Received: 18 October 2019 | Revised: 19 March 2020 | Accepted: 30 April 2020 DOI: 10.1111/gcb.15153 PRIMARY RESEARCH ARTICLE Growth and resilience responses of Scots pine to extreme droughts across Europe depend on predrought growth conditions Arun K. Bose1,2 | Arthur Gessler1,3,4 | Andreas Bolte5 | Alessandra Bottero1,4 | Allan Buras6 | Maxime Cailleret7 | J. Julio Camarero8 | Matthias Haeni1 | Ana-Maria Hereş9,10 | Andrea Hevia11 | Mathieu Lévesque3 | Juan C. Linares12 | 13,14 15 16,17 Jordi Martinez-Vilalta | Luis Matías | Annette Menzel | 12 1 Raúl Sánchez-Salguero | Matthias Saurer | Michel Vennetier | Daniel Ziche5,18 7 | 1,3,4 Andreas Rigling 1 WSL Swiss Federal Institute for Forest, Snow and Landscape Research, Birmensdorf, Switzerland 2 Forestry and Wood Technology Discipline, Khulna University, Khulna, Bangladesh 3 Institute of Terrestrial Ecosystems, ETH Zurich, Zurich, Switzerland 4 SwissForestLab, Birmensdorf, Switzerland 5 Thünen Institute of Forest Ecosystems, Eberswalde, Germany 6 Land Surface-Atmosphere Interactions, Technische Universitat München, Freising, Germany 7 UMR RECOVER/Ecosystèmes Méditerranéens et Risques, INRAE, Aix-en-Provence cedex 5, France 8 Instituto Pirenaico de Ecologıa (IPE-CSIC), Zaragoza, Spain 9 Department of Forest Sciences, Transilvania University of Braşov, Braşov, Romania 10 BC3 - Basque Centre for Climate Change, Scientific Campus of the University of the Basque Country, Leioa, Spain 11 Departamento de Ciencias Agroforestales, Universidad de Huelva, Palos de la Frontera, Spain 12 Depto. Sistemas Físicos, Químicos y Naturales, Universidad Pablo de Olavide, Sevilla, Spain 13 CREAF, Bellaterra (Cerdanyola del Vallès), Spain 14 Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), Spain 15 Departamento de Biología Vegetal y Ecología, Facultad de Biología, Universidad de Sevilla, Sevilla, Spain 16 Ecoclimatology, Technische Universität München, Freising, Germany 17 Institute for Advanced Study, Technische Universität München, Garching, Germany 18 Faculty of Forest and Environment, Eberswalde University for Sustainable Development, Eberswalde, Germany Correspondence Arun K. Bose, WSL Swiss Federal Institute Abstract for Forest, Snow and Landscape Research, Global climate change is expected to further raise the frequency and severity of ex- Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland. treme events, such as droughts. The effects of extreme droughts on trees are dif- Email: arun.bose@wsl.ch ficult to disentangle given the inherent complexity of drought events (frequency, Funding information severity, duration, and timing during the growing season). Besides, drought effects Marie Skłodowska-Curie, Grant/Award might be modulated by trees’ phenotypic variability, which is, in turn, affected by Number: 749051-REFOREST; FEDER, Grant/ Award Number: IJCI-2015-25845; Ministry long-term local selective pressures and management legacies. Here we investigated of Science, Innovation and Universities, This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2020 The Authors. Global Change Biology published by John Wiley & Sons Ltd Glob Change Biol. 2020;26:4521–4537.  wileyonlinelibrary.com/journal/gcb | 4521 4522 | BOSE et al. Grant/Award Number: RTI2018-096884- B-C31 and RTI2018-096884-B-C33; the magnitude and the temporal changes of tree-level resilience (i.e., resistance, re- VULBOS, Grant/Award Number: UPO- covery, and resilience) to extreme droughts. Moreover, we assessed the tree-, site-, 1263216; PinCaR, Grant/Award Number: UHU-1266324; Bavarian Climate Research and drought-related factors and their interactions driving the tree-level resilience Network; German Waldklimafond, Grant/ to extreme droughts. We used a tree-ring network of the widely distributed Scots Award Number: FKZ 28W-C-4-077- 01; Bavarian State Ministry for Food, pine (Pinus sylvestris) along a 2,800 km latitudinal gradient from southern Spain to Agriculture, and Forestry, Grant/Award northern Germany. We found that the resilience to extreme drought decreased in Number: ST327 mid-elevation and low productivity sites from 1980–1999 to 2000–2011 likely due to more frequent and severe droughts in the later period. Our study showed that the impact of drought on tree-level resilience was not dependent on its latitudinal location, but rather on the type of sites trees were growing at and on their growth performances (i.e., magnitude and variability of growth) during the predrought period. We found significant interactive effects between drought duration and tree growth prior to drought, suggesting that Scots pine trees with higher magnitude and vari- ability of growth in the long term are more vulnerable to long and severe droughts. Moreover, our results indicate that Scots pine trees that experienced more frequent droughts over the long-term were less resistant to extreme droughts. We, therefore, conclude that the physiological resilience to extreme droughts might be constrained by their growth prior to drought, and that more frequent and longer drought periods may overstrain their potential for acclimation. KEYWORDS acclimation, latitudinal gradient, Pinus sylvestris, predisposition, tree rings 1 | I NTRO D U C TI O N droughts (McDowell et al., 2008; Smith, 2011), because phenotypic acclimation to such extreme events may depend on a multitude of Climate change effects are broadly characterized by elevated tem- factors and their interactions, including drought characteristics perature, changed precipitation regimes, and increased interannual (Anderegg et al., 2015; Gazol et al., 2018), drought history of the variability, often resulting in more frequent and intense climate ex- growing environment (Vicente-Serrano et al., 2013), species-specific tremes such as severe droughts (Dai, 2012; Spinoni, Vogt, Naumann, functional traits and life-history strategies (Anderegg et al., 2016; Barbosa, & Dosio, 2018). The increased frequency and severity of Greenwood et al., 2017; Lévesque et al., 2013), provenance (Sánchez- droughts can significantly impact tree growth by reducing their pho- Salguero et al., 2018; Seidel, Schunk, Matiu, & Menzel, 2016), tree size tosynthetic activity (Flexas & Medrano, 2002; Reddy, Chaitanya, and age (Granda, Gazol, & Camarero, 2018; Magnani, Mencuccini, & & Vivekanandan, 2004) and altering their cambial activity (Gruber, Grace, 2000; Serra-Maluquer, Mencuccini, & Martínez-Vilalta, 2018), Strobl, Veit, & Oberhuber, 2010). In addition, severe drought events tree-to-tree competition (Linares, Camarero, & Carreira, 2010), nu- have been associated to forest decline either through direct abiotic trient imbalances (Hevia et al., 2019), nutrient availability (Gessler, effects leading to hydraulic failure and/or carbon starvation (Adams Schaub, & McDowell, 2017), species composition and stocking of the et al., 2017; Choat et al., 2018; McDowell et al., 2008) or mediated by forest stand (Bottero et al., 2017; Forrester et al., 2016; Grossiord biotic factors, such as insects (Rouault et al., 2006), fungi (Giordano, et al., 2014), trees’ neighbourhood composition (Grossiord, 2019), Gonthier, Varese, Miserere, & Nicolotti, 2009), and mistletoes microclimatic conditions related to forest edge and interior (Buras (Rigling, Eilmann, Koechli, & Dobbertin, 2010). These effects may ul- et al., 2018), and growth trends prior to drought (Zang, Hartl-Meier, timately induce shifts in forest composition (Buras & Menzel, 2019; Dittmar, Rothe, & Menzel, 2014). In the longer term, acclimation is Walther et al., 2002) and reduction in forest productivity (Ciais often complemented by evolutionary genotypic adaptation (Bose et al., 2005). et al., 2020; Hamrick, 2004; Sánchez-Salguero et al., 2018) leading Growing recognition of the impacts of extreme droughts on forest to differentiation of populations and ecotypes with varying adaptive ecosystems has spurred on a number of long-term experiments and capacities to drought, often observed for marginal populations at dry observational studies (e.g., Breshears et al., 2005; Jentsch et al., 2011; species range margins (Bolte et al., 2016; Hampe & Petit, 2005). Seidel, Matiu, & Menzel, 2019). The results of these studies revealed Moreover, the effects of past drought and growing condi- a large variability in pattern and magnitude of responses to extreme tions (legacy effects), can remain for several years and modify the BOSE et al. | 4523 tree growth and physiological responses to the current drought as well as the ability to return to predisturbance growth levels (Anderegg et al., 2015; Kannenberg et al., 2019; Seidel et al., 2019). (recovery; Lloret, Keeling, & Sala, 2011). Specifically, we asked An important question in the debate on drought and acclimation is four research questions: (a) How does the impact of the climatic whether individuals will be able to acclimate fast enough to cope water balance (CWB; i.e., precipitation minus potential evapo- with increased frequency and severity of droughts (Dai, 2012; transpiration) of different seasons on tree growth vary along a Szejner, Belmecheri, Ehleringer, & Monson, 2020). It is therefore latitudinal gradient? (b) How do radial growth rates of Scots pine important to understand how tree growth responses to extreme during drought and nondrought years vary across sites? (c) Has droughts vary across sites with different productivity (Valladares, tree growth resilience to extreme drought changed over the past Gianoli, & Gómez, 2007; Valladares et al., 2014), since site productiv- decades due to an increased frequency and severity of droughts ity can modify trees’ phenotypic strategies such as tree height, root (Serra-Maluquer et al., 2018; Szejner et al., 2020)? (d) How do to shoot ratio, and crown development for efficient conservation drought characteristics, site conditions, and tree growth-related and utilization of water (Vanninen & Mäkelä, 2005). For example, variables modulate the tree growth resilience to extreme drought tree height which is commonly used as an indicator of site produc- events? For this last research question, we considered a list of tivity (e.g., Westoby, Falster, Moles, Vesk, & Wright, 2002) was biological hypotheses based on a literature review (see Table S1: reported to be the strongest predictor of tree mortality in south- e.g., Gazol et al., 2017, 2018; Sánchez-Salguero et al., 2018; Vitali, western United States where 1.8 million trees were studied (Stovall, Büntgen, & Bauhus, 2017; Zang et al., 2014). Shugart, & Yang, 2019). Several recent studies conducted in southern and central Europe have reported drought-induced dieback of Scots pine (Buras 2 | M ATE R I A L S A N D M E TH O DS et al., 2018; Camarero, Gazol, Sangüesa-Barreda, Oliva, & Vicente- Serrano, 2015; Etzold et al., 2019; Galiano, Martínez-Vilalta, & 2.1 | Study sites and tree-ring data Lloret, 2010; Hereş, Martínez-Vilalta, & Claramunt López, 