Glob Change Biol. 2020;26:4521–4537.
|
4521wileyonlinelib a y.com/jou nal/gcb
Recei ed: 18 Oc obe 2019
|
Re ised: 19 Ma ch 2020
|
Accep ed: 30 Ap il 2020
DOI: 10.1111/gcb.15153
PRIMARY RESEARCH ARTICLE
G ow h and esilience esponses o Sco s pine o ex eme
d ough s ac oss Eu ope depend on p ed ough g ow h
condi ions
A un K. Bose1,2 | A hu Gessle 1,3,4 | And eas Bol e5 | Alessand a Bo e o1,4 |
Allan Bu as6 | Maxime Caille e 7 | J. Julio Cama e o8 | Ma hias Haeni1 |
Ana-Ma ia He eş9,10 | And ea He ia11 | Ma hieu Lé esque3 | Juan C. Lina es12 |
Jo di Ma inez-Vilal a13,14 | Luis Ma ías15 | Anne e Menzel16,17 |
Raúl Sánchez-Salgue o12 | Ma hias Sau e 1 | Michel Venne ie 7 | Daniel Ziche5,18 |
And eas Rigling1,3,4
1WSL Swiss Fede al Ins i u e o Fo es , Snow and Landscape Resea ch, Bi mensdo , Swi ze land
2Fo es y and Wood Technology Discipline, Khulna Uni e si y, Khulna, Bangladesh
3Ins i u e o Te es ial Ecosys ems, ETH Zu ich, Zu ich, Swi ze land
4SwissFo es Lab, Bi mensdo , Swi ze land
5Thünen Ins i u e o Fo es Ecosys ems, Ebe swalde, Ge many
6Land Su ace-A mosphe e In e ac ions, Technische Uni e si a München, F eising, Ge many
7UMR RECOVER/Ecosys èmes Médi e anéens e Risques, INRAE, Aix-en-P o ence cedex 5, F ance
8Ins i u o Pi enaico de Ecologıa (IPE-CSIC), Za agoza, Spain
9Depa men o Fo es Sciences, T ansil ania Uni e si y o B aşo , B aşo , Romania
10BC3 - Basque Cen e o Clima e Change, Scien i ic Campus o he Uni e si y o he Basque Coun y, Leioa, Spain
11Depa amen o de Ciencias Ag o o es ales, Uni e sidad de Huel a, Palos de la F on e a, Spain
12Dep o. Sis emas Físicos, Químicos y Na u ales, Uni e sidad Pablo de Ola ide, Se illa, Spain
13CREAF, Bella e a (Ce danyola del Vallès), Spain
14Uni e si a Au ònoma de Ba celona, Bella e a (Ce danyola del Vallès), Spain
15Depa amen o de Biología Vege al y Ecología, Facul ad de Biología, Uni e sidad de Se illa, Se illa, Spain
16Ecoclima ology, Technische Uni e si ä München, F eising, Ge many
17Ins i u e o Ad anced S udy, Technische Uni e si ä München, Ga ching, Ge many
18Facul y o Fo es and En i onmen , Ebe swalde Uni e si y o Sus ainable De elopmen , Ebe swalde, Ge many
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i s License, which pe mi s use and dis ibu ion in
any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modi ica ions o adap a ions a e made.
© 2020 The Au ho s. Global Change Biology published by John Wiley & Sons L d
Co espondence
A un K. Bose, WSL Swiss Fede al Ins i u e
o Fo es , Snow and Landscape Resea ch,
Zü che s asse 111, CH-8903 Bi mensdo ,
Swi ze land.
Email: a un.b[email p o ec ed]
Funding in o ma ion
Ma ie Skłodowska-Cu ie, G an /Awa d
Numbe : 749051-REFOREST; FEDER, G an /
Awa d Numbe : IJCI-2015-25845; Minis y
o Science, Inno a ion and Uni e si ies,
Abs ac
Global clima e change is expec ed o u he aise he equency and se e i y o ex-
eme e en s, such as d ough s. The e ec s o ex eme d ough s on ees a e di -
icul o disen angle gi en he inhe en complexi y o d ough e en s ( equency,
se e i y, du a ion, and iming du ing he g owing season). Besides, d ough e ec s
migh be modula ed by ees’ pheno ypic a iabili y, which is, in u n, a ec ed by
long- e m local selec i e p essu es and managemen legacies. He e we in es iga ed
4522
|
BOSE E al.
1 | INTRODUCTION
Clima e change e ec s a e b oadly cha ac e ized by ele a ed em-
pe a u e, changed p ecipi a ion egimes, and inc eased in e annual
a iabili y, o en esul ing in mo e equen and in ense clima e ex-
emes such as se e e d ough s (Dai, 2012; Spinoni, Vog , Naumann,
Ba bosa, & Dosio, 2018). The inc eased equency and se e i y o
d ough s can signi ican ly impac ee g ow h by educing hei pho-
osyn he ic ac i i y (Flexas & Med ano, 2002; Reddy, Chai anya,
& Vi ekanandan, 2004) and al e ing hei cambial ac i i y (G ube ,
S obl, Vei , & Obe hube , 2010). In addi ion, se e e d ough e en s
ha e been associa ed o o es decline ei he h ough di ec abio ic
e ec s leading o hyd aulic ailu e and/o ca bon s a a ion (Adams
e al., 2017; Choa e al., 2018; McDowell e al., 2008) o media ed by
bio ic ac o s, such as insec s (Rouaul e al., 2006), ungi (Gio dano,
Gon hie , Va ese, Mise e e, & Nicolo i, 2009), and mis le oes
(Rigling, Eilmann, Koechli, & Dobbe in, 2010). These e ec s may ul-
ima ely induce shi s in o es composi ion (Bu as & Menzel, 2019;
Wal he e al., 2002) and educ ion in o es p oduc i i y (Ciais
e al., 2005).
G owing ecogni ion o he impac s o ex eme d ough s on o es
ecosys ems has spu ed on a numbe o long- e m expe imen s and
obse a ional s udies (e.g., B eshea s e al., 2005; Jen sch e al., 2011;
Seidel, Ma iu, & Menzel, 2019). The esul s o hese s udies e ealed
a la ge a iabili y in pa e n and magni ude o esponses o ex eme
d ough s (McDowell e al., 2008; Smi h, 2011), because pheno ypic
acclima ion o such ex eme e en s may depend on a mul i ude o
ac o s and hei in e ac ions, including d ough cha ac e is ics
(Ande egg e al., 2015; Gazol e al., 2018), d ough his o y o he
g owing en i onmen (Vicen e-Se ano e al., 2013), species-speci ic
unc ional ai s and li e-his o y s a egies (Ande egg e al., 2016;
G eenwood e al., 2017; Lé esque e al., 2013), p o enance (Sánchez-
Salgue o e al., 2018; Seidel, Schunk, Ma iu, & Menzel, 2016), ee size
and age (G anda, Gazol, & Cama e o, 2018; Magnani, Mencuccini, &
G ace, 2000; Se a-Maluque , Mencuccini, & Ma ínez-Vilal a, 2018),
ee- o- ee compe i ion (Lina es, Cama e o, & Ca ei a, 2010), nu-
ien imbalances (He ia e al., 2019), nu ien a ailabili y (Gessle ,
Schaub, & McDowell, 2017), species composi ion and s ocking o he
o es s and (Bo e o e al., 2017; Fo es e e al., 2016; G ossio d
e al., 2014), ees’ neighbou hood composi ion (G ossio d, 2019),
mic oclima ic condi ions ela ed o o es edge and in e io (Bu as
e al., 2018), and g ow h ends p io o d ough (Zang, Ha l-Meie ,
Di ma , Ro he, & Menzel, 2014). In he longe e m, acclima ion is
o en complemen ed by e olu iona y geno ypic adap a ion (Bose
e al., 2020; Ham ick, 2004; Sánchez-Salgue o e al., 2018) leading
o di e en ia ion o popula ions and eco ypes wi h a ying adap i e
capaci ies o d ough , o en obse ed o ma ginal popula ions a d y
species ange ma gins (Bol e e al., 2016; Hampe & Pe i , 2005).
