1
1The e ec o ee decline o e soil mic oclima e la gely con ols soil espi a ion
2dynamics in a Medi e anean woodland
3Au ho s: Alexand a Rod íguez1*, Jo ge Du án1,2, Jo ge Cu iel Yus e3,4, Fe nando
4Vallada es5,6,7, Ana Rey5
5
6*Co esponding au ho
71Cen e o Func ional Ecology, CFE, Uni e si y o Coimb a, 3000-456 Coimb a, Po ugal
82Misión Biolóxica de Galicia, Consejo Supe io de In es igaciones Cien í icas, 36143
9Pon e ed a, Spain
10 3BC3 - Basque Cen e o Clima e Change, Scien i ic Campus o he Uni e si y o he Basque
11 Coun y, 48940 Leioa, Spain
12 4IKERBASQUE - Basque Founda ion o Science, Ma ia Diaz de Ha o 3, 6 solai ua, 48013
13 Bilbao, Bizkaia, Spain
14 5Depa men o Biogeog aphy and Global Change, Na ional Museum o Na u al Sciences,
15 MNCN, CSIC, 28006 Mad id, Spain
16 6LINCGlobal, Mad id, Spain
17 7A ea o Biodi e si y and Conse a ion, ESCET, Rey Juan Ca los Uni e si y, 28933 Mós oles,
18 Mad id, Spain
19
20 P esen pos al add ess o he co esponding au ho :
21 Cen e o Func ional Ecology, CFE, Depa men o Li e Sciences, Uni e si y o Coimb a,
22 Calçada Ma im de F ei as, 3000-456 Coimb a, Po ugal
23 Full elephone: +351 239240752; E-mail: [email p o ec ed]
24
25 Type o a icle: Resea ch pape .
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
2
26 Abs ac
27 As d ough -induced ee de olia ion and mo ali y (i.e. ee decline) in he Medi e anean is
28 expec ed o wo sen wi h ongoing clima e change, i is o pa amoun impo ance o unde s and
29 how, why, and when ee decline a ec s soil espi a ion (Rs) and i s ela ionship wi h soil
30 mic oclima e and o he po en ial con ols. We ca ied ou a no el s udy explo ing he
31 in e ac ing e ec s o clima ic a iabili y and ee decline on soil mic oclima e and Rs empo al
32 a iabili y in a Medi e anean holm oak woodland. The s udy u he explo es he e ec s o
33 ee decline on he main con ols o Rs a he s and scale. We moni o ed Rs, soil empe a u e
34 (Tsoil), and soil olume ic wa e con en (SWC) unde he canopy o 30 holm oak ees wi h
35 di e en de olia ion deg ees (heal hy, a ec ed, and dead) du ing wo yea s o con as ing
36 p ecipi a ion pa e ns. We es ima ed di e en plan s uc u al a iables (e.g. DBH, heigh , and
37 canopy diame e ) on hose selec ed ees unde whose canopies we also collec ed soil samples
38 o analyze di e en soil physicochemical a iables. Ou esul s s ess he impo an ole o ee
39 heal h as a modula o o he esponse o Rs o SWC, wi h s onge esponses o Rs o a ia ions
40 in he amoun and dis ibu ion o p ecipi a ion unde heal hy han unde declining ees. They
41 also sugges ha ee decline can signi ican ly inc ease SWC and dec ease Rs bu la gely
42 depending on he declining s age, he yea , and he season. Finally, ee decline also a ec ed
43 he ela ionship o Rs wi h soil mic oclima ic a iables, pa icula ly SWC, and he ela i e
44 impo ance o he di e en d i e s o Rs, wi h mic oclima e a iables gaining impo ance as
45 ees de olia e and die. Al oge he , ou esul s poin owa ds a nega i e impac o d ough -
46 induced ee decline on soil C con en and cycling, pa icula ly unde o ecas ed clima e change
47 scena ios wi h d ye and mo e in ense p ecipi a ion egimes.
48
49 Keywo ds: Que cus ilex; o es die-o ; clima e change; soil unc ioning; soil CO2 e lux;
50 en i onmen al con ol.
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
3
51 1. In oduc ion
52 Clima e models in Medi e anean ecosys ems o ecas inc eases in empe a u e and mo e
53 in ensi e and ex ensi e d ough s, wi h mo e equen and in ense ex eme empe a u e and
54 p ecipi a ion e en s (IPCC, 2021). These changes will a ec he s uc u e, composi ion, and
55 unc ioning o o es s in s ill unknown ways, wi h impo an implica ions o he ca bon (C)
56 balance (Liu e al., 2016; Reichs ein e al., 2013; Wang e al., 2012). Fo ecas ed changes in he
57 equency, in ensi y, and dis ibu ion o ain all e en s (IPCC, 2021) will likely inc ease he
58 impo ance o ain all pulses on he C cycle o Medi e anean en i onmen s (Rey e al., 2021,
59 2017; Song e al., 2017; Wang e al., 2016), la gely in luenced by he in a-annual a ia ion in
60 soil wa e con en (Galla do e al., 2009). Also, he inc easingly d y and ho clima ic condi ions
61 will exace ba e he d ough -induced ee de olia ion and mo ali y (he eina e “ o es die-o ”)
62 obse ed o e he pas wo decades (Ca nice e al., 2011; Llo e e al., 2004).
63 D ough -induced o es die-o impac s o es ecosys ems by educing ee
64 p oduc i i y, educing he inpu o li e and oo exuda es, and modi ying soil mic oclima ic
65 condi ions, and in u n, soil mic obial communi ies (Schlesinge e al., 2016). Acco dingly,
66 Rod íguez e al. (2017, 2020) ound e idence o a cascade e ec o ongoing d ough -induced
67 o es die-o in a Medi e anean holm oak (Que cus ilex) woodland ul ima ely dec easing he
68 con en and labili y o soil C. Al hough e o s o unde s and he e ec s o o es die-o on
69 impo an soil p ocesses ela ed o C cycling ha e inc eased la ely (A ila e al., 2019, 2016;
