Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Disentangling the role of sex dimorphism and forest structure as drivers of growth and wood density in expanding Juniperus thurifera L. woodlands

Disentangling the role of sex dimorphism and forest structure as drivers of growth and wood... Key message The dioecious tree species Juniperus thurifera L. is undergoing a spontaneous process of forest expan- sion in southwest Europe. We investigated how growth, climate sensitivity, and wood density varied simultaneously between sexes and among stages of expansion while accounting for the variability of forest structure. We found few sex-based differences but detected lower wood density, greater growth rates, and higher sensitivity to drought in expanding fronts compared to long-existing forests. Context Juniperus thurifera L. (Spanish juniper) is a dioecious tree species undergoing a natural process of forest expan- sion in southwest Europe. Aims To assess how radial growth and wood density are simultaneously shaped by sex-based differences, the stage of forest expansion (long-existing forests, transition zones, and expanding fronts), variability in forest structure, and climate (in the case of radial growth). Methods We measured forest structure characteristics, tree rings, and wood density in 17 plots dominated by Spanish juniper in three stages of forest expansion in central Spain. We used linear mixed models (LMMs) to explore the main drivers of variability in radial growth and wood density and sex- and stage-based differences in climate-growth sensitivity. Results Rather than by sex, growth and wood density were mainly shaped by the stage of forest expansion, forest structure variables that characterize these stages, and climate variables (in the case of growth). Conclusion Sexual dimorphism had a minimal effect in growth and wood density in expanding Spanish juniper woodlands. Expanding fronts could be benefiting from land-use legacies in the abandoned fields they are colonizing, as reflected in higher growth rates and lower wood density, especially during years with less summer drought stress. However, this pattern could be reversed in the event of an increase in drought episodes. Keywords Dioecy · Juniperus thurifera L. · Tree rings · Forest expansion · Spontaneous tree establishment · Abandoned fields · Mediterranean region 1 Introduction Handling Editor: Shuguang (Leo) Liu In recent decades, southwest Europe has experienced a sig- This article is part of the topical collection on Establishment nificant expansion of Mediterranean and temperate tree spe- of second-growth forests in human landscapes: ecological mechanisms and genetic consequences cies that differ in their tolerance to harsh environmental con- ditions: Quercus robur L., Quercus ilex L., Fagus sylvatica Contribution of the co-authors Conceptualization: Irene Martín- L., and Juniperus thurifera L. (Gimeno et al. 2012c; Başnou Forés, Raquel Alfaro-Sánchez; Methodology: Raquel Alfaro- et al. 2013; Vilà-Cabrera et al. 2017; Valdés-Correcher et al. Sánchez, Irene Martín-Forés; Formal analysis and investigation: 2019; Acuña-Míguez et al. 2020; Martín-Forés et al. 2020). Raquel Alfaro-Sánchez; Writing—original draft preparation: Raquel Alfaro-Sánchez; Writing—review and editing: Josep European forests have undergone natural expansion—pri- Maria Espelta, Belén Acuña-Míguez, Irene Martín-Forés; Funding marily into former cultivated areas—as a consequence of a acquisition: Fernando Valladares; Supervision: Irene Martín-Forés rural exodus and the abandonment of traditional agricultural practices (Keenan 2015; Palmero-Iniesta et al. 2020; Hampe Extended author information available on the last page of the article Vol.:(0123456789) 1 3 86 Page 2 of 19 Annals of Forest Science (2021) 78: 86 et al. 2020). The vegetation dynamics of abandoned fields is reported in many dioecious trees, including Spanish juniper driven by soil characteristics, climate conditions, and prop- (Gauquelin et al. 2002; Montesinos et al. 2006; Rozas et al. agule availability (Tasser et al. 2007). Land-use legacies 2009; Juvany and Munné-Bosch 2015). Previous studies of from previous agricultural practices including greater nutri- Spanish juniper woodlands die ff r regarding the presence of ent content (De Schrijver et al. 2011) or more decomposer sexual dimorphism in radial growth and climate-growth sen- activity (Freschet et al. 2014) facilitate the establishment sitivity: differences between sexes were found for these traits of trees and shrubs during the initial period after coloni- by Montesinos et al. (2006), Rozas et al. (2009), Olano et al. zation and give rise to increased tree growth (Lambin and (2015), and DeSoto et al. (2016), but no evidence was found Meyfroidt 2011; Gerstner et al. 2014). Recently expanding by Gimeno et al. (2012c). In fact, as a response to environmen- forests in Europe have reported greater growth rates than tal heterogeneity, in dioecious plants, sex-based variation in long-existing forests—even after accounting for the effects non-reproductive traits may be more limited than intraspecific of age and competition—but lower wood density (Pretzsch variation (e.g., Anderson et al. 2014). Thereby, the inclusion of et al. 2018; Alfaro-Sánchez et al. 2019, 2020). A reduction forest structure variables as covariates in experimental design in wood density implies a greater number and surface area may improve our understanding of the trade-offs between of conductive vessels that can modify the growth-climate growth and reproduction (Obeso 2002; McKown et al. 2017). sensitivity of trees (Greenwood et  al. 2017). However, In this study, we assessed whether or not radial growth expanding forests of broadleaf species such as F. sylvatica and wood density in juniper woodlands undergoing a nat- have lower wood density but no significant differences in ural process of forest expansion in southwest Europe are growth-climate sensitivity when compared to long-existing simultaneously influenced by (i) sex-based differences, (ii) forests, probably because of the better growth conditions the stage of forest expansion, and (iii) the variables of for- present in the former croplands they are colonizing (Alfaro- est structure that characterize each stage (e.g., tree density, Sánchez et al. 2019). Yet, little is known about the response age, and size). We also explored sex- and stage-based differ - in growth, wood density, or climate sensitivity in expanding ences with regard to climate-growth sensitivity. We studied tree species of other functional types such as the evergreen a unique mosaic of patches of Spanish juniper in central Mediterranean conifer J. thurifera (but see Gimeno et al. Spain that, to the best of our knowledge, contains the larg- 2012c). est number of Spanish juniper trees in this hotspot for this The J. thurifera (Spanish juniper) is a dioecious spe- species (90% of its world distribution is located in Spain, cies of tree that is endemic to North Africa and the Ibe- Blanco and Castro 1997). In these patches, a mix of young rian Peninsula (Blanco and Castro 1997), recognized under and mature junipers grow along a gradient of forest expan- the European Habitats Directive. It is drought-tolerant and sion, from low-density areas in expanding fronts to rela- under continental climates tends to grow in low densities tively dense areas in long-existing stands. We hypothesized (Olano et al. 2008). It is well adapted to rocky, poorly devel- that sex-based differences in growth would increase with oped soils (Gauquelin et al. 1999) and has been shown to cambial age due to the increase with age of reproductive increase in secondary growth under harsh climatic condi- costs in females (Montesinos et al. 2012). Given the find- tions (Granda et al. 2014). The Spanish juniper is currently ings of previous studies on the effect of land-use legacies undergoing wide-ranging expansion and densification pro- on growth, wood density, and climate-growth sensitivity cesses in certain areas of central Spain (Blanco and Castro (Alfaro-Sánchez et al. 2019), we expected greater growth, 1997; Thompson 2005), whereas in other areas, it is facing lower wood density, and higher climate sensitivity in both competition from other fast-growing tree species including female and male junipers in the expanding fronts that are pines and oaks (Olano et al. 2012). Since the middle of the colonizing former croplands. Overall, our results should help nineteenth century, the expansion of Spanish juniper wood- understand how variation in forest structure—in this case, lands has been facilitated by the progressive abandonment of mainly derived from the processes of forest expansion into the traditional management of wood-pasture systems and the abandoned fields—can influence the expression of second- existence of a diverse dispersal community that ensures seed ary die ff rences between sexes in Spanish juniper woodlands, availability and the frugality of the species (Thompson 2005; which ultimately will help their future preservation. Escribano-Avila et  al. 2012). The expansion of Spanish juniper woodlands in central Spain has created a mosaic of forest patches ranging from long-existing stands to expand-2 Methods ing fronts. Most of these expanding fronts are composed of young juniper trees, and their performance and survival 2.1 Study area under current and future climate hazards remain unknown. Sexual dimorphism in secondary (non-reproductive) sexual This study was conducted in central Spain in the Alto Tajo characteristics (e.g., growth, vigor, and physiology) has been Natural Park and surrounding areas (Fig. 1). The climate of 1 3 Annals of Forest Science (2021) 78: 86 Page 3 of 19 86 this area is continental Mediterranean, characterized by hot to the 1950s (Escribano-Avila et al. 2012; Gimeno et al. dry summers and cold winters. Mean annual temperature 2012c; Villellas et al. 2020; Acuña-Míguez et al. 2020). We and precipitation were 10.7 °C and 462 mm for the period defined long-existing forests as patches containing cores of 1950–2017, respectively (KNMI Climate Explorer; http:// well-preserved juniper woodlands that existed in 1956. The climex p.k nmi.n l/). We selected three sites with unmanaged expanding fronts correspond to areas of recent colonization and well-preserved juniper woodlands, in which the characterized by scattered trees located on former Spanish juniper is the dominant tree species. Maranchón agricultural land. The transition zones were intermediate is the northernmost site, followed by Ribarredonda and forests between these two extremes of gradient. All three then Huertahernando to the south (Fig. 1). The maximum forest stages were represented at all three sites (Fig. 1). It is distance between sites was 30  km along an elevation important to note that the three age categories were defined gradient of 1000–1300  m a.s.l. Seventeen plots were by the land-use age and not by the age of trees in the stand, established at these three sites along a gradient of forest as in previous studies (Başnou et al. 2013, Alfaro-Sánchez expansion: seven plots in Maranchón and five plots in both et  al. 2019). Therefore, long-existing juniper patches in Ribarredonda and Huertahernando (Fig. 1). The gradient the area do not necessarily have a tree age structure that is of forest expansion was separated into three stages, namely, characteristic of old-growth forests and may even have been long-existing forests, expanding fronts, and transition zones. disturbed by thinning or fires recently, so young trees are The stages were identified by previous studies of the area also present in these patches. based on the comparison of land-cover maps dating back Fig. 1 Location of the three sampling sites (Maranchón, Ribarre- cated for each plot (EXP, expanding front; TRAN, transition zone; donda, and Huertahernando) in central Spain in the Alto Tajo Natural LON, long-existing forest) Park and surrounding areas. The gradient of forest expansion is indi- 1 3 86 Page 4 of 19 Annals of Forest Science (2021) 78: 86 et al. 2019, 2020). We used the second cores to count and 2.2 Field sampling measure annual ring widths. They were air-dried, glued onto wooden mounts, and polished using sandpaper of progres- Exhaustive field sampling was conducted in autumn 2017. In total, we georeferenced 816 juniper trees in the study plots. sively finer grain until tree rings were visible. The cores were dated with a stereomicroscope and scanned at 2400 Due to the differences in tree density among forest stages, plot areas were flexible to be able to sample a minimum of d.p.i. We measured ring widths to an accuracy of 0.001 mm using the CooRecorder v9.3 software (Cybis Elektronik 35 adult trees per plot (Table 1). We visually identified the sex of all the selected individuals when they had male flow - 2018). The dataset is available in Alfaro-Sánchez et al. 2020. The cross-dating of individual series was checked using the ers