Phenological shifts in climatic response of secondary growth allow Juniperus sabina L. to cope with altitudinal and temporal climate variability

Agricultural and Forest Meteorology 217 (2016) 35–45

Contents lists available at ScienceDirect

Agricultural and Forest Meteorology journal homepage: www.elsevier.com/locate/agrformet

Phenological shifts in climatic response of secondary growth allow Juniperus sabina L. to cope with altitudinal and temporal climate variability Alberto Arzac a , Ana I. García-Cervigón b , Sergio M. Vicente-Serrano c , Javier Loidi a , José M. Olano b,∗ a

Departamento de Biología Vegetal y Ecología, Facultad de Ciencia y Tecnología, Universidad del País Vasco, Barrio Sarriena s/n, Leioa E-48940, Spain Área de Botánica, Departamento de Ciencias Agroforestales, EU de Ingenierías Agrarias, Universidad de Valladolid, Campus Duques de Soria, Soria E-42004, Spain c Instituto Pirenaico de Ecología, Consejo Superior de Investigaciones Científicas, Avda Monta˜ nana 1005, Zaragoza E-50059, Spain b

a r t i c l e

i n f o

Article history: Received 8 September 2015 Received in revised form 9 November 2015 Accepted 14 November 2015 Available online 6 December 2015 Keywords: Altitudinal gradient Global warming Juniperus sabina Mediterranean high mountains Shrubs Tree rings

a b s t r a c t The long-term persistence of Mediterranean high mountain flora is challenged by global warming. Understanding plant responses to different climatic drivers along altitudinal gradients is necessary in order to predict the response of Mediterranean mountain systems to the ongoing increase in drought intensity and severity. We selected the prostrate shrub Juniperus sabina L. as a model species and explored its growth response to climatic variability over the last six decades through dendrochronological methods at four sites along a 750 m altitudinal gradient. Secondary growth was maximal at lower altitudes, where growing seasons are longer. Water availability was the main factor controlling secondary growth variability along the whole gradient, although the timing and strength of climatic variables affecting growth shifted with altitude. Earlier and stronger signals were detected in lower sites, where a combination with late summer signals suggested the existence of a second growth pulse in response to longer growing seasons. J. sabina adjusted its secondary growth to the changing climatic conditions by shifting the timing of its climatic response to favorable climatic windows, with earlier responses to spring rainfall in lower sites and an expanded growing season at the highest site during the second half of the study period. Our results show that at least some Mediterranean plants can grow faster at the drier edge of their range even under intense climate pressure, indicating that the response to drought stress may be highly idiosyncratic. The plastic nature of J. sabina secondary growth combined with its role as a nurse plant in Mediterranean mountains may be key to maintaining the high diversity levels of these particular ecosystems. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Mediterranean mountains are a global hot spot of plant biodiversity (Médail and Quézel, 1997). Isolation and a singular climatic combination of low winter temperatures and summer drought stress have shaped a highly endemic flora (Médail and Diadema, 2009) whose persistence is threatened by current global climate change (Pauli et al., 2012). Projections of above average warming for the Western Mediterranean (Giorgi and Lionello, 2008) pose a serious potential threat for plant species persistence, especially since the low altitude of mountains constrains the possibilities for altitudinal ascent as a way to keep within the same climatic

∗ Corresponding author. Tel.: +34 975129485. E-mail address: [email protected] (J.M. Olano). http://dx.doi.org/10.1016/j.agrformet.2015.11.011 0168-1923/© 2015 Elsevier B.V. All rights reserved.

˜ envelopes (Jump and Penuelas, 2005; Walther et al., 2005). One the one hand, most of these mountains are relatively low (usually below 2500 m), providing little space for altitudinal displacement. Furthermore, both niche models and different empirical evidences suggest that under the current climate change scenario the altitudinal ascent that is occurring in other European mountains may not suffice to provide adequate conditions for Mediterranean high mountain plants (Olano et al., 2013; Pauli et al., 2012; Thuiller et al., 2005). Understanding plant responses to different climatic drivers along altitudinal gradients, and the mechanisms underlying these responses, is thus particularly relevant to predict the response of Mediterranean mountain systems in this context of global biodiversity loss. Climatic factors constraining plant growth in temperate and boreal environments shift along altitudinal gradients, moving from water shortage at lower altitudes to temperature limitation

36

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

at higher altitudes (Babst et al., 2013). However, under certain environments, precipitation may constrain growth all along the altitudinal gradient (Yang et al., 2013), even at the timberline (Galván et al., 2015; Liang et al., 2014). This is the case in Mediterranean mountains (Babst et al., 2013), at least in drier locations (Linares and Tíscar, 2010). Actually, water availability has been shown to be the most critical climatic factor controlling Mediterranean plant performance (Lebourgeois et al., 2012; Martín-Benito et al., 2010; Pasho et al., 2011a). The fact that water availability determines plant performance along the whole altitudinal gradient (Olano et al., 2013) may be essential to our understanding the differential behavior of Mediterranean mountain flora to climatic change. Water shortage affects secondary growth at multiple scales, impacting both xylogenesis and resource acquisition (GarcíaCervigón et al., 2015; Olano et al., 2013). Drought decreases the cambial division rate, reduces cell expansion and constrains secondary wall formation, which eventually may arrest secondary growth (Camarero et al., 2010; Olano et al., 2014). The combination of water shortages with rising temperatures may exacerbate the impact on plant water budgets, by increasing evapotranspirative demand (Christensen et al., 2007; Vicente-Serrano et al., 2014a). At the same time, this combination may trigger more frequent and severe drought episodes in the Mediterranean (Vicente-Serrano et al., 2014b), which could ultimately reduce growth rates and increase mortality events (Camarero et al., 2015). In addition, decreasing temperatures along altitudinal gradients lead to lower growth rates due to shorter optimal periods for cambial activity (Griˇcar et al., 2014; King et al., 2013; Prislan et al., 2013), limiting the ability of plants to invest reserves in secondary growth with increasing altitude. In alpine environments, secondary growth shows higher sensitivity to temperature than photosynthesis (Körner, 2003), and limitations in the ability to invest available photosynthates in secondary growth may lead to an excess in carbon reserves (Hoch et al., 2002). However, low water availability may negatively impact photosynthetic capacity, decreasing the carbon pools available for growth (Galiano et al., 2011). As a consequence, the impact of climatic variables on carbon reserve levels and their relation with plant growth in Mediterranean mountains is still unclear. Annual growth rings in perennial plants have been used to interpret the biological response to climate (Speer, 2010). However, most of the literature on this topic has focused on tree species (Dittmar and Elling, 1999; Fonti et al., 2007; Fritts et al., 1965), with work on forbs and shrubs being more scarce (Gazol and Camarero, 2012; Olano et al., 2013; Palombo et al., 2014; von Arx et al., 2006). This knowledge gap may be particularly relevant in Mediterranean high mountains where shrubs are a critical functional component (Ojeda et al., 2000). In these systems, the large contribution of shrubs in terms of diversity and biomass is amplified by their role as nurse plants (Castro et al., 2002; Gómez-Aparicio et al., 2004; Michalet et al., 2014). Shrubs facilitate the establishment and performance of a wide array of mountain plants (García-Cervigón et al., 2013; Gómez-Aparicio et al., 2008), increasing their potential range and thereby promoting the maintenance of high biodiversity levels (Cavieres et al., 2014; Schöb et al., 2013). In addition, shrubs also preserve soil fertility by means of the reduction of erosion rates (García-Ruiz and Lana-Renault, 2011; García-Ruiz et al., 1996). In spite of this engineering role of high mountain shrubs, very little is known about their response to climatic factors along altitudinal gradients in terms of growth and carbon pools (García-Cervigón et al., 2012). To gain knowledge of the response of Mediterranean mountain systems to global warming, we selected the prostrate shrub Juniperus sabina L. as a model species. J. sabina is a keystone species in Mediterranean mountains, creating islands of fertility with higher