70 Cu iel Yus e e al., 2019; Ga cía-Angulo e al., 2020), esul s a e s ill sca ce and con adic o y.
71 Fo ins ance, some s udies ha e epo ed dec eases in soil espi a ion (Rs) as a esul o educed
72 oo ac i i y (A ila e al., 2016), whe eas o he s ha e ound no changes (Ba ba e al., 2016) o
73 e en inc eased a es (Ba ba e al., 2013; Cu iel Yus e e al., 2019; Edbu g e al., 2012)
74 ollowing changes in mic oclima e (e.g., inc eases in soil mois u e), inc easing li e inpu s and
75 seconda y successional p ocesses. Thus, he impac s o o es die-o on Rs a e complex and
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
4
76 la gely depend on he balance o i s e ec s on oo s (au o ophic) and mic obial (he e o ophic)
77 componen s (A ila e al., 2016). Mo eo e , ecen s udies sugges ha o es die-o can la gely
78 modula e he esponse o mic obial espi a ion o clima e change and ha o es die-o e ec s
79 on di e en soil a ibu es and p ocesses a e seasonal (A ila e al., 2019, 2016; Rod íguez e
80 al., 2019). S ill, he in e ac ing e ec s o ee decline and clima ic a iabili y on Rs ha e been
81 sca cely explo ed, speci ically conside ing Medi e anean o es s and d ough as he main
82 d i e o ee decline (Ba ba e al., 2016).
83 Soil espi a ion is he la ges sou ce o CO2 om e es ial ecosys ems and, he e o e,
84 a cen al piece o he global C balance (Schlesinge and And ews, 2000). Howe e , i is also a
85 e y complex p ocess con olled by se e al physiological, phenological, and en i onmen al
86 p ocesses ha a y bo h in ime and space (Rey e al., 2021, 2011; Tang and Baldocchi, 2005).
87 A he global scale, i s empo al and spa ial a iabili y is mainly con olled by ai empe a u e
88 and p ecipi a ion, whe eas a mic o and s and scales, o he ac o s ela ed o plan communi y
89 and soil o ganic ma e become mo e impo an (Ba ba e al., 2013; Raich and Schlesinge ,
90 1992). Mo eo e , whe eas au o ophic espi a ion (RA) is mainly con olled by ee physiology
91 and p oduc i i y (Högbe g e al., 2001; Ma eucci e al., 2015), he e o ophic espi a ion (RH)
92 is la gely con olled by soil mic oclima ic (i.e. wa e con en and empe a u e) and
93 biogeochemical ac o s (e.g. o ganic ma e con en and quali y, mic obial communi y
94 s uc u e) (Cu iel Yus e e al., 2007; Tang and Baldocchi, 2005; Zhao e al., 2016). This
95 complexi y is a leas pa ially esponsible o he la ge unce ain ies associa ed wi h
96 p edic ions o global a es o Rs and Rs esponses o clima e change (Wa ne e al., 2019). Thus,
97 he e is an u gen need o con inue o s udy his c i ical ecosys em p ocess unde di e en
98 scena ios and a di e en empo al and spa ial scales.
99 We ca ied ou a no el wo-yea ield s udy o explo e he in e ac ing e ec s o clima ic
100 a iabili y and di e en s ages o ee decline (i.e. heal hy, a ec ed, and dead) on soil
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
5
101 mic oclima e and Rs in a Que cus ilex Medi e anean woodland. Fu he mo e, we explo ed he
102 e ec s o ee decline alone on he main con ols o Rs a he s and scale (i.e. plan a iables
103 and soil mic oclima ic and physicochemical a iables). We hypo hesized ha i) ee decline
104 modi ies soil mic oclima ic condi ions and, in u n, Rs a es, bu ii) hese e ec s would depend
105 on he season and yea . Mo eo e , assuming changes in he ela i e con ibu ion o RA and RH
106 associa ed wi h ee decline (Cu iel Yus e e al., 2019), we hypo hesized ha iii) he ela i e
107 impo ance o plan , soil mic oclima ic, and physicochemical a iables con olling Rs would
108 change conside ing all ee decline s ages (i.e. heal hy, a ec ed and dead) oge he and
109 sepa a ely.
110
111 2. Ma e ial and Me hods
112 2.1 S udy si e
113 The s udy was ca ied ou in a holm oak woodland loca ed in he cen al pa o he Ibe ian
114 Peninsula, sou hwes o Mad id (40°23′N, 4°11′W; 630-660 m abo e sea le el). The clima e is
115 con inen al Medi e anean wi h mean annual empe a u e o 13 °C and mean annual
116 p ecipi a ion o 601 mm (Felicísimo e al., 2011). Mos ain all concen a es om au umn o
117 sp ing, while summe s a e wa m and d y. Soil is a Cambisol, sandy and sligh ly acid (pH~6.3),
118 wi h low o al C and N con en (Table S1). Abo eg ound ege a ion is cha ac e ized by a ee
119 densi y o ~180 ees ha−1, mos ly composed o Que cus ilex L. ssp. ballo a [Des .] Samp (holm
120 oak) wi h sca ce Junipe us oxyced us Sib h. & Sm (ceda ). The unde s o y is domina ed by
121 Re ama sphae oca pa L., La andula s oechas ssp. peduncula a (Mill.) Samp. ex Rozei a, and
122 di e se pas u e species (Rod íguez e al., 2017). In 2005, his egion su e ed a s ong e en o
123 holm oak de olia ion (a ound 20–30% o he o al popula ion) and mo ali y (15%) due o a
124 se e e d ough . Since hen, he holm oaks o his woodland show di e en decline s ages
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
6
125 (heal hy, a ec ed, and dead) wi h con as ing soil a ibu es unde nea h (Rod íguez e al., 2020,
126 2017) (Table S1).
127
128 2.2 Expe imen al design
129 In sp ing 2013, we selec ed 30 holm oak adul ees o simila size (38 cm and 4 m o diame e
130 a b eas heigh and heigh on a e age, espec i ely, Table S1) bu wi h di e en le els o
131 canopy decline (10 heal hy, 10 a ec ed, and 10 dead), and consequen ly wi h signi ican ly
132 di e en canopy diame e (Table S1; Rod íguez e al. 2017), sepa a ed a leas 10 m om o he
133 ees. Canopy decline s a us o he di e en ees was isually de e mined – heal hy <10%
134 de olia ed; a ec ed > 50% de olia ed; dead 100% de olia ed.