Mo eo e , he e ec s o pas d ough and g owing condi-
ions (legacy e ec s), can emain o se e al yea s and modi y he
G an /Awa d Numbe : RTI2018-096884-
B-C31 and RTI2018-096884-B-C33;
VULBOS, G an /Awa d Numbe : UPO-
1263216; PinCaR, G an /Awa d Numbe :
UHU-1266324; Ba a ian Clima e Resea ch
Ne wo k; Ge man Waldklima ond, G an /
Awa d Numbe : FKZ 28W-C-4-077-
01; Ba a ian S a e Minis y o Food,
Ag icul u e, and Fo es y, G an /Awa d
Numbe : ST327
he magni ude and he empo al changes o ee-le el esilience (i.e., esis ance, e-
co e y, and esilience) o ex eme d ough s. Mo eo e , we assessed he ee-, si e-,
and d ough - ela ed ac o s and hei in e ac ions d i ing he ee-le el esilience
o ex eme d ough s. We used a ee- ing ne wo k o he widely dis ibu ed Sco s
pine (Pinus syl es is) along a 2,800 km la i udinal g adien om sou he n Spain o
no he n Ge many. We ound ha he esilience o ex eme d ough dec eased in
mid-ele a ion and low p oduc i i y si es om 1980–1999 o 2000–2011 likely due
o mo e equen and se e e d ough s in he la e pe iod. Ou s udy showed ha
he impac o d ough on ee-le el esilience was no dependen on i s la i udinal
loca ion, bu a he on he ype o si es ees we e g owing a and on hei g ow h
pe o mances (i.e., magni ude and a iabili y o g ow h) du ing he p ed ough pe iod.
We ound signi ican in e ac i e e ec s be ween d ough du a ion and ee g ow h
p io o d ough , sugges ing ha Sco s pine ees wi h highe magni ude and a i-
abili y o g ow h in he long e m a e mo e ulne able o long and se e e d ough s.
Mo eo e , ou esul s indica e ha Sco s pine ees ha expe ienced mo e equen
d ough s o e he long- e m we e less esis an o ex eme d ough s. We, he e o e,
conclude ha he physiological esilience o ex eme d ough s migh be cons ained
by hei g ow h p io o d ough , and ha mo e equen and longe d ough pe iods
may o e s ain hei po en ial o acclima ion.
KEYWORDS
acclima ion, la i udinal g adien , Pinus syl es is, p edisposi ion, ee ings
|
4523
BOSE E al.
ee g ow h and physiological esponses o he cu en d ough
(Ande egg e al., 2015; Kannenbe g e al., 2019; Seidel e al., 2019).
An impo an ques ion in he deba e on d ough and acclima ion is
whe he indi iduals will be able o acclima e as enough o cope
wi h inc eased equency and se e i y o d ough s (Dai, 2012;
Szejne , Belmeche i, Ehle inge , & Monson, 2020). I is he e o e
impo an o unde s and how ee g ow h esponses o ex eme
d ough s a y ac oss si es wi h di e en p oduc i i y (Vallada es,
Gianoli, & Gómez, 2007; Vallada es e al., 2014), since si e p oduc i -
i y can modi y ees’ pheno ypic s a egies such as ee heigh , oo
o shoo a io, and c own de elopmen o e icien conse a ion
and u iliza ion o wa e (Vanninen & Mäkelä, 2005). Fo example,
ee heigh which is commonly used as an indica o o si e p oduc-
i i y (e.g., Wes oby, Fals e , Moles, Vesk, & W igh , 2002) was
epo ed o be he s onges p edic o o ee mo ali y in sou h-
wes e n Uni ed S a es whe e 1.8 million ees we e s udied (S o all,
Shuga , & Yang, 2019).
Se e al ecen s udies conduc ed in sou he n and cen al
Eu ope ha e epo ed d ough -induced dieback o Sco s pine (Bu as
e al., 2018; Cama e o, Gazol, Sangüesa-Ba eda, Oli a, & Vicen e-
Se ano, 2015; E zold e al., 2019; Galiano, Ma ínez-Vilal a, &
Llo e , 2010; He eş, Ma ínez-Vilal a, & Cla amun López, 2012;
Sánchez-Salgue o, Na a o-Ce illo, Cama e o, & Fe nández-
Cancio, 2012) causing a shi owa d he dominance o oak (Que cus
spp.) species (Ca nice e al., 2014; Galiano e al., 2010; Rigling
e al., 2013). Al hough he impac o a ious ee- and si e-le el ac-
o s on ee g ow h du ing d ough has been s udied om local o
global scales (e.g., Ande egg e al., 2015; Bu as e al., 2018; Gazol
e al., 2018; Zang e al., 2014), hei in e ac i e e ec s a e s ill no
clea ly unde s ood (Maes e al., 2019). Fo example, some la ge-scale
s udies ound a low o mode a e in luence o d ough se e i y on
ee g ow h esponse (e.g., Gazol, Cama e o, Ande egg, & Vicen e-
Se ano, 2017; Sánchez-Salgue o e al., 2018), possibly because hey
did no conside in e ac i e e ec s be ween d ough cha ac e is ics
and long- e m ee g ow h pe o mances. In addi ion, la ge-scale
s udies o en cha ac e ize d ough acco ding o a p ede ined me eo-
ological season (e.g., d ough in sp ing–summe ) i espec i e o local
si e condi ions, soil mois u e con en , and geog aphic loca ion (e.g.,
Bo e o e al., 2017; Gao e al., 2018; Gazol e al., 2018). As a conse-
quence, si e-speci ic clima e-g ow h signals migh be o e looked i
a pa icula s udied season is no he mos ele an pe iod o ee
adial g ow h (Pasho, Cama e o, de Luis, & Vicen e-Se ano, 2011;
Sánchez-Salgue o e al., 2015).
He e we combined Sco s pine ee- ing wid h da a om 30
si es in o a ne wo k o de e mine how g ow h esponses o ex-
eme d ough a ied along a la i udinal g adien ac oss Eu ope
s e ching om sou he n Spain o no he n Ge many. T ee g ow h
esponse was assessed o e ospec i ely quan i y sho - and
long- e m d ough e ec s on g ow h o nume ous indi iduals,
si es, and species a annual esolu ion. T ee g ow h esilience was
de ined as he capaci y o a ee o each g ow h a es simila o
hose p io o a gi en d ough e en . Thus, esilience encompasses
he capaci y o bu e he impac o a dis u bance ( esis ance),
as well as he abili y o e u n o p edis u bance g ow h le els
( eco e y; Llo e , Keeling, & Sala, 2011). Speci ically, we asked
ou esea ch ques ions: (a) How does he impac o he clima ic
wa e balance (CWB; i.e., p ecipi a ion minus po en ial e apo-
anspi a ion) o di e en seasons on ee g ow h a y along a
la i udinal g adien ? (b) How do adial g ow h a es o Sco s pine
du ing d ough and nond ough yea s a y ac oss si es? (c) Has
ee g ow h esilience o ex eme d ough changed o e he pas
decades due o an inc eased equency and se e i y o d ough s
(Se a-Maluque e al., 2018; Szejne e al., 2020)? (d) How do
d ough cha ac e is ics, si e condi ions, and ee g ow h- ela ed
a iables modula e he ee g ow h esilience o ex eme d ough
e en s? Fo his las esea ch ques ion, we conside ed a lis o
biological hypo heses based on a li e a u e e iew (see Table S1:
e.g., Gazol e al., 2017, 2018; Sánchez-Salgue o e al., 2018; Vi ali,
Bün gen, & Bauhus, 2017; Zang e al., 2014).