nutrient and humidity levels (García-Cervigón et al., 2012, 2015; Verdú and García-Fayos, 2003) and enhancing the establishment and fitness of protégée plants (García-Cervigón et al., unpublished results). We analyzed ring width and carbohydrate pools along a 750 m gradient comprising the whole altitudinal range of the species to test the following hypotheses: (a) secondary growth would be maximal at intermediate locations, since growth would be constrained by drought at the lower altitudinal limit and by drought and low temperatures at the upper altitudinal limit; (b) temperature limitation of cambial activity at the upper altitudinal limit would lead to higher carbohydrate contents due to sink limitation; (c) the effect of drought would be stronger at lower sites, where the combination with higher temperatures would increase the evapotranspirative demand; (d) the timing of climatic factors controlling secondary growth will shift with altitude, with earlier signals in lower sites; and (e) rising temperatures during the 20th century will have led to enhanced and earlier drought signals. 2. Materials and methods 2.1. Study species J. sabina (Savin juniper) is a prostrate shrub inhabiting mountains from Western Europe to Eastern Asia (Adams et al., 2007; Wesche and Ronnenberg, 2004). It is usually less than 1 m tall, but with a large lateral extent, with individual plants covering surfaces as large as 1900 m2 (García-Cervigón et al., 2012). It prefers calcareous substrates and has a wide altitudinal range, from 1000 to 2750 m asl, with an optimum distribution in the oromediterranen belt where it often coexists with other conifers (Mateo Sanz, 1997). 2.2. Study area The study area was located over an altitudinal gradient of 750 m in the Javalambre mountain range (Teruel), Eastern Spain. Four sampling sites were selected along an elevation gradient, comprising the whole juniper altitudinal range in this mountain (S1: 1250 m, S2: 1400 m, S3: 1700 m and S4: 2000 m, see Fig. 1). In the lowest locality, Savin juniper co-occurs with other juniper species (Juniperus thurifera L., J. phoenicea L., J. oxycedrus L.). In the second locality (S2) Savin juniper occurs in open grasslands with scattered Pinus nigra Arnold. and Quercus faginea Lam. individuals. The habitat is similar in S3, but with the added presence of the more mesic Pinus sylvestris L. The upper site (S4) is above the tree-line and the landscape is dominated by Savin juniper (García-Cervigón et al., 2012). Parent material is predominantly limestone, leading to calcareous soils. The climate is Mediterranean continental with a marked summer drought, although water deficit intensity decreases with altitude. Maximum temperatures combined with minimum rainfall occur between July and August and are associated with a pronounced dry period that can start as early as May at the lowest altitudes (Fig. 2A). The frost period can be extended from early October to early June at higher altitudes, whereas at lower altitudes frost duration is shorter and may range from the end of October to May (Mateo Sanz et al., 2013). Climatic data for total monthly precipitation were obtained from the neighboring weather station of Torrijas (1400 m asl), located 15 km from the study site. Monthly maximum and minimum temperatures were available from the station of Segorbe (364 m asl), located 25 km from the study site and corrected for altitude. Climate data was quality controlled and homogenised to remove any artificial non-climate noise in the series (see details in El Kenawy et al., 2013 and Vicente-Serrano et al., 2010a). Climate analysis revealed that mean temperature has increased during the period 1953–2009, in particular during

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

37

Fig. 1. Location of Javalambre mountain range (Teruel), Eastern Spain and images of the four sampling sites (S1, 1250 m asl; S2, 1400 m asl; S3, 1700 m asl; S4, 2000 m asl).

the growing season (May to July) (Fig. 2B), but there have been no observed changes in precipitation patterns (Fig. 2C).

2.3. Field and laboratory work Sampling was carried out in 2010. 35–45 juniper individuals were selected in each site and a wood disk was sawed from a central branch. In the lab, wood disks were polished using sand papers of increasing grit until the cellular structure could be seen. Afterwards, samples were visually cross-dated and ring widths were measured using a Velmex sliding stage micrometer interfaced with a computer. Cross-dating quality was checked using the COFECHA software (Grissino-Mayer, 2001). For standardization, raw series were fitted to a spline function with a 50% frequency response of 32 years, which was flexible enough to reduce the non-climatic variance by preserving high-frequency climatic information (Cook and Peters, 1981). The residuals obtained by dividing each raw ring width by its fitted spline values were prewhitened by autoregressive modeling, giving dimensionless indices that represent independent records of annual growth for each measured series. Growth indices were averaged on an annual basis into a site chronology using a biweight robust mean. Ring-width series were detrended, standardized and averaged with the ARSTAN computer program (Cook and Holmes, 1996). First order autocorrelation (AC1), mean sensitivity (msx ), mean inter-tree correlation (rbt ), signal-to-noise ratio (SNR), and expressed population signal (EPS) were calculated to assess chronology properties.

In order to assess non structural carbohydrate contents (NSC), an additional sampling took place in February 2014, prior to the growing season. Apical branches from 60 individuals per locality were collected and bagged in a cool box immediately after sampling. To stop enzymatic activity and avoid carbohydrate degradation, branches were first microwaved at 600 W for at least 90 s the day after their collection (Popp et al., 1996), and subsequently stored at −20 ◦ C. Before the extraction of NSCs, bark was removed and samples were oven-dried at 60 ◦ C for 72 h. Dry samples were finely ground with a mixer mill (Retsch MM 400). NSCs were measured using the anthrone method, extracting soluble sugars in a first step, and then starch after its chemical conversion to glucose (Olano et al., 2006). Sugars and starch concentrations were expressed as percentage of dry matter (% DM). 2.4. Statistical analyses To test differences in secondary growth between localities, raw ring-width (RW) series were compared along the gradient by means of a generalized additive mixed model (GAMM). Altitude (site) and age were considered as fixed factors, modeling the relationship between secondary growth and age with a spline curve, and tree identity was included as the random component. The model was allowed to hold different variances per group (i.e. per site). Differences in stem NSC levels were evaluated with a one-way ANOVA including site as a factor. We used Pearson’s correlation to identify the relationships between residual ring-width chronologies and climatic factors for

38

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

Fig. 2. Climatic data and trends for the study area for the period 1953–2009. (A) Mean monthly precipitation and reference evapotranspiration (ETo). Black line: average ETo, gray line: average precipitation, light gray surface: water surplus, dark gray surface: water deficit. Vertical lines represent one standard deviation for average precipitation and ETo. (B) Annual and main growing season (May–July) mean temperature trends for the study period. (C) Precipitation trends for the study period.

