimenez-Ama , P., Zahn, R., G imal , J.O., 2014. Simila i ies and dissim-
ila i ies be ween he las wo deglacia ions and in e glacia ions in he No h
A lan ic egion. Qua . Sci. Re . 99, 122e134. h ps://doi.o g/10.1016/
j.quasci e .2014.06.016.
Masson-Delmo e, V., Jouzel, J., Landais, A., S ie ena d, M., Johnsen, S.J.,
Whi e, J.W.C., We ne , M., S einbjo nsdo i , A., Fuh e , K., 2005. GRIP Deu e-
ium Excess Re eals Rapid and O bi al-Scale Changes in G eenland Mois u e
O igin. Science 309 (5731), 118e121. h ps://doi.o g/10.1126/science.1108575.
Ma e o, I.S.O., G egoi e, L.J., I ano ic, R.F., Tindall, J.C., Haywood, A.M., 2017. The 8.2
ka cooling e en caused by Lau en ide ice saddle collapse. Ea h Plane Sci. Le .
473, 205e214. h ps://doi.o g/10.1016/j.epsl.2017.06.011.
Mau i, A., Da is, B.A.S., Collins, P.M., Kaplan, J.O., 2015. The clima e o Eu ope du ing
he Holocene: a g idded pollen-based econs uc ion and i s mul i-p oxy
e alua ion. Qua . Sci. Re . 112, 109e127. h ps://doi.o g/10.1016/
j.quasci e .2015.01.013.
McManus, J.F., F ancois, R., Ghe a di, J.-M., Keigwin, L.D., B own-Lege , S., 2004.
Collapse and apid esump ion o A lan ic me idional ci cula ion linked o
deglacial clima e changes. Na u e 428 (6985), 834e837. h ps://doi.o g/10.1038/
na u e02494.
Mesa-Fe n
andez, J.M., Jim
enez-Mo eno, G., Rod igo-G
amiz, M., Ga cía-Alix, A.,
Jim
enez-Espejo, F.J., Ma ínez-Ruiz, F., Ande son, R.S., Camue a, J., Ramos-
Rom
an, M.J., 2018. Vege a ion and geochemical esponses o Holocene apid
clima e change in he Sie a Ne ada (sou heas e n Ibe ia): he Laguna Honde a
eco d. Clim. Pas 14 (11), 1687e1706. h ps://doi.o g/10.5194/cp-14-1687-2018.
Mo ell
on, M., A anba i, J., Mo eno, A., Gonz
alez-Samp
e iz, P., Vale o-Ga c
es, B.L.,
2018. Ea ly Holocene humidi y pa e ns in he Ibe ian Peninsula econs uc ed
om lake, pollen and speleo hem eco ds. Qua . Sci. Re . 181, 1e18. h ps://
doi.o g/10.1016/j.quasci e .2017.11.016.
Mo ell
on, M., Vale o-Ga c
es, B., Vegas-Vila úbia, T., Gonz
alez-Samp
e iz, P.,
Rome o,

O., Delgado-Hue as, A., Ma a, P., Mo eno, A., Rico, M., Co ella, J.P.,
2009. La eglacial and Holocene palaeohyd ology in he wes e n Medi e anean
egion: he Lake Es anya eco d (NE Spain). Qua . Sci. Re . 28 (25e26),
2582e2599. h ps://doi.o g/10.1016/j.quasci e .2009.05.014.
Mo eno, A., Iglesias, M., Azo in-Molina, C., P
e ez-Mejías, C., Ba olom
e, M.,
Sancho, C., S oll, H., Cacho, I., F igola, J., Os
aca , C., Mu~
noz, A., Delgado-
Hue as, A., Blad
e, I., Vimeux, F., 2021. Measu emen epo : spa ial a iabili y o
no he n Ibe ian ain all s able iso ope alues ein es iga ing a mosphe ic
con ols on daily and mon hly imescales. A mos. Chem. Phys. 21 (13),
10159e10177. h ps://doi.o g/10.5194/acp-21-10159-2021.