2 | MATERIALS AND METHODS
2.1 | S udy si es and ee- ing da a
We compiled ee- ing wid h da a o Sco s pine om 30 si es
(Table S2) along an app oxima ely 2,800 km long la i udinal g adi-
en om sou he n Spain (Baza; 37.2°N, 4.0°W) o no h-eas e n
Ge many (To gelow; 53.6°N, 14°E; Figu e 1). To a oid age- ela ed
g ow h e ec s only ees olde han 30 yea s a he ime o exam-
ined d ough we e selec ed, esul ing in 615 adul Sco s pine ees
(6–60 ees pe si e). F om each ee, wo o ou ee ing wid h
se ies we e included, measu ed om inc emen co es ex ac ed a
b eas heigh (1.3 m heigh ) and c oss-da ed ollowing s anda d den-
d och onological p ocedu es (G issino-Maye , 2001).
Conside ing he la ge di e ences in p oduc i i y among s udy
si es along his long g adien , he si es we e g ouped using a hie -
a chical clus e analysis (Kau man & Rousseeuw, 1990). The clas-
si ica ion was based on si e p oduc i i y index (i.e., dominan ee
heigh a 50 yea s o s and age) and si e ele a ion. Dominan ee
heigh has been commonly used as an indica o o si e p oduc i -
i y (e.g., Bugmann, 1996; Diéguez-A anda, Bu kha , & Rod íguez-
Soallei o, 2005; Wes oby e al., 2002) including Sco s pine si es
(e.g., Diéguez-A anda, Ál a ez González, Ma cos Ba io, & Albe o
Rojo, 2005; Hökkä & Ojansuu, 2004; Mäkinen, Yue, & Kohnle, 2017;
Palahı, Tomé, Pukkala, T asoba es, & Mon e o, 2004). The hie a chi-
cal clus e ing was done using he hclus unc ion and wa d.D me hod
in R (R De elopmen Co e Team, 2018). Based on he hie a chical
clus e analysis (Figu e S1), h ee g oups (i.e., si e ypes) we e cha -
ac e ized as (a) high-ele a ion si es (1,865–2,011 m a.s.l.) wi h low
p oduc i i y (6.0–14.0 m in s and dominan ee heigh ) e e ed as
“HELP”, (b) mid-ele a ion si es (600–1,450 m a.s.l.) wi h low p oduc-
i i y (7.5–11.0 m in s and dominan ee heigh ) e e ed as “MELP,”
and (c) low-ele a ion si es (33–326 m a.s.l.) wi h high p oduc i -
i y (15–23.7 m in s and dominan ee heigh ) e e ed as “LEHP”
(Figu e S1).
4524
|
BOSE E al.
2.2 | Analy ical app oaches
Add essing ou ou esea ch ques ions, he analy ical app oach
in ol ed wo s eps: da a p epa a ion and da a analysis. The da a
p epa a ion s ep embodied ou subs eps, (a) quan i ica ion o
ee- ing wid h indices; (b) quan i ica ion o d ough indices; (c)
iden i ica ion o d ough , p ed ough , and pos d ough pe iods
(i.e., yea s); and (d) quan i ica ion o ee g ow h esilience indices.
The da a analysis s ep embodied ou s eps, ha is, one o each
esea ch ques ion.
2.3 | Quan i ica ion o ee- ing wid h indices
We aimed a quan i ying g ow h esponses o ex eme d ough
e en s o e he ecen 50 yea s pe iod oughly om yea 1960 o
yea 2011. Howe e , ou s udied ees la gely di e ed in age ac oss
si es (Table S2). Hence, ing wid h da a we e ans o med in o di-
mensionless ing wid h indices (RWI) wi h bo h age- ela ed g ow h
ends and lowe equency a ia ion emo ed om he ime se ies
(Cook & Kai iuks is, 1990). Fo his, ing wid h da a we e de ended
by i ing a nega i e exponen ial cu e o using a 30 yea cubic spline
wi h a 50% equency cu o (Cook & Kai iuks is, 1990). In addi ion
o hese de ending me hods, we also con e ed he aw ing wid h
da a in o basal a ea inc emen (cm2 pe yea ; Biondi & Quedan, 2008)
using he dplR package in R (Bunn e al., 2018). We assessed he
sui abili y o hese app oaches o disen angle he d ough e ec s
on ee g ow h by compu ing he co ela ion coe icien wi h he
d ough indices (c . nex sec ion) and by cha ac e izing he end
o e a 50 yea pe iod (Table S3; Figu e S2). The esul s showed
ha he nega i e exponen ial de ending me hod pe o med bes
among he used app oaches in e ms o he magni ude o co ela ion
wi h he d ough and o cap u ing he long- e m ends (Table S3;
Figu e S2). We hus used he de ended nega i e exponen ial RWI
(he ea e e e ed o as RWI) o he analysis.
To build he si e-le el ee- ing ch onology, we a e aged he de-
ended indi idual RWI se ies wi h a Tukey's biweigh obus mean
(Cook & Kai iuks is, 1990; F i s, 2001). The RWI and a e age ee-
le el ch onology we e calcula ed using he de end and ch on unc-
ions, espec i ely, a ailable om he dplR R package (Bunn e al.,
2018; R De elopmen Co e Team, 2018).
2.4 | Quan i ica ion o d ough indices
Mon hly mean empe a u e (°C) and o al p ecipi a ion (mm) da a
we e ob ained o each si e om di e en clima e da a sou ces
(Table S4). To compu e he co ela ion coe icien be ween d ough
indices and he RWI, we conside ed a 50 yea pe iod o all si es.
Howe e , he ange o yea s o he 50 yea pe iod a ied ac oss
si es due o di e ences in iming o da a collec ion.
Fo d ough index, we ini ially conside ed he De Ma onne
Index (De Ma onne, 1926), he S anda dized P ecipi a ion Index
(McKee, Doesken, & Kleis , 1993), and he S anda dized P ecipi a ion
E apo anspi a ion Index (SPEI; Vicen e-Se ano, Begue ía, & López-
Mo eno, 2010). The SPEI had a s onge co ela ion wi h RWI han
he o he indices examined o mos o he si es (see Table S5). Hence,
SPEI was used o de ining he d ough and nond ough yea s.
The SPEI is a uni less d ough index, which akes in o accoun
bo h p ecipi a ion and po en ial e apo anspi a ion e ec s in he
calcula ion o he CWB, and is commonly used in he li e a u e o
iden i ying and cha ac e izing d ough and nond ough yea s (e.g.,
Bo e o e al., 2017; Gazol e al., 2018). The po en ial e apo ans-
pi a ion was calcula ed using he Tho n hwai e unc ion o he R
package SPEI (Begue ia & Vicen e-Se ano, 2013). The SPEI was
hen calcula ed om CWB using he spei unc ion o he R package
SPEI (Begue ia & Vicen e-Se ano, 2013). Fo each si e, we calcu-
la ed SPEI o a ious imescales ha is, in eg a ed o e 1–15 mon hs
in o de o ep esen di e en leng hs o he g owing season o a
leas di e en g ow h sensi i e pe iods wi hin he cu en and he
FIGURE 1 Loca ion o he 30 Sco s
pine s udy si es dis ibu ed along a
la i udinal g adien ha anged om
sou he n Spain o no he n Ge many.
HELP, high-ele a ion si es wi h low
p oduc i i y; LEHP, low-ele a ion si es
wi h high p oduc i i y; MELP, mid-
ele a ion si es wi h low p oduc i i y.
The g ey shade used as a backg ound
wi hin he map ep esen s he na u al
dis ibu ion o Sco s pine adap ed om
Má yás, Ackzell, & Samuel (2004)
|
4525
BOSE E al.
p e ious g owing season. We assessed he Pea son co ela ion be-
ween RWI and SPEIs (i.e. he di e en ime in e als) o iden i ying
he mos ele an SPEI (i.e., mos sensi i e ime in e al) o each
si e o de ine he d ough and nond ough yea s (see Table S6). The
esul ing SPEIs (i.e., hose bes co ela ed wi h RWI) a e p esen ed
in he Table S7.