the common period (1953–2010) for each site. Monthly climatic data (mean temperature and total precipitation) were selected for a time window spanning from May of the previous year to November of the current year, in order to consider current and previous growing seasons. Multiple regressions were performed to determine significant climatic parameters at each site. Variation in climatic response along the studied period was assessed by using moving correlations of 25 years from 1953 to 2010 for April to August rainfall. To evaluate the different effects of hydric conditions at different time-scales on residual ring-width chronologies over the gradient, we employed the Standardized Precipitation–Evaporation index (SPEI). The SPEI is a multiscalar drought index, calculated including the influence of precipitation and atmospheric evaporative demand on drought severity (Vicente-Serrano et al., 2010b). The advantages of SPEI over other drought indices are that the effect of evaporative demand is considered in its calculation and that it can be calculated for different time scales, which is essential in order to record

the strong diversity in the response of vegetation types to drought (Vicente-Serrano et al., 2013). We calculated correlations between residual ring width chronologies and SPEI index over different time scales (1–18 months). Reference evapotranspiration used to calculate the SPEI was obtained by means of the Hargreaves equation (Hargreaves and Samani, 1985) using maximum and minimum temperature and the extraterrestrial solar radiation. 3. Results 3.1. Variation of secondary growth and NSC content along the gradient A total of 170 juniper disks were analyzed and 36,153 annual rings measured. Ages of the sampled disks ranged between 86 and 223 years (134 ± 31 years; mean ± SD) and mean ring-width was 0.26 mm ± 0.03 mm. Ring-width chronologies (Table 1) showed values of mean sensitivity (msx ) between 0.23 and 0.27, mean

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

39

Table 1 Summary statistics calculated for the period 1953–2009 for Juniperus sabina ring-width chronologies. S1, 1250 m asl; S2, 1400 m asl; S3, 1700 m asl; S4, 2000 m asl. Parameter/site

S1

S2

S3

S4

Altitude (m asl) No. Trees (radii) ChLeng RW (␮m) ± SD msx rbt EPS AC1

1250 33 (58) 1821–2011 297 ± 2a 0.248 0.366 0.952 0.222

1400 44 (89) 1843–2011 263 ± 2a 0.240 0.339 0.959 0.168

1700 33 (48) 1830–2011 219 ± 2b 0.270 0.327 0.938 0.201

2000 33 (73) 1720–2010 227 ± 1b 0.230 0.422 0.975 0.223

ChLeng, chronology length; RW, ring-width; msx , mean sensitivity; rbt , mean correlation between plants; EPS, expressed population signal; AC1, first order autocorrelation. Superscripts indicate significant differences between groups at P < 0.05. Different letters for RW values indicate significant differences at P.

Table 2 Estimated parameters for fixed effects of the generalized additive mixed model for secondary growth. S1 (1250 m asl) is the baseline model. Value indicates the coefficient for the additive effect at other altitudes (S2—1400 m asl, S3—1700 m asl and S4—2000 m asl). s(age) models the curvilinear relationship between secondary growth and age.

Intercept S2 S3 S4 s(age)

Value

SE

DF

t

P

308.963 −16.194 −83.066 −66.902 −21.279

15.502 20.485 22.298 21.335 20.543

35982 166 166 166 35982

19.931 −0.791 −3.725 −3.136 −1.036

<0.001 0.430 <0.001 0.002 0.300

SE, standard error; DF, degrees of freedom.

Table 3 Pairwise Pearson’s correlations between residual chronologies of ring-width and sites. S1, 1250 m asl; S2, 1400 m asl; S3, 1700 m asl; S4, 2000 m asl. S2 S1 S2 S3

0.838

S3 0.650 0.762

S4 0.143 0.345 0.632

Table 4 Results of multiple forward stepwise regressions between the four Juniperus sabina residual chronologies and monthly climatic parameters for average temperature (T) and accumulated precipitation (P) comprising the 1953–2009 period. Time-window for climatic response ranges from May of the previous year (uppercase letters) to September of the target year (lowercase letters). Standardized beta coefficients are shown. Coefficients in bold type are significant at P < 0.01. Only months included in at least one of the models are shown. S1, 1250 m asl; S2, 1400 m asl; S3, 1700 m asl; S4, 2000 m asl. Site

S1

S2

S3

R2 adj. F P May P Jun P Jul P Aug P Oct T Nov P Feb T May P Jun P Aug P Sep P

0.486 14.025 <0.001 −0.328

0.483 11.264 <0.001

0.338 10.374 <0.001

S4 0.144 5.615 0.006

−0.256 −0.229 −0.321 −0.295 0.321 −0.348 0.508

0.391 0.209 0.284

0.475

0.330

0.220

Coefficients in bold are significant at P < 0.001 and underlined at P < 0.01.

correlation between individuals (rbt ) between 0.33 and 0.42, and expressed population signal (EPS) over 0.90 in all cases. These values suggest adequate replication and a strong common signal shared by individuals in each locality. The generalized additive mixed model indicated that junipers at the lower sites (S1 and S2) had significantly higher growth rates than at the higher sites (S3 and S4, Tables 1 and 2). Age did not affect secondary growth rate and adjusted significantly to the spline curve (P < 0.001). Standard deviation of secondary growth data decreased with altitude, in a proportion of 1: 0.94: 0.75: 0.80 for sites S1, S2, S3 and S4, respectively. The disparity of behaviors in cambial activity along the altitudinal gradient was reflected by progressive reduction in synchrony between residual chronologies at different altitudes. In fact, low and high juniper populations showed uncorrelated growth patterns, in spite of being located just 11 km apart (Table 3). NSC content ranged from 5.17 ± 0.67% at S4 (2000 m) to 6.26 ± 0.81% at S2 (1400). Values for S1 (5.23 ± 0.68%) and for S3 (5.20 ± 0.67%) were similar to the highest site. The effect of site on NSC content was significant (F = 7.671; P < 0.001), albeit posthoc tests revealed that only S2 (1400 m) was significantly different from the other localities. 3.2. Variation of the intensity and timing of climatic factors controlling secondary growth Pearson’s correlations and multiple regressions revealed strong links between climatic parameters and residual chronologies of secondary growth. The main factor controlling growth along the entire length of the gradient was precipitation, although

temperature of some months also affected growth in sites S2 and S4. Climatic control was maximal at the low sites (S1 and S2) and decreased at the upper sites (Table 4; Fig. 3). The timing of climatic signals also shifted with altitude. At the lowest site (S1) growth was favored by rainfall from the end of the previous growing season to the end of the current growing season (previous year November rainfall, late winter rainfall—February, late spring rainfall – May and June – and current year October rainfall). Interestingly, rainfall in May and June exerted a negative effect on growth during the following growing season. Multiple regression model for S1 (Table 4) reduced the number of significant factors, including positive effects of current year May and September rainfall, previous year November rainfall and a negative effect of the previous May’s rainfall. Climatic response at S2 was very similar in intensity, although it showed some small shifts in the timing of climatic signals. Pearson’s correlation indicated a positive effect of May, June and August rainfall, a negative effect of previous year May, June and October rainfall and of previous year June and October temperatures. Multiple regression model for S2 included positive effects from May and June rainfall, whereas positive late summer September rainfall signal from S1 shifted to August, and previous year rainfall signals moved from May to June (negative) and from November to October (positive). At S3, secondary growth was positively affected by current year May, June and August rainfall and negatively affected by previous May and late winter (February) temperature. Multiple regression model for S3 showed a positive effect of current June rainfall, a negative effect of previous July rainfall and a negative effect of current year February temperature.