135 On he no h ace o each ee, we es ablished a sampling poin below he in luence o
136 he ee canopy, 0.3 m away om he unk (Rod íguez e al., 2017). Then poly inylchlo ide
137 soil colla s (15 cm in diame e and 7 cm in heigh ) we e inse ed 3.5 cm in o he soil a each
138 sampling poin , whe e hey emained o wo consecu i e annual pe iods: om June 2013 o
139 May 2014 (yea 1) and om June 2014 o May 2015 (yea 2).
140 Bo h annual pe iods we e ex emely wa m acco ding o he e e ence pe iod (1971-
141 2000 and 1981-2010 un il and om Janua y 2015, espec i ely; AEMET), wi h simila mean
142 annual ai empe a u es (15.9 oC and 15.8 oC, espec i ely) bu lowe p ecipi a ion alues in
143 yea 1 (288.8 mm) han in yea 2 (343.3 mm), pa icula ly in la e summe and all (Fig. 1).
144
145 2.3 Soil mic oclima e and soil espi a ion measu emen s
146 We moni o ed Rs, along wi h soil empe a u e (Tsoil) and soil olume ic wa e con en (SWC),
147 in all sampling poin s om June 2013 o May 2015 a a equency o app oxima ely wice a
148 mon h du ing sp ing and all and once a mon h du ing summe and win e (30 campaigns in
149 o al; Figu e 1). To a oid s ong diu nal luc ua ions, measu emen s we e done du ing he
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
7
150 midday pe iod (be ween 10:00 and 14:00 h). Mo eo e , special ca e was aken o ully
151 andomize he sampling sequence ac oss ee decline s ages in each measu emen campaign.
152 Small plan s, excessi e li e , insec s, and g asses we e ca e ully emo ed om each colla
153 be o e each Rs measu emen . A po able in a ed gas analyze (IRGA) connec ed o a soil
154 espi a ion chambe (EGM-4 and SRC-1; PP Sys ems, USA) was used o measu e Rs a es in
155 si u du ing 90 s o ensu e eliable measu emen s. A he ime o each Rs measu emen , Tsoil and
156 SWC we e also measu ed adjacen o each soil colla a 4 cm dep h ( he mos ac i e) inse ing
157 e ically a soil digi al he mome e (Max ech, Lokhna h En e p ise, India) and a ime-domain
158 e lec ome e (Fieldscou TDR 300, Spec um Technologies, Inc., Plain ield, IL, USA),
159 espec i ely. We es ima ed seasonal, annual, and wo-yea a e age alues o Rs, Tsoil, and SWC
160 conside ing all measu es ca ied ou wi hin each espec i e pe iod. Whe eas ou expe imen al
161 design ha dly p o ides accu a e daily, seasonal o annual Rs es ima ions, i is pe ec ly alid in
162 e ms o compa ison among he di e en decline s ages and co ela ions wi h he di e en
163 p edic o a iables.
164
165 2.4 Plan and soil physicochemical a iables
166 A he end o May 2013, we measu ed ee heigh , diame e a b eas heigh (DBH), and he
167 canopy diame e o he 30 s udied ees by using a clinome e , a DBH ape, and a e aging wo
168 pe pendicula measu emen s o canopy diame e , espec i ely. We also es ima ed a dis ance-
169 dependen compe i ion index be ween ees o each one o he s udied ees (Hegyi 1974),
170 conside ing all he compe ing holm oaks and ceda s wi hin a 5.5 m adius, he DBH o he
171 subjec ee and ha o i s compe i o s as well as he dis ance be ween he subjec ee and i s
172 compe i o s (Table S1; Rod íguez e al. 2017). On he same da e, we also measu ed soil
173 compac ion in all sampling poin s by using a soil compac ion me e (FieldScou SC 900, USA).
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
8
174 We ca ied ou wo di e en soil samplings (May 2013 and Ma ch 2015) o es ima e
175 a ious physicochemical a iables o each decline s age (heal hy, a ec ed, and dead). Soil
176 samples we e collec ed close o he soil colla s and om he i s 10 cm o he soil p o ile by
177 using a 5 cm (i.d.) me al co e and hen kep a 4 °C un il analysis. Be o e analysis, soil samples
178 we e sie ed (2 mm mesh size), homogenized in ield-mois condi ions and analyzed o
179 g a ime ic mois u e by o en-d ying a subsample a 60 °C o cons an mass. Soil o al C and
180 N con en , and soil pH, we e analyzed in soil samples collec ed in May 2013 by d y combus ion
181 wi h an elemen al analyze (LECO T uSpec CN, S . Joseph, MI, USA) and using a soil- o-
182 wa e a io o 1:2.5 (m/ ), espec i ely. The possible occu ence o ino ganic C was checked
183 by a Die ich-F uhling olume ic calcime e . Gi en i s absence, o al C was conside ed equal
184 o soil o ganic C. Labile C was de e mined in he 2015 samples by using he High G adien
185 Magne ic Sepa a ion (HGMS) me hod as desc ibed in Rod íguez e al. (2020). B ie ly, 10
186 g ams o each sample unde wen HGMS by a F an z Isodynamic Sepa a o (Model L-1, SG
187 F an z Co, Inc., T en on, New Je sey, USA) ha sepa a ed each soil sample in wo ac ions
188 wi h di e en u no e imes, a non-magne ic and a magne ic ac ion (Chi i e al., 2019). The
189 magne ic ac ion has la ge con ibu ions om ela i ely ecen C o ms han he non-magne ic
190 ac ion and hus, is supposed o be mo e labile. We analyzed o al C in he MA ac ion (CMA)
191 by d y combus ion (see abo e) and conside ed i as labile C. Soil bulk densi y was also
192 es ima ed in he soil samples collec ed in 2015 (Table S1).
193
194 2.5 S a is ical analyses
195 Linea models we e pe o med o ela e seasonal ai empe a u e and p ecipi a ion alues wi h
196 ou es ima ed Tsoil and SWC seasonal alues, espec i ely, as indica i e o he sui abili y o ou
197 seasonal measu emen s.
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
9
198 We used gene alized leas squa es (GLS) models o es he e ec s o soil mic oclima ic
199 (in e - and in a-annual) a iabili y and ee decline s a us, as well as hei in e ac ions, on
200 SWC, Tsoil, and Rs. Pai wise s a is ical di e ences among he means o he ac o le els we e
201 es ed using simul aneous es s o gene al linea hypo heses (mul iple compa isons o means:
202 Tukey con as s). Fo all he GLS models, he ee was used as a andom ac o o accoun o
203 empo al dependencies and, when necessa y, no-no mali y and he e oscedas ici y o he
204 esiduals we e co ec ed by log- ans o ming he dependen a iable and/o using he a gumen
205 weigh s, espec i ely.