Fo iden i ying he ex eme d ough yea o a si e, we selec ed
he yea wi h he lowes SPEI alue. Fo each si e, we i s selec ed
he ex eme d ough yea s o he pe iod o 1980–2011. We hen
selec ed he ex eme d ough yea o he pe iod o 1980–1999 and
o he pe iod o 2000–2011.
2.5 | Iden i ica ion o d ough , p ed ough , and
pos d ough pe iods
We cha ac e ized d ough pe iods by single o mul iple yea s based
on SPEI ≤ −1.00 and p ed ough o pos d ough pe iods (i.e., wi hou
d ough ) based on SPEI ≥ −0.99. We limi ed he p ed ough and pos -
d ough pe iods o a maximum o 3 yea s, bu o d ough pe iods we
conside ed all consecu i e yea s wi h SPEI ≤ −1.00 (see Table S6).
We iden i ied he mos ex eme d ough s du ing 1980–1999, and
du ing 2000–2011 o all s udy si es (see Table S6) o compa ing
he ee g ow h esponses o ex eme d ough s du ing he ecen
decade (2000–2011) wi h he p e ious wo decades (1980–1999).
Since many si es had no d ough du ing 1990–1999, we decided o
enla ge he ea lie pe iod back un il 1980.
2.6 | T ee g ow h esilience indices
Fo ee g ow h esilience, we compu ed h ee esilience indices
as sugges ed by Llo e e al. (2011): esis ance, eco e y, and esil-
ience. The esis ance quan i ies he a io be ween g ow h du ing a
d ough pe iod and g ow h du ing he p eceding nond ough pe iod,
ep esen ing hus he capaci y o he ees o bu e he s ess and
main ain g ow h du ing d ough . The eco e y quan i ies he g ow h
eac ion ollowing he d ough pe iod and is de ined by he a io be-
ween g ow h du ing he pos d ough pe iod and g ow h du ing he
d ough pe iod. The esilience quan i ies he a io be ween g ow h
du ing he pos d ough pe iod and g ow h du ing he p ed ough
pe iod, which ep esen s he capaci y o ees o eco e and e-
gain he g ow h o he p ed ough pe iod. We quan i ied esis ance,
eco e y, and esilience o all ees o all si es du ing he mos ex-
eme d ough s in 1980–1999 and in 2000–2011 (see Table S6).
2.7 | Resea ch ques ion 1: Impac o seasonal
d ough (SPEI) on ee g ow h
Based on he esul s o p elimina y analysis (i.e., co ela ion be ween
RWI and di e en SPEIs), we iden i ied he eigh bes co ela ed
SPEIs o unde s anding he magni ude (i.e., deg ee o co ela ion)
and pa e n (i.e., ype o co ela ion) o in luences o d ough on
RWI, and how ha magni ude and pa e n o co ela ion a ied
ac oss he la i udinal g adien examined in his s udy. The selec ed
SPEI imescales we e Augus 15 (i.e., om p e ious June o cu en
Augus ), May 12 (i.e., om p e ious June o cu en May), May 9 (i.e.,
om p e ious Sep embe o cu en May), May 6 (i.e., om p e ious
Decembe o cu en May), May 3 (i.e., sp ing, om cu en Ma ch o
cu en May), Augus 6 (i.e., om cu en Ma ch o cu en Augus ),
Augus 3 (i.e., summe , om cu en June o cu en Augus ), and
No embe 6 (i.e., om cu en June o cu en No embe ).
2.8 | Resea ch ques ion 2: T ee g ow h a e in
d ough and nond ough yea s
Fo unde s anding he absolu e ee adial g ow h pe o mances
du ing d ough and nond ough yea s, we modeled absolu e ee
adial g ow h (non-de ended ee ing wid h) as a unc ion o si e
ypes ( h ee le els: LEHP, MELP, and HELP), d ough s a us ( wo
le els: d ough yea s and nond ough yea s), and he in e ac ion
be ween si e ypes and d ough s a us. Fo unde s anding he po-
en ial ole o ee age on absolu e ee adial g ow h, we conside ed
ee age as a co a ia e in his analysis.
2.9 | Resea ch ques ion 3: Tempo al change in ee
g ow h esilience o ex eme d ough s
We modeled esis ance, eco e y, and esilience as a unc ion o
ime pe iod ( wo le els: 1980–1999 and 2000–2011), si e ypes
( h ee le els: LEHP, MELP, and HELP), and he in e ac ion be ween
ime pe iod and si e ypes.
2.10 | Resea ch ques ion 4: Fac o s a ec ing ee
g ow h esilience o ex eme d ough
Fo his esea ch ques ion, we selec ed he mos ex eme d ough
du ing he en i e 1980–2011 s udy pe iod and used he co espond-
ing esis ance, eco e y, and esilience indices as esponse a iables
in a mixed-e ec s model (c . nex sec ion). We conside ed se e al
ee-, si e-, and d ough -le el explana o y a iables and a ious wo-
way in e ac ion e ms (see Table S1). The a iables included ee size
(i.e., ee diame e a b eas heigh [DBH] inside ba k a he d ough
yea ), ee g ow h, and ee g ow h a iabili y p io o d ough ep e-
sen ing he a e age and s anda d de ia ion o RWI, espec i ely o 10
consecu i e yea s p io o he ex eme d ough excluding he yea s
conside ed as p ed ough pe iod o quan i ying he h ee esilience
indices, si e ypes, ele a ion, la i ude, d ough se e i y (measu ed
by he a e age SPEI du ing he d ough pe iod), d ough du a ion
(measu ed by he leng h o he d ough pe iod in yea s), and d ough
equency (measu ed by he numbe o d ough yea s (SPEI ≤ −1.00)
wi hin 10 yea s p eceding he maximum d ough pe iod).
4526
|
BOSE E al.
FIGURE 2 Seasonal co ela ions be ween Sco s pine ee- ing wid h indices and he S anda dized P ecipi a ion E apo anspi a ion Index
(SPEI) o he pe iod o app oxima ely 1960–2011 ac oss he la i udinal g adien . Only he seasons ha exhibi ed he s onges e ec on
ee- ing wid h indices a e plo ed (see Sec ion 2). No e. ‘p e ious’ e e s o he yea p e ious o ee ing o ma ion, while ‘cu en ’ e e s
o he cu en yea o ing o ma ion, summe : June, July, and Augus , sp ing: Ma ch, Ap il, and May, au umn: Sep embe , Oc obe , and
No embe , and win e : Decembe , Janua y, and Feb ua y. HELP, high-ele a ion si es wi h low p oduc i i y; LEHP, low-ele a ion si es wi h
high p oduc i i y; MELP, mid-ele a ion si es wi h low p oduc i i y. Pea son's p oduc -momen co ela ion wi h a h eshold <0.05 was used
o s a is ical signi icance. Co ela ion magni ude: he la ge he ci cles, he s onge he co ela ions. See Table S7 o co ela ion sco es
ha a e displayed in his igu e
|
4527
BOSE E al.
2.11 | S a is ical analyses
We used a linea mixed-e ec modeling app oach o esea ch ques-
ion 2, 3, and 4 in which ou a iables o in e es we e conside ed
as ixed e ec s and ees nes ed wi hin si es we e conside ed as
andom e ec s. The modeling was pe o med using he unc ion
lme o he R package nlme (Pinhei o & Ba es, 2000; Pinhei o, Ba es,
DebRoy, & Sa ka , 2014). The esponse a iables we e log- ans-
o med o no malize esiduals and homogenize a iances and we
checked he assump ions o no mali y o he esiduals and homo-
genei y o he a iances. P elimina y analysis indica ed ha an ad-
di ional e o s uc u e o accoun o plo spa ial au oco ela ion did
no imp o e model pe o mance and hus was no inco po a ed in o
he inal model. We also assessed po en ial mul icollinea i y among
explana o y a iables using he Va iance In la ion Fac o (VIF) and
disca ded a iables when VIF > 2.0. The VIF was calcula ed using he
unc ion i o he R package ca (Fox & Weisbe g, 2011). The pos
hoc Tukey mul iple compa ison es was pe o med o de ec he
s a is ical di e ences (Ho ho n, B e z, & Wes all, 2008).