40

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45 0.6

* *

S1 S2 S3

S4

0.4

*

Correlation coefficient

*

*

Precipitation

*

0.2

0.0

-0.2

-0.4

*

*

AY JUN JUL UG SEP CT OV EC D O M A N 0.6

r b fe m a Month

n ja

r y ap m a

n ju

l ju aug sep

t oc

Temperature

Correlation coefficient

0.4

0.2

0.0

-0.2

*

*

-0.4

AY JUN JUL UG SEP CT OV EC D O M A N

n ja

r b fe m a

r y ap m a

n ju

jul aug sep

t oc

Month Fig. 3. Correlations (Pearson’s coefficients) calculated between tree ring width and monthly climate variables (mean temperature and total precipitation) for precipitation and mean temperature for the period 1953–2009 in the four study sites. The correlations were calculated from May of the previous year (uppercase letters) to November of the current year (lowercase letters) of tree ring formation. Dashed lines indicated P < 0.05 and dotted lines indicate P < 0.01. Asterisks represents months included in multiple regression models. S1, 1250 m asl; S2, 1400 m asl; S3, 1700 m asl; S4, 2000 m asl.

Ring-width at the highest site (S4) was positively correlated with current year June and August rainfall, negatively with previous year August rainfall and also a negative effect of previous year May temperature was observed. Multiple regression model for S4 showed a positive effect of current June rainfall and a negative effect of previous August rainfall. 3.3. Altitudinal variation of the effect of drought on secondary growth Correlations between residual ring-width series and SPEI showed sharp differences in the timing, intensity and length of the effect of drought (Fig. 4). The lowest altitude population showed a strong (r = 0.65), long-term (12 mo.) response to SPEI (November to October), reflecting a persistent water deficit. Intensity was similar (r = 0.65) at S2 (1400 m), but time lag decreased to just 5 months during the current year growth season (April to August). Site 3 (1700 m) showed a decrease in intensity (r = 0.47), and a one month

lag based on current year June SPEI. High altitude juniper population growth (S4 = 2000 m) showed an even weaker level of control by drought (r = 0.34), but a longer temporal lag (June to August) restricted to the period where wood formation occurs. Interestingly, S4 also showed a negative response in April of the growth year, with two maxima at lags of 2 and 11 months. Previous year SPEI in spring and summer also exerted a significant negative effect at all four study sites, with negative June effects with a lag of 2 months for S1 and one month for S2 and S3, this signal moving to August with a 3-month lag in S4. 3.4. Effect of climatic trends during the 20th century on secondary growth along the gradient Temperature increase was strong in our study area during the study period, with annual mean temperature augmenting at a pace of 0.25 ◦ C per decade between 1953 and 2010, and these rates being doubled during the growing season, from May to July. At the same

1250 m

nov oct sep aug jul jun may apr mar feb jan DEC NOV OCT SEP AUG JUL JUN MAY

0.7 0.6 0.5 0.4

1400 m

nov oct sep aug jul jun may apr mar feb jan DEC NOV OCT SEP AUG JUL JUN MAY

0.3 0.2 0.1 0.0 -0.1 -0.2 -0.3 -0.4

2000 m 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

1700 m

Pearson r

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Months

nov oct sep aug jul jun may apr mar feb jan DEC NOV OCT SEP AUG JUL JUN MAY

41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

nov oct sep aug jul jun may apr mar feb jan DEC NOV OCT SEP AUG JUL JUN MAY

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Months

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

Time scale (months)

Time scale (months)

Fig. 4. Mean correlation coefficients between ring-width chronologies and monthly Standardized Precipitation–Evaporation index (SPEI) series at different time scales for the period 1953–2009 in the four study sites. The correlations were calculated from May of the previous year (uppercase letters) to November of the current year (lowercase letters) of tree ring formation. S1, 1250 m asl; S2, 1400 m asl; S3, 1700 m asl; S4, 2000 m asl.

time, rainfall remained constant or slightly declined, altogether resulting in enhanced drought intensity. Moving correlations indicated shifts in the timing and intensity of the climatic response along the analyzed period (Fig. 5). Sensitivity to April rainfall increased in the lowest (S1) locality, becoming significant for the 1984–1994 period. Sensitivity to May rainfall remained highly significant for the lowest (S1) and lowintermediate locality (S2), whereas it showed a sharp increase in the high-intermediate locality (S3), becoming significant. The effect of June rainfall increased and became highly significant from S1 to S3, whereas it increased but was not significant at the highest locality (S4). July rainfall did not have a significant effect for any locality (data not shown), whereas August rainfall signal increased for S4, becoming significant since 1983. 4. Discussion Secondary growth variability along the whole altitudinal range was mainly controlled by water availability. The predominance of water control along the whole gradient supports the idea that growth in Mediterranean high mountains is drought constrained (García-Cervigón et al., 2012) and echoes the findings from other arid mountains in the world (He et al., 2013; Liang et al., 2014; Yang et al., 2013), but differs from the observed behavior in temperate and alpine environments (Dang et al., 2013; King et al., 2013). The timing and strength of climatic variables affecting growth shifted with altitude, with earlier and stronger signals in lower sites, and the intensity of the effect of summer variables related to water availability decreasing with altitude and increasing over

the study period. Despite the stronger climatic control at the lower end of the gradient and contrary to our expectations, secondary growth was maximum at lower altitudes and unrelated to lower NSC levels. These results mean that at least some Mediterranean plants can grow faster at their drier edge under intense climateforcing, indicating that the response to drought stress may be highly idiosyncratic (Camarero et al., 2015). Differences in growth response to climatic variables along the gradient might reflect the alleviation of drought stress with altitude as well as the existence of different phenological patterns of cambial activity determined by altitudinal shifts in temperature. Cambial activity initiation requires a minimal temperature threshold (Rossi et al., 2008) with the timing of the onset of secondary growth showing a gradual delay with altitude (Moser et al., 2010). The exact altitudinal delay depends on the mountain characteristics and on the plant species, but it has been estimated to be between 3 and 4 days for every 100 m of altitude (Moser et al., 2010; Wang et al., 2014). The expected difference in cambial onset for our altitudinal gradient (750 m) would therefore range between 19.5 and 26 days. This might explain our observation of a one month shift in the initiation of the ring-width response to spring precipitation (from May to June) between the low- to the high-altitude juniper populations, since an earlier cambial activation (Table 4; Fig. 3) would allow an earlier response of secondary growth to precipitation. Otherwise, in Mediterranean environments, a temporal arrest of secondary growth is a characteristic response to summer drought (De Luis et al., 2011; Olano et al., 2015; Rozas et al., 2011). Positive responses to summer (September/August) rainfall at low altitudes might reflect the reactivation of cambial activity in response to late

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45 1250 m 1400 m 1700 m 2000 m

-0.4

-0.4

-0.6

-0.6

1996

1994

1990

1992

1988

1986

1982

1984

1978

1980

1972

1968

1970

1980

1996

1992

1994

1990

1988

1986

1982

August rainfall

1984

-0.2

1976

1996

1994

1992

1990

1988

1986

1984

1982

-0.2

1980

0 1978

0 1974

0.2

1976

0.2

1972

0.4

1970

0.4

1966

0.6

1968

0.6

1978

0.8

June rainfall

1974

0.8

1976

-0.6

1974

-0.6

1972

-0.4

1970

-0.4

1968

-0.2

1964

1996

1994

1992

1988

1990

1986

1984

1980

1982

1978

1976

1972

1974

0 1970

0 1966

0.2

1968

0.2

1964

0.4

-0.2

May rainfall

0.6

0.4

1964

Correlaon coefficient

0.6

0.8

1966

April rainfall

1964

0.8

1966

42

Fig. 5. Temporal shifts of correlations between Juniperus sabina ring-width chronologies at different altitudes and April, May, June and August monthly precipitation. Moving correlations were calculated for 25-year intervals for the period 1953–2009, and year under each correlation corresponds to the middle of the interval. Solid horizontal lines indicate P < 0.05 and dashed lines indicate P < 0.01.