or female cones and considered these trees to be repro- ductive individuals. If the tree did not have o fl wers or cones, CooRecorder and COFECHA programs (Holmes 1983). For subsequent climate-growth analyses, individual tree ring we scored it as a tree of unknown sex. The 451 trees identi- fied as reproductive at the time of sampling had a minimum width series were detrended with cubic smoothing splines of 10 years to remove non-climatic growth trends related size threshold of trunk diameter at breast height of ≥ 3 cm and total height of > 1.40 m. Above this size threshold, we to the increase in tree age and size (Cook and Kairiukstis 1990). A comparison between detrended methods is shown found that only 5.5% of trees were of unknown sex. We determined the mean tree density per plot and meas- in Figs. 6 and 7. ured the quadratic diameter (QD, calculated as the square root of the sum of square diameter at breast height of each 2.4 Climate sensitivity stem of a tree; Stewart and Salazar 1992), the maximum tree height (measured with Haglöf Vertex IV hypsometer), and Sums of the monthly mean temperatures and precipitation were accessed for the period 1950–2017 from the homog- the average crown diameter calculated as the mean of the projection of two perpendicular axes passing through the enized and quality-checked E-OBS v.17.0 dataset (Haylock et  al. 2008) in the KNMI Climate Explorer ( http:// clime axis of the trunk (measured with a Haglöf DME distance measurer). xp. knmi. nl/). Our study sites are spatially located in two different E-OBS v.17.0 grid cells, Maranchón in one and 2.3 Wood density and tree growth measurements Ribarredonda and Huertahernando in another. Given the proximity of the sites, we averaged the climate data from We only measured the wood density and tree growth of the these two grid cells at 0.25° spatial resolution for use in subsequent analyses. We calculated the drought index SPEI 451 reproductive trees. We extracted two increment cores 50 cm from ground height using a Pressler increment borer (standardized precipitation-evapotranspiration) using the R package SPEI (Vicente-Serrano et al. 2010) with a time scale (0.5 cm; Haglöf, Långsele, Sweden). One of these two cores −3 was used to estimate wood density (g cm ), following Wil- of 3 months, based on temperature and precipitation data from E-OBS v.17.0. Lower values of SPEI correspond to liamson and Wiemann (2010), which is calculated as the dry weight of the full core (including heartwood and softwood) greater drought stress. divided by its saturated volume (see also Alfaro-Sánchez Table 1 Plot and tree characteristics at different sites and stages along the forest expansion gradient Plot area Sex Tree den- Tree age* QD (cm) Height Crown Number of (ha) sity (trees (m) diameter stems −1 ha ) (m) Site (lat., long.) Gradient Plots n Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Maranchón Long 2 59 0.58 0.17 0.6 0.5 115 14 50 19 24 15 6 1.9 5.5 2 2.4 1.8 (41.06, − 2.20) Transition 3 86 0.84 0.32 0.6 0.5 71 29 33 13 15 9 4.1 1.2 4.2 1.4 2.4 1.7 Expanding 1 30 1.35 0.7 0.5 32 26 8 11 5 3.8 0.7 3.8 1 2 1.1 Ribarredonda Long 2 56 0.42 0.03 0.6 0.5 125 9 46 14 21 10 5.7 1.4 4.9 1.7 2.4 1.9 (40.87, − 2.30) Transition 1 28 0.35 0.7 0.4 123 26 10 12 8 4.6 1.5 3.6 1.8 1.5 1.1 Expanding 2 48 0.83 0.2 0.8 0.4 49 8 28 12 12 7 4.4 1.2 3.8 1.3 1.6 1.1 Huertahernando Long 1 29 0.6 0.6 0.5 89 33 11 23 10 5.5 1.1 6 1.4 3 2 (40.83, − 2.28) Transition 2 58 1 0.5 0.6 0.5 68 34 30 14 18 9 4.9 1.3 5 1.5 2.5 1.5 Expanding 2 57 1.8 0.05 0.6 0.5 22 1 23 7 14 7 4.1 1.1 4.4 1.5 1.7 1.1 QD: quadratic mean diameter; sex: female = 0, male = 1. *Tree age is estimated as the number of rings measured per core. The total tree age of trees is expected to be older. 1 3 Annals of Forest Science (2021) 78: 86 Page 5 of 19 86 To assess the effect of climate on growth, we ran linear with age, we included the interactions between cambial age regression models (LMs) at tree level between detrended with sex and stages of forest expansion. Cambial age is the ring width series and climate variables. The considered cli- age of the sampled cores, where the first year corresponds to mate variables were monthly and seasonal (combination of the first tree ring outside the pith. To determine if climate- several consecutive months) temperatures and SPEI-3 from growth sensitivity varied between sexes and among stages, September of the previous growing season to September we included the interactions between February–April tem- of the year in which the ring was formed. We calculated perature and August SPEI-3 with sex and stages. To allow the percentage of trees with significant slopes for each of for comparisons among stages of forest expansion, annual the climate variables considered in the LMs to identify growth was restricted to the first 35 years of life of the trees, the month or season with the highest temperature-growth the maximum period of time in which the three stages over- and drought-growth sensitivity, that is, the average Febru- lapped with a minimum of five trees at each site (see Fig.  8). ary–April temperatures and August SPEI-3 (see “Results” Tree individuals nested in plot was included as a random section). We used these two climate variables as explanatory ee ff ct to compensate for the repeated measures taken from an variables in the linear mixed effects models (LMMs) in the individual tree. A first-order autocorrelation structure (AR1) subsequent analyses of climate-growth sensitivity (see the was also included in the growth LMM to control for the following sections). temporal autocorrelation of growth measures. In both the wood density and growth models, the explana- 2.5 Statistical analyses tory variables were standardized to eliminate differences in scale measurements. Crown diameter, tree height, and QD We used ANOVAs to test for differences in tree density and were highly correlated, and so to avoid collinearity, we only the proportion of sexes in stages of forest expansion and at included QD in the models. We compared alternative full- sites (significance level was set at p < 0.05). We used LMMs saturated models including linear or spline adjustments (up and generalized linear mixed effects models (GLMMs) to to two degrees of freedom) for each continuous predictor and assess how tree characteristics varied between sexes and selected the adjustment used in the model with the lowest among stages of forest expansion and sites. Specifically, AIC. We then checked for collinearity within the predictor LMMs were used to model the following dependent vari- variables included in the two selected full-saturated models ables: tree age, QD, height, and crown diameter; a general- (including linear or spline adjustments for the continuous ized linear mixed effects model (GLMM) with a Poisson variables) using variance inflation factors (VIF) with the regression was used to model the number of stems. For all performance R package (Lüdecke et al. 2020); we found models (LMMs and GLMM), the explanatory variables were VIF scores ≤ 10, which indicates a lack of collinearity (Kock sex, stages of forest expansion, and site. Plot was included as and Lynn 2012). A full model for wood density (WD) would a random effect. We used Tukey–Kramer post-hoc analysis look like this in R syntax (considering linear adjustments for for multiple comparisons in the R package lsmeans (Lenth the continuous variables, Eq. 1): 2016) to identify specic fi die ff rences in tree age, QD, height, lme(WD ∼ Sex + Stage + Site + QD crown diameter, and number of stems among the three stages + nstems + TD + Age + Age × Sex of forest expansion and at sites and between sexes for each (1) stage and site. + Age × Stage, random =∼ 1Plot, We used LMMs to assess how wood density and annual � data = dataset, method= ML ) growth varies between sexes, among stages of forest expan- sion, and is associated with forest structure. Specifically, A full model for log-transformed growth (log[Growth]) for the wood density LMM, the explanatory variables were would look like this in R syntax (considering linear adjust- sex, stage of forest expansion, site, QD, number of stems ments for the continuous variables, Eq. 2): (nstems), tree density (TD), and tree age. To determine lme(log[Growth] ∼ Sex + Stage + Site + QD if wood density varied between sexes or among stages in + nstems + TD + Cambial age + T + SPEI terms of tree age, we included in the model the interactions between tree age with sex and stages of forest expansion. − 3 + Cambial age × Sex + Cambial age× Plot was included as a random effect. For the growth LMM, Stage + T × Sex + T × Stage + SPEI the dependent variable, annual growth, was transformed − 3 × Sex + SPEI − 3 × Stage, with a natural logarithm to conform to normality. The random =∼ 1Plot∕Tree, explanatory variables were sex, stage of forest expansion, site, QD, number of stems, tree density, cambial age, Febru- correlation = corCAR1(form =∼ 1Plot∕Tree), � � ary–April temperature (T), and August SPEI-3 (SPEI-3). To data = dataset, method = ML (2) determine if growth varied between sexes or among stages 1 3 86 Page 6 of 19 Annals of Forest Science (2021) 78: 86 Next, the models were reduced to those with the lowest per tree than the expanding fronts; the transition zone Akaike Information Criterion (AIC) (the best or most parsi- showed no significant differences when compared to the monious models) using the dredge function from the MuMIn other two stages (Fig. 10b, Table 1). We found no significant R package (Barton 2019). The method was set to maxi- differences among sites for the variables tree age, number of mum likelihood (ML) during the fixed effect model selec- stems, QDs, or height, although larger crown diameters were tion phase, although the final models are presented using found in Huertahernando, the southernmost site (Fig. 10f–h, restricted maximum likelihood (REML) (Kuznetsova et al. Table 1). 2017). Model fits were checked visually to ensure that they conformed to model assumptions. We calculated marginal 3.1 Wood density patterns (i.e., the proportion of variance explained by fixed effects) and conditional (i.e., the proportion of variance explained Wood density increased with tree age, decreased with QD, by fixed and random effects) r with the sjPlot R package and varied among stages of forest expansion and sites. Post- (Lüdecke 2021). All statistical analyses were performed hoc analysis confirmed significantly lower wood density in using R version 3.5.1 (R Core Team 2019). the expanding front and at the southernmost site (Huerta- hernando) compared to the transition zone and the north- ernmost site (Maranchón), respectively. The long-existing 3 Results forests and Ribarredonda showed no significant differences in wood density compared to the other two stages and sites, Tree density increased gradually from the sparse expand- respectively (Table 2, Fig. 3). −1 ing fronts (28 trees ha ) to the relatively dense (for juniper −1 woodland) long-existing forests (116 trees ha ). The transi- tion zone showed no significant differences in tree density −1 (83 trees ha ) with either the expanding fronts or the long- existing forests (Fig. 2a). No significant differences were found in tree density among sites (Fig. 9a). Table 2 Wood density LMM coefficients. The marginal and con- The proportion of male trees doubled that of female trees ditional r2 are also shown. The level site: Maranchón and the level in the three stages of forest expansion and at all three sites, stage: expanding is included in the intercept. SE standard error. Plot i.e., ~ 60% of trees were male and ~ 30% females (Fig. 2b, was included as a random effect in the model, but the within-group variance (or residual variance, σ2) and the between-group variance 9b). The highest proportion of unknown sex trees (9%) was (or random intercept variance, τ00 Plot) were effectively 0 detected in the expanding front (Fig.  2b) and at Ribarre- donda (Fig.  9b), probably due to a higher proportion of Predictors Estimates SE p younger trees (Fig. 10a, f). (Intercept) 0.545 0.004 < 0.001 Female trees were significantly older and taller and had Quadratic diameter − 0.003 0.002 0.097 greater QDs, more stems per tree, and larger crown diam- Age 0.010 0.002 < 0.001 eters than male trees for each stage of forest expansion and Site [Ribarredonda] − 0.005 0.004 0.218 site (Fig.  10). The long-existing forests had significantly Site [Huertahernando] − 0.015 0.004 < 0.001 older and taller trees, greater QDs, and larger crown diam- Stage [Transition] 0.012 0.004 0.003 eters than the trees in the transition zone and the expanding Stage [Long] 0.006 0.005 0.202 front (Fig. 10a, c–e, Table 1; Acuña-Míguez et al. 2020). 