206 We explo ed he ela ionship be ween Rs and Tsoil o he whole sampling pe iod and
207 each decline s age sepa a ely using a e aged alues o each empe a u e deg ee. We iden i ied
208 wo di e en pe iods de e mined by a empe a u e h eshold om which he ela ionship o Rs
209 wi h Tsoil changes (Pe iod I – below he empe a u e h eshold, mos o he yea excep summe ;
210 and Pe iod II – om he h eshold, summe ). Fo Pe iod I, soil espi a ion was conside ed
211 dependen on soil empe a u e acco ding o an exponen ial ela ionship:
212
213 (1)
𝑅
𝑠
=
𝑅
𝑏𝑎𝑠𝑎𝑙
∗
𝑒
𝑏
∗
𝑇
214
215 whe e is he soil espi a ion (µmol C m-2 s-1), is soil empe a u e (ºC) measu ed a a dep h
𝑅
𝑠
𝑇
216 o 4 cm, and ( he basal espi a ion a e) and b a e i ed pa ame e s. The Q10 ( he inc ease
𝑅
𝑏𝑎𝑠𝑎𝑙
217 in he lux a e o a 10 ºC inc ease in empe a u e) was calcula ed as ollows:
218 (2)
𝑄
10
=
𝑒
10
∗
𝑏
219
220 Fo Pe iod II, Rs was be e explained by soil wa e con en h ough a loga i hmic unc ion:
221 (3)
𝑅
𝑠
=
𝑎
+
𝑏
∗
𝑙𝑛
(
𝑆𝑊𝐶
)
222
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
16
372 lesse ex en o soil C quali y and/o RA, as d i e s o ha esponse. In a p e ious s udy on he
373 same si e, Rod íguez e al. (2017) ound ha while po en ial RH was up o 66% lowe unde
374 a ec ed han unde heal hy ees, dead ees showed simila alues o RH o hose o heal hy
375 ees (Rod íguez e al., 2017). Thus, al hough hese p e ious esul s do no suppo he
376 hypo hesis o highe RH unde dead han unde heal hy ees, hey do sugges a po en ial
377 eco e y o RH a e ee dea h. Al hough we did no analyze he di e en componen s o Rs
378 sepa a ely, and he RH alues om Rod iguez e al. (2017) do no ake in o accoun all he
379 clima ic a iabili y o ou wo-yea s udy, all abo e sugges ha he he e o ophic componen
380 could ha e a key ole in he obse ed esponse o Rs o ee decline in his Medi e anean
381 woodland.
382 The appa en lack o e ec o o es die-o on soil C and nu ien cycling du ing sp ing
383 could be la gely explained by he he baceous coloniza ion unde declining ees (Rod íguez e
384 al. 2017) and he peak ac i i y o he baceous oo s du ing his season, which may ha e helped
385 o se he nega i e e ec o ee decline o e he ee and soil mic obial ac i i y (A ila e al.,
386 2016; Rod íguez e al., 2019; Tang and Baldocchi, 2005). On he o he hand, whe eas we did
387 no ind di e ences in win e Tsoil and summe SWC among decline s ages, we did ind hem
388 in Rs in bo h win e s and summe s, sugges ing a high esilience o he RA componen o
389 un a o able clima ic condi ions in heal hy bu no in declining ees. In any case, ou s udy
390 shows ha he soil e ec s o ee decline a e de ec able be o e ee dea h. Fu he , ou esul s
391 sugges ha , as mo ali y inc eases in his ype o woodlands, and p o ided ha condi ions o
392 mic obial unc ioning a e me , hey could s ill emi a signi ican amoun o CO2 while likely
393 ixing much less C om he a mosphe e han heal hy woodlands. This si ua ion could lead, a
394 leas in he sho - (i.e. yea s) o medium- e m (i.e. decades), o a swi in he ole o hese o es s
395 om sinks o sou ces o C (Baldocchi e al., 2004; Na e e al., 2011; Xiong e al., 2011). The
396 s ong nega i e e ec s o o es die-o on bo h soil C con en and labili y obse ed in his
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
17
397 sys em by Rod íguez e al. (2020) suppo his hypo hesis. Finally, hese esul s ein o ce he
398 need o conside bo h he spa ial (i.e. di e en declining s ages) and clima ic (seasonal and
399 annual) a iabili y, wo componen s s ill sca cely explo ed oge he , when ying o ully
400 unde s and he complex e ec s o ee decline on soil a ibu es and unc ioning (A ila e al.,
401 2016).
402
403 4.2 E ec o ee decline on he ela ionship o Rs wi h soil mic oclima ic condi ions
404 As expec ed (e.g. Rey e al. 2011; Ba ba e al. 2016), Rs was be e explained by Tsoil up o a
405 de e mined h eshold (Pe iod I), which co esponded wi h s ill op imal Tsoil (~17oC) bu low
406 SWC (< 7%) condi ions. Thus, once SWC eached limi ing alues, Rs was no longe be e
407 explained by Tsoil bu by SWC (Pe iod II), highligh ing he key ole o wa e a ailabili y in he
408 unc ioning o Medi e anean ecosys ems (Reichs ein e al., 2002; Rey e al., 2011, 2002). The
409 basal espi a ion alues suppo ed he lowe espi a ion ac i i y unde a ec ed han unde
410 heal hy ees discussed abo e. In e es ingly, he highe Q10 alues and a iances o Rs explained
411 by Tsoil and SWC unde declining han unde heal hy ees, along wi h he signi ican
412 co ela ions ound be ween he seasonal means o Rs and hose o p ecipi a ion only unde
413 declining ees migh sugges ha he con ol o Rs by soil mic oclima ic a iables inc eases
414 wi h ee decline. As au o ophic and he e o ophic espi a ions a he s and le el a e mainly
415 con olled by plan s and soil mic oclima e, espec i ely (Chen e al., 2019; Högbe g e al.,
416 2001; Ma eucci e al., 2015), ou esul s migh indica e an inc ease in he ela i e con ibu ion
417 o he he e o ophic espi a ion unde ee decline. In any case, hese esul s s ess he capaci y
418 o ee decline o sha ply modula e he ela ionship o Rs wi h soil mic oclima ic a iables.