summer storms, producing a second pulse of growth that might extend enough to benefit from autumn rainfall. At higher altitudes, however, lower temperatures may preclude the existence of a secondary pulse of growth due to an earlier cessation of the growing season. Although sensitivity of cambial cessation to temperature along altitudinal gradients has not been fully supported (Moser et al., 2010; Wang et al., 2014), regional studies of plant-climate response do support a longer growing season (He et al., 2015) and a delay of cambial activity cessation associated with warmer temperatures (Camarero et al., 2010; De Luis et al., 2011). Finally, higher secondary growth at low altitudes did not fit with our expectations of increased performance in the central range of the species’ ecological distribution, where climatic conditions are less stressful (Olano et al., 2013; Matías and Jump, 2015). In contrast, our results agree with previous studies suggesting that growth rates are directly related to the growing season length (Coomes and Allen, 2007; Dang et al., 2013; Wang et al., 2014) and the temperaturecontrolled cambial division rate (Wang et al., 2015). The highest juniper population (2000 m) was negatively affected by precipitation during early spring. At this altitude, early spring precipitation frequently occurs as snow, thus this signal would suggest that a longer-lasting snowpack has a negative effect on growth. Presence of snow has been considered favorable for plant growth in drought-limited mountains (Olano et al., 2013) due to thermal insulation and to the positive effect of snow infiltration on soil water replenishment, thus delaying summer drought initiation. However, a prolonged snowpack, covering totally or partially prostrate plants, may potentially limit their photosynthetic activity and delay the onset of cambial activity, leading to narrower rings (Bär et al., 2007; Franklin, 2013). Climatic factors leading to higher investment in secondary growth during a given season have been previously associated with a reduction in following year growth in conifers (Andreu et al., 2007; García-Cervigón et al., 2012). In our case, rainfall during the

growing season enhanced growth but had a negative effect on following year growth, suggesting a trade-off between current and future year growth. A possible explanation is that climatic factors promoting cambial growth in one year might deplete NSC levels, leading to a reduction in growth potential for the next year. This hypothesis contrasts with previous works indicating that in alpine environments secondary growth is constrained by climatic limitations on cambial activity, leading to an excess of carbohydrates (Körner, 2003). In our study, however, NSC patterns thus did not meet the expectations of the cambial limitation hypothesis: concentrations peaked at the low-intermediate location rather than at the highest location where secondary growth rates were at a minimum. Despite the limitations of our data (NSC levels were evaluated only on one occasion and in different individuals from those for which secondary growth was measured), our results seem to agree with other studies that support a positive effect of NSC levels on secondary growth rates (Galiano et al., 2011; Pérez de Lis et al., unpublished results), suggesting the existence of a double constraint of cambial activity due to NSC shortage and climatic control by low temperatures in winter and drought during the growing season. Apart from variations in cambial activity related to climatic differences along the altitudinal gradient, the timing of the climatic response of J. sabina secondary growth also shifted over the last 60 years as an adjustment to the changing underlying climatic conditions. Mean temperature increase during the past and current centuries has intensified the effect of water limitation in Mediterranean environments, even at high elevations (Galván et al., 2015). As a consequence, increased drought stress should be considered as a potential determinant of decline at the drier distribution limit of J. sabina (Voltas et al., 2013). However, our results did not fit these expectations and radial growth remained stable at all sites (Fig. S1). Changes in the timing of climatic sensitivity during the study period might have counterbalanced the increasing negative

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

effect of drought, explaining this stability over time. An earlier climatic response to rainfall in April at the lowest locality and in May at the 1700 m locality herald an earlier onset of cambial activity (Lugo et al., 2012) as an adjustment to the new climatic conditions, whereas at the highest locality, increased sensitivity to August rainfall might indicate an expanding growing season. Our results show that the plastic nature of juniper secondary growth might allow this species to adjust its cambial phenology to benefit from favorable hydric levels along a wide range of climatic conditions. In fact, the large plasticity observed in J. sabina secondary growth may be comparable to that observed for a multispecific assemblage at a regional scale gradient (Pasho et al., 2011b). This is particularly relevant under the current climate warming scenario and has key implications at the community level. Adjusting the timing of the climatic response seems to be a suitable mechanism for J. sabina to keep performing well within the same altitudinal limits with no need to move upwards. Otherwise, the improved microenvironmental conditions associated with J. sabina canopies (García-Cervigón et al., 2015; Verdú and García-Fayos, 2003) may allow the persistence of a set of species whose distribution limits might otherwise be compromised by increasing temperature and drought stress in Mediterranean mountains. The highly plastic behavior of J. sabina secondary growth with respect to climatic variations, combined with its facilitative role in Mediterranean mountains, may thus represent a mechanism that is critical to maintaining the high diversity levels of these particular ecosystems (Cavieres et al., 2014). Acknowledgements We are especially grateful to Miguel García Hidalgo for his invaluable labor with laboratory tasks. We thank David Brown for English language advice. This work was supported by a FPI-EHU grant to A.A., a FPI-MICINN grant to A.I.G.-C., project CGL2012-34209 and Ecometas Network (CGL2014-53840-REDT) both financed by Spanish Ministry of Economy and Competitivity and FEDER funds. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.agrformet.2015. 11.011. References Adams, R., Scharwzbach, A., Nguyen, S., Morris, J., Liu, J.-Q., 2007. Geographic variation in Juniperus sabina L., J. sabina var. arenaria (E.H. Wilson) Farjon, J. sabina var. davurica (Pall.) Farjon and J. sabina var. monogolensis R. P. Adams. Phytologia 89, 153–166. Andreu, L., Gutiérrez, E., Macias, M., Ribas, M., Bosch, O., Camarero, J.J., 2007. Climate increases regional tree-growth variability in Iberian pine forests. Global Change Biol. 13, 1–12, http://dx.doi.org/10.1111/j.1365-2486.2007. 01322.x. Babst, F., Poulter, B., Trouet, V., Tan, K., Neuwirth, B., Wilson, R., Carrer, M., Grabner, M., Tegel, W., Levanic, T., Panayotov, M., Urbinati, C., Bouriaud, O., Ciais, P., Frank, D., 2013. Site- and species-specific responses of forest growth to climate across the European continent. Global Ecol. Biogeogr. 22, 706–717, http://dx. doi.org/10.1111/geb.12023. Bär, A., Pape, R., Bräuning, A., Löffler, J., 2007. Growth-ring variations of dwarf shrubs reflect regional climate signals in alpine environments rather than topoclimatic differences. J. Biogeogr. 35, 625–636, http://dx.doi.org/10.1111/j. 1365-2699.2007.01804.x. Camarero, J.J., Gazol, A., Sangüesa-Barreda, G., Oliva, J., Vicente-Serrano, S., 2015. To die or not to die: early warnings of tree dieback in response to a severe drought. J. Ecol. 103, 44–57, http://dx.doi.org/10.1111/1365-2745.12295. Camarero, J.J., Olano, J.M., Parras, A., 2010. Plastic bimodal xylogenesis in conifers from continental Mediterranean climates. New Phytol. 185, 471–480, http:// dx.doi.org/10.1111/j.1469-8137.2009.03073.x. Castro, J., Zamora, R., Hódar, J., Gómez, J., 2002. Use of shrubs as nurse plants: a new technique for reforestation in Mediterranean Mountains. Restor. Ecol. 10, 297–305, http://dx.doi.org/10.1046/j.1526-100X.2002.01022.x.