Mo eno, A., Mo ell
on, M., Ma ín-Pue as, C., F igola, J., Canals, M., Cacho, I.,
Co ella, J., P
e ez, A., Belmon e, A., Vegas-Vila úbia, T., Gonz
alez-Samp
e iz, P.,
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
16

Vale o-Ga c
es, B., 2011. Was he e a common hyd ological pa e n in he Ibe ian
Peninsula egion du ing he medie al clima e anomaly? PAGES News 19 (1),
16e17. h ps://doi.o g/10.22498/pages.19.1.16.
Mo eno, A., P
e ez-Mejías, C., Ba olom
e, M., Sancho, C., Cacho, I., S oll, H., Delgado-
Hue as, A., Hells om, J., Edwa ds, R.L., Cheng, H., 2017. New speleo hem da a
om Molinos and Ejul e ca es e eal Holocene hyd ological a iabili y in
no heas Ibe ia. Qua . Res. 88 (2), 223e233. h ps://doi.o g/10.1017/
qua.2017.39.
Mo eno, A., S oll, H., Jim
enez-S
anchez, M., Cacho, I., Vale o-Ga c
es, B., I o, E.,
Edwa ds, R.L., 2010. A speleo hem eco d o glacial (25e11.6ky BP) apid cli-
ma ic changes om no he n Ibe ian Peninsula. Global Plane . Change 71 (3e4),
218e231. h ps://doi.o g/10.1016/j.gloplacha.2009.10.002.
Mo eno, A., S ensson, A., B ooks, S.J., Conno , S., Engels, S., Fle che , W., Gen y, D.,
Hei i, O., Labuhn, I., Pe s¸oiu, A., Pey on, O., Sado i, L., Vale o-Ga c
es, B., Wul , S.,
Zanche a, G., 2014. A compila ion o Wes e n Eu opean e es ial eco ds
60e8 ka BP: owa ds an unde s anding o la i udinal clima ic g adien s. Qua .
Sci. Re . 106, 167e185. h ps://doi.o g/10.1016/j.quasci e .2014.06.030.
Naugh on, F., Cos as, S., Gomes, S.D., Desp a , S., Rod igues, T., Sanchez Go~
ni, M.F.,
Renssen, H., T igo, R., B onk-Ramsey, C., Oli ei a, D., Salguei o, E., Voelke , A.H.L.,
Ab an es, F., 2019. Coupled ocean and a mosphe ic changes du ing G eenland
s adial 1 in sou hwes e n Eu ope. Qua . Sci. Re . 212, 108e120. h ps://doi.o g/
10.1016/j.quasci e .2019.03.033.
Nielsen, L.C., De Yo eo, J.J., DePaolo, D.J., 2013. Gene al model o calci e g ow h
kine ics in he p esence o impu i y ions. Geochem. Cosmochim. Ac a 115,
100e114. h ps://doi.o g/10.1016/j.gca.2013.04.001.
O'B ien, S.R., Mayewski, P.A., Meeke , L.D., Meese, D.A., Twickle , M.S., Whi low, S.I.,
1995. Complexi y o Holocene clima e as econs uc ed om a G eenland ice
co e. Science 270 (5244), 1962e1964. h ps://doi.o g/10.1126/
science.270.5244.1962.
Peck, V.L., Hall, I.R., Zahn, R., Elde field, H., G ousse , F., Hemming, S.R., Scou se, J.D.,
2006. High esolu ion e idence o linkages be ween NW Eu opean ice shee
ins abili y and A lan ic Me idional O e u ning Ci cula ion. Ea h Plane Sci.
Le . 243 (3e4), 476e488. h ps://doi.o g/10.1016/j.epsl.2005.12.023.
P
e ez-Mejías, C., Mo eno, A., Sancho, C., Ba olom
e, M., S oll, H., Os
aca , M.C.,
Cacho, I., Delgado-Hue as, A., 2018. T ans e ence o iso opic signal om ain all
o d ipwa e s and a med calci e in Medi e anean semi-a id ka s . Geochem.