Fo esea ch ques ion 4, we used he in o ma ion- heo e ic ap-
p oach (Bu nham & Ande son, 2002; Johnson & Omland, 2004),
which p o ides a measu e o s eng h o each candida e model ha
ep esen s a plausible hypo hesis ela i e o he en i e se o candida e
models conside ed (Maze olle, 2006). In he con ex o ou esea ch
ques ion (i.e., wha a e he ac o s d i ing he ee g ow h esilience
o ex eme d ough ?), we conside ed 16 hypo heses (i.e., candida e
models; Table S1), which we e de eloped based on he cu en unde -
s anding esul ed om di e en s udies ha examined ee g ow h
esilience o ex eme d ough s. Model selec ion was pe o med using
he AICcmoda g package o R (Maze olle, 2011). Candida e mod-
els we e compa ed using Akaike's in o ma ion c i e ion co ec ed
o small sample sizes (AICc). Akaike weigh s we e compu ed o as-
sess he suppo o each model. We used mul imodel in e ence o
compu e he model-a e aged es ima es o he explana o y a iables
and hei 95% con idence in e als (Bu nham & Ande son, 2002). A
con idence in e al excluding 0 indica ed ha he co esponding ex-
plana o y a iable had an e ec on he esponse a iable (Bu nham &
Ande son, 2002; Maze olle, 2006). In addi ion o ou candida e mod-
els we also conside ed a null model and a ull model. The coe icien o
a ia ion (R2) o ixed and andom e ec s we e calcula ed using he
unc ion .squa edGLMM o he MuMIn package in R (Ba oń, 2013).
The backg ound map o Figu e 2 was downloaded using he unc ion
map_da a om ggmap package in R (Kahle & Wickham, 2013).
3 | RESULTS
3.1 | Impac o seasonal d ough (SPEI) on ee g ow h
Ou esul s showed signi ican di e ences in he esponse o ee
g ow h o he di e en ime pe iods o SPEI. The cu en yea summe
o au umn (June–No embe ) SPEI signi ican ly con olled ee g ow h a
LEHP si es o no he n Ge many (Figu e 2), while ee g ow h a MELP
si es was d i en by SPEI o sp ing (Ma ch–May), summe (June–Augus ),
and sp ing and summe combined. T ee g ow h in HELP si es was ei he
non ela ed o nega i ely co ela ed wi h di e en ime pe iods o he
SPEI (Figu e 2; Table S3). O e all, he magni ude o co ela ion be ween
RWI and di e en SPEIs was highe o MELP han he wo o he si e
ypes (Table S3). Th ee si es o HELP si e ype had a nega i e co ela-
ion wi h SPEIs while one si e o HELP si e ype was no signi ican ly
co ela ed wi h any SPEI conside ed in ou analysis (Table S3).
3.2 | T ee g ow h a e in d ough and nond ough
yea s
In d ough and nond ough yea s, ee adial g ow h was highe a LEHP
han a HELP and a MELP si es (Figu e 3). The MELP si es had signi i-
can ly lowe ee adial g ow h in d ough yea s han in nond ough
yea s (Table S8; Figu e 3). Con a y o MELP, ee g ow h was no signi -
ican ly di e en be ween d ough and nond ough yea s a LEHP and
a HELP si es (Table S8; Figu e 3). T ee age was nega i ely associa ed
wi h he adial g ow h (p < .0001) i espec i e o si e ypes (Table S8).
3.3 | Tempo al change in ee g ow h esilience o
ex eme d ough s
T ee g ow h esis ance o ex eme d ough o all si e ypes (i.e.,
HELP, LEHP, and MELP) did no change o e he wo pe iods (i.e.,
FIGURE 3 Mean annual adial g ow h in d ough and non-
d ough yea s o he pe iod o app oxima ely 1980–2011 ac oss
he h ee si e ypes (i.e., LEHP, low-ele a ion si es wi h high
p oduc i i y; MELP, mid-ele a ion si es wi h low p oduc i i y;
HELP, high-ele a ion si es wi h low p oduc i i y). E o ba s
ep esen he mean ± s anda d e o (n = 615). Le e s on op o
he ba s show he esul s (a < b < c) o he pos hoc Tukey mul iple
compa ison es wi h a h eshold <0.05 o s a is ical signi icance
indica ing he di e ences among he h ee si e ypes and be ween
non-d ough yea s and d ough yea s wi hin each si e ype
4528
|
BOSE E al.
1980–1999 and 2000–2011; Figu e 4a). Ne e heless, esis ance
was highe a HELP han a LEHP, and highe in he la e compa ed
o MELP, i espec i e o he pe iod (Table S9; Figu e 4a).
T ee g ow h eco e y changed signi ican ly o e he wo pe i-
ods o all si e ypes, whe e eco e y dec eased om 1980–1999
o 2000–2011 a MELP and HELP si es, while inc eased om
1980–1999 o 2000–2011 a LEHP si es (Table S9; Figu e 4b).
In 1980–1999, eco e y was signi ican ly highe a MELP com-
pa ed o he wo o he si e ypes i espec i e o pe iod (Table S9;
Figu e 4b).
T ee g ow h esilience changed signi ican ly o e he wo pe-
iods o LEHP and MELP si es, bu no o HELP si es. Resilience
dec eased om 1980–1999 o 2000–2011 a MELP si es, while i
inc eased om 1980–1999 o 2000–2011 a LEHP si es (Table S9;
Figu e 4c).
3.4 | Fac o s a ec ing ee g ow h esilience o
ex eme d ough
The model ha included addi i e and in e ac ion e ec s o all a i-
ables conside ed in he analysis had ull suppo o Akaike weigh o
esis ance (Table 1). A lowe esis ance was associa ed wi h highe
p ed ough g ow h a e (Table 2). In addi ion, a lowe esis ance was
associa ed wi h g ea e d ough equency, and wi h longe d ough
bu depending upon p ed ough g ow h a e (Table 2). Resis ance was
highe a HELP and LEHP si es han a MELP si es (Table 2; Figu e 5a).
The model ha included d ough se e i y and si e ypes, and he
in e ac ion be ween he wo a iables had he highes suppo o
Akaike weigh o eco e y (0.74; Table 1). Reco e y was lowe a
HELP and LEHP si es han a MELP si es (Table 2; Figu e 5b). In ad-
di ion, he eco e y was highe whe e ees expe ienced a highe
equency o d ough s (Table 2).
The model ha included p ed ough g ow h a e and d ough du-
a ion, and he in e ac ion be ween he wo a iables had he high-
es suppo o Akaike weigh o esilience (Table 1). Resilience was
nega i ely associa ed wi h p ed ough g ow h a e and p ed ough
g ow h a iabili y and he e was no di e ence ac oss he h ee si e
ypes (Table 2; Figu e 5c).