419
420 4.3 Main d i e s o Rs a he s and scale
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
18
421 Ou esul s con i med ou hi d hypo hesis abou changes in he ela i e impo ance o plan ,
422 soil mic oclima ic, and physicochemical a iables con olling Rs conside ing he ee decline
423 s ages oge he and sepa a ely. Ou esul s e ealed ha canopy diame e (a good p oxy o ee
424 decline; see me hods sec ion) is he mos impo an p edic o o Rs, indica ing ha , a he s and
425 le el, heal hie ees wi h la ge canopies suppo highe Rs a es. These esul s highligh bo h
426 he impo an ole o ee decline (i.e. heal h s a us) as a d i e o he ecosys em unc ioning
427 and he impo an con ibu ion o he au o ophic componen o Rs in ou ecosys em. Along
428 wi h canopy diame e , o al N was also a e y impo an p edic o o Rs, and he mos impo an
429 p edic o unde heal hy ees, suppo ing he no ion o N as a key nu ien o hese ecosys ems'
430 unc ioning (Högbe g e al., 2001). Fu he mo e, he obse ed nega i e co ela ion be ween
431 canopy diame e and he C:N a io (Figu e S1) suppo s he p e iously obse ed nega i e e ec
432 o o es die-o on soil o ganic ma e quali y and N a ailabili y (Rod íguez e al., 2020, 2017).
433 Al oge he , hese esul s wa n abou a likely posi i e eedback be ween o es die-o and N
434 limi a ion (Gessle e al., 2016), wi h impo an implica ions o he ecosys em C and N
435 balance.
436 The bes models and mos impo an p edic o s ob ained o heal hy as compa ed o
437 a ec ed and dead ees demons a e ha ee decline changes he ela i e impo ance o he
438 di e en Rs, wi h mic oclima e a iables gaining impo ance as he de olia ion deg ee
439 inc eases. T ee de olia ion igge s a cascade e ec on plan unde s o y and soil mic obial
440 communi ies wi h impo an implica ions o ecosys em C and N budge s, including a dec ease
441 in soil o ganic ma e labili y (Rod íguez e al. (2017, 2020), which could la gely a ec RH
442 (Rod íguez e al., 2019; Rui e al., 2016). As discussed abo e, C:N a io inc eased as canopy
443 diame e dec eased, suppo ing he nega i e e ec o ee de olia ion and mo ali y on soil
444 o ganic ma e quali y. Acco dingly, he bes models o bo h a ec ed and dead ees included
445 labile C as one o he mos impo an Rs p edic o s. Bulk densi y and Tsoil we e he second and
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
19
446 hi d mos impo an p edic o s o Rs unde a ec ed ees, whe eas he bes model o Rs unde
447 dead ees included Tsoil and SWC as he mos impo an p edic o s. Bulk densi y is widely
448 ecognized o be nega i ely ela ed o soil o ganic ma e con en (Pé ié and Ouime , 2008) and
449 oo s mass (Daddow and Wa ing on, 1983), which a e among he main con ols o RH and RA,
450 espec i ely. These esul s would ag ee wi h he abo e-discussed pa allel dec ease in RA and
451 RH unde a ec ed ees and almos comple e subs i u ion o Rs o RH unde dead ees. Finally,
452 whe eas he bes models o heal hy and a ec ed ees ne e su passed 50% o he a iance
453 explained, he bes model o dead ees eached 93% o he a iance explained wi h jus bulk
454 densi y and bo h soil mic oclima ic a iables as p edic o s. This esul migh sugges a
455 simpli ica ion o he con ol o he soil espi a ion p ocess unde ee decline, which could
456 jeopa dize i s esilience o he global change d i e s, including clima e change (Hong e al.,
457 2022).
458
459 5. Conclusions
460 Ou s udy adds obus and no el obse a ional e idence o in e ac ing e ec s be ween clima ic
461 a iabili y and d ough -induced ee decline on he spa ial- empo al a iabili y o Rs in
462 Medi e anean holm-oak o es s. The highe esponses o Rs o a ia ions in he amoun and
463 dis ibu ion o p ecipi a ion unde heal hy han unde declining ees s ess he impo an ole
464 o ee heal h as a modula o o he esponse o Rs o soil mois u e condi ions. We also show
465 ha ee decline s ongly in luences soil mic oclima e and Rs in hese Medi e anean o es s,
466 bu he magni ude o his e ec depends on o he ac o s such as he declining s age (i.e.
467 a ec ed o dead), he yea , and he season. Fu he , ou s udy demons a es ha ee decline
468 also changes he ela i e impo ance o he di e en d i e s o Rs, wi h mic oclima e a iables
469 gaining impo ance as he de olia ion deg ee inc eases. Finally, ou s udy exposed a likely
470 posi i e eedback be ween o es die-o and N limi a ion and a simpli ica ion o he soil
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
20
471 espi a ion p ocess wi h ee decline. Al oge he , ou esul s poin owa ds nega i e impac s o
472 d ough -induced ee decline on soil C con en and cycling, pa icula ly unde o ecas ed
473 clima e change scena ios wi h d ye and mo e in ense p ecipi a ion egimes. Thus, mo e s udies
474 in es iga ing he join e ec s o d ough -induced o es die-o and clima ic a iabili y on soil
475 espi a ion and i s main componen s a di e en empo al and spa ial scales a e needed o ully
476 unde s and he a e o his impo an ecosys em p ocess.
477
478 Acknowledgmen s
479 This esea ch was suppo ed by he Spanish Na ional Resea ch Council (CSIC) in he JAE-doc
480 modali y co- inanced by he Eu opean Social Fund (ESF), he ATLANTIS (PID2020-
481 113244GB-C21) p ojec s unded by he Spanish Go e nmen , he Basque Go e nmen h ough
482 he BERC 2022-2025 p og am, and he Spanish Minis y o Science and Inno a ion h ough
483 he BC3 Ma ía de Maez u excellence acc edi a ion (MDM-2017-0714). J.D. and A.R.
484 acknowledge suppo om he FCT (2020.03670.CEECIND and SFRH/BDP/108913/2015,
485 espec i ely), as well as om he MCTES, FSE, UE, and he CFE (UIDB/04004/2021) esea ch
486 uni inanced by FCT/MCTES h ough na ional unds (PIDDAC). The au ho s a e g a e ul o
487 all he people ha a some poin helped wi h ieldwo k, pa icula ly o Da id López. Also,
488 special hanks o Ma ia José Fe nández Alonso and Luis Ma ia Ca ascal o hei ad ice on
489 s a is ics.
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
21
490 Re e ences
491 Ande egg, W.R.L., Ande egg, L.D.L., She man, C., Ka p, D.S., 2012. E ec s o widesp ead
492 d ough -induced aspen mo ali y on unde s o y plan s. Conse . Biol. 26, 1082–1090.