Cosmochim. Ac a 243, 66e98. h ps://doi.o g/10.1016/j.gca.2018.09.014.
P
e ez-Sanz, A., Gonz
alez-Samp
e iz, P., Mo eno, A., Vale o-Ga c
es, B., Gil-Rome a, G.,
Rie ade all, M., Ta a s, P., Lashe as-

Al a ez, L., Mo ell
on, M., Belmon e, A.,
Sancho, C., Se illa-Callejo, M., Na as, A., 2013. Holocene clima e a iabili y,
ege a ion dynamics and fi e egime in he cen al Py enees: he Basa de la
Mo a sequence (NE Spain). Qua . Sci. Re . 73, 149e169. h ps://doi.o g/10.1016/
j.quasci e .2013.05.010.
Plunke , G., Swindles, G.T., 2008. De e mining he Sun's influence on La eglacial
and Holocene clima es: a ocus on clima e esponse o cen ennial-scale sola
o cing a 2800cal.BP. Qua . Sci. Re . 27 (1e2), 175e184. h ps://doi.o g/10.1016/
j.quasci e .2007.01.015.
Railsback, L.B., Liang, F., Vidal Romaní, J.R., G andal-d’Anglade, A., Vaquei o
Rod íguez, M., San os Fidalgo, L., Fe n
andez Mosque a, D., Cheng, H.,
Edwa ds, R.L., 2011. Pe og aphic and iso opic e idence o Holocene long- e m
clima e change and sho e - e m en i onmen al shi s om a s alagmi e om
he Se a do Cou el o no hwes e n Spain, and implica ions o clima ic his o y
ac oss Eu ope and he Medi e anean. Palaeogeog . Palaeoclima ol. Palaeoecol.
305 (1e4), 172e184. h ps://doi.o g/10.1016/j.palaeo.2011.02.030.
Rasmussen, S.O., Vin he , B.M., Clausen, H.B., Ande sen, K.K., 2007. Ea ly Holocene
clima e oscilla ions eco ded in h ee G eenland ice co es. Qua e na y Science
Re iews 26 (15e16), 1907e1914. h ps://doi.o g/10.1016/j.quasci e .2007.06.
015.
Renssen, H., Sepp€
a, H., Hei i, O., Roche, D.M., Goosse, H., Fiche e , T., 2009. The
spa ial and empo al complexi y o he Holocene he mal maximum. Na u e
Geoscience 2 (6), 411e414. h ps://doi.o g/10.1038/ngeo513.
Robe s, C.N., Woodb idge, J., Palmisano, A., Be an, A., Fy e, R., Shennan, S., 2019.
Medi e anean landscape change du ing he Holocene: syn hesis, compa ison
and egional ends in popula ion, land co e and clima e. Holocene 29 (5),
923e937. h ps://doi.o g/10.1177/0959683619826697.
Rod igues, T., G imal , J.O., Ab an es, F., Naugh on, F., Flo es, J.-A., 2010. The las
glacialein e glacial ansi ion (LGIT) in he wes e n mid-la i udes o he No h
A lan ic: ab up sea su ace empe a u e change and sea le el implica ions.
Qua . Sci. Re . 29 (15e16), 1853e1862. h ps://doi.o g/10.1016/
j.quasci e .2010.04.004.
Rossi, C., Bajo, P., Lozano, R.P., Hells om, J., 2018. Younge D yas o Ea ly Holocene
paleoclima e in Can ab ia (N Spain): cons ain s om speleo hem Mg, annual
fluo escence banding and s able iso ope eco ds. Qua . Sci. Re . 192, 71e
85.
h ps://doi.o g/10.1016/j.quasci e .2018.05.025.