4 | DISCUSSION
Using ee ing wid h da a om 30 si es along a 2,800 km la i udinal
g adien ac oss Eu ope we analyzed whe he ee g ow h esilience o
ex eme d ough depended on he geog aphical loca ion o he ee
(Isaac-Ren on e al., 2018) and i esilience o ex eme d ough de-
c eased o e ime due o mo e equen d ough e en s in ecen yea s
(Se a-Maluque e al., 2018; Spinoni, Naumann, Ca ao, Ba bosa, &
Vog , 2014). We examined hese ques ions on Sco s pine, one o he
mos widely dis ibu ed ee species in he wo ld which is also con-
side ed ulne able o ex eme d ough condi ions (Cama e o, Gazol,
Sangüesa-Ba eda, e al., 2015; Galiano e al., 2010; Ma ías, Lina es,
Sánchez-Mi anda, & Jump, 2017; Rigling e al., 2013). Ou s udy
shows ha ee-le el esilience o d ough was no dependen on he
la i udinal loca ion, bu a he on he ype o si e hey we e g owing
a and hei g ow h pe o mance (i.e., magni ude and a iabili y o
g ow h) du ing he p ed ough pe iod. Ou esul s indica e ha ees
wi h highe magni ude and a iabili y in g ow h a e mo e ulne able
o long and se e e d ough s. In addi ion, we ound ha ee g ow h
FIGURE 4 T ee-le el esis ance (a), eco e y (b), and esilience
(c) o he mos ex eme d ough du ing 1980–1999 and du ing
2000–2011 o h ee si e ypes. E o ba s ep esen he
mean ± s anda d e o (n = 615). Le e s on op o he ba s show
he esul s (a < b < c) o he pos hoc Tukey mul iple compa ison
es wi h a h eshold <0.05 o s a is ical signi icance indica ing
he di e ences among he h ee si e ypes and be ween he wo
pe iods wi hin each si e ype. HELP, high-ele a ion si es wi h low
p oduc i i y; LEHP, low-ele a ion si es wi h high p oduc i i y;
MELP, mid-ele a ion si es wi h low p oduc i i y
|
4529
BOSE E al.
TABLE 1 Resul s o he bes models explaining ee g ow h esis ance, eco e y, and esilience o Sco s pine ees along he s udied g adien . F om he 16 es ed models, only he h ee
wi h he highes Akaike's in o ma ion c i e ion (AICc) weigh a e p esen ed
Models Hypo heses Re e ences AICc ∆AICc AICc weigh R2 ( ixed)
R2 ( ixed and
andom)
Resis ance (RT) T ee esis ance o d ough is a ec ed
by
RT~all a iables Full model 117.9 0.0 1.00 .33 .49
RT~D_INT*PGR+D_INT*ST The in ensi y o he d ough , bu
depending upon he g ow h p io o
d ough and si e ypes
Adap ed om Gazol e al. (2018) 130.4 12.5 0.00 .22 .46
RT~D_FRE*PGR The equency o he d ough , bu
depending upon he g ow h p io o
d ough
Adap ed om Gao e al. (2018) 130.5 12.6 0.00 .14 .47
Reco e y (RC) T ee eco e y a e d ough
RC~D_INT*ST Is a ec ed by he in ensi y o he
d ough , bu depending upon he si e
ypes
Adap ed om Gazol e al. (2018) 301.9 0.0 0.74 .16 .36
RC~ST Dec eased wi h si e ypes Sánchez-Salgue o e al. (2018) 307.0 5.1 0.06 .07 .37
RC~D_INT+D_FRE+D_DUR Is a ec ed combinedly by in ensi y
o d ough , du a ion o d ough , and
equency o d ough
Gao e al. (2018) 308.4 6.5 0.03 .08 .37
Resilience (RS) T ee esilience o d ough is
RS~D_DUR*PGR A ec ed by he du a ion o he
d ough , bu depending upon he
g ow h p io o d ough
Adap ed om Taege e al. (2013) 342.4 0.0 0.45 .03 .35
RS~PGR Nega i ely associa ed wi h he g ow h
p io o d ough
Zang e al. (2014); Ruij en and
Be endse (2010)
344.3 1.9 0.17 .02 .35
RS~D_FRE*PGR A ec ed by he equency o he
d ough , bu depending upon he
g ow h p io o d ough
Adap ed om Gao e al. (2018) 344.6 2.2 0.15 .02 .36
No e: PGR = a e age ee g ow h ( ing wid h indices) p io o d ough , ST = si e ype (LEHP (low-ele a ion si es wi h high p oduc i i y), MELP (mid-ele a ion si es wi h low p oduc i i y), and HELP
(high-ele a ion si es wi h low p oduc i i y)), D_FRE = d ough equency measu ed by he numbe o d ough yea s wi hin he pas 10 yea s om he s udied d ough , D_INT = in ensi y o d ough , and
D_DUR = du a ion o d ough , *indica es an in e ac ion e m and +indica es an addi i e e m, PGR was quan i ied om ee g ow h du ing he 10 consecu i e yea s p io o d ough excluding he yea s
conside ed as p ed ough pe iod quan i ying he h ee indices (i.e., esis ance, eco e y, and esilience).
4536
|
BOSE E al.
Lé esque, M., Sau e , M., Siegwol , R., Eilmann, B., B ang, P., Bugmann,
H., & Rigling, A. (2013). D ough esponse o i e coni e species
unde con as ing wa e a ailabili y sugges s high ulne abili y o
No way sp uce and Eu opean la ch. Global Change Biology, 19(10),
3184–3199. h ps://doi.o g/10.1111/gcb.12268
Lina es, J. C., Cama e o, J. J., & Ca ei a, J. A. (2010). Compe i ion
modula es he adap a ion capaci y o o es s o clima ic s ess:
Insigh s om ecen g ow h decline and dea h in elic s ands o he
Medi e anean i Abies pinsapo. Jou nal o Ecology, 98(3), 592–603.
Llo e , F., Keeling, E. G., & Sala, A. (2011). Componen s o ee e-
silience: E ec s o successi e low-g ow h episodes in old pon-
de osa pine o es s. Oikos, 120(12), 1909–1920. h ps://doi.o g/
10.1111/j.1600-0706.2011.19372.x
Maes, S. L., Pe ing, M. P., Vanhellemon , M., Depauw, L., Van den Bulcke,
J., B ūmelis, G., … Ve heyen, K. (2019). En i onmen al d i e s in e ac-
i ely a ec indi idual ee g ow h ac oss empe a e Eu opean o -
es s. Global Change Biology, 25(1), 201–217. h ps://doi.o g/10.1111/
gcb.14493
Magnani, F., Mencuccini, M., & G ace, J. (2000). Age- ela ed decline in
s and p oduc i i y: The ole o s uc u al acclima ion unde hyd au-
lic cons ain s. Plan , Cell & En i onmen , 23(3), 251–263. h ps://doi.
o g/10.1046/j.1365-3040.2000.00537.x
Mäkinen, H., Yue, C., & Kohnle, U. (2017). Si e index changes o Sco s
pine, No way sp uce and la ch s ands in sou he n and cen al Finland.
Ag icul u al and Fo es Me eo ology, 237–238, 95–104. h ps://doi.
o g/10.1016/j.ag o me .2017.01.017
Ma ínez-Vilal a, J., Cocha d, H., Mencuccini, M., S e ck, F., He e o, A.,
Ko honen, J. F. J., … Zwei el, R. (2009). Hyd aulic adjus men o Sco s
pine ac oss Eu ope. New Phy ologis , 184(2), 353–364. h ps://doi.
o g/10.1111/j.1469-8137.2009.02954.x
Ma ínez-Vilal a, J., López, B. C., Loep e, L., & Llo e , F. (2012). S and-
and ee-le el de e minan s o he d ough esponse o Sco s pine
adial g ow h. Oecologia, 168(3), 877–888. h ps://doi.o g/10.1007/
s0044 2-011-2132-8
Ma ías, L., Lina es, J. C., Sánchez-Mi anda, Á., & Jump, A. S. (2017).