2 2 Marginal r /conditional r 0.14/0.14 The long-existing forests also had significantly more stems Fig. 2 Variations in tree density (violin and boxplots) for each stage of forest expansion (a). Sex percentage for each stage of forest expansion (b). All sampled juniper trees were considered (n = 816). The significant differences among stages of forest expansion (a) and between sexes at each stage of forest expansion (b) are denoted by small letters 1 3 Annals of Forest Science (2021) 78: 86 Page 7 of 19 86 Fig. 3 Post-hoc differences among stages of forest expansion (a) and interval of the LS mean. EXP, expanding front; TRAN, transition sites (b) derived from the wood density LMM. Black circles indicate zone; LON, long-existing forest; MAR, Maranchón; RIB, Ribarre- the least square (LS) mean. Error bars indicate the 95% confidence donda; HUE, Huertahernando 3.2 Climate‑growth sensitivity and annual growth patterns We found a higher percentage of trees showing a positive response in tree ring growth in years with high Febru- ary–April temperatures (~ 41% of trees) and August SPEI-3 (~ 52% of trees; Fig. 4). Growth in trees increased with warmer temperatures, under low drought stress, and with greater QD and cambial age during approximately the first 20 years of life but showed a slightly negative trend during the final 15 years of the study period. Growth decreased with tree density and the number of stems per tree. We found greater growth in Spanish junipers in the expanding front than in the transition zone and the long- existing forests, particularly after the first 20 years of life of the trees. In the expanding front, Spanish junipers also showed higher growth under low drought stress conditions compared to the trees growing in the other two stages. We found a minimal difference in climate-growth sensitivity between sexes. Males showed higher growth sensitivity to early spring temperatures (February–April) than females; i.e., males grew more during years with warmer early spring temperatures. Females showed higher growth sensitivity to summer drought; that is, females grew more during years with less drought stress. Sex-based differences were almost indiscernible when we plotted the model predictions (Table 3, Fig. 5). Climate variables, QD, and the interaction between cambial age and stages of forest expansion were the best explanatory variables for the growth LMM (Table 3, Fig. 5a, d). Fig. 4 Percentage of trees showing positive (POS, red bars) or nega- tive (NEG, blue bars) responses in growth to monthly and seasonal temperature and SPEI-3 variables. The positive or negative responses correspond to the positively or negatively significant slopes (p < 0.05) 4 Discussion obtained for each individual tree in linear regression models between detrended growth and climate variables In this study, we investigated how radial growth and wood density were influenced by sex, the stage of forest expan- in forest structure found among stages—i.e., tree density, sion, and variability in forest structure. We also explored age, and tree diameter—had a much greater effect on growth sex-based and stage-based differences in climate-growth and wood density of junipers than sex. Specifically, trees sensitivity. The stage of forest expansion and the variability 1 3 86 Page 8 of 19 Annals of Forest Science (2021) 78: 86 Table 3 Growth LMM Predictors Estimates SE p coefficients. Marginal and conditional r are also given. (Intercept) − 0.341 0.118 0.004 The level stage: expanding is QD [1st degree] 1.205 0.147 < 0.001 included in the intercept. SE QD [2nd degree] 0.475 0.222 0.033 standard error. σ is the residual variance. τ Tree and τ Plot Number of stems [1st degree] − 0.584 0.11 < 0.001 00 00 are the between-group variance Number of stems [2nd degree] − 0.058 0.196 0.768 for the random effects tree and Tree density − 0.113 0.066 0.112 plot, respectively Cambial age [1st degree] 1.879 0.103 < 0.001 Cambial age [2nd degree] 0.417 0.087 < 0.001 FEB-APR T 0.072 0.005 < 0.001 AUG SPEI-3 0.108 0.006 < 0.001 Sex [male] 0.006 0.037 0.867 Gradient [transition] 0.133 0.133 0.338 Gradient [long] 0.39 0.177 0.048 FEB-APR T × sex [male] 0.013 0.006 0.027 AUG SPEI-3 × sex [male] − 0.011 0.005 0.047 AUG SPEI-3 × gradient [transition] − 0.014 0.006 0.026 AUG SPEI-3 × gradient [long] − 0.034 0.006 < 0.001 Cambial age [1st degree] × gradient [transition] − 0.515 0.138 < 0.001 Cambial age [2nd degree] × gradient [transition] − 0.396 0.108 < 0.001 Cambial age [1st degree] × gradient [long] − 1.363 0.143 < 0.001 Cambial age [2nd degree] × gradient [long] − 0.472 0.102 < 0.001 Random effects   σ 0.25   τ 0.15 00 Tree   τ 0.27 00 Plot 2 2   Marginal r /conditional r 0.22/0.43 in the expanding fronts had higher growth rates and lower environments (Ortiz et al. 2002; Barrett et al. 2010) and wood density than junipers in the transition zone and long- have been associated with the greater cost of reproduction existing forests. The positive response in growth found in the for females trees (Vasiliauskas and Aarssen 1992; Montes- expanding fronts increased more during summers with low inos et al. 2012). Compensatory mechanisms for the higher drought stress than in the other two stages. reproductive cost in females such as greater photosynthetic capacity or water-use efficiency (WUE; Dawson and Bliss 4.1 Sex‑based differences in expanding juniper 1989; Olano et al. 2015; Rozas et al. 2009) have been identi- woodlands fied as the most likely explanation for this paradox (Tozawa et al. 2009). Nevertheless, no evidence of sex-related differ - It is important to know whether sex effects on growth and ences in WUE was found in a previous study conducted at density exist in Spanish juniper stands to improve our our study sites (Acuña-Míguez et al. 2020). understanding of compensatory mechanisms (or trade-offs) Females outperformed males in height, QD, number of between growth and reproduction. A higher proportion of stems, crown diameter, and age, irrespectively of the stage male trees were found in all three stages of forest expan- and site considered. Females invested more resources in veg- sion and at all three sites, suggesting that Spanish juniper etative height growth than males (Gauquelin et al. 2002) stands in central Spain are male-biased. Similar results have as their larger canopies, consisting of a greater number of been found in other studies conducted in juniper stands in stems, allow them to bear a large number of cones that may central Spain (Gimeno et al. 2012c), Morocco, and the Pyr- enhance their reproductive capacity. Our results showed enees (Gauquelin et al. 2002). In young populations, these minimal sex-based differences in wood density and growth. results can be explained simply because males begin to Specifically, we found that males may grow slightly more flower earlier than females and consequently can be identi- than females during years with warmer early spring tem- fied more readily (Gauquelin et al. 2002). However, male- peratures, whereas females may grow more than males biased sex ratios are also commonly found in more stressful during years with low summer drought stress. However, 1 3 Annals of Forest Science (2021) 78: 86 Page 9 of 19 86 Fig. 5 Growth LMM predic- tions as a function of QD (a), tree density (b), and the number of stems per tree (c). Growth LMM predictions for the three stages of forest expansion dur- ing the first 35 years of life of the trees (cambial age) (d) and for August SPEI-3 (f). Growth LMM predictions for female and male individuals in terms of February–April temperatures (e) and August SPEI-3 (g). Indi- vidual growth values are shown with circles. Note that the lower values of SPEI correspond to greater drought stress when plotting the predictions of our models, sex-based dif- and cambial age was not selected in the most parsimoni- ferences were almost indiscernible (Fig. 5e, g). Hence, we ous growth model, suggesting that there are no significant suggest taking into account tree size or other relative growth differences in annual growth rates before or after the trees response variables in forest structure in sex dimorphism become reproductive. During early stages, negligible differ - models for a better understanding of the impact of sex on ences in reproductive resource investment is evident for the the performance of dioecious species (Obeso 2002). lack of sex-based differences that we found (McKown et al. Previous studies reporting sex-based differences in Span- 2017). The Spanish junipers in our study system, although ish juniper radial growth and climate-growth sensitivity have reproductive, are relatively young. Thus, we cannot rule reached divergent conclusions regarding which sex performs out the possibility that sex-related differences in secondary best in each of these traits. For instance, some authors report growth may be modified during ontogeny due to physiologi- greater growth in males than in females (Gauquelin et al. cal adjustments (Rozas et al. 2009). 2002; Montesinos et al. 2006), whereas other studies found greater growth and higher summer precipitation-sensitivity 4.2 Main drivers of growth and wood density in females (Rozas et al. 2009). Sex-based differences with in expanding juniper woodlands regard to climate may be site-dependent (Olano et al. 2015; DeSoto et al. 2016), with females growing more than males Less trait divergence has been reported in dioecious species under less restrictive environmental conditions, or age- adapting to new environmental conditions (Arbuthnott et al. dependent (Rozas et al. 2009) as young females are more 2014), which is consistent with the expansion that the Span- sensitive to drought conditions. Given the great reproduc- ish juniper woodlands are currently undergoing in central tive effort of Spanish juniper females (Montesinos et al. Spain. Rather than sex, the main drivers of radial growth 2012), we hypothesized that sexual dimorphism in second- and wood density were the specific stage of forest expan - ary growth in Spanish junipers should occur from early sion (and the variables of forest structure that characterize stages onwards. By contrast, the interaction between sex each stage) and climate (in the case of growth). We found 1 3 86 Page 10 of 19 Annals of Forest Science (2021) 78: 86 greater growth rates at the expanding front, followed by the greater WUE than in long-existing forests owing to changes transition zone and the long-existing forest (differences that in certain functional attributes (e.g., higher leaf mass per became evident at approximately 10 years of age, Fig. 5d). area) and greater nitrogen availability stemming from the Growth decreased with tree age, higher tree density (Rozas former agricultural use (Guerrieri et al. 2021). Indeed, in et al. 2009; Gimeno et al. 2012c), and the number of stems our study area, Acuña-Míguez et al. (2020) found increased per tree. The Spanish juniper is a multi-stemmed species, WUE in the expanding fronts that was mainly related to their and we show here that annual growth increased in trees with lower vegetation cover and younger age. As such, our results fewer stems. In unmanaged stands, long-existing forests agree with the general pattern that trees growing under low tend to have more stems because, given the lack of logging competitive stress have higher growth rates, greater WUE, or browsing by livestock, older trees have more stems. A and a better response in growth to high water availability reduction in growth in multi-stemmed individuals has been (e.g., Linares et al. 2009; Sánchez-Salguero et al. 2015). observed in other resprouting species and is attributed to Rozas et al. (2009) showed that growth-climate sensitivity the preferential investment of resources in height growth in Spanish junipers is higher in earlier life stages. Here, we owing to the competition for light among stems (see Espelta accounted for a possible age effect when comparing sexes et al. 2003). Favorable climatic conditions, i.e., warm spring (females are significantly older than males) and among temperatures and wet summers, enhance annual growth in stages of forest expansion (the expanding fronts and transi- Spanish junipers (Rozas et al. 2009; Gimeno et al. 2012a). tion zone were significantly younger than the long-existing Relatively high temperatures at the beginning of the grow- forests) by restricting our analyses to the first 35 years of ing season stimulate earlier cambium reactivation and so cambial age. Thus, we can rule out any possibility that sex- enhance growth (Begum et al., 2008). Despite the excellent and stage-based differences in the growth response to cli - adaptation in Mediterranean tree species to drought, con- mate variability are caused by an age effect. straints in water availability during the growing season can Tree and cambial age also affect wood density (Frances- result in markedly less growth that lasts for several years chini et al. 2013). Specifically, our results showed a decrease in (e.g., Anderegg et al. 2015; Gazol et al. 2018). Therefore, mean wood density with tree age. Tree ring density chronolo- growth in Spanish junipers could be severely reduced by the gies show that larger rings are associated with a decrease