2012; Sánchez-Salguero, Navarro-Cerrillo, Camarero, & Fernández- We compiled tree-ring width data of Scots pine from 30 sites Cancio, 2012) causing a shift toward the dominance of oak (Quercus (Table S2) along an approximately 2,800 km long latitudinal gradi- spp.) species (Carnicer et al., 2014; Galiano et al., 2010; Rigling ent from southern Spain (Baza; 37.2°N, 4.0°W) to north-eastern et al., 2013). Although the impact of various tree- and site-level fac- Germany (Torgelow; 53.6°N, 14°E; Figure 1). To avoid age-related tors on tree growth during drought has been studied from local to growth effects only trees older than 30 years at the time of exam- global scales (e.g., Anderegg et al., 2015; Buras et al., 2018; Gazol ined drought were selected, resulting in 615 adult Scots pine trees et al., 2018; Zang et al., 2014), their interactive effects are still not (6–60 trees per site). From each tree, two to four tree ring width clearly understood (Maes et al., 2019). For example, some large-scale series were included, measured from increment cores extracted at studies found a low to moderate influence of drought severity on breast height (1.3 m height) and cross-dated following standard den- tree growth response (e.g., Gazol, Camarero, Anderegg, & Vicente- drochronological procedures (Grissino-Mayer, 2001). Serrano, 2017; Sánchez-Salguero et al., 2018), possibly because they Considering the large differences in productivity among study did not consider interactive effects between drought characteristics sites along this long gradient, the sites were grouped using a hier- and long-term tree growth performances. In addition, large-scale archical cluster analysis (Kaufman & Rousseeuw, 1990). The clas- studies often characterize drought according to a predefined meteo- sification was based on site productivity index (i.e., dominant tree rological season (e.g., drought in spring–summer) irrespective of local height at 50 years of stand age) and site elevation. Dominant tree site conditions, soil moisture content, and geographic location (e.g., height has been commonly used as an indicator of site productiv- Bottero et al., 2017; Gao et al., 2018; Gazol et al., 2018). As a conse- ity (e.g., Bugmann, 1996; Diéguez-Aranda, Burkhart, & Rodríguez- quence, site-specific climate-growth signals might be overlooked if Soalleiro, 2005; Westoby et al., 2002) including Scots pine sites a particular studied season is not the most relevant period for tree (e.g., Diéguez-Aranda, Álvarez González, Marcos Barrio, & Alberto radial growth (Pasho, Camarero, de Luis, & Vicente-Serrano, 2011; Rojo, 2005; Hökkä & Ojansuu, 2004; Mäkinen, Yue, & Kohnle, 2017; Sánchez-Salguero et al., 2015). Palahı́, Tomé, Pukkala, Trasobares, & Montero, 2004). The hierarchi- Here we combined Scots pine tree-ring width data from 30 cal clustering was done using the hclust function and ward.D method sites into a network to determine how growth responses to ex- in R (R Development Core Team, 2018). Based on the hierarchical treme drought varied along a latitudinal gradient across Europe cluster analysis (Figure S1), three groups (i.e., site types) were char- stretching from southern Spain to northern Germany. Tree growth acterized as (a) high-elevation sites (1,865–2,011 m a.s.l.) with low response was assessed to retrospectively quantify short- and productivity (6.0–14.0 m in stand dominant tree height) referred as long-term drought effects on growth for numerous individuals, “HELP”, (b) mid-elevation sites (600–1,450 m a.s.l.) with low produc- sites, and species at annual resolution. Tree growth resilience was tivity (7.5–11.0 m in stand dominant tree height) referred as “MELP,” defined as the capacity of a tree to reach growth rates similar to and (c) low-elevation sites (33–326 m a.s.l.) with high productiv- those prior to a given drought event. Thus, resilience encompasses ity (15–23.7 m in stand dominant tree height) referred as “LEHP” the capacity to buffer the impact of a disturbance (resistance), (Figure S1). 4524 | BOSE et al. F I G U R E 1 Location of the 30 Scots pine study sites distributed along a latitudinal gradient that ranged from southern Spain to northern Germany. HELP, high-elevation sites with low productivity; LEHP, low-elevation sites with high productivity; MELP, mid- elevation sites with low productivity. The grey shade used as a background within the map represents the natural distribution of Scots pine adapted from Mátyás, Ackzell, & Samuel (2004) 2.2 | Analytical approaches To build the site-level tree-ring chronology, we averaged the de- trended individual RWI series with a Tukey's biweight robust mean Addressing our four research questions, the analytical approach (Cook & Kairiukstis, 1990; Fritts, 2001). The RWI and average tree- involved two steps: data preparation and data analysis. The data level chronology were calculated using the detrend and chron func- preparation step embodied four substeps, (a) quantification of tions, respectively, available from the dplR R package (Bunn et al., tree-ring width indices; (b) quantification of drought indices; (c) 2018; R Development Core Team, 2018). identification of drought, predrought, and postdrought periods (i.e., years); and (d) quantification of tree growth resilience indices. The data analysis step embodied four steps, that is, one for each 2.4 | Quantification of drought indices research question. Monthly mean temperature (°C) and total precipitation (mm) data were obtained for each site from different climate data sources 2.3 | Quantification of tree-ring width indices (Table S4). To compute the correlation coefficient between drought indices and the RWI, we considered a 50 year period for all sites. We aimed at quantifying growth responses to extreme drought However, the range of years for the 50 year period varied across events over the recent 50 years period roughly from year 1960 to sites due to differences in timing of data collection. year 2011. However, our studied trees largely differed in age across For drought index, we initially considered the De Martonne sites (Table S2). Hence, ring width data were transformed into di- Index (De Martonne, 1926), the Standardized Precipitation Index mensionless ring width indices (RWI) with both age-related growth (McKee, Doesken, & Kleist, 1993), and the Standardized Precipitation trends and lower frequency variation removed from the time series Evapotranspiration Index (SPEI; Vicente-Serrano, Beguería, & López- (Cook & Kairiukstis, 1990). For this, ring width data were detrended Moreno, 2010). The SPEI had a stronger correlation with RWI than by fitting a negative exponential curve or using a 30 year cubic spline the other indices examined for most of the sites (see Table S5). Hence, with a 50% frequency cutoff (Cook & Kairiukstis, 1990). In addition SPEI was used for defining the drought and nondrought years. to these detrending methods, we also converted the raw ring width The SPEI is a unitless drought index, which takes into account data into basal area increment (cm2 per year; Biondi & Q uedan, 2008) both precipitation and potential evapotranspiration effects in the using the dplR package in R (Bunn et al., 2018). We assessed the calculation of the CWB, and is commonly used in the literature for suitability of these approaches to disentangle the drought effects identifying and characterizing drought and nondrought years (e.g., on tree growth by computing the correlation coefficient with the Bottero et al., 2017; Gazol et al., 2018). The potential evapotrans- drought indices (cf. next section) and by characterizing the trend piration was calculated using the Thornthwaite function of the R over a 50 year period (Table S3; Figure S2). The results showed package SPEI (Begueria & Vicente-Serrano, 2013). The SPEI was that the negative exponential detrending method performed best then calculated from CWB using the spei function of the R package among the used approaches in terms of the magnitude of correlation SPEI (Begueria & Vicente-Serrano, 2013). For each site, we calcu- with the drought and of capturing the long-term trends (Table S3; lated SPEI of various timescales that is, integrated over 1–15 months Figure S2). We thus used the detrended negative exponential RWI in order to represent different lengths of the growing season or at (hereafter referred to as RWI) for the analysis. least different growth sensitive periods within the current and the BOSE et al. | 4525 previous growing season. We assessed the Pearson correlation be- and pattern (i.e., type of correlation) of influences of drought on tween RWI and SPEIs (i.e. the different time intervals) for identifying RWI, and how that magnitude and pattern of correlation varied the most relevant SPEI (i.e., most sensitive time interval) for each across the latitudinal gradient examined in this study. The selected site to define the drought and nondrought years (see Table S6). The SPEI timescales were August 15 (i.e., from previous June to current resulting SPEIs (i.e., those best correlated with RWI) are presented August), May 12 (i.e., from previous June to current May), May 9 (i.e., in the Table S7. from previous September to current May), May 6 (i.e., from previous For identifying the extreme drought year of a site, we selected December to current May), May 3 (i.e., spring, from current March to the year with the lowest SPEI value. For each site, we first selected current May), August 6 (i.e., from current March to current August), the extreme drought years for the period of 1980–2011. We then August 3 (i.e., summer, from current June to current August), and selected the extreme drought year for the period of 1980–1999 and November 6 (i.e., from current June to current November). for the period of 2000–2011. 