Con as ing g ow h o ecas s ac oss he geog aphical ange o
Sco s pine due o al i udinal and la i udinal di e ences in clima ic
sensi i i y. Global Change Biology, 23(10), 4106–4116. h ps://doi.
o g/10.1111/gcb.13627
Má yás, C., Ackzell, L., & Samuel, C. J. A. (2004). EUFORGEN echni-
cal guidelines o gene ic conse a ion and use o Sco s pine (Pinus
syl es is). In e na ional Plan Gene ic Resou ces Ins i u e, Rome, I aly, 6
Maze olle, M. J. (2006). Imp o ing da a analysis in he pe ology: Using
Akaike’s In o ma ion C i e ion (AIC) o assess he s eng h o bio-
logical hypo heses. Amphibia-Rep ilia, 27, 169–180. h ps://doi.
o g/10.1163/15685 38067 77239922
Maze olle, M. J. (2011). AICcmoda g: Model selec ion and mul imodel
in e ence based on (Q)AIC(c). R package e sion 1.17. Re ie ed om
h p://c an. -p oje c .o g/web/packa ges/AICcm oda g /index.h ml
McDowell, N., Pockman, W. T., Allen, C. D., B eshea s, D. D., Cobb, N.,
Kolb, T., … Yepez, E. A. (2008). Mechanisms o plan su i al and
mo ali y du ing d ough : Why do some plan s su i e while o he s
succumb o d ough ? New Phy ologis , 178(4), 719–739. h ps://doi.
o g/10.1111/j.1469-8137.2008.02436.x
McKee, T. B., Doesken, N. J., & Kleis , J. (1993). The ela ionship o d ough
equency and du a ion o ime scales (pp. 179–184). P ep in s, Eigh
Con e ence on Applied Clima ology. Anaheim, CA: Ame ican Me eo
Socie y.
Ogle, K., Whi ham, T. G., & Cobb, N. S. (2000). T ee- ing a ia ion in pinyon
p edic s likelihood o dea h ollowing se e e d ough . Ecology, 81(11),
3237–3243. h ps://doi.o g/10.1890/0012-9658(2000)081[3237:
TRVIP P]2.0.CO;2
Olson, M. E., So iano, D., Rosell, J. A., An odillo, T., Donoghue, M. J.,
Edwa ds, E. J., … Méndez-Alonzo, R. (2018). Plan heigh and hy-
d aulic ulne abili y o d ough and cold. P oceedings o he Na ional
Academy o Sciences o he Uni ed S a es o Ame ica, 115(29), 7551.
h ps://doi.o g/10.1073/pnas.17217 28115
Palahı, M., Tomé, M., Pukkala, T., T asoba es, A., & Mon e o, G. (2004).
Si e index model o Pinus syl es is in no h-eas Spain. Fo es
Ecology and Managemen , 187(1), 35–47. h ps://doi.o g/10.1016/
S0378 -1127(03)00312 -8
Pasho, E., Cama e o, J. J., de Luis, M., & Vicen e-Se ano, S. M. (2011).
Impac s o d ough a di e en ime scales on o es g ow h ac oss a
wide clima ic g adien in no h-eas e n Spain. Ag icul u al and Fo es
Me eo ology, 151(12), 1800–1811. h ps://doi.o g/10.1016/j.ag o
me .2011.07.018
Peñuelas, J., Hun , J. M., Ogaya, R., & Jump, A. S. (2008). Twen ie h cen-
u y changes o ee- ing δ13C a he sou he n ange-edge o Fagus
syl a ica: Inc easing wa e -use e iciency does no a oid he g ow h
decline induced by wa ming a low al i udes. Global Change Biology,
14(5), 1076–1088.
Pinhei o, J. C., & Ba es, D. M. (2000). Mixed e ec s models in S and S-PLUS.
New Yo k, NY: Sp inge Ve lag.
Pinhei o, J., Ba es, D., DebRoy, S., & Sa ka , D. (2014). nlme: Linea and
nonlinea mixed e ec s models. R package e sion 3.1-117. Re ie ed
om h ps://c an. p oj ec .o g/web/packa ges/nlme/index.h ml
P e zsch, H., Schü ze, G., & Uhl, E. (2013). Resis ance o Eu opean ee
species o d ough s ess in mixed e sus pu e o es s: E idence
o s ess elease by in e -speci ic acili a ion. Plan Biology, 15(3),
483–495.
R De elopmen Co e Team. (2018). R: A language and en i onmen o
s a is ical compu ing. Vienna, Aus ia: R Founda ion o S a is ical
Compu ing. Re ie ed om www. -p oje c .o g
Reddy, A. R., Chai anya, K. V., & Vi ekanandan, M. (2004). D ough -
induced esponses o pho osyn hesis and an ioxidan me abolism in
highe plan s. Jou nal o Plan Physiology, 161(11), 1189–1202. h ps://
doi.o g/10.1016/j.jplph.2004.01.013
Rigling, A., Bigle , C., Eilmann, B., Feldmeye -Ch is e, E., Gimmi, U.,
Ginzle , C., … Dobbe in, M. (2013). D i ing ac o s o a ege a ion
shi om Sco s pine o pubescen oak in d y Alpine o es s. Global
Change Biology, 19(1), 229–240. h ps://doi.o g/10.1111/gcb.12038
Rigling, A., Eilmann, B., Koechli, R., & Dobbe in, M. (2010). Mis le oe-
induced c own deg ada ion in Sco s pine in a xe ic en i onmen . T ee
Physiology, 30, 845–852. h ps://doi.o g/10.1093/ eep hys/ pq038
Rouaul , G., Candau, J.-N., Lieu ie , F., Nageleisen, L.-M., Ma in, J.-C.,
& Wa zée, N. (2006). E ec s o d ough and hea on o es insec
popula ions in ela ion o he 2003 d ough in Wes e n Eu ope.
Annals o Fo es Science, 63(6), 613–624. h ps://doi.o g/10.1051/
o es :2006044
Ruij en, J., & Be endse, F. (2010). Di e si y enhances communi y e-
co e y, bu no esis ance, a e d ough . Jou nal o Ecology, 98(1),
81–86.
Ryan, M. G., & Yode , B. J. (1997). Hyd aulic limi s o ee heigh and ee
g ow h. BioScience, 47, 235–242. h ps://doi.o g/10.2307/1313077
Sánchez-Salgue o, R., Cama e o, J. J., He ia, A., Mad igal-González,
J., Lina es, J. C., Balles e os-Cano as, J. A., … Rigling, A. (2015).
Wha d i es g ow h o Sco s pine in con inen al Medi e anean cli-
ma es: D ough , low empe a u es o bo h? Ag icul u al and Fo es
Me eo ology, 206, 151–162. h ps://doi.o g/10.1016/j.ag o me .
2015.03.004
Sánchez-Salgue o, R., Cama e o, J. J., Rozas, V., Géno a, M., Olano,
J. M., A zac, A., … Lina es, J. C. (2018). Resis , eco e o bo h?
G ow h plas ici y in esponse o d ough is geog aphically s uc-
u ed and linked o in aspeci ic a iabili y in Pinus pinas e . Jou nal
o Biogeog aphy, 45(5), 1126–1139. h ps://doi.o g/10.1111/
jbi.13202
Sánchez-Salgue o, R., Na a o-Ce illo, R. M., Cama e o, J. J., &
Fe nández-Cancio, Á. (2012). Selec i e d ough -induced decline o
pine species in sou heas e n Spain. Clima ic Change, 113(3), 767–785.
h ps://doi.o g/10.1007/s1058 4-011-0372-6
|
4537
BOSE E al.
Scholz, F., Phillips, N., Bucci, S., & Golds ein, G. (2011). Hyd aulic ca-
paci ance: Biophysics and unc ional signi icance o in e nal wa e
sou ces in ela ion o ee size. In F. C. Meinze , B. Lachenb uch &
T. E. Dawson (Eds.), Size- and age- ela ed changes in ee s uc u e
and unc ion (Vol. 4, pp. 341–361). Do d ech , The Ne he lands:
Sp inge .