in increase in the frequency and severity of drought episodes mean wood density (Lundgren 2004). As such, years with high projected for southwest Europe, particularly in the highly water availability increased tree ring growth and decreased sensitive expanding fronts. mean ring density (Franceschini et al. 2013), particularly in As hypothesized, the expanding fronts had greater posi- younger trees. The expanding fronts had less wood density tive responses in growth during years with greater water than the transition zone and the long-existing forests, even after availability than the other stages of forest expansion accounting for age effects. Greater densities are thought to be (Gimeno et al. 2012c). Similar results have been found for less vulnerable to cavitation (Hacke et al. 2001). However, the other tree species developing on former agricultural land rapid growth of expanding forests enables the development of (Alfaro-Sánchez et  al. 2019). We suggest that growth is large vessels that could increase the risk of cavitation during mediated by nutrient limitations (Forrester 2015) deriving periods of continuous drought (Lambers et al. 2008). from the tree density and land-use legacies found at each of the stages of forest expansion. Indeed, the expanding fronts 4.3 Land‑use legacies had lower tree densities, followed by the transition zone and the long-existing forests. Furthermore, Gimeno et al. Expanding juniper woodlands are mainly colonizing adjacent (2012c) found no clear nucleation in some of the expand- former agricultural land that was abandoned in recent ing fronts studied here but did detect clumped patterns in decades as a result of the rural exodus towards urban nuclei. the long-existing forests. The clumping patterns, mostly Spontaneous tree establishment in abandoned fields in the attributable to perching and nursing effects in adult trees, Mediterranean region has proved to benefit from increased could have increased competition for resources in junipers soil nitrogen levels deriving from their previous agricultural growing in these relatively dense long-existing patches. By use (Nadal-Romero et  al. 2018; Guerrieri et  al. 2021). contrast, the lower tree density in the expanding front (~ 28 Recently expanding forests have been reported to enhance −1 trees ha ) has the effect of reducing intraspecific compe- growth rates but lower wood density in comparison with long- tition and facilitating greater exposure to light, which is existing forests, even after accounting for age and competition reflected in higher growth rates. Increased or stable growth effects (Alfaro-Sánchez et al. 2019). Similarly, the expanding at increased intrinsic water use efficiency (WUE) has been fronts of Spanish juniper are probably benefiting from land- reported in recent decades in juniper woodlands elsewhere use legacies inherited from the former croplands (De Schrijver (Granda et al. 2014). Previous studies analyzing the spon- et al. 2011; Vilà-Cabrera et al. 2017; Alfaro-Sánchez et al. taneous establishment of secondary forests have reported 2019), as is confirmed by the greater survival rate of saplings 1 3 Annals of Forest Science (2021) 78: 86 Page 11 of 19 86 (Gimeno et al. 2012b), higher WUE (Granda et al. 2014; of cambial age, we ruled out the possibility that stage-based Acuña-Míguez et al. 2020), and the greater growth rates and differences in growth and climatic sensitivity are caused by lower wood density values reported at the expanding fronts in an age effect. Instead, we suggest that expanding fronts of this study. Greater vigor and survival rates in the expanding Spanish juniper are benefitting from a combination of lower fronts of Spanish junipers have been related to environmental intraspecific competition found in the expanding patches and differences between expanding and long-existing forests land-use legacies stemming from the abandoned e fi lds they (Gimeno et  al. 2012b), i.e., better soil water-retention colonize, as is shown by higher growth rates and lower tree capacity in former agricultural fields due to plowing (Flinn densities. Our results also reveal that the different intraspecific and Marks 2007). However, our study lacked soil nutrient competition found in the stages of forest expansion mediate content information for each of the stages of forest expansion. tree growth response to climate variability under adverse Further studies should assess differences in nitrogen content weather conditions. As such, the greater intraspecific com- among stages of forest expansion in Spanish junipers to help petition in long-existing forests (which, as reported in other understand the specific drivers of the reported stage-based studies, show clumping patterns and lower WUE) is causing differences in growth, wood density, and climate sensitivity. a lower positive response in growth during years with low drought stress compared to the other two stages. Assessing how growth, wood density, and climate-growth sensitivity 5 Conclusions varied among three stages of forest expansion—i.e., the long- existing forests, expanding fronts, and transitions zones— In this study, we assessed the simultaneous effect of sex, for - will improve our understanding of the dynamics of Spanish est expansion stage, and forest structural characteristics on junipers undergoing a natural process of forest expansion. It radial growth and wood density and explored sex- and stage- will also ultimately help in their future preservation and, in based differences in climate-growth sensitivity. We found that particular, prevent risks associated with any increase in the sex had only a minimal effect on the first stages of growth severity and number of drought episodes in the Mediterranean in Spanish junipers (the first 35 years of cambial age) and region. showed that radial growth, wood density, and climate sensitiv- ity varied among stages of forest expansion, mainly due to the different structural characteristics of the forests found in each of the stages. By restricting our analyses to the first 35 years  Appendix Fig. 6 Detrending growth at individual level using cubic smoothing splines of 10  years for the three studied sites and a negative exponential method 1 3 86 Page 12 of 19 Annals of Forest Science (2021) 78: 86 Fig. 7 Detrending growth at individual level using the negative exponential method for the three studied sites 1 3 Annals of Forest Science (2021) 78: 86 Page 13 of 19 86 Fig. 8 Sample depth for the three stages of forest expansion and site. The dashed line indicates the 35 years of cambial age. The gray area indicates a sample depth of 5 trees 1 3 86 Page 14 of 19 Annals of Forest Science (2021) 78: 86 Fig. 9 Tree density variations (violin and boxplots) for each site (a). Percentage of trees per sex for each site (b); all sampled juniper trees were considered (n = 816). Significant differ ences between sexes at each site (b) are denoted by small letters 1 3 Annals of Forest Science (2021) 78: 86 Page 15 of 19 86 Fig. 10 Violins and boxplots displaying differences in repro- ductive trees (n = 451) among stages of forest expansion and at sites for the following vari- ables: tree age (a, f), number of stems per tree (b, g), quadratic diameter (QD, c, h), height (d, i), and crown diameter (e, j). The capital letters indicate significant differences among stages of forest expansion or at sites, while the lowercase letters indicate significant differences between sexes obtained in post- hoc tests from LMMs (Table 4) 1 3 86 Page 16 of 19 Annals of Forest Science (2021) 78: 86 Acknowledgements We are especially grateful for the help, advice, and support provided by David López-Quiroga, José Miguel Olano, Adrián Escudero, Pablo Álvarez, Esteban Manrique, Eduardo Serna and Miguel Díaz. We are also thankful to José Antonio Lozano, direc- tor of the Alto Tajo Natural Park, for facilitating the research at the park. Funding This study was funded by the grants SPONFOREST (Biodi- vERsA3-2015–58), PCIN-2016–055 (financed by the Spanish Research Agency (AEI) and the Spanish Ministry of Economy, Industry and Competitiveness (MINECO)), COMEDIAS (MINECO, CGL2017- 83170-R), and REMEDINAL TE (Ref. TE-CM. S2018/EMT-4338, 2019–2023-Comunidad de Madrid). Data availability The dataset used in the current study will be available on Figshare after a 1-month embargo period in this link: https:// doi. org/ 10. 6084/ m9. figsh are. 14806 683. v1. The provisional link to download the data during the embargo period is https:// figsh are. com/s/ d42a2 9f95d a711d 586a5. Declarations Consent for publication All authors gave their informed consent to this publication and its content. Conflict of interest The authors declare no competing interests. References Acuña-Míguez B, Valladares F, Martín-Forés I (2020) Both mature patches and expanding areas of Juniperus thurifera forests are vul- nerable to climate change but for different reasons. Forests 11:960. https:// doi. org/ 10. 3390/ f1109 0960 Alfaro-Sánchez R, Jump AS, Pino J et al (2019) Land use legacies drive higher growth, lower wood density and enhanced climatic sensitivity in recently established forests. Agric for Meteorol 276– 277:107630. https:// doi. org/ 10. 1016/j. agrfo rmet. 2019. 107630 Alfaro-Sánchez R, Valdés-Correcher E, Espelta JM, et al (2020) How do social status and tree architecture influence radial growth, wood density and drought response in spontaneously established oak forests? Annals of Forest Science 77. https://doi. or g/10. 1007/ s13595- 020- 00949-x Alfaro Sánchez R, Espelta, JM, Valladares F, Acuña-Míguez B, Martín- Forés I (2021) Disentangling the role of sex dimorphism and for- est structure as drivers of growth and wood density in expanding Juniperus thurifera L. woodlands. figshare. Dataset. 10.6084/ m9.figshare.14806683.v1 Anderegg WRL, Schwalm C, Biondi F et al (2015) Pervasive drought legacies in forest ecosystems and their implications for carbon cycle models. Science 349:528. https://do i. org/10 . 1126/ science . aab18 33 Anderson JT, Wagner MR, Rushworth CA et al (2014) The evolution of quantitative traits in complex environments. Heredity (edinb) 112:4–12. https:// doi. org/ 10. 1038/ hdy. 2013. 33 Arbuthnott D, Dutton EM, Agrawal AF, Rundle HD (2014) The ecol- ogy of sexual conflict: ecologically dependent parallel evolution of male harm and female resistance in Drosophila melanogaster. Ecol Lett 17:221–228. https:// doi. org/ 10. 1111/ ele. 12222 Barrett SCH, Yakimowski SB, Field DL, Pickup M (2010) Ecological genetics of sex ratios in plant populations. Philos Trans R Soc Lond B Biol Sci 365:2549–2557. https:// doi. org/ 10. 1098/ rs tb. 2010. 0002 1 3 Table 4 LMM coefficients f or the following dependent variables: tree age, number of stems per tree, quadratic mean diameter (QD), height, and crown diameter using only reproductive trees (n = 451). The variable QD is log-transformed. The marginal and conditional r is also provided (delta for number of stems). The levels sex: female, site: Maranchón, and gradient: long-existing are included in the intercept QD [log] Number of stems Crown diameter Tree age Height Predictors Estimates SE p Estimates SE p Estimates SE p Estimates SE p Estimates SE p (Intercept) 3.08 0.10 < 0.001 2.90 0.27 < 0.001 5.54 0.23 < 0.001 45.64 2.90 < 0.001 5.99 0.24 < 0.001 Gradient [transition] − 0.39 0.11 < 0.001 0.88 0.09 0.179 − 0.99 0.24 < 0.001 − 13.15 3.81 0.001 − 1.29 0.30 < 0.001 Gradient [expanding] − 0.53 0.11 < 0.001 0.71 0.08 0.002 − 1.38 0.26 < 0.001 − 18.34 3.99 < 0.001 − 1.50 0.32 < 0.001 Sex [male] − 0.22 0.05 < 0.001 0.85 0.06 0.012 − 0.51 0.15 0.001 − 2.83 1.17 0.015 − 0.42 0.13 0.001 Site [Ribarredonda] − 0.06 0.11 0.564 0.81 0.09 0.050 − 0.27 0.25 0.277 Site [Huertahernando] 0.18 0.11 0.101 1.04 0.10 0.696 0.84 0.24 < 0.001 Random effects σ 0.28 0.38 2.33 131.82 1.56 τ 0.02 0.01 0.07 34.92 0.20 00 Plot 2 2 Marginal r /conditional r 0.18/0.24 0.09/0.11 0.18/0.21 0.26/0.42 0.22/0.30 Annals of Forest Science (2021) 78: 86 Page 17 of 19 86 Barton K (2019) MuMIn: Multi-Model Inference Gimeno TE, Escudero A, Delgado A, Valladares F (2012b) Previous land use alters the effect of climate change and facilitation on Başnou C, Álvarez E, Bagaria G et al (2013) Spatial patterns of land expanding woodlands of Spanish Juniper. Ecosystems 15:564– use changes across a Mediterranean metropolitan landscape: 579. https:// doi. org/ 10. 1007/ s10021- 012- 9529-z implications for biodiversity management. Environ Manage Gimeno TE, Pías B, Martínez-Fernández J et al (2012c) The decreased 52:971–980. https:// doi. org/ 10. 1007/ s00267- 013- 0150-5 competition in expanding versus mature juniper woodlands is Blanco BC, Castro EB (1997) Los bosques ibéricos: una interpretación counteracted by adverse climatic effects on growth. Eur J Forest geobotánica. Planeta Res 131:977–987. https:// doi. org/ 10. 