2.8 | Research question 2: Tree growth rate in 2.5 | Identification of drought, predrought, and drought and nondrought years postdrought periods For understanding the absolute tree radial growth performances We characterized drought periods by single or multiple years based during drought and nondrought years, we modeled absolute tree on SPEI ≤ −1.00 and predrought or postdrought periods (i.e., without radial growth (non-detrended tree ring width) as a function of site drought) based on SPEI ≥ −0.99. We limited the predrought and post- types (three levels: LEHP, MELP, and HELP), drought status (two drought periods to a maximum of 3 years, but for drought periods we levels: drought years and nondrought years), and the interaction considered all consecutive years with SPEI ≤ −1.00 (see Table S6). between site types and drought status. For understanding the po- We identified the most extreme droughts during 1980–1999, and tential role of tree age on absolute tree radial growth, we considered during 2000–2011 for all study sites (see Table S6) for comparing tree age as a covariate in this analysis. the tree growth responses to extreme droughts during the recent decade (2000–2011) with the previous two decades (1980–1999). Since many sites had no drought during 1990–1999, we decided to 2.9 | Research question 3: Temporal change in tree enlarge the earlier period back until 1980. growth resilience to extreme droughts We modeled resistance, recovery, and resilience as a function of 2.6 | Tree growth resilience indices time period (two levels: 1980–1999 and 2000–2011), site types (three levels: LEHP, MELP, and HELP), and the interaction between For tree growth resilience, we computed three resilience indices time period and site types. as suggested by Lloret et al. (2011): resistance, recovery, and resil- ience. The resistance quantifies the ratio between growth during a drought period and growth during the preceding nondrought period, 2.10 | Research question 4: Factors affecting tree representing thus the capacity of the trees to buffer the stress and growth resilience to extreme drought maintain growth during drought. The recovery quantifies the growth reaction following the drought period and is defined by the ratio be- For this research question, we selected the most extreme drought tween growth during the postdrought period and growth during the during the entire 1980–2011 study period and used the correspond- drought period. The resilience quantifies the ratio between growth ing resistance, recovery, and resilience indices as response variables during the postdrought period and growth during the predrought in a mixed-effects model (cf. next section). We considered several period, which represents the capacity of trees to recover and re- tree-, site-, and drought-level explanatory variables and various two- gain the growth of the predrought period. We quantified resistance, way interaction terms (see Table S1). The variables included tree size recovery, and resilience for all trees of all sites during the most ex- (i.e., tree diameter at breast height [DBH] inside bark at the drought treme droughts in 1980–1999 and in 2000–2011 (see Table S6). year), tree growth, and tree growth variability prior to drought repre- senting the average and standard deviation of RWI, respectively of 10 consecutive years prior to the extreme drought excluding the years 2.7 | Research question 1: Impact of seasonal considered as predrought period for quantifying the three resilience drought (SPEI) on tree growth indices, site types, elevation, latitude, drought severity (measured by the average SPEI during the drought period), drought duration Based on the results of preliminary analysis (i.e., correlation between (measured by the length of the drought period in years), and drought RWI and different SPEIs), we identified the eight best correlated frequency (measured by the number of drought years (SPEI ≤ −1.00) SPEIs for understanding the magnitude (i.e., degree of correlation) within 10 years preceding the maximum drought period). 4526 | BOSE et al. F I G U R E 2 Seasonal correlations between Scots pine tree-ring width indices and the Standardized Precipitation Evapotranspiration Index (SPEI) for the period of approximately 1960–2011 across the latitudinal gradient. Only the seasons that exhibited the strongest effect on tree-ring width indices are plotted (see Section 2). Note. ‘previous’ refers to the year previous to tree ring formation, while ‘current’ refers to the current year of ring formation, summer: June, July, and August, spring: March, April, and May, autumn: September, October, and November, and winter: December, January, and February. HELP, high-elevation sites with low productivity; LEHP, low-elevation sites with high productivity; MELP, mid-elevation sites with low productivity. Pearson's product-moment correlation with a threshold <0.05 was used for statistical significance. Correlation magnitude: the larger the circles, the stronger the correlations. See Table S7 for correlation scores that are displayed in this figure BOSE et al. | 4527 2.11 | Statistical analyses sites was driven by SPEI of spring (March–May), summer (June–August), and spring and summer combined. Tree growth in HELP sites was either We used a linear mixed-effect modeling approach for research ques- nonrelated or negatively correlated with different time periods of the tion 2, 3, and 4 in which our variables of interest were considered SPEI (Figure 2; Table S3). Overall, the magnitude of correlation between as fixed effects and trees nested within sites were considered as RWI and different SPEIs was higher for MELP than the two other site random effects. The modeling was performed using the function types (Table S3). Three sites of HELP site type had a negative correla- lme of the R package nlme (Pinheiro & Bates, 2000; Pinheiro, Bates, tion with SPEIs while one site of HELP site type was not significantly DebRoy, & Sarkar, 2014). The response variables were log-trans- correlated with any SPEI considered in our analysis (Table S3). formed to normalize residuals and homogenize variances and we checked the assumptions of normality of the residuals and homo- geneity of the variances. Preliminary analysis indicated that an ad- 3.2 | Tree growth rate in drought and nondrought ditional error structure to account for plot spatial autocorrelation did years not improve model performance and thus was not incorporated into the final model. We also assessed potential multicollinearity among In drought and nondrought years, tree radial growth was higher at LEHP explanatory variables using the Variance Inflation Factor (VIF) and than at HELP and at MELP sites (Figure 3). The MELP sites had signifi- discarded variables when VIF > 2.0. The VIF was calculated using the cantly lower tree radial growth in drought years than in nondrought function vif of the R package car (Fox & Weisberg, 2011). The post years (Table S8; Figure 3). Contrary to MELP, tree growth was not signif- hoc Tukey multiple comparison test was performed to detect the icantly different between drought and nondrought years at LEHP and statistical differences (Hothorn, Bretz, & Westfall, 2008). at HELP sites (Table S8; Figure 3). Tree age was negatively associated For research question 4, we used the information-theoretic ap- with the radial growth (p < .0001) irrespective of site types (Table S8). proach (Burnham & Anderson, 2002; Johnson & Omland, 2004), which provides a measure of strength for each candidate model that represents a plausible hypothesis relative to the entire set of candidate 3.3 | Temporal change in tree growth resilience to models considered (Mazerolle, 2006). In the context of our research extreme droughts question (i.e., what are the factors driving the tree growth resilience to extreme drought?), we considered 16 hypotheses (i.e., candidate Tree growth resistance to extreme drought for all site types (i.e., models; Table S1), which were developed based on the current under- HELP, LEHP, and MELP) did not change over the two periods (i.e., standing resulted from different studies that examined tree growth resilience to extreme droughts. Model selection was performed using the AICcmodavg package of R (Mazerolle, 2011). Candidate mod- els were compared using Akaike's information criterion corrected for small sample sizes (AICc). Akaike weights were computed to as- sess the support for each model. We used multimodel inference to compute the model-averaged estimates of the explanatory variables and their 95% confidence intervals (Burnham & Anderson, 2002). A confidence interval excluding 0 indicated that the corresponding ex- planatory variable had an effect on the response variable (Burnham & Anderson, 2002; Mazerolle, 2006). In addition to our candidate mod- els we also considered a null model and a full model. The coefficient of variation (R2) for fixed and random effects were calculated using the function r.squaredGLMM of the MuMIn package in R (Bartoń, 2013). The background map of Figure 2 was downloaded using the function map_data from ggmap package in R (Kahle & Wickham, 2013). F I G U R E 3 Mean annual radial growth in drought and non- 3 | R E S U LT S drought years for the period of approximately 1980–2011 across the three site types (i.e., LEHP, low-elevation sites with high 3.1 | Impact of seasonal drought (SPEI) on tree growth productivity; MELP, mid-elevation sites with low productivity; HELP, high-elevation sites with low productivity). Error bars represent the mean ± standard error (n = 615). Letters on top of Our results showed significant differences in the response of tree the bars show the results (a < b < c) of the post hoc Tukey multiple growth to the different time periods of SPEI. The current year summer comparison test with a threshold <0.05 for statistical significance to autumn (June–November) SPEI significantly controlled tree growth at indicating the differences among the three site types and between LEHP sites of northern Germany (Figure 2), while tree growth at MELP non-drought years and drought years within each site type 4528 | BOSE et al. 1980–1999 and 2000–2011; Figure 4a). Nevertheless, resistance In 1980–1999, recovery was significantly higher at MELP com- was higher at HELP than at LEHP, and higher in the latter compared pared to the two other site types irrespective of period (Table S9; to MELP, irrespective of the period (Table S9; Figure 4a). Figure 4b). Tree growth recovery changed significantly over the two peri- Tree growth resilience changed significantly over the two pe- ods for all site types, where recovery decreased from 1980–1999 riods for LEHP and MELP sites, but not for HELP sites. Resilience to 2000–2011 at MELP and HELP sites, while increased from decreased from 1980–1999 to 2000–2011 at MELP sites, while it 1980–1999 to 2000–2011 at LEHP sites (Table S9; Figure 4b). increased from 1980–1999 to 2000–2011 at LEHP sites (Table S9; Figure 4c). 3.4 | Factors affecting tree growth resilience to extreme drought The model that included additive and interaction effects of all vari- ables considered in the analysis had full support of Akaike weight for resistance (Table 1). A lower resistance was associated with higher predrought growth rate (Table 2). In addition, a lower resistance was associated with greater drought frequency, and with longer drought but depending upon predrought growth rate (Table 2). Resistance was higher at HELP and LEHP sites than at MELP sites (Table 2; Figure 5a). The model that included drought severity and site types, and the interaction between the two variables had the highest support of Akaike weight for recovery (0.74; Table 1). Recovery was lower at HELP and LEHP sites than at MELP sites (Table 2; Figure 5b). In ad- dition, the recovery was higher where trees experienced a higher frequency of droughts (Table 2). The model that included predrought growth rate and drought du- ration, and the interaction between the two variables had the high- est support of Akaike weight for resilience (Table 1). Resilience was negatively associated with predrought growth rate and predrought growth variability and there was no difference across the three site types (Table 2; Figure 5c). 