43

Cavieres, L.A., Brooker, R.W., Butterfield, B.J., Cook, B.J., Kikvidze, Z., Lortie, C.J., Michalet, R., Pugnaire, F.I., Schöb, C., Xiao, S., Anthelme, F., Björk, R.G., Dickinson, K.J.M., Cranston, B.H., Gavilán, R., Gutiérrez-Girón, A., Kanka, R., Maalouf, J.-P., Mark, A.F., Noroozi, J., Parajuli, R., Phoenix, G.K., Reid, A.M., Ridenour, W.M., Rixen, C., Wipf, S., Zhao, L., Escudero, A., Zaitchik, B.F., Lingua, E., Aschehoug, E.T., Callaway, R.M., 2014. Facilitative plant interactions and climate simultaneously drive alpine plant diversity. Ecol. Lett. 17, 193–202, http://dx.doi.org/10.1111/ele.12217. Cook, E., Holmes, R., 1996. Guide for computer program ARSTAN. In: Grissino-Mayer, H., Holmes, R., Fritts, H. (Eds.), The International Tree-Ring Data Bank Program Library Version 2.0 User’s Manual. Laboratory of Tree-Ring Research, University of Arizona, Tucson, pp. 75–87. Cook, E., Peters, K., 1981. The smoothing spline: a new approach to standardizing forest interior tree-ring width series for dendroclimatic studies. Tree-ring Bull. 41, 45–53. Coomes, D., Allen, R., 2007. Effects of size, competition and altitude on tree growth. J. Ecol. 95, 1084–1097, http://dx.doi.org/10.1111/j.1365-2745.2007.01280.x. Christensen, J.H., Hewitson, B., Busuioc, A., Chen, A., Gao, X., Held, I., Jones, R., Kolli, ˜ Rueda, V., Mearns, L., Menéndez, C.G., R.K., Kwon, W.-T., Laprise, R., Magana Räisänen, J., Rinke, A., Sarr, A., Whetton, P., 2007. Regional climate projections. In: Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tingor, M., Miller, H.L. (Eds.), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, New York, NY, pp. 847–940. Dang, H., Zhang, Y., Zhang, K., Jiang, M., Zhang, Q., 2013. Climate-growth relationships of subalpine fir (Abies fargesii) across the altitudinal range in the Shennongjia Mountains, central China. Clim. Change 117, 903–917, http://dx. doi.org/10.1007/s10584-012-0611-5. ˇ K., 2011. Climate de Luis, M., Novak, K., Raventós, J., Griˇcar, J., Prislan, P., Cufar, factors promoting intra-annual density fluctuations in Aleppo pine (Pinus halepensis) from semiarid sites. Dendrochronologia 29, 163–169, http://dx.doi. org/10.1016/j.dendro.2011.01.005. Dittmar, C., Elling, W., 1999. Radial growth of Norway spruce and European beech in relation to weather and altitude. Forstwiss. Cent. Bl. 118, 251–270, http://dx. doi.org/10.1007/BF02768990. El Kenawy, A., López-Moreno, J., Stepanek, P., Vicente-Serrano, S., 2013. An assessment of the role of homogenization protocols in the performance of daily temperature series and trends: application to Northeastern Spain. Int. J. Climatol. 33, 87–108, http://dx.doi.org/10.1002/joc.3410. Fonti, P., Solomonoff, N., García-González, I., 2007. Earlywood vessels of Castanea sativa record temperature before their formation. New Phytol. 173, 562–570, http://dx.doi.org/10.1111/j.1469-8137.2006.01945.x. Franklin, R., 2013. Growth response of the alpine shrub, Linanthus pungens, to snowpack and temperature at a rock glacier site in the eastern Sierra Nevada of California, USA. Quat. Int. 310, 20–33, http://dx.doi.org/10.1016/j.quaint.2012. 07.018. Fritts, H., Smith, D., Cardis, J., Budelsky, C., 1965. Tree-ring characteristics along a vegetation gradient in northern Arizona. Ecology 46, 393–401, http://dx.doi. org/10.2307/1934872. Galiano, L., Martínez-Vilalta, J., Lloret, F., 2011. Carbon reserves and canopy defoliation determine the recovery of Scots pine 4 yr after a drought episode. New Phytol. 190, 750–759, http://dx.doi.org/10.1111/j.1469-8137.2010.03628. x. Galván, D.J., Büntgen, U., Ginzler, C., Grudd, H., Gutiérrez, E., Labuhn, I., Camarero, J.J., 2015. Drought-induced weakening of growth-temperature associations in high-elevation Iberian pines. Global Planet Change 124, 95–106, http://dx.doi. org/10.1016/j.gloplacha.2014.11.011. García-Cervigón, A.I., Gazol, A., Sanz, V., Camarero, J.J., Olano, J.M., 2013. Intraspecific competition replaces interspecific facilitation as abiotic stress decreases: the shifting nature of plant-plant interactions. Perspect. Plant Ecol. Evol. Syst. 15, 226–236, http://dx.doi.org/10.1016/j.ppees.2013.04.001. García-Cervigón, A.I., Linares, J., Aibar, P., Olano, J.M., 2015. Facilitation promotes changes in leaf economics traits of a perennial forb. Oecologia, http://dx.doi. org/10.1007/s00442-015-3312-8. García-Cervigón, A.I., Olano, J.M., Gozalbo, M., Camarero, J.J., 2012. Arboreal and prostrate conifers coexisting in Mediterranean high mountains differ in their climatic responses. Dendrochronologia 30, 249–310, http://dx.doi.org/10. 1016/j.dendro.2012.02.004. García-Ruiz, J., Lana-Renault, N., 2011. Hydrological and erosive consequences of farmland abandonment in Europe, with special reference to the Mediterranean region—a review. Agric. Ecosyst. Environ. 140, 317–338, http://dx.doi.org/10. 1016/j.agee.2011.01.003. García-Ruiz, J.M., Lasanta, T., Ruiz-Flano, P., Ortigosa, L., White, S., González, C., Martí, C., 1996. Land-use changes and sustainable development in mountain areas: a case study in the Spanish Pyrenees. Landsc. Ecol. 11, 267–277, http:// dx.doi.org/10.1007/BF02059854. Gazol, A., Camarero, J., 2012. The performance of Mediterranean subshrubs depends more on microsite than on regional climate conditions. J. Veg. Sci. 23, 1062–1070, http://dx.doi.org/10.1111/j.1654-1103.2012.01429.x. Giorgi, F., Lionello, P., 2008. Climate change projections for the Mediterranean region. Global Planet Change 63, 90–104, http://dx.doi.org/10.1016/j. gloplacha.2007.09.005. Gómez-Aparicio, L., Zamora, R., Castro, J., Hódar, J., 2008. Facilitation of tree saplings by nurse plants: microhabitat amelioration or protection against herbivores? J. Veg. Sci. 19, 161–172, http://dx.doi.org/10.3170/2008-8-18347.