493 h ps://doi.o g/10.1111/j.1523-1739.2012.01913.x
494 A ila, J.M., Galla do, A., Gómez-Apa icio, L., 2019. Pa hogen-induced ee mo ali y
495 in e ac s wi h p edic ed clima e change o al e soil espi a ion and nu ien
496 a ailabili y in Medi e anean sys ems. Biogeochemis y 142, 53–71.
497 h ps://doi.o g/10.1007/s10533-018-0521-3
498 A ila, J.M., Galla do, A., Ibáñez, B., Gómez-Apa icio, L., 2016. Que cus sube dieback
499 al e s soil espi a ion and nu ien a ailabili y in Medi e anean o es s. J Ecol 104,
500 1441–1452. h ps://doi.o g/10.1111/1365-2745.12618
501 Baldocchi, D.D., Xu, L., Kiang, N., 2004. How plan unc ional- ype, wea he , seasonal
502 d ough , and soil physical p ope ies al e wa e and ene gy luxes o an oak–g ass
503 sa anna and an annual g assland. Ag icul u al and Fo es Me eo ology 123, 13–39.
504 h ps://doi.o g/10.1016/j.ag o me .2003.11.006
505 Ba ba, J., Cu iel Yus e, J., Ma ínez-Vilal a, J., Llo e , F., 2013. D ough -induced ee species
506 eplacemen is e lec ed in he spa ial a iabili y o soil espi a ion in a mixed
507 Medi e anean o es . Fo es Ecology and Managemen 306, 79–87.
508 h ps://doi.o g/10.1016/j. o eco.2013.06.025
509 Ba ba, J., Cu iel Yus e, J., Poya os, R., Janssens, I.A., Llo e , F., 2016. S ong esilience o
510 soil espi a ion componen s o d ough -induced die-o esul ing in o es seconda y
511 succession. Oecologia 1–15. h ps://doi.o g/10.1007/s00442-016-3567-8
512 Bi ch, H.F., 1958. The e ec o soil d ying on humus decomposi ion and ni ogen
513 a ailabili y. Plan Soil 10, 9–31. h ps://doi.o g/10.1007/BF01343734
514 B e z, F., Ho ho n, T., Wes all, P., 2011. Mul iple Compa isons Using R. CRC P ess.
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
22
515 Bu nham, K.P., Ande son, D.R., 2010. Model Selec ion and Mul imodel In e ence: A
516 P ac ical In o ma ion-Theo e ic App oach, 2nd edi ion. ed. Sp inge , New Yo k,
517 USA.
518 Ca nice , J., Coll, M., Ninye ola, M., Pons, X., Sánchez, G., Peñuelas, J., 2011. Widesp ead
519 c own condi ion decline, ood web dis up ion, and ampli ied ee mo ali y wi h
520 inc eased clima e change- ype d ough . PNAS 108, 1474–1478.
521 h ps://doi.o g/10.1073/pnas.1010070108
522 Chen, Z., Xu, Y., Cas ellano, M.J., Fon aine, S., Wang, W., Ding, W., 2019. Soil Respi a ion
523 Componen s and hei Tempe a u e Sensi i i y Unde Chemical Fe ilize and
524 Compos Applica ion: The Role o Ni ogen Supply and Compos Subs a e Quali y.
525 Jou nal o Geophysical Resea ch: Biogeosciences 124, 556–571.
526 h ps://doi.o g/10.1029/2018JG004771
527 Chi i, T., Ce ini, G., Ma zaioli, F., D’Acqui, L.P., Fo e, C., Cas aldi, S., Valen ini, R., 2019.
528 Composi ion and u no e ime o o ganic ma e in soil ac ions wi h di e en
529 magne ic suscep ibili y. Geode ma 349, 88–96.
530 h ps://doi.o g/10.1016/j.geode ma.2019.04.042
531 Cu iel Yus e, J., Baldocchi, D.D., Ge shenson, A., Golds ein, A., Misson, L., Wong, S., 2007.
532 Mic obial soil espi a ion and i s dependency on ca bon inpu s, soil empe a u e and
533 mois u e. Global Change Biology 13, 2018–2035. h ps://doi.o g/10.1111/j.1365-
534 2486.2007.01415.x
535 Cu iel Yus e, J., Flo es-Ren e ía, D., Ga cía-Angulo, D., He eş, A.-M., B agă, C., Pe i an,
536 A.-M., Pe i an, I.C., 2019. Cascading e ec s associa ed wi h clima e-change-induced
537 coni e mo ali y in moun ain empe a e o es s esul in ho -spo s o soil CO2
538 emissions. Soil Biology and Biochemis y 133, 50–59.
539 h ps://doi.o g/10.1016/j.soilbio.2019.02.017
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
23
540 Daddow, R.L., Wa ing on, G., 1983. G ow h-limi ing soil bulk densi ies as in luenced by
541 soil ex u e. Wa e shed Sys ems De elopmen G oup, USDA Fo es Se ice.
542 Edbu g, S., Hicke, J., B ooks, P., Pendall, E., Ewa s, B., No on, U., Gochis, D., Gu man, E.,
543 Meddens, A., 2012. Cascading impac s o ba k bee le-caused ee mo ali y on
544 coupled biogeophysical and biogeochemical p ocesses. F on ie s in Ecology and he
545 En i onmen 10, 416–424. h ps://doi.o g/10.1890/110173
546 Felicísimo, Á.M., Muñoz, J., Villalba, C.J., Ma eo, R.G., 2011. Impac os, ulne abilidad y
547 adap ación al cambio climá ico de la biodi e sidad española. 1. Flo a y ege ación.
548 O icina Española de Cambio Climá ico, Minis e io de Medio Ambien e y Medio
549 Ru al y Ma ino, Mad id.
550 Galla do, A., Co elo, F., Mo illas, L., Delgado, M., 2009. Ciclos de nu ien es y p ocesos
551 edá icos en los ecosis emas e es es: especi icidades del caso medi e áneo y sus
552 implicaciones pa a las elaciones suelo-plan a. Re is a Ecosis emas 18.
553 h ps://doi.o g/10.7818/ e.2014.18-2.00
554 Ga cía-Angulo, D., He eş, A.-M., Fe nández-López, M., Flo es, O., Sanz, M.J., Rey, A.,
555 Vallada es, F., Cu iel Yus e, J., 2020. Holm oak decline and mo ali y exace ba es
556 d ough e ec s on soil biogeochemical cycling and soil mic obial communi ies ac oss
557 a climac ic g adien . Soil Biology and Biochemis y 107921.