4 | D I S CU S S I O N Using tree ring width data from 30 sites along a 2,800 km latitudinal gradient across Europe we analyzed whether tree growth resilience to extreme drought depended on the geographical location of the tree (Isaac-Renton et al., 2018) and if resilience to extreme drought de- creased over time due to more frequent drought events in recent years (Serra-Maluquer et al., 2018; Spinoni, Naumann, Carrao, Barbosa, & Vogt, 2014). We examined these questions on Scots pine, one of the most widely distributed tree species in the world which is also con- sidered vulnerable to extreme drought conditions (Camarero, Gazol, F I G U R E 4 Tree-level resistance (a), recovery (b), and resilience (c) to the most extreme drought during 1980–1999 and during Sangüesa-Barreda, et al., 2015; Galiano et al., 2010; Matías, Linares, 2000–2011 for three site types. Error bars represent the Sánchez-Miranda, & Jump, 2017; Rigling et al., 2013). Our study mean ± standard error (n = 615). Letters on top of the bars show shows that tree-level resilience to drought was not dependent on the the results (a < b < c) of the post hoc Tukey multiple comparison latitudinal location, but rather on the type of site they were growing test with a threshold <0.05 for statistical significance indicating at and their growth performance (i.e., magnitude and variability of the differences among the three site types and between the two periods within each site type. HELP, high-elevation sites with low growth) during the predrought period. Our results indicate that trees productivity; LEHP, low-elevation sites with high productivity; with higher magnitude and variability in growth are more vulnerable MELP, mid-elevation sites with low productivity to long and severe droughts. In addition, we found that tree growth BOSE et al. TA B L E 1 Results of the best models explaining tree growth resistance, recovery, and resilience of Scots pine trees along the studied gradient. From the 16 tested models, only the three with the highest Akaike's information criterion (AICc) weight are presented R2 (fixed and 2 Models Hypotheses References AICc ∆AICc AICc weight R (fixed) random) Resistance (RT) Tree resistance to drought is affected by RT~all variables Full model 117.9 0.0 1.00 .33 .49 RT~D_INT*PGR+D_INT*ST The intensity of the drought, but Adapted from Gazol et al. (2018) 130.4 12.5 0.00 .22 .46 depending upon the growth prior to drought and site types RT~D_FRE*PGR The frequency of the drought, but Adapted from Gao et al. (2018) 130.5 12.6 0.00 .14 .47 depending upon the growth prior to drought Recovery (RC) Tree recovery after drought RC~D_INT*ST Is affected by the intensity of the Adapted from Gazol et al. (2018) 301.9 0.0 0.74 .16 .36 drought, but depending upon the site types RC~ST Decreased with site types Sánchez-Salguero et al. (2018) 307.0 5.1 0.06 .07 .37 RC~D_INT+D_FRE+D_DUR Is affected combinedly by intensity Gao et al. (2018) 308.4 6.5 0.03 .08 .37 of drought, duration of drought, and frequency of drought Resilience (RS) Tree resilience to drought is RS~D_DUR*PGR Affected by the duration of the Adapted from Taeger et al. (2013) 342.4 0.0 0.45 .03 .35 drought, but depending upon the growth prior to drought RS~PGR Negatively associated with the growth Zang et al. (2014); Ruijven and 344.3 1.9 0.17 .02 .35 prior to drought Berendse (2010) RS~D_FRE*PGR Affected by the frequency of the Adapted from Gao et al. (2018) 344.6 2.2 0.15 .02 .36 drought, but depending upon the growth prior to drought Note: PGR = average tree growth (ring width indices) prior to drought, ST = site type (LEHP (low-elevation sites with high productivity), MELP (mid-elevation sites with low productivity), and HELP (high-elevation sites with low productivity)), D_FRE = drought frequency measured by the number of drought years within the past 10 years from the studied drought, D_INT = intensity of drought, and D_DUR = duration of drought, *indicates an interaction term and +indicates an additive term, PGR was quantified from tree growth during the 10 consecutive years prior to drought excluding the years considered as predrought period quantifying the three indices (i.e., resistance, recovery, and resilience). | 4529 4530 | BOSE et al. TA B L E 2 Log-transformed estimates Types of Estimate Lower Upper of predictor variables and 95% confidence Parameters effect (β) 95% CI 95% CI intervals (CI) based on model averaging for Resistance the three response variables resistance, Predrought growth Additive −0.13 −0.22 −0.04 recovery, and resilience. Only predictor variables that had a strong effect (i.e., a HELP versus MELP Additive 0.31 0.09 0.52 95% confidence interval excluding 0) are LEHP versus MELP Additive 0.19 0.02 0.36 presented Drought intensity*predrought growth Interaction 0.005 0.002 0.008 Drought duration*predrought growth Interaction −0.22 −0.35 −0.09 Drought frequency Additive −0.13 −0.20 −0.06 Recovery HELP versus MELP Additive −0.23 −0.45 −0.02 LEHP versus MELP Additive −0.16 −0.34 −0.01 Drought frequency Additive 0.08 0.01 0.16 Resilience Predrought growth Additive −0.18 −0.28 −0.07 Predrought growth variability Additive −0.22 −0.43 −0.01 HELP, high-elevation sites with low productivity; LEHP, low-elevation sites with high productivity; MELP, mid-elevation sites with low productivity. resilience to extreme drought was lower during 2000–2011 than dur- changes in precipitation or evaporative demand during these periods ing 1980–1999 at mid-elevation and lower productivity sites. These re- will directly affect the water availability of the trees and thus their sults may indicate that more frequent drought events that occurred in physiological responses, resulting in lower growth (Cabon, Peters, 2000–2011 than in earlier period make Scots pine trees more vulner- Fonti, Martínez-Vilalta, & De Cáceres, 2020). A tree at a less dry site able to extreme droughts. However, we found high variability in tree- in contrast is growing mostly without water limitations. During such level responses (Figure 5) as detected by previous studies (e.g., Gazol periods with plenty of water supply there will be no strong direct et al., 2018; Maes et al., 2019). This high tree-level variability (Figure 5) effect of precipitation (and at least within a certain range of evapo- and low marginal R2 values (Table 2) may indicate that combining tree rative demand) on tree physiology and on growth (Martínez-Vilalta information from different sites without direct measurements of local et al., 2009; Sterck, Zweifel, Sass-Klaassen, & Chowdhury, 2008). site-related factors (e.g., soil water content, nutrient availability, stand Thus, it is reasonable to consider that the climatic control of tree stocking, and rooting depth) compromises the predictive power of the growth is stronger at MELP which are relatively drier than LEHP sites models (DeSoto et al., 2020; Gazol et al., 2018; Grossiord et al., 2014). (Table S2). However, we need to improve our understanding on local drought characteristics because the CWB and SPEI do not consider the soil water holding capacity and the depth of water table which 4.1 | Impact of seasonal drought on tree growth are important parameters for estimating the soil water available for trees (Zang et al., 2020). Our results reveal important seasonal differences across lower and higher productivity sites in terms of SPEI-RWI correlations. Scots pine trees growing at mid-elevation lower productivity (MELP) 4.2 | Tree growth rate in drought and sites showed a greater sensitivity to spring–summer SPEI while nondrought years trees growing at low-elevation higher productivity (LEHP) sites of northern Germany were more sensitive to summer–autumn SPEI We found higher annual growth among trees at lower-elevation sites (Figure 2). This is consistent with results from Pasho et al. (2011), than trees at higher-elevation sites (Figure 3). Trees at low-elevation who studied eight tree species including Scots pine for the period of sites were relatively younger than trees at high-elevation sites and 1950–2005 using the Standardized Precipitation Index at timescales thus more likely in their full growth ontogenetic stage. However, from 1 to 48 months. Our results showed that Scots pine trees grow- Scots pine trees at high-elevation sites had no growth difference ing in MELP are more sensitive to drought (measured by SPEI) than between our selected drought and nondrought years (Figure 3) ir- trees growing in LEHP and HELP sites (Figure 2; Table S3), which is respective of the fact that they experienced similar drought severity consistent with the findings of other studies on Scots pine (Lévesque, and frequency as the two other site types in terms of SPEI (Figure 6). Rigling, Bugmann, Weber, & Brang, 2014; Pasho et al., 2011; Therefore, tree growth at high-elevation Scots pine forests is proba- Sánchez-Salguero et al., 2015). Trees at MELP sites are likely to grow bly controlled by other factors than changes in water availability dur- under water limitation during extended periods of the year, and ing drought (Carrer, Nola, Eduard, Motta, & Urbinati, 2007; Cudlín BOSE et al. | 4531 F I G U R E 6 Drought severity (measure by the SPEI [Standardized Precipitation and Evapotranspiration Index]) (a) and drought frequency (measured by the number of droughts that occurred within 10 years preceding extreme drought) (b) during and prior to the examined extreme drought, respectively. Error bars represent the mean ± standard error (n = 30). Letters on top of the bars show the results (a < b < c) of the post hoc Tukey multiple comparison test with a threshold <0.05 for statistical significance. HELP, high- elevation sites with low productivity; LEHP, low-elevation sites with high productivity; MELP, mid-elevation sites with low productivity F I G U R E 5 Tree growth resistance (a), recovery (b), and resilience (c) to the most extreme drought during 1980–2011 with than to water availability (Hagedorn et al., 2014). Our high-elevation 95% confidence intervals. Note. Average pre-drought growth site type has three sites with trees that were older than 150 years (ring width indices) was quantified from tree growth during the 10 during our selected drought period and older than the trees at any consecutive years prior to drought excluding the years considered other site examined in our analysis. Besides, the mean annual tem- as pre-drought period for quantifying the three indices (resistance, recovery, and resilience). HELP, high-elevation sites with low perature at four out of six high-elevation sites was <7.0°C which is productivity; LEHP, low-elevation sites with high productivity; almost 2°C cooler than at any other site examined in our analysis. MELP, mid-elevation sites with low productivity. See