Ruan, J., Khe bouche, F., Gen y, D., Blama , D., Cheng, H., Dewilde, F., Hachi, S.,
Edwa ds, R.L., R
egnie , E., Michelo , J.-L., 2016. E idence o a p olonged d ough
ca. 4200 y BP co ela ed wi h p ehis o ic se lemen abandonmen om he
Gueldaman GLD1 Ca e, No he n Alge ia. Clim. Pas 12 (1), 1e14. h ps://
doi.o g/10.5194/cp-12-1-2016.
Sancho, C., Belmon e,

A., Ba olom
e, M., Mo eno, A., Leunda, M., L
opez-Ma ínez, J.,
2018. Middle- o-la e Holocene palaeoen i onmen al econs uc ion om he
A294 ice-ca e eco d (Cen al Py enees, no he n Spain). Ea h Plane Sci. Le .
484, 135e144. h ps://doi.o g/10.1016/j.epsl.2017.12.027.
Scholz, D., Ho mann, D.L., 2011. S alAge ean algo i hm designed o cons uc ion o
speleo hem age models. Qua . Geoch onol. 6 (3e4), 369e382. h ps://doi.o g/
10.1016/j.quageo.2011.02.002.
Skinne , L.C., Shackle on, N.J., 2006. Decons uc ing Te mina ions I and II: e isi ing
he glacioeus a ic pa adigm based on deep-wa e empe a u e es ima es. Qua .
Sci. Re . 25 (23e24), 3312e3321. h ps://doi.o g/10.1016/
j.quasci e .2006.07.005.
Sma , P.L., F iede ich, H., 1987. Wa e Mo emen and S o age in he Unsa u a ed
Zone o a Ma u ely Ka s ified Ca bona e Aqui e . P oceedings o he con e ence
on En i onmena l P oblems in Ka s Te anes and hei Solu ions, Dublin, Ohio,
pp. 59e87.
Smi h, A.C., Wynn, P.M., Ba ke , P.A., Leng, M.J., Noble, S.R., Tych, W., 2016. No h
A lan ic o cing o mois u e deli e y o Eu ope h oughou he Holocene. Sci.
Rep. 6 (1). h ps://doi.o g/10.1038/s ep24745.
Spe anza, A., an Geel, B., an de Plich , J., 2003. E idence o sola o cing o
clima e change a ca. 850 cal BC om a Czech pea sequence. Global Plane .
Change 35 (1e2), 51e65. h ps://doi.o g/10.1016/S0921-8181(02)00091-7.
Sp€
o l, C., 2011. Long- e m pe o mance o he Gasbench iso ope a io mass spec-
ome y sys em o he s able iso ope analysis o ca bona e mic osamples.
Rapid Commun. Mass Spec om. 25 (11), 1683e1685. h ps://doi.o g/10.1002/
cm.5037.
S oll, H.M., Cacho, I., Gasson, E., Sliwinski, J., Kos , O., Mo eno, A., Iglesias, M.,
To ne , J., Pe ez-Mejias, C., Haghipou , N., Cheng, H., Edwa ds, R.L., 2022. Rapid
no he n hemisphe e ice shee mel ing du ing he penul ima e deglacia ion.
Na . Commun. 13 (1), 3819. h ps://doi.o g/10.1038/s41467-022-31619-3.
S oll, H.M., Mo eno, A., Mendez-Vicen e, A., Gonzalez-Lemos, S., Jimenez-
Sanchez, M., Dominguez-Cues a, M.J., Edwa ds, R.L., Cheng, H., Wang, X., 2013.
Paleoclima e and g ow h a es o speleo hems in he no hwes e n Ibe ian
Peninsula o e he las wo glacial cycles. Qua . Res. 80 (2), 284e290. h ps://
doi.o g/10.1016/j.yq es.2013.05.002.
S oll, H.M., Mülle , W., P ie o, M., 2012. I-STAL, a model o in e p e a ion o Mg/Ca,
S /Ca and Ba/Ca a ia ions in speleo hems and i s o wa d and in e se appli-
ca ion on seasonal o millennial scales: I-s al speleo hem ace elemen model.