1007/ s10342- 011- 0569-2 Cook ER, Kairiukstis LA (eds) (1990) Methods of dendrochronology. Granda E, Rossatto DR, Camarero JJ et al (2014) Growth and carbon Springer Netherlands, Dordrecht isotopes of Mediterranean trees reveal contrasting responses to Cybis Elektronik (2018) CDendro and CooRecorder increased carbon dioxide and drought. Oecologia 174:307–317. Dawson TE, Bliss LC (1989) Patterns of water use and the tissue https:// doi. org/ 10. 1007/ s00442- 013- 2742-4 water relations in the dioecious shrub, Salix arctica: the physi- Greenwood S, Ruiz-Benito P, Martínez-Vilalta J et  al (2017) Tree ological basis for habitat partitioning between the sexes. Oecolo- mortality across biomes is promoted by drought intensity, lower gia 79:332–343. https:// doi. org/ 10. 1007/ BF003 84312 wood density and higher specific leaf area. Ecol Lett 20:539–553. De Schrijver A, De Frenne P, Ampoorter E et al (2011) Cumulative https:// doi. org/ 10. 1111/ ele. 12748 nitrogen input drives species loss in terrestrial ecosystems. Glob Guerrieri MR, Correia M, Martin-Forés I, et al (2021) Land-use lega- Ecol Biogeogr 20:803–816. https://doi. or g/10. 1111/j. 1466- 8238. cies influence tree water-use efficiency and nitrogen availability 2011. 00652.x in recently established European forests. Functional Ecology. In DeSoto L, Olano JM, Rozas V (2016) Secondary growth and carbo- press hydrate storage patterns differ between sexes in Juniperus thu- Hacke UG, Sperry JS, Pockman WT et al (2001) Trends in wood den- rifera. Frontiers in Plant Science 7. https:// doi. org/ 10. 3389/ fpls. sity and structure are linked to prevention of xylem implosion 2016. 00723 by negative pressure. Oecologia 126:457–461. https://doi. or g/10. Escribano-Avila G, Sanz-Pérez V, Pías B et al (2012) Colonization of 1007/ s0044 20100 628 abandoned land by Juniperus thurifera is mediated by the inter- Hampe A, Alfaro-Sánchez R, Martín-Forés I (2020) Establishment of action of a diverse dispersal assemblage and environmental het- second-growth forests in human landscapes: ecological mecha- erogeneity. PLoS ONE 7:e46993. https:// doi. org/ 10. 1371/ journ nisms and genetic consequences. Ann For Sci 77:87, s13595–020– al. pone. 00469 93 00993–7. https:// doi. org/ 10. 1007/ s13595- 020- 00993-7 Espelta JM, Retana J, Habrouk A (2003) Resprouting patterns after fire Haylock MR, Hofstra N, Klein Tank AMG, et al (2008) A European and response to stool cleaning of two coexisting Mediterranean daily high-resolution gridded data set of surface temperature and oaks with contrasting leaf habits on two different sites. For Ecol precipitation for 1950–2006. Journal of Geophysical Research Manage 179:401–414. https:// doi. org/ 10. 1016/ S0378- 1127(02) 113. https:// doi. org/ 10. 1029/ 2008J D0102 01 00541-8 Holmes RL (1983) Computer-assisted quality control in tree-ring dat- Flinn KM, Marks PL (2007) Agricultural legacies in forest environ- ing and measurement. Tree-Ring Bulletin 43:69–78 ments: tree communities, soil properties, and light availability. Juvany M, Munné-Bosch S (2015) Sex-related differences in stress Ecol Appl 17:452–463. https:// doi. org/ 10. 1890/ 05- 1963 tolerance in dioecious plants: a critical appraisal in a physiologi- Forrester DI (2015) Transpiration and water-use efficiency in mixed- cal context. J Exp Bot 66:6083–6092. https://doi. or g/10. 1093/ jxb/ species forests versus monocultures: effects of tree size, stand erv343 density and season. Tree Physiol 35:289–304. https:// doi. org/ 10. Keenan RJ (2015) Climate change impacts and adaptation in forest 1093/ treep hys/ tpv011 management: a review. Ann for Sci 72:145–167. https:// doi. org/ Franceschini T, Longuetaud F, Bontemps J-D et al (2013) Effect of 10. 1007/ s13595- 014- 0446-5 ring width, cambial age, and climatic variables on the within-ring Kock N, Lynn G (2012) Lateral collinearity and misleading results in wood density profile of Norway spruce Picea abies (L.) Karst. variance-based SEM: an illustration and recommendations. JAIS Trees 27:913–925. https:// doi. org/ 10. 1007/ s00468- 013- 0844-6 13:546–580. https:// doi. org/ 10. 17705/ 1jais. 00302 Freschet GT, Östlund L, Kichenin E, Wardle DA (2014) Aboveground Kuznetsova A, Brockhoff PB, Christensen RHB (2017) lmerTest pack - and belowground legacies of native Sami land use on boreal for- age: tests in linear mixed effects models. Journal of Statistical est in northern Sweden 100 years after abandonment. Ecology Software 82:1–26. https:// doi. org/ 10. 18637/ jss. v082. i13 95:963–977. https:// doi. org/ 10. 1890/ 13- 0824.1 Lambers H, Chapin FS, Pons TL (2008) Plant physiological ecology, Gauquelin T, Bertaudiere V, Montes N et al (1999) Endangered stands 2nd ed. Springer, New York of thuriferous juniper in the western Mediterranean basin: eco- Lambin EF, Meyfroidt P (2011) Global land use change, economic logical status, conservation and management. Biodivers Conserv globalization, and the looming land scarcity. Proc Natl Acad Sci 8:1479–1498 108:3465–3472. https:// doi. org/ 10. 1073/ pnas. 11004 80108 Gauquelin T, Bertaudière-Montès A, Badri W, Montès N (2002) Sex Lenth RV (2016) Least-Squares Means: The R Package lsmeans. J Stat ratio and sexual dimorphism in mountain dioecious thuriferous Softw 69:1–33. https:// doi. org/ 10. 18637/ jss. v069. i01 juniper (Juniperus thurifera L., Cupressaceae). Bot J Linn Soc Linares J-C, Delgado-Huertas A, Julio Camarero J et al (2009) Com- 138:237–244. https:// doi. org/ 10. 1046/j. 1095- 8339. 2002. 13800 petition and drought limit the response of water-use efficiency 2237.x to rising atmospheric carbon dioxide in the Mediterranean fir Gazol A, Camarero JJ, Vicente-Serrano SM et  al (2018) Forest Abies pinsapo. Oecologia 161:611–624. https:// doi. org/ 10. 1007/ resilience to drought varies across biomes. Glob Change Biol s00442- 009- 1409-7 24:2143–2158. https:// doi. org/ 10. 1111/ gcb. 14082 Lüdecke D (2021) sjPlot: data visualization for statistics in social Gerstner K, Dormann CF, Stein A et al (2014) Effects of land use on science plant diversity - a global meta-analysis. J Appl Ecol 51:1690– Lüdecke D, Makowski D, Waggoner P, Patil I (2020) performance: 1700. https:// doi. org/ 10. 1111/ 1365- 2664. 12329 assessment of regression models performance Gimeno TE, Camarero JJ, Granda E et al (2012a) Enhanced growth Lundgren C (2004) Microfibril angle and density patterns of fertilized of Juniperus thurifera under a warmer climate is explained by a and irrigated Norway spruce. Silva Fenn 38. https:// doi. org/ 10. positive carbon gain under cold and drought. Tree Physiol 32:326– 14214/ sf. 439 336. https:// doi. org/ 10. 1093/ treep hys/ tps011 1 3 86 Page 18 of 19 Annals of Forest Science (2021) 78: 86 Martín-Forés I, Magro S, Bravo-Oviedo A, et al (2020) Spontaneous Rozas V, DeSoto L, Olano JM (2009) Sex-specific, age-dependent forest regrowth in South-West Europe: consequences for nature’s sensitivity of tree-ring growth to climate in the dioecious tree contributions to people. People and Nature. In press Juniperus thurifera. New Phytol 182:687–697. https://doi. or g/10. McKown AD, Klápště J, Guy RD, et al (2017) Sexual homomorphism 1111/j. 1469- 8137. 2009. 02770.x in dioecious trees: extensive tests fail to detect sexual dimor- Sánchez-Salguero R, Linares JC, Camarero JJ et al (2015) Disentan- phism in Populus. Scientific Reports 7. https:// doi. org/ 10. 1038/ gling the effects of competition and climate on individual tree s41598- 017- 01893-z growth: a retrospective and dynamic approach in Scots pine. For Montesinos D, De Luís M, Verdú M et al (2006) When, how and how Ecol Manage 358:12–25. https:// doi. org/ 10. 1016/j. foreco. 2015. much: gender-specific resource-use strategies in the dioecious tree 08. 034 Juniperus thurifera. Ann Bot 98:885–889. https://doi. or g/10. 1093/ Stewart JL, Salazar R (1992) A review of measurement options for aob/ mcl172 multipurpose trees. Agrofor Syst 19:173–183. https:// doi. org/ 10. Montesinos D, García-Fayos P, Verdú M (2012) Masting uncoupling: 1007/ BF001 38507 mast seeding does not follow all mast flowering episodes in a Tasser E, Walde J, Tappeiner U et al (2007) Land-use changes and dioecious juniper tree. Oikos 121:1725–1736. https:// doi. org/ 10. natural reforestation in the Eastern Central Alps. Agr Ecosyst 1111/j. 1600- 0706. 2011. 20399.x Environ 118:115–129. https://doi. or g/10. 1016/j. ag ee.2006. 05. 004 Nadal-Romero E, Otal-Laín I, Lasanta T et al (2018) Woody encroach- Thompson JD (2005) Plant Evolution in the Mediterranean. Oxford ment and soil carbon stocks in subalpine areas in the Central University Press, Oxford Spanish Pyrenees. Sci Total Environ 636:727–736. https:// doi. Tozawa M, Ueno N, Seiwa K (2009) Compensatory mechanisms for org/ 10. 1016/j. scito tenv. 2018. 04. 324 reproductive costs in the dioecious tree Salix integra. Botany Obeso JR (2002) The costs of reproduction in plants. New Phytol 87:315–323. https:// doi. org/ 10. 1139/ B08- 125 155:321–348. https://doi. or g/10. 1046/j. 1469- 8137. 2002. 00477.x Valdés-Correcher E, van Halder I, Barbaro L et al (2019) Insect her- Olano JM, García-Cervigón AI, Arzac A, Rozas V (2015) Intra-annual bivory and avian insectivory in novel native oak forests: divergent effects of stand size and connectivity. For Ecol Manage 445:146– wood density fluctuations and tree-ring width patterns are sex- 153. https:// doi. org/ 10. 1016/j. foreco. 2019. 05. 018 and site-dependent in the dioecious conifer Juniperus thurifera L. Vasiliauskas SA, Aarssen LW (1992) Sex ratio and neighbor effects in Trees 29:1341–1353. https://doi. or g/10. 1007/ s00468- 015- 1212-5 monospecific stands of Juniperus Virginiana. Ecology 73:622– Olano JM, Rozas V, Bartolomé D, Sanz D (2008) Effects of changes in 632. https:// doi. org/ 10. 2307/ 19407 68 traditional management on height and radial growth patterns in a Vicente-Serrano SM, Beguería S, López-Moreno JI (2010) A multi- Juniperus thurifera L. woodland. For Ecol Manage 255:506–512. scalar drought index sensitive to global warming: the standard- https:// doi. org/ 10. 1016/j. foreco. 2007. 09. 015 ized precipitation evapotranspiration index. J Clim 23:1696–1718. Olano JM, Zavala MA, Rozas V (2012) Disruption of Juniperus thu- https:// doi. org/ 10. 1175/ 2009J CLI29 09.1 rifera woodland structure in its northwestern geographical range: Vilà-Cabrera A, Espelta JM, Vayreda J, Pino J (2017) “New Forests” potential drivers and limiting factors. Eur J Forest Res 131:563– from the twentieth century are a relevant contribution for c storage 570. https:// doi. org/ 10. 1007/ s10342- 011- 0531-3 in the Iberian peninsula. Ecosystems 20:130–143. https://d oi.o rg/ Ortiz PL, Arista M, Talavera S (2002) Sex ratio and reproductive effort 10. 1007/ s10021- 016- 0019-6 in the dioecious Juniperus communis subsp. alpina (Suter) Celak. Villellas J, Martín-Forés I, Mariette S, et al (2020) Functional distance (Cupressaceae) along an altitudinal gradient. Ann Bot 89:205– is driven more strongly by environmental factors than by genetic 211. https:// doi. org/ 10. 1093/ aob/ mcf028 relatedness in Juniperus thurifera L. expanding forest stands. Ann Palmero-Iniesta M, Espelta JM, Gordillo J, Pino J (2020) Changes For Sci 77:66. https:// doi. org/ 10. 1007/ s13595- 020- 00973-x in forest landscape patterns resulting from recent afforestation Williamson GB, Wiemann MC (2010) Measuring wood specific grav - in Europe (1990–2012): defragmentation of pre-existing forest ity…Correctly. Am J Bot 97:519–524. https:// doi. org/ 10. 3732/ versus new patch proliferation. Ann For Sci ajb. 09002 43 Pretzsch H, Biber P, Schütze G et al (2018) Wood density reduced while wood volume growth accelerated in Central European for- ests since 1870. For Ecol Manage 429:589–616. https:// doi. org/ Publisher's note Springer Nature remains neutral with regard to 10. 1016/j. foreco. 2018. 07. 045 jurisdictional claims in published maps and institutional affiliations. 1 3 Annals of Forest Science (2021) 78: 86 Page 19 of 19 86 Authors and Affiliations 1 2 3 3 4 Raquel Alfaro‑Sánchez  · Josep Maria Espelta  · Fernando Valladares  · Belén Acuña‑Míguez  · Irene Martín‑Forés * Raquel Alfaro-Sánchez Department of Biology, Wilfrid Laurier University, 75 r.alfarosanchez@gmail.com University Avenue W, Waterloo, ON N2L 3C5, Canada Josep Maria Espelta Centre de Recerca Ecològica I Aplicacions Forestals, josep.espelta@uab.cat CREAF, Bellaterra (Cerdanyola de Vallès), 08193 Catalonia, Spain Fernando Valladares valladares@ccma.csic.es Departament of Biogeography and Global Change, National Museum of Natural Sciences, Spanish Council for Scientific Belén Acuña-Míguez Research, CSIC, C/Serrano, 115dpdo, 28006 Madrid, Spain belacumig@gmail.com School of Biological Sciences, The University of Adelaide, Irene Martín-Forés Adelaide, South Australia 5005, Australia imfores@pdi.ucm.es 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Forest Science Springer Journals

Disentangling the role of sex dimorphism and forest structure as drivers of growth and wood density in expanding Juniperus thurifera L. woodlands

Download PDF
19 pages

Loading next page...