statistics for Trees in our high-elevation sites are perhaps well adapted to such the fitted line in Tables 1 and 2 low mean annual temperature and thus short growing season, con- firming the findings of other studies conducted in southern Europe et al., 2017). In our dataset, three out of six high-elevation sites had (Peñuelas, Hunt, Ogaya, & Jump, 2008) and in southwestern United negative correlation with SPEI (Table S3) indicating that tree radial States (Adams & Kolb, 2004). It is also important to mention that our growth at high-elevation sites examined in our study was more re- drought intensity index may not fully capture the absolute water bal- lated to other factors such as ontogeny and temperature (Camarero, ance differences between drought and nondrought years because of Gazol, Galván, Sangüesa-Barreda, & Gutiérrez, 2015; Körner, 2003) the elevational differences between the sites and climate stations 4532 | BOSE et al. (Table S4), which may partly explain the lack of growth differences Our results showed that the top-ranked model (according to between drought and nondrought years for our high-elevation sites. Akaike weights) for all the three resilience indices (resistance, re- covery, and resilience) included interaction effects between tree and drought-related variables (Table 1). This means that the im- 4.3 | Temporal change in tree growth resilience to pact of drought on tree-level resilience is not independent, but extreme droughts rather dependent on how the trees were growing (i.e., magnitude and variability of growth) during the predrought period and on the Although Scots pine is a relatively drought-tolerant species, our re- type of site they were growing at. In addition to drought severity, sults showed a decreased growth recovery and resilience to extreme which was usually found to significantly affect tree-level resilience drought at mid-elevation and lower productivity sites. This is largely at various spatial scales variables (Gazol et al., 2017, 2018; Zang consistent with the patterns reported by Serra-Maluquer et al. (2018) et al., 2014), we also considered drought duration and drought fre- in three pine species (including Scots pine) growing in NE Spain. In our quency as explanatory variables as suggested by Gao et al. (2018). case, this result likely reflects the higher frequency and severity of We indeed found a significant negative effect of drought frequency droughts events prior to the extreme drought occurring in 2000–2011 on tree resistance (Table 2), suggesting that trees that experienced than prior to the extreme drought occurring in 1980–1999 (Figure 6). more frequent droughts were less resistant to extreme droughts. Moreover, trees became older and larger over time. A larger tree car- Therefore, trees that display higher variability in growth are not ries higher nonphotosynthetic biomass, which requires a greater in- only sensitive to extreme drought but also to frequent drought oc- vestment for defense and maintenance (Ryan & Yoder, 1997; Scholz, currence as for example, shown by Seidel et al. (2016) for different Phillips, Bucci, & Goldstein, 2011) and increasing tree height may in- Scots pine provenances. Higher aboveground biomass growth can crease hydraulic constraints and xylem cavitation risks under drought be related to lower or at least nonproportional biomass allocation (Olson et al., 2018). This reduced tree resilience over time may chal- to roots (Gessler et al., 2017) and might consequently increase tree- lenge the physiological potential for acclimation to more intensive level sensitivity to upcoming drought periods (Martínez-Vilalta, and frequent drought events that we expect in the future (Dai, 2012). López, Loepfe, & Lloret, 2012). Long or repeated droughts may re- The decreased tree growth resilience at MELP sites may indicate that duce the number of living branches and, because of needle multi- drought hardening (i.e., physiological processes by which a tree be- year life span, the leaf area per branch (Galiano, Martínez-Vilalta, comes more acclimate to drought conditions; Villar-Salvador, Peñuelas, & Lloret, 2011; Vennetier et al., 2013). This holds back short-term & Jacobs, 2013) is not very important and does not allow trees to ac- tree leaf area recovery and may drive tree growth to very low lev- climate to frequent and prolonged drought events irrespective of their els during and after drought. Although a reduced leaf area limits drought experiences. water stress during and after droughts and may favor resilience, it In contrast to MELP sites, tree growth recovery and resilience to also hampers carbohydrate reserves buildup and may lead to car- extreme droughts increased at LEHP sites during the recent period bon starvation. Accordingly, higher growth variability may be re- (i.e., 2000–2011) compared to the previous period (i.e., 1980–1999). lated to a higher vulnerability to the upcoming stresses (Cailleret Although drought severity and drought frequency did not change et al., 2019; DeSoto et al., 2020; McDowell et al., 2008; Ogle, significantly at LEHP sites over these two periods (Figure 6), the Whitham, & Cobb, 2000). For instance, Ogle et al., (2000) noticed duration of the extreme drought was longer in 1980–1999 than in a 1.5 times higher growth variability in dead pinyon pine trees rel- 2000–2011. For example, seven sites of the LEHP site type expe- ative to surviving ones in drought years preceding mortality across rienced a 2 year long drought during 1980–1999, while none of the forests of southwestern United States. droughts occurring during 2000–2011 was longer than 1 year at that Although the predrought growth of Scots pine had a strong site type (Table S6). relationship with resistance, it had no-relationship with recovery. This lack of relationship between predrought growth and recov- ery resulted in a relatively weak relationship (although statistically 4.4 | Factors affecting tree growth resilience to significant) between predrought growth and resilience (Figure 5), extreme drought because the resilience is mathematically related to both resistance and recovery (Lloret et al., 2011), and in the context of our study, Several studies conducted at local to global scales provided un- recovery was more closely related to resilience than the relation- derstanding on tree, site, and drought-related variables influenc- ship between resistance and resilience (Figure S3). Scots pine trees ing tree-level resilience to extreme drought (e.g., Gao et al., 2018; growing at MELP sites displayed lower resistance but higher recov- Gazol et al., 2017, 2018; Sánchez-Salguero et al., 2018; Taeger, ery than trees growing at LEHP and HELP sites (Table 2; Figure 5), Zang, Liesebach, Schneck, & Menzel, 2013; Vitali et al., 2017; Zang suggesting different growth strategies exercised by trees from dif- et al., 2014). To our knowledge, however, none of these studies com- ferent sites to cope with drought (Sánchez-Salguero et al., 2018). pared explicitly the model performances with and without interac- However, resistance and recovery are relative indices and do not tion effects among all these variables. allow comparison across trees in terms of their absolute growth performances. Although resistance and recovery hold a negative BOSE et al. | 4533 mathematical relationship (Lloret et al., 2011), they provide use- location-specific tree-, site-, and drought-related factors and their ful insights on disentangling trees that tried to remain firm during interactions drive tree growth performances. drought years (i.e., higher resistance and lower recovery) from trees that tried to conform drought impact (i.e., lower resistance and AC K N OW L E D G E M E N T S higher recovery; Gazol et al., 2017). These two different strategies The study was supported by a Marie Skłodowska-Curie Individual to extreme droughts by trees from high productivity and low pro- Fellowship (PROJECT ID: 749051-REFOREST) to A.K.B. R.S-.S. ductivity sites are also reported by other local (Vitali et al., 2017; was supported by Postdoctoral grant (IJCI-2015-25845, FEDER Zang et al., 2014) and global-scale studies (Gazol et al., 2017). For funds), J.J.C., J.C.L., and R.S-.S. were supported by the RTI2018- instance, a multispecies comparison across Spain found higher 096884-B-C31 and RTI2018-096884-B-C33 projects (Ministry of resistance for species dominating mesic sites than for species Science, Innovation and Universities, Spain) and VULBOS project from more xeric sites, which presented a higher recovery (Gazol (UPO-1263216, FEDER Funds, Andalusia Regional Government, et al., 2018). At our LEHP, Scots pine trees were on average taller Consejería de Economía, Conocimiento, Empresas y Universidad and larger than MELP sites (Table S2). These larger trees in mesic 2014-2020), A.H. by PinCaR project (UHU-1266324, FEDER sites most likely developed larger crowns, which might be able to Funds, Andalusia Regional Government, Consejería de Economía, still provide sufficient photosynthates to sustain growth during the Conocimiento, Empresas y Universidad 2014-2020), and A.M. by drought period. Alternatively, these larger trees had sufficient re- the Bavarian Ministry of Science via the Bavarian Climate Research serves to compensate for the drought-induced reduction in pho- Network (bayklif). Part of the sampling was funded within the pro- tosynthesis. Both mechanisms would result in a higher resistance ject DENDROKLIMA by the German Waldklimafond (FKZ 28W- compared to trees in lower productivity sites (Martínez-Vilalta C-4-077-01) and ST327 by the Bavarian State Ministry for Food, et al., 2009). However, taller trees associated with larger crowns Agriculture, and Forestry. We thank Kerstin Treydte, Richard Peters, and concomitantly greater water demand can be more vulnerable and Stefan Klesse for providing valuable supports on tree ring to drought-induced hydraulic failure (McDowell et al., 2008; Olson width detrending methods, David Forrester for providing informa- et al., 2018) and the required postdrought investment of assimilates tion on site productivity index, Tanja Sanders for providing valuable for restoring their hydraulic system could slow down their recovery comments on earlier version of the manuscript, and Landesforst process (Brodribb, Bowman, Nichols, Delzon, & Burlett, 2010). Mecklenburg-Vorpommern, Landeskompetenzzentrum Forst In this analysis, we considered tree, site, and drought level Eberswalde, and Nordwestdeutsche Forstliche Versuchsanstalt for variables and their interactions, but available data did not allow to maintaining the intensive forest monitoring sites (Level II) used in quantify the effects of forest stand-level variables including tree- this study, Javier Donés (Director of Montes de Valsaín), National to-tree competition (Serra-Maluquer et al., 2018), site conditions Parks Autonomous Agency (OAPN), Junta de Castilla y León