G-cubed 13 (9). h ps://doi.o g/10.1029/2012GC004183.
Swindles, G.T., Plunke , G., Roe, H.M., 2007. A delayed clima ic esponse o sola
o cing a 2800 cal. BP: mul ip oxy e idence om h ee I ish pea lands. Holo-
cene 17 (2), 177e182. h ps://doi.o g/10.1177/0959683607075830.
Tan, L., Shen, C.-C., Cai, Y., Lo, L., Cheng, H., An, Z., 2014. T ace-elemen a ia ions in
an annually laye ed s alagmi e as eco de s o clima ic changes and an h o-
pogenic pollu ion in Cen al China. Qua . Res. 81 (2), 181e188. h ps://doi.o g/
10.1016/j.yq es.2013.12.001.
Tang, J., K€
ohle , S.J., Die zel, M., 2008. S 2þ/Ca2þand 44Ca/40Ca ac iona ion
du ing ino ganic calci e o ma ion: I. S inco po a ion. Geochem. Cosmochim.
Ac a 72 (15), 3718e3732. h ps://doi.o g/10.1016/j.gca.2008.05.031.
Ta a s, P., Hei i, O., Vale o-Ga c
es, B., Ca~
nedo-A güelles, M., P a , N., Rie ade all, M.,
Gonz
alez-Samp
e iz, P., 2018. Chi onomid-in e ed Holocene empe a u e
econs uc ion in Basa de la Mo a Lake (Cen al Py enees). Holocene 28 (11),
1685e1696. h ps://doi.o g/10.1177/0959683618788662.
Teso ie o, A.J., Pankow, J.F., 1996. Solid solu ion pa i ioning o S 2þ, Ba2þ, and
Cd2þ o calci e. Geochem. Cosmochim. Ac a 60 (6), 1053e1063. h ps://doi.o g/
10.1016/0016-7037(95)00449-1.
Tha che , D.L., Wanamake , A.D., Dennis on, R.F., Asme om, Y., Polyak, V.J.,
Fullick, D., Ummenho e , C.C., Gillikin, D.P., Haws, J.A., 2020. Hyd oclima e
a iabili y om wes e n Ibe ia (Po ugal) du ing he Holocene: insigh s om a
composi e s alagmi e iso ope eco d. Holocene 30 (7), 966e981. h ps://doi.o g/
10.1177/0959683620908648.
T emaine, D.M., F oelich, P.N., Wang, Y., 2011. Speleo hem calci e a med in si u:
Mode n calib a ion o
d
18
O and
d
13
C paleoclima e p oxies in a con inuously-
moni o ed na u al ca e sys em. Geochimica e Cosmochimica Ac a 75 (17),
4929e4950. h ps://doi.o g/10.1016/j.gca.2011.06.005.
Usoskin, I.G., Solanki, S.K., Ko al so , G.A., 2007. G and minima and maxima o sola
ac i i y: new obse a ional cons ain s. As on. As ophys. 471 (1), 301e309.
h ps://doi.o g/10.1051/0004-6361:20077704.
Van Geel, B., Buu man, J., Wa e bolk, H.T., 1996. A chaeological and palaeoecological
indica ions o an ab up clima e change in The Ne he lands, and e idence o
clima ological eleconnec ions a ound 2650 BP. J. Qua . Sci. 11 (6), 451e460.
h ps://doi.o g/10.1002/(SICI)1099-1417(199611/12)11:6<451::AID-
JQS275>3.0.CO;2-9.
Vegas, J., Ruiz-Zapa a, B., O iz, J.E., Gal
an, L., To es, T., Ga cía-Co 
es,

A., Gil-
Ga cía, M.J., P
e ez-Gonz
alez, A., Galla do-Mill
an, J.L., 2010. Iden ifica ion o a id
phases du ing he las 50 cal. ka BP om he Fuen illejo maa -lacus ine eco d
(Campo de Cala a a Volcanic Field, Spain): a id phases du ing he las 50 ka
om he uen illejo maa -lake eco d, Spain. J. Qua . Sci. 25 (7), 1051e1062.
h ps://doi.o g/10.1002/jqs.1262.