558 h ps://doi.o g/10.1016/j.soilbio.2020.107921
559 Gessle , A., Schaub, M., McDowell, N.G., 2016. The ole o nu ien s in d ough -induced
560 ee mo ali y and eco e y. New Phy ol n/a-n/a. h ps://doi.o g/10.1111/nph.14340
561 Högbe g, P., No dg en, A., Buchmann, N., Taylo , A.F.S., Ekblad, A., Högbe g, M.N.,
562 Nybe g, G., O osson-Lö enius, M., Read, D.J., 2001. La ge-scale o es gi dling
563 shows ha cu en pho osyn hesis d i es soil espi a ion. Na u e 411, 789–792.
564 h ps://doi.o g/10.1038/35081058
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
24
565 Hong, P., Schmid, B., De Laende , F., Eisenhaue , N., Zhang, X., Chen, H., C a en, D., De
566 Boeck, H.J., Hau ie , Y., Pe chey, O.L., Reich, P.B., S eudel, B., S iebel, M., Thaku ,
567 M.P., Wang, S., 2022. Biodi e si y p omo es ecosys em unc ioning despi e
568 en i onmen al change. Ecology Le e s 25, 555–569.
569 h ps://doi.o g/10.1111/ele.13936
570 IPCC, 2021. Clima e Change 2021: The Physical Science Basis. Con ibu ion o Wo king
571 G oup I o he Six h Assessmen Repo o he In e go e nmen al Panel on Clima e
572 Change, Masson-Delmo e, V., P. Zhai, A. Pi ani, S. L. Conno s, C. Péan, S. Be ge ,
573 N. Caud, Y. Chen, L. Gold a b, M. I. Gomis, M. Huang, K. Lei zell, E. Lonnoy, J. B.
574 R. Ma hews, T. K. Maycock, T. Wa e ield, O. Yelekçi, R. Yu and B. Zhou (eds.).
575 ed.
576 Ja is, P., Rey, A., Pe sikos, C., Winga e, L., Raymen , M., Pe ei a, J., Banza, J., Da id, J.,
577 Miglie a, F., Bo ghe i, M., Manca, G., Valen ini, R., 2007. D ying and we ing o
578 Medi e anean soils s imula es decomposi ion and ca bon dioxide emission: he
579 “Bi ch e ec .” T ee Physiol. 27, 929–940.
580 Liu, Y., Liu, S., Wan, S., Wang, J., Luan, J., Wang, H., 2016. Di e en ial esponses o soil
581 espi a ion o soil wa ming and expe imen al h ough all educ ion in a ansi ional
582 oak o es in cen al China. Ag icul u al and Fo es Me eo ology 226–227, 186–198.
583 h ps://doi.o g/10.1016/j.ag o me .2016.06.003
584 Llo e , F., Sisca , D., Dalmases, C., 2004. Canopy eco e y a e d ough dieback in holm-
585 oak Medi e anean o es s o Ca alonia (NE Spain). Global Change Biology 10,
586 2092–2099. h ps://doi.o g/10.1111/j.1365-2486.2004.00870.x
587 Lozano-Pa a, J., Pulido, M., Lozano-Fondón, C., Schnabel, S., 2018. How do Soil Mois u e
588 and Vege a ion Co e s In luence Soil Tempe a u e in D ylands o Medi e anean
589 Regions? Wa e 10, 1747. h ps://doi.o g/10.3390/w10121747
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
25
590 Ma eucci, M., G uening, C., Goded Balla in, I., Seu e , G., Cesca i, A., 2015. Componen s,
591 d i e s and empo al dynamics o ecosys em espi a ion in a Medi e anean pine
592 o es . Soil Biology and Biochemis y 88, 224–235.
593 h ps://doi.o g/10.1016/j.soilbio.2015.05.017
594 Moo e, D.J.P., T ahan, N.A., Wilkes, P., Quai e, T., S ephens, B.B., Elde , K., Desai, A.R.,
595 Neg on, J., Monson, R.K., 2013. Pe sis en educed ecosys em espi a ion a e insec
596 dis u bance in high ele a ion o es s. Ecology Le e s 16, 731–737.
597 h ps://doi.o g/10.1111/ele.12097
598 Na e, L.E., Gough, C.M., Mau e , K.D., Boh e , G., Ha diman, B.S., Le Moine, J., Munoz,
599 A.B., Nadelho e , K.J., Spa ks, J.P., S ahm, B.D., Vogel, C.S., Cu is, P.S., 2011.
600 Dis u bance and he esilience o coupled ca bon and ni ogen cycling in a no h
601 empe a e o es . J. Geophys. Res. 116, G04016.
602 h ps://doi.o g/10.1029/2011JG001758
603 Pé ié, C., Ouime , R., 2008. O ganic ca bon, o ganic ma e and bulk densi y ela ionships in
604 bo eal o es soils. Can. J. Soil. Sci. 88, 315–325. h ps://doi.o g/10.4141/CJSS06008
605 Pinhei o, J., Ba es, D., DebRoy, S., Sa ka , D., R Co e Team, 2020. nlme: Linea and
606 Nonlinea Mixed E ec s Models. R package e sion 3.1-148.
607 R Co e Team, 2017. R: A language and en i onmen o s a is ical compu ing. R Founda ion
608 o S a is ical Compu ing,. Vienna, Aus ia.
609 Raich, J.W., Schlesinge , W.H., 1992. The global ca bon dioxide lux in soil espi a ion and
610 i s ela ionship o ege a ion and clima e. Tellus B 44, 81–99.
611 h ps://doi.o g/10.1034/j.1600-0889.1992. 01-1-00001.x
612 Rangel, T.F., Diniz-Filho, J.A.F., Bini, L.M., 2010. SAM: a comp ehensi e applica ion o
613 Spa ial Analysis in Mac oecology. Ecog aphy 33, 46–50.
614 h ps://doi.o g/10.1111/j.1600-0587.2009.06299.x
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
32
725
726 Figu e 1
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
33
727
728 Figu e 2
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
34
729
730 Figu e 3
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
35
731 Supplemen a y ma e ial
732 Table S1. Plan s uc u e and soil physicochemical a iables o he di e en ee decline
733 s ages (n = 10 in all cases). Values ep esen he mean ± 1SE. Di e en le e s nex o alues
734 wi hin each a iable ep esen signi ican di e ences among de olia ion deg ees (P < 0.05; one-
735 way ANOVA). Canopy = canopy diame e ; DBH = diame e a b eas heigh ; CI = compe i ion
736 index; TC, o al C; TN, o al N; CMA, o al ca bon in he magne ic ac ion (conside ed as labile
737 C). Plan a iables, TC, TN, C:N a io and pH a e da a om Rod íguez e al. (2017). Bulk
738 densi y and CMA a e modi ied da a om Rod íguez e al. (2020).