forest including soil and topography (Vennetier, Ripert, & Rathgeber, 2018; guards and all the participants involved in the “International Tree- Zalloni, Battipaglia, Cherubini, Saurer, & De Micco, 2019), species Ring Summer School” celebrated in 2012 in Valsaín (Segovia-Spain), mixtures (Pretzsch, Schütze, & Uhl, 2013), trees neighborhood com- and Ramzi Touchan and Dave Meko. We also thank all colleagues position (Grossiord, 2019), and stand stocking (Bottero et al., 2017; who were involved in field sampling and laboratory measurements D'Amato, Bradford, Fraver, & Palik, 2013), which will be the sub- for the other sites. ject of subsequent analyses. Moreover, we did not include possible subspecies variation due to local evolutionary adaptation induced C O N FL I C T O F I N T E R E S T by drought-related selection (Hampe & Petit, 2005). We detected None is declared. large tree-level variability within and across sites, and a larger pro- portion of the variance of our models was explained by the random DATA AVA I L A B I L I T Y S TAT E M E N T effects (i.e., trees nested within sites) than by the fixed effect vari- The data of this study are available from the corresponding author ables (Table 1). Therefore, these points need to be considered when upon reasonable request. interpreting our results. To conclude, we show that tree radial growth responses of Scots ORCID pine to extreme drought depend on site condition, tree growth prior Arun K. Bose https://orcid.org/0000-0001-8581-1651 to drought, and the number of droughts that a tree experienced Arthur Gessler https://orcid.org/0000-0002-1910-9589 within the 10 years before the selected drought. Our study iden- Andreas Bolte https://orcid.org/0000-0003-4106-0387 tified a reduced tree growth resilience to extreme drought in Scots Allan Buras https://orcid.org/0000-0003-2179-0681 pine trees growing at mid-elevation and low productivity sites likely Maxime Cailleret https://orcid.org/0000-0001-6561-1943 driven by the more frequent and severe drought events that occurred J. Julio Camarero https://orcid.org/0000-0003-2436-2922 at those sites in recent years. We show that the assessment of tree Ana-Maria Hereş https://orcid.org/0000-0002-1839-1770 growth responses in terms of resistance, recovery, and resilience Andrea Hevia https://orcid.org/0000-0001-6352-1442 to extreme drought using radial growth data is challenging along Mathieu Lévesque https://orcid.org/0000-0003-0273-510X large ecological and biogeographical gradients, since a multitude of Juan C. Linares https://orcid.org/0000-0001-8375-6353 4534 | BOSE et al. Jordi Martinez-Vilalta https://orcid.org/0000-0002-2332-7298 Bunn, A., Korpela, M., Biondi, F., Campelo, F., Merian, P., Qeadan, F., & Zang, C. (2018). dplR: Dendrochronology program library in R. R package ver- Luis Matías https://orcid.org/0000-0001-5603-5390 sion 1.6.4. Retrieved from https://CRAN.R-proje​ct.org/packa​ge=dplR Annette Menzel https://orcid.org/0000-0002-7175-2512 Buras, A., & Menzel, A. (2019). Projecting tree species composition Raúl Sánchez-Salguero https://orcid.org/0000-0002-6545-5810 changes of European forests for 2061–2090 under RCP 4.5 and Matthias Saurer https://orcid.org/0000-0003-3977-2166 RCP 8.5 scenarios. Frontiers in Plant Science, 9, 1986. https://doi. org/10.3389/fpls.2018.01986 Daniel Ziche https://orcid.org/0000-0001-9031-051X Buras, A., Schunk, C., Zeiträg, C., Herrmann, C., Kaiser, L., Lemme, H., Andreas Rigling https://orcid.org/0000-0003-1944-4042 … Menzel, A. (2018). Are Scots pine forest edges particularly prone to drought-induced mortality? Environmental Research Letters, 13, REFERENCES 025001. https://doi.org/10.1088/1748-9326/aaa0b4 Burnham, K. P., & Anderson, D. R. (2002). Model selection and multimodel Adams, H. D., & Kolb, T. E. (2004). Drought responses of conifers in ecotone inference: A practical information-theoretic approach (2nd ed.). New forests of northern Arizona: Tree ring growth and leaf δ13C. Oecologia, York: Springer-Verlag. 140(2), 217–225. https://doi.org/10.1007/s0044​2-004-1585-4 Cabon, A., Peters, R. L., Fonti, P., Martínez-Vilalta, J., & De Cáceres, Adams, H. D., Zeppel, M. J. B., Anderegg, W. R. L., Hartmann, H., M. (2020). Temperature and water potential co-limit stem cambial Landhäusser, S. M., Tissue, D. T., … McDowell, N. G. (2017). A activity along a steep elevational gradient. New Phytologist, 226(5), multi-species synthesis of physiological mechanisms in drought-in- 1325–1340. https://doi.org/10.1111/nph.16456 duced tree mortality. Nature Ecology & Evolution, 1(9), 1285–1291. Cailleret, M., Dakos, V., Jansen, S., Robert, E. M. R., Aakala, T., Amoroso, https://doi.org/10.1038/s4155​9-017-0248-x M. M., … Martínez-Vilalta, J. (2019). Early-warning signals of individ- Anderegg, W. R. L., Klein, T., Bartlett, M., Sack, L., Pellegrini, A. F. A., ual tree mortality based on annual radial growth. Frontiers in Plant Choat, B., & Jansen, S. (2016). Meta-analysis reveals that hydraulic Science, 9, 1964. https://doi.org/10.3389/fpls.2018.01964 traits explain cross-species patterns of drought-induced tree mortal- Camarero, J. J., Gazol, A., Galván, J. D., Sangüesa-Barreda, G., & ity across the globe. Proceedings of the National Academy of Sciences Gutiérrez, E. (2015). Disparate effects of global-change drivers of the United States of America, 113(18), 5024–5029. https://doi. on mountain conifer forests: Warming-induced growth enhance- org/10.1073/pnas.15256​78113 ment in young trees vs. CO2 fertilization in old trees from wet sites. Anderegg, W. R. L., Schwalm, C., Biondi, F., Camarero, J. J., Koch, G., Global Change Biology, 21(2), 738–749. https://doi.org/10.1111/ Litvak, M., … Pacala, S. (2015). Pervasive drought legacies in forest gcb.12787 ecosystems and their implications for carbon cycle models. Science, Camarero, J. J., Gazol, A., Sangüesa-Barreda, G., Oliva, J., & Vicente- 349(6247), 528–532. Serrano, S. M. (2015). To die or not to die: Early warnings of tree Bartoń, K. (2013). MuMIn: multi-model inference. R package version, 1(5). dieback in response to a severe drought. Journal of Ecology, 103(1), Retrieved from https://cran.r-proje​c t.org/web/packa​ges/MuMIn​/ 44–57. https://doi.org/10.1111/1365-2745.12295 index.html Carnicer, J., Coll, M., Pons, X., Ninyerola, M., Vayreda, J., & Peñuelas, J. Begueria, S., & Vicente-Serrano, S. M. (2013). SPEI: Calculation of the stan- (2014). Large-scale recruitment limitation in Mediterranean pines: dardised precipitation-evapotranspiration index. R package version 1.6. The role of Quercus ilex and forest successional advance as key re- Vienna, Austria: CRAN. gional drivers. Global Ecology and Biogeography, 23(3), 371–384. Biondi, F., & Quedan, F. (2008). A theory-driven approach to tree- Carrer, M., Nola, P., Eduard, J. L., Motta, R., & Urbinati, C. (2007). Regional ring standardization: Defining the biological trend from expected variability of climate–growth relationships in Pinus cembra high el- basal area increment. Tree Ring Research, 64, 81–96. https://doi. evation forests in the Alps. Journal of Ecology, 95(5), 1072–1083. org/10.3959/2008-6.1 https://doi.org/10.1111/j.1365-2745.2007.01281.x Bolte, A., Czajkowski, T., Cocozza, C., Tognetti, R., de Miguel, M., Pšidová, Choat, B., Brodribb, T. J., Brodersen, C. R., Duursma, R. A., López, R., E., … Müller, J. (2016). Desiccation and mortality dynamics in seed- & Medlyn, B. E. (2018). Triggers of tree mortality under drought. lings of different European beech (Fagus sylvatica L.) populations Nature, 558(7711), 531–539. under extreme drought conditions. Frontiers in Plant Science, 7, 751. Ciais, P. H., Reichstein, M., Viovy, N., Granier, A., Ogée, J., Allard, V., … https://doi.org/10.3389/fpls.2016.00751 Valentini, R. (2005). Europe-wide reduction in primary productiv- Bose, A. K., Moser, B., Rigling, A., Lehmann, M. M., Milcu, A., Peter, M., … ity caused by the heat and drought in 2003. Nature, 437, 529–533. Gessler, A. (2020). Memory of environmental conditions across gen- https://doi.org/10.1038/natur​e 03972 erations affects the acclimation potential of scots pine. Plant, Cell & Cook, E. R., & Kairiukstis, L. A. (1990). Methods of dendrochronology: Environment, 43(5), 1288–1299. https://doi.org/10.1111/pce.13729 Applications in the environmental sciences. Dordrecht, The Netherlands: Bottero, A., D'Amato, A. W., Palik, B. J., Bradford, J. B., Fraver, S., Kluwer. ISBN-13: 978-0-7923-0586-6. https://doi.org/10.1007/978- Battaglia, M. A., & Asherin, L. A. (2017). Density-dependent vulner- 94-015-7879-0 ability of forest ecosystems to drought. Journal of Applied Ecology, Cudlín, P., Cudlín, P., Cudlín, P., Tognetti, R., Malis, F., Alados, C. L., 54(6), 1605–1614. https://doi.org/10.1111/1365-2664.12847 … Wielgolaski, F. E. (2017). Drivers of treeline shift in different Breshears, D. D., Cobb, N. S., Rich, P. M., Price, K. P., Allen, C. D., Balice, European mountains. Climate Research, 73(1–2), 135–150. https:// R. G., … Meyer, C. W. (2005). Regional vegetation die-off in response doi.org/10.3354/cr01465 to global-change-type drought. Proceedings of the National Academy Dai, A. (2012). Increasing drought under global warming in observations of Sciences of the United States of America, 102(42), 15144–15148. and models. Nature Climate Change, 3, 52. https://doi.org/10.1073/pnas.05057​3 4102 D'Amato, A. W., Bradford, J. B., Fraver, S., & Palik, B. J. (2013). Effects of Brodribb, T. J., Bowman, D. J. M. S., Nichols, S., Delzon, S., & Burlett, R. thinning on drought vulnerability and climate response in north tem- (2010). Xylem function and growth rate interact to determine recov- perate forest ecosystems. Ecological Applications, 23(8), 1735–1742. ery rates after exposure to extreme water deficit. New Phytologist, https://doi.org/10.1890/13-0677.1 188(2), 533–542. https://doi.org/10.1111/j.1469-8137.2010.03 De Martonne, E. (1926). Une nouvelle fanction climatologique. L’indice 393.x D’aridite La Meteriologie, 449–458. Bugmann, H. (1996). Functional types of trees in temperate and boreal DeSoto, L., Cailleret, M., Sterck, F., Jansen, S., Kramer, K., Robert, E. M. forests: Classification and testing. Journal of Vegetation Science, 7(3), R., … Martínez-Vilalta, J. (2020). Low growth resilience to drought 359–370. https://doi.org/10.2307/3236279 BOSE et al. | 4535 is related to future mortality risk in trees. Nature Communications, Grossiord, C. (2019). Having the right neighbors: How tree species diver- 11(1), 545. https://doi.org/10.1038/s4146​7-020-14300​-5 sity modulates drought impacts on forests. New Phytologist, in press. Diéguez-Aranda, U., Álvarez González, J. G., Marcos Barrio, A., & Alberto https://doi.org/10.1111/nph.15667 Rojo, A. (2005). Site quality equations for Pinus sylvestris L. planta- Grossiord, C., Granier, A., Ratcliffe, S., Bouriaud, O., Bruelheide, H., tions in Galicia (northwestern Spain). Annals of Forest Science, 