Walczak, I.W., Baldini, J.U.L., Baldini, L.M., McDe mo , F., Ma sden, S., S andish, C.D.,
Richa ds, D.A., And eo, B., Sla e , J., 2015. Recons uc ing high- esolu ion
clima e using CT scanning o unsec ioned s alagmi es: a case s udy iden i-
ying he mid-Holocene onse o he Medi e anean clima e in sou he n Ibe ia.
Qua . Sci. Re . 127, 117e128. h ps://doi.o g/10.1016/j.quasci e .2015.06.013.
Walke , M., Head, M.J., Lowe, J., Be kelhamme , M., Bj
€
O ck, S., Cheng, H.,
Cwyna , L.C., Fishe , D., Gkinis, V., Long, A., Newnham, R., Rasmussen, S.O.,
Weiss, H., 2019. Subdi iding he Holocene Se ies/Epoch: Fo maliza ion o
s ages/ages and subse ies/subepochs, and designa ion o GSSPs and auxilia y
s a o ypes. Jou nal o Qua e na y Science 34 (3), 173e186. h ps://doi.o g/10.
1002/jqs.3097.
Wanne , H., Bee , J., Bü iko e , J., C owley, T.J., Cubasch, U., Flückige , J., Goosse, H.,
G osjean, M., Joos, F., Kaplan, J.O., Kü el, M., Mülle , S.A., P en ice, I.C.,
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
17
Solomina, O., S ocke , T.F., Ta aso , P., Wagne , M., Widmann, M., 2008. Mid- o
La e Holocene clima e change: an o e iew. Qua . Sci. Re . 27 (19e20),
1791e1828. h ps://doi.o g/10.1016/j.quasci e .2008.06.013.
Wa ken, S.F., Fohlmeis e , J., Sch €
ode -Ri z au, A., Cons an in, S., Sp€
o l, C., Ge des, A.,
Espe , J., F ank, N., A ps, J., Te en e, M., Riechelmann, D.F.C., Mangini, A.,
Scholz, D., 2018. Recons uc ion o la e Holocene au umn/win e p ecipi a ion
a iabili y in SW Romania om a high- esolu ion speleo hem ace elemen
eco d. Ea h Plane Sci. Le . 499, 122e133. h ps://doi.o g/10.1016/
j.epsl.2018.07.027.
Wi h, S.B., Glu , L., Gilli, A., Anselme i, F.S., 2013. Holocene flood equency ac oss
he Cen al Alps esola o cing and e idence o a ia ions in No h A lan ic
a mosphe ic ci cula ion. Qua . Sci. Re . 80, 112e128. h ps://doi.o g/10.1016/
j.quasci e .2013.09.002.
Zielho e , C., Fle che , W.J., Mischke, S., De Ba is , M., Campbell, J.F.E., Joannin, S.,
Tjallingii, R., El Hamou i, N., Junginge , A., S ele, A., Bussmann, J., Schneide , B.,
Laue , T., Spi ze , K., S uple , M., B ache , T., Mikdad, A., 2017. A lan ic o cing
o Wes e n Medi e anean win e ain minima du ing he las 12,000 yea s.
Qua . Sci. Re . 157, 29e51. h ps://doi.o g/10.1016/j.quasci e .2016.11.037.
Zielho e , C., K€
ohle , A., Mischke, S., Benkaddou , A., Mikdad, A., Fle che , W.J., 2019.
Wes e n Medi e anean hyd o-clima ic consequences o Holocene ice- a ed
deb is (Bond) e en s. Clim. Pas 15 (2), 463e475. h ps://doi.o g/10.5194/cp-
15-463-2019.
J.L. Be nal-Wo mull, A. Mo eno, M. Ba olom
e e al. Qua e na y Science Re iews 305 (2023) 108006
18