Heal hy
A ec ed
Dead
Plan a iables
Heigh (m)
4.54 ± 0.20
3.81 ± 0.15
3.82 ± 0.32
Canopy (m)
5.97 ± 0.31a
4.46 ± 0.30b
4.46 ± 0.37b
DBH (cm)
45.8 ± 3.86
33.8 ± 5.17
33.8 ± 5.17
CI (x 10-3)
0.8 ± 0.4a
5.5 ± 2.7ab
8.7 ± 3.3b
Soil physicochemical a iables
TC (%)
2.78 ± 0.44
2.27 ± 0.21
2.77 ± 0.31
TN (%)
0.22 ± 0.03
0.17 ± 0.02
0.21 ± 0.02
C:N a io
12.1 ± 0.38
13.4 ± 0.43
13.1 ± 0.51
CMA (%)
4.86 ± 1.07
2.79 ± 0.40
2.97 ± 0.53
pH (uni less)
6.58 ± 0.11
6.30 ± 0.10
6.56 ± 0.10
Compac ion (uni less)
54 ± 15
111 ± 26
114 ± 31
Bulk densi y (g cm-3)
1.03 ± 0.08
1.09 ± 0.04
1.06 ± 0.06
739
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
36
740 Table S2. Seasonal means o p ecipi a ion (P), ai empe a u e (Tai ), soil wa e con en (SWC),
741 soil empe a u e (Tsoil) and soil espi a ion (Rs) o he wo s udied yea s. Values ep esen he
742 mean ± 1SE. Bold P alues ep esen s a is ically signi ican e ec s o yea (Y), season (S), as
743 well as signi ican in e ac ions o bo h ac o s. Values wi h di e en le e s wi hin each yea
744 ep esen signi ican di e ences among seasons (P < 0.05).
Yea
Season
P (mm)*
Tai (°C)*
n
SWC (%)
Tsoil (°C)
Rs
yea 1
Summe
6.45
25.0
30
0.9 ± 0.1c
30.3 ± 0.6a
2.05 ± 0.24c
Fall
67.3
16.0
30
3.4 ± 0.3b
16.7 ± 0.2c
2.02 ± 0.11c
Win e
147.7
7.10
30
7.7 ± 0.4a
13.5 ± 0.3d
2.41 ± 0.15b
Sp ing
67.3
15.3
30
4.0 ± 0.2b
19.3 ± 0.3b
3.03 ± 0.16a
yea 2
Summe
21.2
24.3
29
0.9 ± 0.1c
29.1 ± 0.4a
1.87 ± 0.21c
Fall
186.8
17.0
29
9.6 ± 0.4a
15.6 ± 0.1c
3.40 ± 0.17a
Win e
66.8
6.6
29
11.1 ± 0.7a
7.2 ± 0.2d
1.40 ± 0.10d
Sp ing
68.5
15.3
29
4.7 ± 0.3b
17.1 ± 0.2b
2.86 ± 0.18b
Yea
χ2
77.8
186.2
8.07
P
< 0.001
< 0.001
< 0.01
Season
χ2
1527.3
7142.6
218.2
GLS models
P
< 0.001
< 0.001
< 0.001
Y x S
χ2
151.7
568.6
251.5
P
< 0.001
< 0.001
< 0.001
745 *Ge a e (40°17’N 3°43’O) and Cua o Vien os (40°22’N 3°47’O) s a ions (AEMET)
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
37
746 Table S3. Bes - i ing eg ession models and p edic o s o he wo-yea a e age alues o soil
747 espi a ion o each ee decline s age sepa a ely (heal hy, a ec ed and dead) and all oge he
748 (All). The bes h ee models (M1-M3), anked acco ding o hei AICc alue, and he h ee
749 mos impo an p edic o s (P ed.), anked acco ding o hei ela i e impo ance (Impo ) a e
750 p esen ed. Colou ed cells indica e a iables ha we e included in a pa icula model (one pe
751 ow). Canopy, ee canopy diame e ; SWC, soil wa e con en ; Lab C, labile C; TN, soil o al
752 ni ogen; BD, bulk densi y.
Bes models
Bes p edic o s
Canopy
SWC
Tsoil
Lab C
TN
BD
R2
AICc
P ed.
Impo .
Heal hy
M1
0.49
29.4
TN
0.65
M2
0.34
31.9
BD
0.20
M3
0.25
33.1
Lab C
0.16
A ec ed
M1
0.25
6.80
Lab C
0.39
M2
0.22
7.16
BD
0.32
M3
0.09
8.72
Tsoil
0.15
Dead
M1
0.93
19.7
Tsoil
0.98
M2
0.77
23.1
Lab C
0.95
M3
0.38
26.8
SWC
0.78
All
M1
0.42
60.9
Canopy
0.96
M2
0.45
62.5
TN
0.82
M3
0.44
62.6
Tsoil
0.30
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
38
753
754 Figu e S1. Hea map ep esen a ion o Spea man’s ank co ela ion ma ix among he wo-yea
755 a e age alues o soil espi a ion (Rs), plan (Heigh , Canopy, DBH, CI), and soil mic oclima ic
756 (SWC and Tsoil) and physicochemical (TC, TN, C:N a io, Labile C, pH, Compac ion, Bulk
757 densi y) a iables o all decline s ages oge he . Only signi ican alues a e shown (P < 0.05).
758 SWC, wo-yea s a e age alues o soil olume ic wa e con en ; Tsoil, wo-yea s a e age
759 alues o soil empe a u e; TC, o al C; TN, o al N.
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
39
760
761 Figu e S2. Rela ionship be ween soil empe a u e and soil espi a ion o each ee decline
762 s age using a e aged alues o each empe a u e deg ee. Ve ical lines show he empe a u e
763 h eshold o each decline s age (i.e., 17C o heal hy and dead and 18C o a ec ed).
This p ep in esea ch pape has no been pee e iewed. Elec onic copy a ailable a : h ps://ss n.com/abs ac =4259956
P ep in no pee e iewed