62(2), Checko, E., … Gessler, A. (2014). Tree diversity does not always im- 143–152. https://doi.org/10.1051/fores​t:2005006 prove resistance of forest ecosystems to drought. Proceedings of the Diéguez-Aranda, U., Burkhart, H. E., & Rodríguez-Soalleiro, R. (2005). National Academy of Sciences of the United States of America, 111(41), Modeling dominant height growth of radiata pine (Pinus radiata D. Don) 14812–14815. https://doi.org/10.1073/pnas.14119​70111 plantations in north-western Spain. Forest Ecology and Management, Gruber, A., Strobl, S., Veit, B., & Oberhuber, W. (2010). Impact of drought 215(1), 271–284. https://doi.org/10.1016/j.foreco.2005.05.015 on the temporal dynamics of wood formation in Pinus sylvestris. Etzold, S., Ziemińska, K., Rohner, B., Bottero, A., Bose, A. K., Ruehr, N. Tree Physiology, 30(4), 490–501. https://doi.org/10.1093/treep​hys/ K., … Rigling, A. (2019). One century of forest monitoring data in tpq003 Switzerland reveals species- and site-specific trends of climate-in- Hagedorn, F., Shiyatov, S. G., Mazepa, V. S., Devi, N. M., Grigor'ev, A. duced tree mortality. Frontiers in Plant Science, 10, 307. https://doi. A., Bartysh, A. A., … Moiseev, P. A. (2014). Treeline advances along org/10.3389/fpls.2019.00307 the Urals mountain range – Driven by improved winter conditions? Flexas, J., & Medrano, H. (2002). Drought-inhibition of photosynthesis in Global Change Biology, 20(11), 3530–3543. https://doi.org/10.1111/ C3 plants: Stomatal and non-stomatal limitations revisited. Annals of gcb.12613 Botany, 89(2), 183–189. https://doi.org/10.1093/aob/mcf027 Hampe, A., & Petit, R. J. (2005). Conserving biodiversity under climate Forrester, D. I., Bonal, D., Dawud, S., Gessler, A., Granier, A., Pollastrini, change: The rear edge matters. Ecology Letters, 8(5), 461–467. https:// M., & Grossiord, C. (2016). Drought responses by individual tree spe- doi.org/10.1111/j.1461-0248.2005.00739.x cies are not often correlated with tree species diversity in European Hamrick, J. L. (2004). Response of forest trees to global environmental forests. Journal of Applied Ecology, 53(6), 1725–1734. https://doi. changes. Forest Ecology and Management, 197(1), 323–335. https:// org/10.1111/1365-2664.12745 doi.org/10.1016/j.foreco.2004.05.023 Fox, J., & Weisberg, S. (2011). An {R} companion to applied regression Hereş, A.-M., Martínez-Vilalta, J., & Claramunt López, B. (2012). Growth (2nd edn). Thousand Oaks, CA: SAGE Publications Inc. patterns in relation to drought-induced mortality at two Scots pine Fritts, H. C. (2001). Tree rings and climate. Blackburn, UK. ISBN-13: (Pinus sylvestris L.) sites in NE Iberian Peninsula. Trees, 26(2), 621–630. 978-1-930665-39-2. Hevia, A., Sánchez-Salguero, R., Camarero, J. J., Querejeta, J. I., Sangüesa- Galiano, L., Martínez-Vilalta, J., & Lloret, F. (2010). Drought-induced mul- Barreda, G., & Gazol, A. (2019). Long-term nutrient imbalances linked tifactor decline of Scots pine in the Pyrenees and potential vegeta- to drought-triggered forest dieback. Science of the Total Environment, tion change by the expansion of co-occurring oak species. Ecosystems, 690, 1254–1267. https://doi.org/10.1016/j.scito​tenv.2019.06.515 13(7), 978–991. https://doi.org/10.1007/s1002​1-010-9368-8 Hökkä, H., & Ojansuu, R. (2004). Height development of Scots pine on Galiano, L., Martínez-Vilalta, J., & Lloret, F. (2011). Carbon reserves and peatlands: Describing change in site productivity with a site index canopy defoliation determine the recovery of Scots pine 4 yr after a model. Canadian Journal of Forest Research, 34(5), 1081–1092. drought episode. New Phytologist, 190(3), 750–759. https://doi.org/10.1139/x03-275 Gao, S., Liu, R., Zhou, T., Fang, W., Yi, C., Lu, R., … Luo, H. (2018). Dynamic Hothorn, T., Bretz, F., & Westfall, P. (2008). Simultaneous inference in responses of tree-ring growth to multiple dimensions of drought. general parametric models. Biometrical Journal, 50(3), 346–363. Global Change Biology, 24(11), 5380–5390. https://doi.org/10.1111/ https://doi.org/10.1002/bimj.20081​0 425 gcb.14367 Isaac-Renton, M., Montwé, D., Hamann, A., Spiecker, H., Cherubini, P., Gazol, A., Camarero, J. J., Anderegg, W. R. L., & Vicente-Serrano, S. M. & Treydte, K. (2018). Northern forest tree populations are physio- (2017). Impacts of droughts on the growth resilience of Northern logically maladapted to drought. Nature Communications, 9(1), 5254. Hemisphere forests. Global Ecology and Biogeography, 26(2), 166–176. https://doi.org/10.1038/s4146​7-018-07701​- 0 https://doi.org/10.1111/geb.12526 Jentsch, A., Kreyling, J., Elmer, M., Gellesch, E., Glaser, B., Grant, K., … Gazol, A., Camarero, J. J., Vicente-Serrano, S. M., Sánchez-Salguero, R., Beierkuhnlein, C. (2011). Climate extremes initiate ecosystem-reg- Gutiérrez, E., de Luis, M., … Galván, J. D. (2018). Forest resilience ulating functions while maintaining productivity. Journal of Ecology, to drought varies across biomes. Global Change Biology, 24(5), 2143– 99(3), 689–702. https://doi.org/10.1111/j.1365-2745.2011.01817.x 2158. https://doi.org/10.1111/gcb.14082 Johnson, J. B., & Omland, K. S. (2004). Model selection in ecology and Gessler, A., Schaub, M., & McDowell, N. G. (2017). The role of nutrients evolution. Trends in Ecology & Evolution, 19(2), 101–108. https://doi. in drought-induced tree mortality and recovery. New Phytologist, org/10.1016/j.tree.2003.10.013 214(2), 513–520. https://doi.org/10.1111/nph.14340 Kahle, D., & Wickham, H. (2013). ggmap: Spatial visualization with gg- Giordano, L., Gonthier, P., Varese, G. C., Miserere, L., & Nicolotti, G. plot2. The R Journal., 5, 144–161. https://doi.org/10.32614​/RJ- (2009). Mycobiota inhabiting sapwood of healthy and declining 2013-014 Scots pine (Pinus sylvestris L.) trees in the Alps. Fungal Diversity, 38, Kannenberg, S. A., Maxwell, J. T., Pederson, N., D'Orangeville, L., Ficklin, 69–83. D. L., & Phillips, R. P. (2019). Drought legacies are dependent on Granda, E., Gazol, A., & Camarero, J. J. (2018). Functional diversity dif- water table depth, wood anatomy and drought timing across the ferently shapes growth resilience to drought for co-existing pine eastern US. Ecology Letters, 22(1), 119–127. https://doi.org/10.1111/ species. Journal of Vegetation Science, 29(2), 265–275. https://doi. ele.13173 org/10.1111/jvs.12617 Kaufman, L., & Rousseeuw, P. (1990). Finding groups in data: An introduc- Greenwood, S., Ruiz-Benito, P., Martínez-Vilalta, J., Lloret, F., Kitzberger, tion to cluster analysis. New York: John Wiley & Sons Inc. T., Allen, C. D., … Jump, A. S. (2017). Tree mortality across biomes is Körner, C. (2003). Carbon limitation in trees. Journal of Ecology, 91(1), promoted by drought intensity, lower wood density and higher spe- 4–17. https://doi.org/10.1046/j.1365-2745.2003.00742.x cific leaf area. Ecology Letters, 20(4), 539–553. https://doi.org/10.1111/ Lévesque, M., Rigling, A., Bugmann, H., Weber, P., & Brang, P. (2014). ele.12748 Growth response of five co-occurring conifers to drought across Grissino-Mayer, H. D. (2001). Evaluating crossdating accuracy: A man- a wide climatic gradient in Central Europe. Agricultural and Forest ual and tutorial for the computer program COFECHA. Tree-Ring Meteorology, 197, 1–12. https://doi.org/10.1016/j.agrfo​rmet.2014. Research, 57, 205–221. 06.001 4536 | BOSE et al. Lévesque, M., Saurer, M., Siegwolf, R., Eilmann, B., Brang, P., Bugmann, Academy of Sciences of the United States of America, 115(29), 7551. H., & Rigling, A. (2013). Drought response of five conifer species https://doi.org/10.1073/pnas.17217​28115 under contrasting water availability suggests high vulnerability of Palahı́, M., Tomé, M., Pukkala, T., Trasobares, A., & Montero, G. (2004). Norway spruce and European larch. Global Change Biology, 19(10), Site index model for Pinus sylvestris in north-east Spain. Forest 3184–3199. https://doi.org/10.1111/gcb.12268 Ecology and Management, 187(1), 35–47. https://doi.org/10.1016/ Linares, J. C., Camarero, J. J., & Carreira, J. A. (2010). Competition S0378​-1127(03)00312​-8 modulates the adaptation capacity of forests to climatic stress: Pasho, E., Camarero, J. J., de Luis, M., & Vicente-Serrano, S. M. (2011). Insights from recent growth decline and death in relict stands of the Impacts of drought at different time scales on forest growth across a Mediterranean fir Abies pinsapo. Journal of Ecology, 98(3), 592–603. wide climatic gradient in north-eastern Spain. Agricultural and Forest Lloret, F., Keeling, E. G., & Sala, A. (2011). Components of tree re- Meteorology, 151(12), 1800–1811. https://doi.org/10.1016/j.agrfo​ silience: Effects of successive low-growth episodes in old pon- rmet.2011.07.018 derosa pine forests. Oikos, 120(12), 1909–1920. https://doi.org/ Peñuelas, J., Hunt, J. M., Ogaya, R., & Jump, A. S. (2008). Twentieth cen- 10.1111/j.1600-0706.2011.19372.x tury changes of tree-ring δ13C at the southern range-edge of Fagus Maes, S. L., Perring, M. P., Vanhellemont, M., Depauw, L., Van den Bulcke, sylvatica: Increasing water-use efficiency does not avoid the growth J., Brūmelis, G., … Verheyen, K. (2019). Environmental drivers interac- decline induced by warming at low altitudes. Global Change Biology, tively affect individual tree growth across temperate European for- 14(5), 1076–1088. ests. Global Change Biology, 25(1), 201–217. https://doi.org/10.1111/ Pinheiro, J. C., & Bates, D. M. (2000). Mixed effects models in S and S-PLUS. gcb.14493 New York, NY: Springer Verlag. Magnani, F., Mencuccini, M., & Grace, J. (2000). Age-related decline in Pinheiro, J., Bates, D., DebRoy, S., & Sarkar, D. (2014). nlme: Linear and stand productivity: The role of structural acclimation under hydrau- nonlinear mixed effects models. R package version 3.1-117. Retrieved lic constraints. Plant, Cell & Environment, 23(3), 251–263. https://doi. from https://cran.rproj​ect.org/web/packa​ges/nlme/index.html org/10.1046/j.1365-3040.2000.00537.x Pretzsch, H., Schütze, G., & Uhl, E. (2013). Resistance of European tree Mäkinen, H., Yue, C., & Kohnle, U. (2017). Site index changes of Scots species to drought stress in mixed versus pure forests: Evidence pine, Norway spruce and larch stands in southern and central Finland. of stress release by inter-specific facilitation. Plant Biology, 15(3), Agricultural and Forest Meteorology, 237–238, 95–104. https://doi. 483–495. org/10.1016/j.agrfo​rmet.2017.01.017 R Development Core Team. (2018). R: A language and environment for Martínez-Vilalta, J., Cochard, H., Mencuccini, M., Sterck, F., Herrero, A., statistical computing. Vienna, Austria: R Foundation for Statistical Korhonen, J. F. J., … Zweifel, R. (2009). Hydraulic adjustment of Scots Computing. Retrieved from www.r-proje​c t.org pine across Europe. New Phytologist, 184(2), 353–364. https://doi. Reddy, A. R., Chaitanya, K. V., & Vivekanandan, M. (2004). Drought- org/10.1111/j.1469-8137.2009.02954.x induced responses of photosynthesis and antioxidant metabolism in Martínez-Vilalta, J., López, B. C., Loepfe, L., & Lloret, F. (2012). Stand- higher plants. Journal of Plant Physiology, 161(11), 1189–1202. https:// and tree-level determinants of the drought response of Scots pine doi.org/10.1016/j.jplph.2004.01.013 radial growth. Oecologia, 168(3), 877–888. https://doi.org/10.1007/ Rigling, A., Bigler, C., Eilmann, B., Feldmeyer-Christe, E., Gimmi, U., s0044​2-011-2132-8 Ginzler, C., … Dobbertin, M. (2013). Driving factors of a vegetation Matías, L., Linares, J. C., Sánchez-Miranda, Á., & Jump, A. S. (2017). shift from Scots pine to pubescent oak in dry Alpine forests. Global Contrasting growth forecasts across the geographical range of Change Biology, 19(1), 229–240. https://doi.org/10.1111/gcb.12038 Scots pine due to altitudinal and latitudinal differences in climatic Rigling, A., Eilmann, B., Koechli, R., & Dobbertin, M. (2010). Mistletoe- sensitivity. Global Change Biology, 23(10), 4106–4116. https://doi. induced crown degradation in Scots pine in a xeric environment. Tree org/10.1111/gcb.13627 Physiology, 30, 845–852. https://doi.org/10.1093/treep​hys/tpq038 Mátyás, C., Ackzell, L., & Samuel, C. J. A. (2004). EUFORGEN techni- Rouault, G., Candau, J.