44

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45

Gómez-Aparicio, L., Zamora, R., Gómez, J.M., Hódar, J.A., Castro, J., Baraza, E., 2004. Applying plant facilitation to forest restoration: a meta-analysis of the use of shrubs as nurse plants. Ecol. Appl. 14, 1128–1138, http://dx.doi.org/10.1890/ 03-5084. ˇ K., 2014. Griˇcar, J., Prislan, P., Vladimír Gryc, V., Vavrˇcík, H., de Luis, M., Cufar, Plastic and locally adapted phenology in cambial seasonality and production of xylem and phloem cells in Picea abies from temperate environments. Tree Physiol. 34, 869–881, http://dx.doi.org/10.1093/treephys/tpu026. Grissino-Mayer, H., 2001. Evaluating crossdating accuracy: a manual and tutorial for the computer program COFECHA. Tree-Ring Res. 57, 205–221, http://hdl.handle.net/10150/251654. Hargreaves, G.L., Samani, Z., 1985. Reference crop evapotranspiration from temperature. Appl. Eng. Agric. 1, 96–99, http://dx.doi.org/10.13031/2013. 26773. He, Z., Du, J., Zhao, W., Yang, J., Chen, L., Zhu, X., Chang, X., Liu, H., 2015. Assessing temperature sensitivity of subalpine shrub phenology insemi-arid mountain regions of China. Agric. For. Meteorol. 213, 42–53, http://dx.doi.org/10.1016/j. agrformet.2015.06.013. He, M., Yang, B., Bräuning, A., 2013. Tree growth–climate relationships of Juniperus tibetica along an altitudinal gradient on the southern Tibetan Plateau. Trees 27, 429–439, http://dx.doi.org/10.1007/s00468-012-0813-5. Hoch, G., Popp, M., Körner, C., 2002. Altitudinal increase of mobile carbon pools in Pinus cembra suggests sink limitation of growth at the Swiss treeline. Oikos 98, 361–374, http://dx.doi.org/10.1034/j.1600-0706.2002.980301.x. ˜ J., 2005. Running to stand still: adaptation and the response of Jump, A., Penuelas, plants to rapid climate change. Ecol. Lett. 8, 1010–1020, http://dx.doi.org/10. 1111/j.1461-0248.2005.00796.x. King, G., Gugerli, F., Fonti, P., Frank, D., 2013. Tree growth response along an elevational gradient: climate or genetics? Oecologia 173, 1587–1600, http:// dx.doi.org/10.1007/s00442-013-2696-6. Körner, C., 2003. Carbon limitation in trees. J. Ecol. 91, 4–17, http://dx.doi.org/10. 1046/j.1365-2745.2003.00742.x. Lebourgeois, F., Mérian, P., Courdier, F., Ladierm, J., Dreyfus, P., 2012. Instability of climate signal in tree-ring width in Mediterranean mountains: a multispecies analysis. Trees 26, 715–729, http://dx.doi.org/10.1007/s00468011-0638-7. Liang, E., Dawadi, B., Pederson, N., Eckstein, D., 2014. Is the growth of birch at the upper timberline in the Himalayas limited by moisture or by temperature? Ecology 95, 2453–2465, http://dx.doi.org/10.1890/13-1904.1. Linares, J., Tíscar, P., 2010. Climate change impacts and vulnerability of the southern populations of Pinus nigra subsp. salzmannii. Tree Physiol. 30, 795–806, http://dx.doi.org/10.1093/treephys/tpq052. Lugo, J., Deslauriers, A., Rossi, S., 2012. Duration of xylogenesis in black spruce lengthened between 1950 and 2010. Ann. Bot. 110, 1099–1108, http://dx.doi. org/10.1093/aob/mcs175. ˜ I., 2010. Black pine (Pinus nigra Arn.) Martín-Benito, D., del Río, M., Canellas, growth divergence along a latitudinal gradient in Western Mediterranean mountains. Ann. For. Sci. 67, 401, http://dx.doi.org/10.1051/forest/2009121. Mateo Sanz, G., 1997. Catálogo de plantas vasculares del Rincón de Ademuz (Valencia). Jardín Botánico de Valencia, Valencia. Mateo Sanz, G., Lozano Terrazas, J., Aguilella Palasí, A., 2013. Catálogo florístico de las sierras de Gúdar y Javalambre (Teruel). Jolube Consultor-Editor Botánico, Jaca. Spain. Matías, L., Jump, A., 2015. Asymmetric changes of growth and reproductive investment herald altitudinal and latitudinal range shifts of two woody species. Global Change Biol. 21, 882–896, http://dx.doi.org/10.1111/gcb. 12683. Médail, F., Diadema, K., 2009. Glacial refugia influence plant diversity patterns in the Mediterranean Basin. J. Biogeogr. 36, 1333–1345, http://dx.doi.org/10. 1111/j.1365-2699.2008.02051.x. Médail, F., Quézel, P., 1997. Hot-spots analysis for conservation of plant biodiversity in the Mediterranean Basin. Ann. Mo. Bot. Gard. 84, 112–127, http://dx.doi.org/10.2307/2399957. Michalet, R., Schob, C., Lortie, C., Brooker, R., Callaway, R., 2014. Partitioning net interactions among plants along altitudinal gradients to study community responses to climate change. Funct. Ecol. 28, 75–86, http://dx.doi.org/10.1111/ 1365-2435.12136. Moser, L., Fonti, P., Büntgen, U., Franzen, J., Luterbacher, J., Frank, D., 2010. Timing and duration of European larch growing season along altitudinal gradients in the Swiss Alps. Tree Physiol. 30, 225–233, http://dx.doi.org/10.1093/treephys/ tpp108. ˜ T., Arroyo, J., 2000. Plant diversity patterns in the Aljibe Ojeda, F., Maranón, Mountains (S. Spain): a comprehensive account. Biodivers. Conserv. 9, 1323–1343, http://dx.doi.org/10.1023/A:1008923213321. Olano, J.M., Almería, I., Eugenio, M., Von Arx, G., 2013. Under pressure: how a Mediterranean high-mountain forb coordinates growth and hydraulic xylem anatomy in response to temperature and water constraints. Funct. Ecol. 27, 1295–1303, http://dx.doi.org/10.1111/1365-2435.12144. Olano, J.M., García-Cervigón, A., Arzac, A., Rozas, V., 2015. Intra-annual wood density fluctuations and tree-ring width pattern are sex- and site-dependent in the dioecious conifer Juniperus thurifera L. Trees, http://dx.doi.org/10.1007/ s00468-015-1212-5. Olano, J.M., Linares, J.C., García-Cervigón, A.I., Arzac, A., Delgado, A., Rozas, V., 2014. Drought-induced increase in water use efficiency reduces tree secondary growth and tracheid wall thickness in a Mediterranean conifer. Oecologia 176, 273–283, http://dx.doi.org/10.1007/s00442-014-2989-4.