 
/lp/springer-journals/disentangling-the-role-of-sex-dimorphism-and-forest-structure-as-rgBUQMmTqp

References (74)

Publisher
Springer Journals
Copyright
Copyright © INRAE and Springer-Verlag France SAS, part of Springer Nature 2021
ISSN
1286-4560
eISSN
1297-966X
DOI
10.1007/s13595-021-01097-6
Publisher site
See Article on Publisher Site

Abstract

Key message The dioecious tree species Juniperus thurifera L. is undergoing a spontaneous process of forest expan- sion in southwest Europe. We investigated how growth, climate sensitivity, and wood density varied simultaneously between sexes and among stages of expansion while accounting for the variability of forest structure. We found few sex-based differences but detected lower wood density, greater growth rates, and higher sensitivity to drought in expanding fronts compared to long-existing forests. Context Juniperus thurifera L. (Spanish juniper) is a dioecious tree species undergoing a natural process of forest expan- sion in southwest Europe. Aims To assess how radial growth and wood density are simultaneously shaped by sex-based differences, the stage of forest expansion (long-existing forests, transition zones, and expanding fronts), variability in forest structure, and climate (in the case of radial growth). Methods We measured forest structure characteristics, tree rings, and wood density in 17 plots dominated by Spanish juniper in three stages of forest expansion in central Spain. We used linear mixed models (LMMs) to explore the main drivers of variability in radial growth and wood density and sex- and stage-based differences in climate-growth sensitivity. Results Rather than by sex, growth and wood density were mainly shaped by the stage of forest expansion, forest structure variables that characterize these stages, and climate variables (in the case of growth). Conclusion Sexual dimorphism had a minimal effect in growth and wood density in expanding Spanish juniper woodlands. Expanding fronts could be benefiting from land-use legacies in the abandoned fields they are colonizing, as reflected in higher growth rates and lower wood density, especially during years with less summer drought stress. However, this pattern could be reversed in the event of an increase in drought episodes. Keywords Dioecy · Juniperus thurifera L. · Tree rings · Forest expansion · Spontaneous tree establishment · Abandoned fields · Mediterranean region 1 Introduction Handling Editor: Shuguang (Leo) Liu In recent decades, southwest Europe has experienced a sig- This article is part of the topical collection on Establishment nificant expansion of Mediterranean and temperate tree spe- of second-growth forests in human landscapes: ecological mechanisms and genetic consequences cies that differ in their tolerance to harsh environmental con- ditions: Quercus robur L., Quercus ilex L., Fagus sylvatica Contribution of the co-authors Conceptualization: Irene Martín- L., and Juniperus thurifera L. (Gimeno et al. 2012c; Başnou Forés, Raquel Alfaro-Sánchez; Methodology: Raquel Alfaro- et al. 2013; Vilà-Cabrera et al. 2017; Valdés-Correcher et al. Sánchez, Irene Martín-Forés; Formal analysis and investigation: 2019; Acuña-Míguez et al. 2020; Martín-Forés et al. 2020). Raquel Alfaro-Sánchez; Writing—original draft preparation: Raquel Alfaro-Sánchez; Writing—review and editing: Josep European forests have undergone natural expansion—pri- Maria Espelta, Belén Acuña-Míguez, Irene Martín-Forés; Funding marily into former cultivated areas—as a consequence of a acquisition: Fernando Valladares; Supervision: Irene Martín-Forés rural exodus and the abandonment of traditional agricultural practices (Keenan 2015; Palmero-Iniesta et al. 2020; Hampe Extended author information available on the last page of the article Vol.:(0123456789) 1 3 86 Page 2 of 19 Annals of Forest Science (2021) 78: 86 et al. 2020). The vegetation dynamics of abandoned fields is reported in many dioecious trees, including Spanish juniper driven by soil characteristics, climate conditions, and prop- (Gauquelin et al. 2002; Montesinos et al. 2006; Rozas et al. agule availability (Tasser et al. 2007). Land-use legacies 2009; Juvany and Munné-Bosch 2015). Previous studies of from previous agricultural practices including greater nutri- Spanish juniper woodlands die ff r regarding the presence of ent content (De Schrijver et al. 2011) or more decomposer sexual dimorphism in radial growth and climate-growth sen- activity (Freschet et al. 2014) facilitate the establishment sitivity: differences between sexes were found for these traits of trees and shrubs during the initial period after coloni- by Montesinos et al. (2006), Rozas et al. (2009), Olano et al. zation and give rise to increased tree growth (Lambin and (2015), and DeSoto et al. (2016), but no evidence was found Meyfroidt 2011; Gerstner et al. 2014). Recently expanding by Gimeno et al. (2012c). In fact, as a response to environmen- forests in Europe have reported greater growth rates than tal heterogeneity, in dioecious plants, sex-based variation in long-existing forests—even after accounting for the effects non-reproductive traits may be more limited than intraspecific of age and competition—but lower wood density (Pretzsch variation (e.g., Anderson et al. 2014). Thereby, the inclusion of et al. 2018; Alfaro-Sánchez et al. 2019, 2020). A reduction forest structure variables as covariates in experimental design in wood density implies a greater number and surface area may improve our understanding of the trade-offs between of conductive vessels that can modify the growth-climate growth and reproduction (Obeso 2002; McKown et al. 2017). sensitivity of trees (Greenwood et  al. 2017). However, In this study, we assessed whether or not radial growth expanding forests of broadleaf species such as F. sylvatica and wood density in juniper woodlands undergoing a nat- have lower wood density but no significant differences in ural process of forest expansion in southwest Europe are growth-climate sensitivity when compared to long-existing simultaneously influenced by (i) sex-based differences, (ii) forests, probably because of the better growth conditions the stage of forest expansion, and (iii) the variables of for- present in the former croplands they are colonizing (Alfaro- est structure that characterize each stage (e.g., tree density, Sánchez et al. 2019). Yet, little is known about the response age, and size). We also explored sex- and stage-based differ - in growth, wood density, or climate sensitivity in expanding ences with regard to climate-growth sensitivity. We studied tree species of other functional types such as the evergreen a unique mosaic of patches of Spanish juniper in central Mediterranean conifer J. thurifera (but see Gimeno et al. Spain that, to the best of our knowledge, contains the larg- 2012c). est number of Spanish juniper trees in this hotspot for this The J. thurifera (Spanish juniper) is a dioecious spe- species (90% of its world distribution is located in Spain, cies of tree that is endemic to North Africa and the Ibe- Blanco and Castro 1997). In these patches, a mix of young rian Peninsula (Blanco and Castro 1997), recognized under and mature junipers grow along a gradient of forest expan- the European Habitats Directive. It is drought-tolerant and sion, from low-density areas in expanding fronts to rela- under continental climates tends to grow in low densities tively dense areas in long-existing stands. We hypothesized (Olano et al. 2008). It is well adapted to rocky, poorly devel- that sex-based differences in growth would increase with oped soils (Gauquelin et al. 1999) and has been shown to cambial age due to the increase with age of reproductive increase in secondary growth under harsh climatic condi- costs in females (Montesinos et al. 2012). Given the find- tions (Granda et al. 2014). The Spanish juniper is currently ings of previous studies on the effect of land-use legacies undergoing wide-ranging expansion and densification pro- on growth, wood density, and climate-growth sensitivity cesses in certain areas of central Spain (Blanco and Castro (Alfaro-Sánchez et al. 2019), we expected greater growth, 1997; Thompson 2005), whereas in other areas, it is facing lower wood density, and higher climate sensitivity in both competition from other fast-growing tree species including female and male junipers in the expanding fronts that are pines and oaks (Olano et al. 2012). Since the middle of the colonizing former croplands. Overall, our results should help nineteenth century, the expansion of Spanish juniper wood- understand how variation in forest structure—in this case, lands has been facilitated by the progressive abandonment of mainly derived from the processes of forest expansion into the traditional management of wood-pasture systems and the abandoned fields—can influence the expression of second- existence of a diverse dispersal community that ensures seed ary die ff rences between sexes in Spanish juniper woodlands, availability and the frugality of the species (Thompson 2005; which ultimately will help their future preservation. Escribano-Avila et  al. 2012). The expansion of Spanish juniper woodlands in central Spain has created a mosaic of forest patches ranging from long-existing stands to expand-2 Methods ing fronts. Most of these expanding fronts are composed of young juniper trees, and their performance and survival 2.1 Study area under current and future climate hazards remain unknown. Sexual dimorphism in secondary (non-reproductive) sexual This study was conducted in central Spain in the Alto Tajo characteristics (e.g., growth, vigor, and physiology) has been Natural Park and surrounding areas (Fig. 1). The climate of 1 3 Annals of Forest Science (2021) 78: 86 Page 3 of 19 86 this area is continental Mediterranean, characterized by hot to the 1950s (Escribano-Avila et al. 2012; Gimeno et al. dry summers and cold winters. Mean annual temperature 2012c; Villellas et al. 2020; Acuña-Míguez et al. 2020). We and precipitation were 10.7 °C and 462 mm for the period defined long-existing forests as patches containing cores of 1950–2017, respectively (KNMI Climate Explorer; http:// well-preserved juniper woodlands that existed in 1956. The climex p.k nmi.n l/). We selected three sites with unmanaged expanding fronts correspond to areas of recent colonization and well-preserved juniper woodlands, in which the characterized by scattered trees located on former Spanish juniper is the dominant tree species. Maranchón agricultural land. The transition zones were intermediate is the northernmost site, followed by Ribarredonda and forests between these two extremes of gradient. All three then Huertahernando to the south (Fig. 1). The maximum forest stages were represented at all three sites (Fig. 1). It is distance between sites was 30  km along an elevation important to note that the three age categories were defined gradient of 1000–1300  m a.s.l. Seventeen plots were by the land-use age and not by the age of trees in the stand, established at these three sites along a gradient of forest as in previous studies (Başnou et al. 2013, Alfaro-Sánchez expansion: seven plots in Maranchón and five plots in both et  al. 2019). Therefore, long-existing juniper patches in Ribarredonda and Huertahernando (Fig. 1). The gradient the area do not necessarily have a tree age structure that is of forest expansion was separated into three stages, namely, characteristic of old-growth forests and may even have been long-existing forests, expanding fronts, and transition zones. disturbed by thinning or fires recently, so young trees are The stages were identified by previous studies of the area also present in these patches. based on the comparison of land-cover maps dating back Fig. 1 Location of the three sampling sites (Maranchón, Ribarre- cated for each plot (EXP, expanding front; TRAN, transition zone; donda, and Huertahernando) in central Spain in the Alto Tajo Natural LON, long-existing forest) Park and surrounding areas. The gradient of forest expansion is indi- 1 3 86 Page 4 of 19 Annals of Forest Science (2021) 78: 86 et al. 2019, 2020). We used the second cores to count and 2.2 Field sampling measure annual ring widths. They were air-dried, glued onto wooden mounts, and polished using sandpaper of progres- Exhaustive field sampling was conducted in autumn 2017. In total, we georeferenced 816 juniper trees in the study plots. sively finer grain until tree rings were visible. The cores were dated with a stereomicroscope and scanned at 2400 Due to the differences in tree density among forest stages, plot areas were flexible to be able to sample a minimum of d.p.i. We measured ring widths to an accuracy of 0.001 mm using the CooRecorder v9.3 software (Cybis Elektronik 35 adult trees per plot (Table 1). We visually identified the sex of all the selected individuals when they had male flow - 2018). The dataset is available in Alfaro-Sánchez et al. 2020. The cross-dating of individual series was checked using the ers or female cones and considered these trees to be repro- ductive individuals. If the tree did not have o fl wers or cones, CooRecorder and COFECHA programs (Holmes 1983). For subsequent climate-growth analyses, individual tree ring we scored it as a tree of unknown sex. The 451 trees identi- fied as reproductive at the time of sampling had a minimum width series were detrended with cubic smoothing splines of 10 years to remove non-climatic growth trends related size threshold of trunk diameter at breast height of ≥ 3 cm and total height of > 1.40 m. Above this size threshold, we to the increase in tree age and size (Cook and Kairiukstis 1990). A comparison between detrended methods is shown found that only 5.5% of trees were of unknown sex. We determined the mean tree density per plot and meas- in Figs. 6 and 7. ured the quadratic diameter (QD, calculated as the square root of the sum of square diameter at breast height of each 2.4 Climate sensitivity stem of a tree; Stewart and Salazar 1992), the maximum tree height (measured with Haglöf Vertex IV hypsometer), and Sums of the monthly mean temperatures and precipitation were accessed for the period 1950–2017 from the homog- the average crown diameter calculated as the mean of the projection of two perpendicular axes passing through the enized and quality-checked E-OBS v.17.0 dataset (Haylock et  al. 2008) in the KNMI Climate Explorer ( http:// clime axis of the trunk (measured with a Haglöf DME distance measurer). xp. knmi. nl/). Our study sites are spatially located in two different E-OBS v.17.0 grid cells, Maranchón in one and 2.3 Wood density and tree growth measurements Ribarredonda and Huertahernando in another. Given the proximity of the sites, we averaged the climate data from We only measured the wood density and tree growth of the these two grid cells at 0.25° spatial resolution for use in subsequent analyses. We calculated the drought index SPEI 451 reproductive trees. We extracted two increment cores 50 cm from ground height using a Pressler increment borer (standardized precipitation-evapotranspiration) using the R package SPEI (Vicente-Serrano et al. 2010) with a time scale (0.5 cm; Haglöf, Långsele, Sweden). One of these two cores −3 was used to estimate wood density (g cm ), following Wil- of 3 months, based on temperature and precipitation data from E-OBS v.17.0. Lower values of SPEI correspond to liamson and Wiemann (2010), which is calculated as the dry weight of the full core (including heartwood and softwood) greater drought stress. divided by its saturated volume (see also Alfaro-Sánchez Table 1 Plot and tree characteristics at different sites and stages along the forest expansion gradient Plot area Sex Tree den- Tree age* QD (cm) Height Crown Number of (ha) sity (trees (m) diameter stems −1 ha ) (m) Site (lat., long.) Gradient Plots n Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD Maranchón Long 2 59 0.58 0.17 0.6 0.5 115 14 50 19 24 15 6 1.9 5.5 2 2.4 1.8 (41.06, − 2.20) Transition 3 86 0.84 0.32 0.6 0.5 71 29 33 13 15 9 4.1 1.2 4.2 1.4 2.4 1.7 Expanding 1 30 1.35 0.7 0.5 32 26 8 11 5 3.8 0.7 3.8 1 2 1.1 Ribarredonda Long 2 56 0.42 0.03 0.6 0.5 125 9 46 14 21 10 5.7 1.4 4.9 1.7 2.4 1.9 (40.87, − 2.30) Transition 1 28 0.35 0.7 0.4 123 26 10 12 8 4.6 1.5 3.6 1.8 1.5 1.1 Expanding 2 48 0.83 0.2 0.8 0.4 49 8 28 12 12 7 4.4 1.2 3.8 1.3 1.6 1.1 Huertahernando Long 1 29 0.6 0.6 0.5 89 33 11 23 10 5.5 1.1 6 1.4 3 2 (40.83, − 2.28) Transition 2 58 1 0.5 0.6 0.5 68 34 30 14 18 9 4.9 1.3 5 1.5 2.5 1.5 Expanding 2 57 1.8 0.05 0.6 0.5 22 1 23 7 14 7 4.1 1.1 4.4 1.5 1.7 1.1 QD: quadratic mean diameter; sex: female = 0, male = 1. *Tree age is estimated as the number of rings measured per core. The total tree age of trees is expected to be older. 