-N., Lieutier, F., Nageleisen, L.-M., Martin, J.-C., cal guidelines for genetic conservation and use for Scots pine (Pinus & Warzée, N. (2006). Effects of drought and heat on forest insect sylvestris). International Plant Genetic Resources Institute, Rome, Italy, 6 populations in relation to the 2003 drought in Western Europe. Mazerolle, M. J. (2006). Improving data analysis in herpetology: Using Annals of Forest Science, 63(6), 613–624. https://doi.org/10.1051/ Akaike’s Information Criterion (AIC) to assess the strength of bio- fores​t:2006044 logical hypotheses. Amphibia-Reptilia, 27, 169–180. https://doi. Ruijven, J., & Berendse, F. (2010). Diversity enhances community re- org/10.1163/15685​38067​77239922 covery, but not resistance, after drought. Journal of Ecology, 98(1), Mazerolle, M. J. (2011). AICcmodavg: Model selection and multimodel 81–86. inference based on (Q)AIC(c). R package version 1.17. Retrieved from Ryan, M. G., & Yoder, B. J. (1997). Hydraulic limits to tree height and tree http://cran.r-proje​c t.org/web/packa​ges/AICcm​odavg​/index.html growth. BioScience, 47, 235–242. https://doi.org/10.2307/1313077 McDowell, N., Pockman, W. T., Allen, C. D., Breshears, D. D., Cobb, N., Sánchez-Salguero, R., Camarero, J. J., Hevia, A., Madrigal-González, Kolb, T., … Yepez, E. A. (2008). Mechanisms of plant survival and J., Linares, J. C., Ballesteros-Canovas, J. A., … Rigling, A. (2015). mortality during drought: Why do some plants survive while others What drives growth of Scots pine in continental Mediterranean cli- succumb to drought? New Phytologist, 178(4), 719–739. https://doi. mates: Drought, low temperatures or both? Agricultural and Forest org/10.1111/j.1469-8137.2008.02436.x Meteorology, 206, 151–162. https://doi.org/10.1016/j.agrfo​r met. McKee, T. B., Doesken, N. J., & Kleist, J. (1993). The relationship of drought 2015.03.004 frequency and duration to time scales (pp. 179–184). Preprints, Eight Sánchez-Salguero, R., Camarero, J. J., Rozas, V., Génova, M., Olano, Conference on Applied Climatology. Anaheim, CA: American Meteor J. M., Arzac, A., … Linares, J. C. (2018). Resist, recover or both? Society. Growth plasticity in response to drought is geographically struc- Ogle, K., Whitham, T. G., & Cobb, N. S. (2000). Tree-ring variation in pinyon tured and linked to intraspecific variability in Pinus pinaster. Journal predicts likelihood of death following severe drought. Ecology, 81(11), of Biogeography, 45(5), 1126–1139. https://doi.org/10.1111/ 3237–3243. https://doi.org/10.1890/0012-9658(2000)081[3237: jbi.13202 TRVIP​P]2.0.CO;2 Sánchez-Salguero, R., Navarro-Cerrillo, R. M., Camarero, J. J., & Olson, M. E., Soriano, D., Rosell, J. A., Anfodillo, T., Donoghue, M. J., Fernández-Cancio, Á. (2012). Selective drought-induced decline of Edwards, E. J., … Méndez-Alonzo, R. (2018). Plant height and hy- pine species in southeastern Spain. Climatic Change, 113(3), 767–785. draulic vulnerability to drought and cold. Proceedings of the National https://doi.org/10.1007/s1058​4-011-0372-6 BOSE et al. | 4537 Scholz, F., Phillips, N., Bucci, S., & Goldstein, G. (2011). Hydraulic ca- France. Forest Ecology and Management, 413, 32–47. https://doi.org/ pacitance: Biophysics and functional significance of internal water 10.1016/j.foreco.2018.01.028 sources in relation to tree size. In F. C. Meinzer, B. Lachenbruch & Vennetier, M., Thabeet, A., Didier, C., Girard, F., Cailleret, M., T. E. Dawson (Eds.), Size- and age-related changes in tree structure Taugourdeau, O., … Caraglio, Y. (2013). Climate change impact on and function (Vol. 4, pp. 341–361). Dordrecht, The Netherlands: tree architectural development and leaf area. In B. R. Singh (Eds.), Springer. Climate change—realities, impacts over ice cap, sea level and risks. Seidel, H., Matiu, M., & Menzel, A. (2019). Compensatory growth of INTECH Open Access Publisher. http://dx.doi.org/10.5772/51510 Scots pine seedlings mitigates impacts of multiple droughts within Vicente-Serrano, S. M., Beguería, S., & López-Moreno, J. I. (2010). A mul- and across years. Frontiers in Plant Science, 10, 519–519. https://doi. tiscalar drought index sensitive to global warming: The Standardized org/10.3389/fpls.2019.00519 Precipitation Evapotranspiration Index. Journal of Climate, 23(7), Seidel, H., Schunk, C., Matiu, M., & Menzel, A. (2016). Diverging drought 1696–1718. https://doi.org/10.1175/2009J​CLI29​09.1 resistance of Scots pine provenances revealed by infrared thermog- Vicente-Serrano, S. M., Gouveia, C., Camarero, J. J., Begueria, S., Trigo, raphy. Frontiers in Plant Science, 7, 1247. https://doi.org/10.3389/ R., Lopez-Moreno, J. I., … Sanchez-Lorenzo, A. (2013). Response fpls.2016.01247 of vegetation to drought time-scales across global land biomes. Serra-Maluquer, X., Mencuccini, M., & Martínez-Vilalta, J. (2018). Changes Proceedings of the National Academy of Sciences of the United States in tree resistance, recovery and resilience across three successive ex- of America, 110(1), 52–57. https://doi.org/10.1073/pnas.12070​ treme droughts in the northeast Iberian Peninsula. Oecologia, 187(1), 68110 343–354. https://doi.org/10.1007/s0044​2-018-4118-2 Villar-Salvador, P., Peñuelas, J. L., & Jacobs, D. F. (2013). Nitrogen nu- Smith, M. D. (2011). An ecological perspective on extreme climatic trition and drought hardening exert opposite effects on the stress events: A synthetic definition and framework to guide future re- tolerance of Pinus pinea L. seedlings. Tree Physiology, 33(2), 221–232. search. Journal of Ecology, 99(3), 656–663. https://doi.org/10.1111/ https://doi.org/10.1093/treep​hys/tps133 j.1365-2745.2011.01798.x Vitali, V., Büntgen, U., & Bauhus, J. (2017). Silver fir and Douglas fir Spinoni, J., Naumann, G., Carrao, H., Barbosa, P., & Vogt, J. (2014). are more tolerant to extreme droughts than Norway spruce in World drought frequency, duration, and severity for 1951–2010. south-western Germany. Global Change Biology, 23(12), 5108–5119. International Journal of Climatology, 34(8), 2792–2804. https://doi. https://doi.org/10.1111/gcb.13774 org/10.1002/joc.3875 Walther, G.-R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, Spinoni, J., Vogt, J. V., Naumann, G., Barbosa, P., & Dosio, A. (2018). T. J., … Bairlein, F. (2002). Ecological responses to recent climate Will drought events become more frequent and severe in Europe? change. Nature, 416(6879), 389–395. International Journal of Climatology, 38(4), 1718–1736. https://doi. Westoby, M., Falster, D. S., Moles, A. T., Vesk, P. A., & Wright, I. J. (2002). org/10.1002/joc.5291 Plant ecological strategies: Some leading dimensions of variation Sterck, F. J., Zweifel, R., Sass-Klaassen, U., & Chowdhury, Q. (2008). between species. Annual Review of Ecology and Systematics, 33(1), Persisting soil drought reduces leaf specific conductivity in Scots pine 125–159. https://doi.org/10.1146/annur​ev.ecols​y s.33.010802.15 (Pinus sylvestris) and pubescent oak (Quercus pubescens). Tree Physiology, 0452 28(4), 529–536. https://doi.org/10.1093/treep​hys/28.4.529 Zalloni, E., Battipaglia, G., Cherubini, P., Saurer, M., & De Micco, V. (2019). Stovall, A. E. L., Shugart, H., & Yang, X. (2019). Tree height explains mor- Wood growth in pure and mixed Quercus ilex L. forests: Drought in- tality risk during an intense drought. Nature Communications, 10(1), fluence depends on site conditions. Frontiers Plant Science, 10, 397. 4385. https://doi.org/10.1038/s4146​7-019-12380​-6 Zang, C., Buras, A., Esquivel-Muelbert, A., Jump, A. S., Rigling, A., & Szejner, P., Belmecheri, S., Ehleringer, J. R., & Monson, R. K. (2020). Rammig, A. (2020). Standardized drought indices in ecological re- Recent increases in drought frequency cause observed multi-year search: Why one size does not fit all. Global Change Biology, 26, 322– drought legacies in the tree rings of semi-arid forests. Oecologia, 192, 324. https://doi.org/10.1111/gcb.14809 241–259. Zang, C., Hartl-Meier, C., Dittmar, C., Rothe, A., & Menzel, A. (2014). Taeger, S., Zang, C., Liesebach, M., Schneck, V., & Menzel, A. (2013). Patterns of drought tolerance in major European temperate forest Impact of climate and drought events on the growth of Scots pine trees: Climatic drivers and levels of variability. Global Change Biology, (Pinus sylvestris L.) provenances. Forest Ecology and Management, 307, 20(12), 3767–3779. https://doi.org/10.1111/gcb.12637 30–42. https://doi.org/10.1016/j.foreco.2013.06.053 Valladares, F., Gianoli, E., & Gómez, J. (2007). Ecological limits to plant phenotypic plasticity. New Phytologist, 176(4), 749–763. https://doi. S U P P O R T I N G I N FO R M AT I O N org/10.1111/j.1469-8137.2007.02275.x Additional supporting information may be found online in the Valladares, F., Matesanz, S., Guilhaumon, F., Araújo, M. B., Balaguer, L., Supporting Information section. Benito-Garzón, M., … Zavala, M. A. (2014). The effects of phenotypic plasticity and local adaptation on forecasts of species range shifts under climate change. Ecology Letters, 17(11), 1351–1364. https://doi. org/10.1111/ele.12348 How to cite this article: Bose AK, Gessler A, Bolte A, et al. Vanninen, P., & Mäkelä, A. (2005). Carbon budget for Scots pine trees: Growth and resilience responses of Scots pine to extreme Effects of size, competition and site fertility on growth allocation and droughts across Europe depend on predrought growth production. Tree Physiology, 25(1), 17–30. https://doi.org/10.1093/ conditions. Glob Change Biol. 2020;26:4521–4537. https://doi. treep​hys/25.1.17 Vennetier, M., Ripert, C., & Rathgeber, C. (2018). Autecology and growth org/10.1111/gcb.15153 of Aleppo pine (Pinus halepensis Mill.): A comprehensive study in