Olano, J.M., Menges, E., Martínez, E., 2006. Carbohydrate storage in five resprouting Florida scrub plants across a fire chronosequence. New Phytol. 170, 99–106, http://dx.doi.org/10.1111/j.1469-8137.2005.01634.x. Palombo, C., Battipaglia, G., Cherubini, P., Chirici, G., Garfì, V., Lasserre, B., Lombardi, F., Marchetti, M., Tognetti, R., 2014. Warming-related growth responses at the southern limit distribution of mountain pine (Pinus mugo Turra subsp. mugo). J. Veg. Sci. 25, 571–583, http://dx.doi.org/10.1111/jvs.12101. Pasho, E., Camarero, J.J., de Luis, M., Vicente-Serrano, S.M., 2011a. Spatial variability in large-scale and regional atmospheric drivers of Pinus halepensis growth in eastern Spain. Agric. For. Meteorol. 151, 1106–1119, http://dx.doi.org/10.1016/ j.agrformet.2011.03.016. Pasho, E., Camarero, J., de Luis, M., Vicente-Serrano, S., 2011b. Impacts of drought at different time scales on forest growth across a wide climatic gradient in north-eastern Spain. Agric. For. Meteorol. 151, 1800–1811, http://dx.doi.org/ 10.1016/j.agrformet.2011.07.018. Pauli, H., Gottfried, M., Dullinger, S., Abdaladze, O., Akhalkatsi, M., Alonso, J., Coldea, G., Dick, J., Erschbamer, B., Fernandez Calzado, R., Goshn, D., Holten, J.I., Kanka, R., Kazakis, G., Kollár, J., Larsson, P., Moiseev, P., Moiseev, D., Molau, U., Molero Mesa, J., Nagy, L., Pelino, G., Puscas, M., Rossi, G., Stanisci, A., Syverhuset, A.O., Theurillat, J.-P., Thomaselli, M., Unterluggauer, P., Villar, L., Vittoz, P., Grabherr, G., 2012. Recent plant diversity changes on Europe’s Mountain summits. Science 336, 353–355, http://dx.doi.org/10.1126/science.1219033. Popp, M., Lied, W., Meyer, A.J., Richter, A., Schiller, P., Schwitte, H., 1996. Sample preservation for determination of organic compounds: microwave versus freeze–drying. J. Exp. Bot. 47, 1469–1473, http://dx.doi.org/10.1093/jxb/47.10. 1469. ˇ K., 2013. Phenological variation in xylem Prislan, P., Griˇcar, J., De Luis, M., Cufar, and phloem formation in Fagus sylvatica from two contrasting sites. Agric. For. Meteorol. 180, 142–151, http://dx.doi.org/10.1016/j.agrformet.2013.06.001. Rossi, S., Deslauriers, A., Griˇcar, J., Seo, J.-W., Rathgeber, C., Anfodillo, T., Morin, H., Levanic, T., Oven, P., Jalkanen, R., 2008. Critical temperatures for xylogenesis in conifers of cold climates. Global Ecol. Biogeogr. 17, 696–707, http://dx.doi.org/ 10.1111/j.1466-8238.2008.00417.x. Rozas, V., García-González, I., Zas, R., 2011. Climatic control of intra-annual wood density fluctuations of Pinus pinaster in NW Spain. Trees 25, 443–453, http:// dx.doi.org/10.1007/s00468-010-0519-5. Schöb, C., Armas, C., Pugnaire, F., 2013. Direct and indirect interactions co-determine species composition in nurse plant systems. Oikos 122, 1371–1379, http://dx.doi.org/10.1111/j.1600-0706.2013.00390.x. Speer, J., 2010. Fundamentals of Tree-Ring Research. The University of Arizona Press, Tucson. Thuiller, W., Lavorel, S., Araújo, M., Sykes, M., Prentice, I., 2005. Climate change threats to plant diversity in Europe. Proc. Natl. Acad. Sci. U.S.A. 102, 8245–8250, http://dx.doi.org/10.1023/A:10106322015572. Verdú, M., García-Fayos, P., 2003. Frugivorous birds mediate sex-biased facilitation in a dioecious nurse plant. J. Veg. Sci. 14, 35–42, http://dx.doi.org/10.1111/j. 1654-1103.2003.tb02125.x. Vicente-Serrano, S.M., Azorin-Molina, C., Sanchez-Lorenzo, A., Morán-Tejeda, E., Lorenzo-Lacruz, J., Revuelto, J., López-Moreno, J.I., Espejo, F., 2013. Temporal evolution of surface humidity in Spain: recent trends and possible physical mechanisms. Clim. Dyn. 42, 2655–2674, http://dx.doi.org/10.1007/s00382013-1885-7. Vicente-Serrano, S.M., Azorin-Molina, C., Sanchez-Lorenzo, A., Revuelto, J., López-Moreno, J.I., González-Hidalgo, J.C., Espejo, F., 2014a. Reference evapotranspiration variability and trends in Spain, 1961–2011. Global Planet Change 121, 26–40, http://dx.doi.org/10.1016/j.gloplacha.2014.06. 005. Vicente-Serrano, S., Beguería, S., López-Moreno, J., 2010b. A multi-scalar drought index sensitive to global warming: the standardized precipitation evapotranspiration index—SPEI. J. Clim. 23, 1696–1718, http://dx.doi.org/10. 1175/2009JCLI2909.1. Vicente-Serrano, S., Beguería, S., López-Moreno, J., García-Vera, M., Stepanek, P., 2010a. A complete daily precipitation database for North-East Spain: reconstruction, quality control and homogeneity. Int. J. Climatol. 30, 1146–1163, http://dx.doi.org/10.1002/joc.1850. Vicente-Serrano, S.M., López-Moreno, J.I., Beguería, S., Lorenzo-Lacruz, J., Sanchez-Lorenzo, A., García-Ruiz, J.M., Azorin-Molina, C., Morán-Tejeda, E., Revuelto, J., Trigo, R.M., Coelho, F., Espejo, F., 2014b. Evidence of increasing drought severity caused by temperature rise in southern Europe. Environ. Res. Lett. 9, 044001, http://dx.doi.org/10.1088/1748-9326/9/4/ 044001. Voltas, J., Camarero, J.J., Carulla, D., Aguilera, M., Ortiz, A., Ferrio, J.P., 2013. A retrospective, dual-isotope approach reveals individual predispositions to winter-drought induced tree dieback in the southernmost distribution limit of Scots pine. Plant Cell Environ. 36, 1435–1448, http://dx.doi.org/10.1111/pce. 12072. von Arx, G., Edwards, P., Dietz, H., 2006. Evidence for life history changes in high-altitude populations of three perennial forbs. Ecology 87, 665–674, http://dx.doi.org/10.1890/05-1041. Walther, G., Beißner, S., Pott, R., 2005. Climate change and high mountain vegetation shifts. In: Broll, G., Keplin, B. (Eds.), Mountain Ecosystems, Studies in Treeline Ecology. Springer Berlin, New York, NY, pp. 77–96. Wang, Y., Liang, E., Ellison, A., Lu, X., Camarero, J., 2015. Facilitation stabilizes moisture-controlled alpine juniper shrublines in the central Tibetan Plateau. Global Planet Change 132, 20–30, http://dx.doi.org/10.1016/j.gloplacha.2015. 06.007.

A. Arzac et al. / Agricultural and Forest Meteorology 217 (2016) 35–45 Wang, Z., Yang, B., Deslauriers, A., Bräuning, A., 2014. Intra-annual stem radial increment response of Qilian juniper to temperature and precipitation along an altitudinal gradient in northwest China. Trees 29, 25–34, http://dx.doi.org/ 10.1007/s00468-014-1037-7. Wesche, K., Ronnenberg, K., 2004. Phytosociological affinities and habitat preferences of Juniperus sabina L. and Artemisia santolinifolia Turcz. ex Bess. in

45

mountain sites of the south-eastern Gobi Altay. Mongolia. Feddes Repert. 115, 585–600, http://dx.doi.org/10.1002/fedr.200411056. Yang, B., He, M., Melvin, T.M., Zhao, Y., Briffa, K., 2013. Climate control on tree growth at the upper and lower treelines: a case study in the Qilian Mountains, Tibetan Plateau. PLoS ONE 8, e69065, http://dx.doi.org/10.1371/journal.pone. 0069065.