Seidel, H., Ma iu, M., & Menzel, A. (2019). Compensa o y g ow h o
Sco s pine seedlings mi iga es impac s o mul iple d ough s wi hin
and ac oss yea s. F on ie s in Plan Science, 10, 519–519. h ps://doi.
o g/10.3389/ pls.2019.00519
Seidel, H., Schunk, C., Ma iu, M., & Menzel, A. (2016). Di e ging d ough
esis ance o Sco s pine p o enances e ealed by in a ed he mog-
aphy. F on ie s in Plan Science, 7, 1247. h ps://doi.o g/10.3389/
pls.2016.01247
Se a-Maluque , X., Mencuccini, M., & Ma ínez-Vilal a, J. (2018). Changes
in ee esis ance, eco e y and esilience ac oss h ee successi e ex-
eme d ough s in he no heas Ibe ian Peninsula. Oecologia, 187(1),
343–354. h ps://doi.o g/10.1007/s0044 2-018-4118-2
Smi h, M. D. (2011). An ecological pe spec i e on ex eme clima ic
e en s: A syn he ic de ini ion and amewo k o guide u u e e-
sea ch. Jou nal o Ecology, 99(3), 656–663. h ps://doi.o g/10.1111/
j.1365-2745.2011.01798.x
Spinoni, J., Naumann, G., Ca ao, H., Ba bosa, P., & Vog , J. (2014).
Wo ld d ough equency, du a ion, and se e i y o 1951–2010.
In e na ional Jou nal o Clima ology, 34(8), 2792–2804. h ps://doi.
o g/10.1002/joc.3875
Spinoni, J., Vog , J. V., Naumann, G., Ba bosa, P., & Dosio, A. (2018).
Will d ough e en s become mo e equen and se e e in Eu ope?
In e na ional Jou nal o Clima ology, 38(4), 1718–1736. h ps://doi.
o g/10.1002/joc.5291
S e ck, F. J., Zwei el, R., Sass-Klaassen, U., & Chowdhu y, Q. (2008).
Pe sis ing soil d ough educes lea speci ic conduc i i y in Sco s pine
(Pinus syl es is) and pubescen oak (Que cus pubescens). T ee Physiology,
28(4), 529–536. h ps://doi.o g/10.1093/ eep hys/28.4.529
S o all, A. E. L., Shuga , H., & Yang, X. (2019). T ee heigh explains mo -
ali y isk du ing an in ense d ough . Na u e Communica ions, 10(1),
4385. h ps://doi.o g/10.1038/s4146 7-019-12380 -6
Szejne , P., Belmeche i, S., Ehle inge , J. R., & Monson, R. K. (2020).
Recen inc eases in d ough equency cause obse ed mul i-yea
d ough legacies in he ee ings o semi-a id o es s. Oecologia, 192,
241–259.
Taege , S., Zang, C., Liesebach, M., Schneck, V., & Menzel, A. (2013).
Impac o clima e and d ough e en s on he g ow h o Sco s pine
(Pinus syl es is L.) p o enances. Fo es Ecology and Managemen , 307,
30–42. h ps://doi.o g/10.1016/j. o eco.2013.06.053
Vallada es, F., Gianoli, E., & Gómez, J. (2007). Ecological limi s o plan
pheno ypic plas ici y. New Phy ologis , 176(4), 749–763. h ps://doi.
o g/10.1111/j.1469-8137.2007.02275.x
Vallada es, F., Ma esanz, S., Guilhaumon, F., A aújo, M. B., Balague , L.,
Beni o-Ga zón, M., … Za ala, M. A. (2014). The e ec s o pheno ypic
plas ici y and local adap a ion on o ecas s o species ange shi s
unde clima e change. Ecology Le e s, 17(11), 1351–1364. h ps://doi.
o g/10.1111/ele.12348
Vanninen, P., & Mäkelä, A. (2005). Ca bon budge o Sco s pine ees:
E ec s o size, compe i ion and si e e ili y on g ow h alloca ion and
p oduc ion. T ee Physiology, 25(1), 17–30. h ps://doi.o g/10.1093/
eep hys/25.1.17
Venne ie , M., Ripe , C., & Ra hgebe , C. (2018). Au ecology and g ow h
o Aleppo pine (Pinus halepensis Mill.): A comp ehensi e s udy in
F ance. Fo es Ecology and Managemen , 413, 32–47. h ps://doi.o g/
10.1016/j. o eco.2018.01.028
Venne ie , M., Thabee , A., Didie , C., Gi a d, F., Caille e , M.,
Taugou deau, O., … Ca aglio, Y. (2013). Clima e change impac on
ee a chi ec u al de elopmen and lea a ea. In B. R. Singh (Eds.),
Clima e change— eali ies, impac s o e ice cap, sea le el and isks.
INTECH Open Access Publishe . h p://dx.doi.o g/10.5772/51510
Vicen e-Se ano, S. M., Begue ía, S., & López-Mo eno, J. I. (2010). A mul-
iscala d ough index sensi i e o global wa ming: The S anda dized
P ecipi a ion E apo anspi a ion Index. Jou nal o Clima e, 23(7),
1696–1718. h ps://doi.o g/10.1175/2009J CLI29 09.1
Vicen e-Se ano, S. M., Gou eia, C., Cama e o, J. J., Begue ia, S., T igo,
R., Lopez-Mo eno, J. I., … Sanchez-Lo enzo, A. (2013). Response
o ege a ion o d ough ime-scales ac oss global land biomes.
P oceedings o he Na ional Academy o Sciences o he Uni ed S a es
o Ame ica, 110(1), 52–57. h ps://doi.o g/10.1073/pnas.12070
68110
Villa -Sal ado , P., Peñuelas, J. L., & Jacobs, D. F. (2013). Ni ogen nu-
i ion and d ough ha dening exe opposi e e ec s on he s ess
ole ance o Pinus pinea L. seedlings. T ee Physiology, 33(2), 221–232.
h ps://doi.o g/10.1093/ eep hys/ ps133
Vi ali, V., Bün gen, U., & Bauhus, J. (2017). Sil e i and Douglas i
a e mo e ole an o ex eme d ough s han No way sp uce in
sou h-wes e n Ge many. Global Change Biology, 23(12), 5108–5119.
h ps://doi.o g/10.1111/gcb.13774
Wal he , G.-R., Pos , E., Con ey, P., Menzel, A., Pa mesan, C., Beebee,
T. J., … Bai lein, F. (2002). Ecological esponses o ecen clima e
change. Na u e, 416(6879), 389–395.
Wes oby, M., Fals e , D. S., Moles, A. T., Vesk, P. A., & W igh , I. J. (2002).
Plan ecological s a egies: Some leading dimensions o a ia ion
be ween species. Annual Re iew o Ecology and Sys ema ics, 33(1),
125–159. h ps://doi.o g/10.1146/annu e .ecols ys.33.010802.15
0452
Zalloni, E., Ba ipaglia, G., Che ubini, P., Sau e , M., & De Micco, V. (2019).
Wood g ow h in pu e and mixed Que cus ilex L. o es s: D ough in-
luence depends on si e condi ions. F on ie s Plan Science, 10, 397.
Zang, C., Bu as, A., Esqui el-Muelbe , A., Jump, A. S., Rigling, A., &
Rammig, A. (2020). S anda dized d ough indices in ecological e-
sea ch: Why one size does no i all. Global Change Biology, 26, 322–
324. h ps://doi.o g/10.1111/gcb.14809
Zang, C., Ha l-Meie , C., Di ma , C., Ro he, A., & Menzel, A. (2014).
Pa e ns o d ough ole ance in majo Eu opean empe a e o es
ees: Clima ic d i e s and le els o a iabili y. Global Change Biology,
20(12), 3767–3779. h ps://doi.o g/10.1111/gcb.12637
SUPPORTING INFORMATION
Addi ional suppo ing in o ma ion may be ound online in he
Suppo ing In o ma ion sec ion.
How o ci e his a icle: Bose AK, Gessle A, Bol e A, e al.
G ow h and esilience esponses o Sco s pine o ex eme
d ough s ac oss Eu ope depend on p ed ough g ow h
condi ions. Glob Change Biol. 2020;26:4521–4537. h ps://doi.
o g/10.1111/gcb.15153