1 3 Annals of Forest Science (2021) 78: 86 Page 5 of 19 86 To assess the effect of climate on growth, we ran linear with age, we included the interactions between cambial age regression models (LMs) at tree level between detrended with sex and stages of forest expansion. Cambial age is the ring width series and climate variables. The considered cli- age of the sampled cores, where the first year corresponds to mate variables were monthly and seasonal (combination of the first tree ring outside the pith. To determine if climate- several consecutive months) temperatures and SPEI-3 from growth sensitivity varied between sexes and among stages, September of the previous growing season to September we included the interactions between February–April tem- of the year in which the ring was formed. We calculated perature and August SPEI-3 with sex and stages. To allow the percentage of trees with significant slopes for each of for comparisons among stages of forest expansion, annual the climate variables considered in the LMs to identify growth was restricted to the first 35 years of life of the trees, the month or season with the highest temperature-growth the maximum period of time in which the three stages over- and drought-growth sensitivity, that is, the average Febru- lapped with a minimum of five trees at each site (see Fig.  8). ary–April temperatures and August SPEI-3 (see “Results” Tree individuals nested in plot was included as a random section). We used these two climate variables as explanatory ee ff ct to compensate for the repeated measures taken from an variables in the linear mixed effects models (LMMs) in the individual tree. A first-order autocorrelation structure (AR1) subsequent analyses of climate-growth sensitivity (see the was also included in the growth LMM to control for the following sections). temporal autocorrelation of growth measures. In both the wood density and growth models, the explana- 2.5 Statistical analyses tory variables were standardized to eliminate differences in scale measurements. Crown diameter, tree height, and QD We used ANOVAs to test for differences in tree density and were highly correlated, and so to avoid collinearity, we only the proportion of sexes in stages of forest expansion and at included QD in the models. We compared alternative full- sites (significance level was set at p < 0.05). We used LMMs saturated models including linear or spline adjustments (up and generalized linear mixed effects models (GLMMs) to to two degrees of freedom) for each continuous predictor and assess how tree characteristics varied between sexes and selected the adjustment used in the model with the lowest among stages of forest expansion and sites. Specifically, AIC. We then checked for collinearity within the predictor LMMs were used to model the following dependent vari- variables included in the two selected full-saturated models ables: tree age, QD, height, and crown diameter; a general- (including linear or spline adjustments for the continuous ized linear mixed effects model (GLMM) with a Poisson variables) using variance inflation factors (VIF) with the regression was used to model the number of stems. For all performance R package (Lüdecke et al. 2020); we found models (LMMs and GLMM), the explanatory variables were VIF scores ≤ 10, which indicates a lack of collinearity (Kock sex, stages of forest expansion, and site. Plot was included as and Lynn 2012). A full model for wood density (WD) would a random effect. We used Tukey–Kramer post-hoc analysis look like this in R syntax (considering linear adjustments for for multiple comparisons in the R package lsmeans (Lenth the continuous variables, Eq. 1): 2016) to identify specic fi die ff rences in tree age, QD, height, lme(WD ∼ Sex + Stage + Site + QD crown diameter, and number of stems among the three stages + nstems + TD + Age + Age × Sex of forest expansion and at sites and between sexes for each (1) stage and site. + Age × Stage, random =∼ 1Plot, We used LMMs to assess how wood density and annual � data = dataset, method= ML ) growth varies between sexes, among stages of forest expan- sion, and is associated with forest structure. Specifically, A full model for log-transformed growth (log[Growth]) for the wood density LMM, the explanatory variables were would look like this in R syntax (considering linear adjust- sex, stage of forest expansion, site, QD, number of stems ments for the continuous variables, Eq. 2): (nstems), tree density (TD), and tree age. To determine lme(log[Growth] ∼ Sex + Stage + Site + QD if wood density varied between sexes or among stages in + nstems + TD + Cambial age + T + SPEI terms of tree age, we included in the model the interactions between tree age with sex and stages of forest expansion. − 3 + Cambial age × Sex + Cambial age× Plot was included as a random effect. For the growth LMM, Stage + T × Sex + T × Stage + SPEI the dependent variable, annual growth, was transformed − 3 × Sex + SPEI − 3 × Stage, with a natural logarithm to conform to normality. The random =∼ 1Plot∕Tree, explanatory variables were sex, stage of forest expansion, site, QD, number of stems, tree density, cambial age, Febru- correlation = corCAR1(form =∼ 1Plot∕Tree), � � ary–April temperature (T), and August SPEI-3 (SPEI-3). To data = dataset, method = ML (2) determine if growth varied between sexes or among stages 1 3 86 Page 6 of 19 Annals of Forest Science (2021) 78: 86 Next, the models were reduced to those with the lowest per tree than the expanding fronts; the transition zone Akaike Information Criterion (AIC) (the best or most parsi- showed no significant differences when compared to the monious models) using the dredge function from the MuMIn other two stages (Fig. 10b, Table 1). We found no significant R package (Barton 2019). The method was set to maxi- differences among sites for the variables tree age, number of mum likelihood (ML) during the fixed effect model selec- stems, QDs, or height, although larger crown diameters were tion phase, although the final models are presented using found in Huertahernando, the southernmost site (Fig. 10f–h, restricted maximum likelihood (REML) (Kuznetsova et al. Table 1). 2017). Model fits were checked visually to ensure that they conformed to model assumptions. We calculated marginal 3.1 Wood density patterns (i.e., the proportion of variance explained by fixed effects) and conditional (i.e., the proportion of variance explained Wood density increased with tree age, decreased with QD, by fixed and random effects) r with the sjPlot R package and varied among stages of forest expansion and sites. Post- (Lüdecke 2021). All statistical analyses were performed hoc analysis confirmed significantly lower wood density in using R version 3.5.1 (R Core Team 2019). the expanding front and at the southernmost site (Huerta- hernando) compared to the transition zone and the north- ernmost site (Maranchón), respectively. The long-existing 3 Results forests and Ribarredonda showed no significant differences in wood density compared to the other two stages and sites, Tree density increased gradually from the sparse expand- respectively (Table 2, Fig. 3). −1 ing fronts (28 trees ha ) to the relatively dense (for juniper −1 woodland) long-existing forests (116 trees ha ). The transi- tion zone showed no significant differences in tree density −1 (83 trees ha ) with either the expanding fronts or the long- existing forests (Fig. 2a). No significant differences were found in tree density among sites (Fig. 9a). Table 2 Wood density LMM coefficients. The marginal and con- The proportion of male trees doubled that of female trees ditional r2 are also shown. The level site: Maranchón and the level in the three stages of forest expansion and at all three sites, stage: expanding is included in the intercept. SE standard error. Plot i.e., ~ 60% of trees were male and ~ 30% females (Fig. 2b, was included as a random effect in the model, but the within-group variance (or residual variance, σ2) and the between-group variance 9b). The highest proportion of unknown sex trees (9%) was (or random intercept variance, τ00 Plot) were effectively 0 detected in the expanding front (Fig.  2b) and at Ribarre- donda (Fig.  9b), probably due to a higher proportion of Predictors Estimates SE p younger trees (Fig. 10a, f). (Intercept) 0.545 0.004 < 0.001 Female trees were significantly older and taller and had Quadratic diameter − 0.003 0.002 0.097 greater QDs, more stems per tree, and larger crown diam- Age 0.010 0.002 < 0.001 eters than male trees for each stage of forest expansion and Site [Ribarredonda] − 0.005 0.004 0.218 site (Fig.  10). The long-existing forests had significantly Site [Huertahernando] − 0.015 0.004 < 0.001 older and taller trees, greater QDs, and larger crown diam- Stage [Transition] 0.012 0.004 0.003 eters than the trees in the transition zone and the expanding Stage [Long] 0.006 0.005 0.202 front (Fig. 10a, c–e, Table 1; Acuña-Míguez et al. 2020). 2 2 Marginal r /conditional r 0.14/0.14 The long-existing forests also had significantly more stems Fig. 2 Variations in tree density (violin and boxplots) for each stage of forest expansion (a). Sex percentage for each stage of forest expansion (b). All sampled juniper trees were considered (n = 816). The significant differences among stages of forest expansion (a) and between sexes at each stage of forest expansion (b) are denoted by small letters 1 3 Annals of Forest Science (2021) 78: 86 Page 7 of 19 86 Fig. 3 Post-hoc differences among stages of forest expansion (a) and interval of the LS mean. EXP, expanding front; TRAN, transition sites (b) derived from the wood density LMM. Black circles indicate zone; LON, long-existing forest; MAR, Maranchón; RIB, Ribarre- the least square (LS) mean. Error bars indicate the 95% confidence donda; HUE, Huertahernando 3.2 Climate‑growth sensitivity and annual growth patterns We found a higher percentage of trees showing a positive response in tree ring growth in years with high Febru- ary–April temperatures (~ 41% of trees) and August SPEI-3 (~ 52% of trees; Fig. 4). Growth in trees increased with warmer temperatures, under low drought stress, and with greater QD and cambial age during approximately the first 20 years of life but showed a slightly negative trend during the final 15 years of the study period. Growth decreased with tree density and the number of stems per tree. We found greater growth in Spanish junipers in the expanding front than in the transition zone and the long- existing forests, particularly after the first 20 years of life of the trees. In the expanding front, Spanish junipers also showed higher growth under low drought stress conditions compared to the trees growing in the other two stages. We found a minimal difference in climate-growth sensitivity between sexes. Males showed higher growth sensitivity to early spring temperatures (February–April) than females; i.e., males grew more during years with warmer early spring temperatures. Females showed higher growth sensitivity to summer drought; that is, females grew more during years with less drought stress. Sex-based differences were almost indiscernible when we plotted the model predictions (Table 3, Fig. 5). Climate variables, QD, and the interaction between cambial age and stages of forest expansion were the best explanatory variables for the growth LMM (Table 3, Fig. 5a, d). Fig. 4 Percentage of trees showing positive (POS, red bars) or nega- tive (NEG, blue bars) responses in growth to monthly and seasonal temperature and SPEI-3 variables. The positive or negative responses correspond to the positively or negatively significant slopes (p < 0.05) 4 Discussion obtained for each individual tree in linear regression models between detrended growth and climate variables In this study, we investigated how radial growth and wood density were influenced by sex, the stage of forest expan- in forest structure found among stages—i.e., tree density, sion, and variability in forest structure. We also explored age, and tree diameter—had a much greater effect on growth sex-based and stage-based differences in climate-growth and wood density of junipers than sex. Specifically, trees sensitivity. The stage of forest expansion and the variability 1 3 86 Page 8 of 19 Annals of Forest Science (2021) 78: 86 Table 3 Growth LMM Predictors Estimates SE p coefficients. Marginal and conditional r are also given. (Intercept) − 0.341 0.118 0.004 The level stage: expanding is QD [1st degree] 1.205 0.147 < 0.001 included in the intercept. SE QD [2nd degree] 0.475 0.222 0.033 standard error. σ is the residual variance. τ Tree and τ Plot Number of stems [1st degree] − 0.584 0.11 < 0.001 00 00 are the between-group variance Number of stems [2nd degree] − 0.058 0.196 0.768 for the random effects tree and Tree density − 0.113 0.066 0.112 plot, respectively Cambial age [1st degree] 1.879 0.103 < 0.001 Cambial age [2nd degree] 0.417 0.087 < 0.001 FEB-APR T 0.072 0.005 < 0.001 AUG SPEI-3 0.108 0.006 < 0.001 Sex [male] 0.006 0.037 0.867 Gradient [transition] 0.133 0.133 0.338 Gradient [long] 0.39 0.177 0.048 FEB-APR T × sex [male] 0.013 0.006 0.027 AUG SPEI-3 × sex [male] − 0.011 0.005 0.047 AUG SPEI-3 × gradient [transition] − 0.014 0.006 0.026 AUG SPEI-3 × gradient [long] − 0.034 0.006 < 0.001 Cambial age [1st degree] × gradient [transition] − 0.515 0.138 < 0.001 Cambial age [2nd degree] × gradient [transition] − 0.396 0.108 < 0.001 Cambial age [1st degree] × gradient [long] − 1.363 0.143 < 0.001 Cambial age [2nd degree] × gradient [long] − 0.472 0.102 < 0.001 Random effects   σ 0.25   τ 0.15 00 Tree   τ 0.27 00 Plot 2 2   Marginal r /conditional r 0.22/0.43 in the expanding fronts had higher growth rates and lower environments (Ortiz et al. 2002; Barrett et al. 2010) and wood density than junipers in the transition zone and long- have been associated with the greater cost of reproduction existing forests. The positive response in growth found in the for females trees (Vasiliauskas and Aarssen 1992; Montes- expanding fronts increased more during summers with low inos et al. 2012). Compensatory mechanisms for the higher drought stress than in the other two stages. reproductive cost in females such as greater photosynthetic capacity or water-use efficiency (WUE; Dawson and Bliss 4.1 Sex‑based differences in expanding juniper 1989; Olano et al. 2015; Rozas et al. 2009) have been identi- woodlands fied as the most likely explanation for this paradox (Tozawa et al. 2009). Nevertheless, no evidence of sex-related differ - It is important to know whether sex effects on growth and ences in WUE was found in a previous study conducted at density exist in Spanish juniper stands to improve our our study sites (Acuña-Míguez et al. 2020). understanding of compensatory mechanisms (or trade-offs) Females outperformed males in height, QD, number of between growth and reproduction. A higher proportion of stems, crown diameter, and age, irrespectively of the stage male trees were found in all three stages of forest expan- and site considered. Females invested more resources in veg- sion and at all three sites, suggesting that Spanish juniper etative height growth than males (Gauquelin et al. 2002) stands in central Spain are male-biased. Similar results have as their larger canopies, consisting of a greater number of been found in other studies conducted in juniper stands in stems, allow them to bear a large number of cones that may central Spain (Gimeno et al. 2012c), Morocco, and the Pyr- enhance their reproductive capacity. Our results showed enees (Gauquelin et al. 2002). In young populations, these minimal sex-based differences in wood density and growth. results can be explained simply because males begin to Specifically, we found that males may grow slightly more flower earlier than females and consequently can be identi- than females during years with warmer early spring tem- fied more readily (Gauquelin et al. 2002). However, male- peratures, whereas females may grow more than males biased sex ratios are also commonly found in more stressful during years with low summer drought stress. However, 1 3 Annals of Forest Science (2021) 78: 86 Page 9 of 19 86 Fig. 5 Growth LMM predic- tions as a function of QD (a), tree density (b), and the number of stems per tree (c). Growth LMM predictions for the three stages of forest expansion dur- ing the first 35 years of life of the trees (cambial age) (d) and for August SPEI-3 (f). Growth LMM predictions for female and male individuals in terms of February–April temperatures (e) and August SPEI-3 (g). Indi- vidual growth values are shown with circles. Note that the lower values of SPEI correspond to greater drought stress when plotting the predictions of our models, sex-based dif- and cambial age was not selected in the most parsimoni- ferences were almost indiscernible (Fig. 5e, g). Hence, we ous growth model, suggesting that there are no significant suggest taking into account tree size or other relative growth differences in annual growth rates before or after the trees response variables in forest structure in sex dimorphism become reproductive. During early stages, negligible differ - models for a better understanding of the impact of sex on ences in reproductive resource investment is evident for the the performance of dioecious species (Obeso 2002). lack of sex-based differences that we found (McKown et al. Previous studies reporting sex-based differences in Span- 2017). The Spanish junipers in our study system, although ish juniper radial growth and climate-growth sensitivity have reproductive, are relatively young. Thus, we cannot rule reached divergent conclusions regarding which sex performs out the possibility that sex-related differences in secondary best in each of these traits. For instance, some authors report growth may be modified during ontogeny due to physiologi- greater growth in males than in females (Gauquelin et al. cal adjustments (Rozas et al. 2009). 2002; Montesinos et al. 2006), whereas other studies found greater growth and higher summer precipitation-sensitivity 4.2 Main drivers of growth and wood density in females (Rozas et al. 2009). Sex-based differences with in expanding juniper woodlands regard to climate may be site-dependent (Olano et al. 2015; DeSoto et al. 2016), with females growing more than males Less trait divergence has been reported in dioecious species under less restrictive environmental conditions, or age- adapting to new environmental conditions (Arbuthnott et al. dependent (Rozas et al. 2009) as young females are more 2014), which is consistent with the expansion that the Span- sensitive to drought conditions. Given the great reproduc- ish juniper woodlands are currently undergoing in central tive effort of Spanish juniper females (Montesinos et al. Spain. Rather than sex, the main drivers of radial growth 2012), we hypothesized that sexual dimorphism in second- and wood density were the specific stage of forest expan - ary growth in Spanish junipers should occur from early sion (and the variables of forest structure that characterize stages onwards. By contrast, the interaction between sex each stage) and climate (in the case of growth). We found 1 3 86 Page 10 of 19 Annals of Forest Science (2021) 78: 86 greater growth rates at the expanding front, followed by the greater WUE than in long-existing forests owing to changes transition zone and the long-existing forest (differences that in certain functional attributes (e.g., higher leaf mass per became evident at approximately 10 years of age, Fig. 5d). area) and greater nitrogen availability stemming from the Growth decreased with tree age, higher tree density (Rozas former agricultural use (Guerrieri et al. 2021). Indeed, in et al. 2009; Gimeno et al. 2012c), and the number of stems our study area, Acuña-Míguez et al. (2020) found increased per tree. The Spanish juniper is a multi-stemmed species, WUE in the expanding fronts that was mainly related to their and we show here that annual growth increased in trees with lower vegetation cover and younger age. As such, our results fewer stems. In unmanaged stands, long-existing forests agree with the general pattern that trees growing under low tend to have more stems because, given the lack of logging competitive stress have higher growth rates, greater WUE, or browsing by livestock, older trees have more stems. A and a better response in growth to high water availability reduction in growth in multi-stemmed individuals has been (e.g., Linares et al. 2009; Sánchez-Salguero et al. 2015). observed in other resprouting species and is attributed to Rozas et al. (2009) showed that growth-climate sensitivity the preferential investment of resources in height growth in Spanish junipers is higher in earlier life stages. Here, we owing to the competition for light among stems (see Espelta accounted for a possible age effect when comparing sexes et al. 2003). Favorable climatic conditions, i.e., warm spring (females are significantly older than males) and among temperatures and wet summers, enhance annual growth in stages of forest expansion (the expanding fronts and transi- Spanish junipers (Rozas et al. 2009; Gimeno et al. 2012a). tion zone were significantly younger than the long-existing Relatively high temperatures at the beginning of the grow- forests) by restricting our analyses to the first 35 years of ing season stimulate earlier cambium reactivation and so cambial age. Thus, we can rule out any possibility that sex- enhance growth (Begum et al., 2008). Despite the excellent and stage-based differences in the growth response to cli - adaptation in Mediterranean tree species to drought, con- mate variability are caused by an age effect. straints in water availability during the growing season can Tree and cambial age also affect wood density (Frances- result in markedly less growth that lasts for several years chini et al. 2013). Specifically, our results showed a decrease in (e.g., Anderegg et al. 2015; Gazol et al. 2018). Therefore, mean wood density with tree age. Tree ring density chronolo- growth in Spanish junipers could be severely reduced by the gies show that larger rings are associated with a decrease in increase in the frequency and severity of drought episodes mean wood density (Lundgren 2004). As such, years with high projected for southwest Europe, particularly in the highly water availability increased tree ring growth and decreased sensitive expanding fronts. mean ring density (Franceschini et al. 2013), particularly in As hypothesized, the expanding fronts had greater posi- younger trees. The expanding fronts had less wood density tive responses in growth during years with greater water than the transition zone and the long-existing forests, even after availability than the other stages of forest expansion accounting for age effects. Greater densities are thought to be (Gimeno et al. 2012c). Similar results have been found for less vulnerable to cavitation (Hacke et al. 2001). However, the other tree species developing on former agricultural land rapid growth of expanding forests enables the development of (Alfaro-Sánchez et  al. 2019). We suggest that growth is large vessels that could increase the risk of cavitation during mediated by nutrient limitations (Forrester 2015) deriving periods of continuous drought (Lambers et al. 2008). from the tree density and land-use legacies found at each of the stages of forest expansion. Indeed, the expanding fronts 4.3 Land‑use legacies had lower tree densities, followed by the transition zone and the long-existing forests. Furthermore, Gimeno et al. Expanding juniper woodlands are mainly colonizing adjacent (2012c) found no clear nucleation in some of the expand- former agricultural land that was abandoned in recent ing fronts studied here but did detect clumped patterns in decades as a result of the rural exodus towards urban nuclei. the long-existing forests. The clumping patterns, mostly Spontaneous tree establishment in abandoned fields in the attributable to perching and nursing effects in adult trees, Mediterranean region has proved to benefit from increased could have increased competition for resources in junipers soil nitrogen levels deriving from their previous agricultural growing in these relatively dense long-existing patches. By use (Nadal-Romero et  al. 2018; Guerrieri et  al. 2021). contrast, the lower tree density in the expanding front (~ 28 Recently expanding forests have been reported to enhance −1 trees ha ) has the effect of reducing intraspecific compe- growth rates but lower wood density in comparison with long- tition and facilitating greater exposure to light, which is existing forests, even after accounting for age and competition reflected in higher growth rates. Increased or stable growth effects (Alfaro-Sánchez et al. 2019). Similarly, the expanding at increased intrinsic water use efficiency (WUE) has been fronts of Spanish juniper are probably benefiting from land- reported in recent decades in juniper woodlands elsewhere use legacies inherited from the former croplands (De Schrijver (Granda et al. 2014). Previous studies analyzing the spon- et al. 2011; Vilà-Cabrera et al. 2017; Alfaro-Sánchez et al. taneous establishment of secondary forests have reported 2019), as is confirmed by the greater survival rate of saplings 1 3 Annals of Forest Science (2021) 78: 86 Page 11 of 19 86 (Gimeno et al. 2012b), higher WUE (Granda et al. 2014; of cambial age, we ruled out the possibility that stage-based Acuña-Míguez et al. 2020), and the greater growth rates and differences in growth and climatic sensitivity are caused by lower wood density values reported at the expanding fronts in an age effect. Instead, we suggest that expanding fronts of this study. Greater vigor and survival rates in the expanding Spanish juniper are benefitting from a combination of lower fronts of Spanish junipers have been related to environmental intraspecific competition found in the expanding patches and differences between expanding and long-existing forests land-use legacies stemming from the abandoned e fi lds they (Gimeno et  al. 2012b), i.e., better soil water-retention colonize, as is shown by higher growth rates and lower tree capacity in former agricultural fields due to plowing (Flinn densities. Our results also reveal that the different intraspecific and Marks 2007). However, our study lacked soil nutrient competition found in the stages of forest expansion mediate content information for each of the stages of forest expansion. tree growth response to climate variability under adverse Further studies should assess differences in nitrogen content weather conditions. As such, the greater intraspecific com- among stages of forest expansion in Spanish junipers to help petition in long-existing forests (which, as reported in other understand the specific drivers of the reported stage-based studies, show clumping patterns and lower WUE) is causing differences in growth, wood density, and climate sensitivity. a lower positive response in growth during years with low drought stress compared to the other two stages. Assessing how growth, wood density, and climate-growth sensitivity 5 Conclusions varied among three stages of forest expansion—i.e., the long- existing forests, expanding fronts, and transitions zones— In this study, we assessed the simultaneous effect of sex, for - will improve our understanding of the dynamics of Spanish est expansion stage, and forest structural characteristics on junipers undergoing a natural process of forest expansion. It radial growth and wood density and explored sex- and stage- will also ultimately help in their future preservation and, in based differences in climate-growth sensitivity. We found that particular, prevent risks associated with any increase in the sex had only a minimal effect on the first stages of growth severity and number of drought episodes in the Mediterranean in Spanish junipers (the first 35 years of cambial age) and region. showed that radial growth, wood density, and climate sensitiv- ity varied among stages of forest expansion, mainly due to the different structural characteristics of the forests found in each of the stages. By restricting our analyses to the first 35 years  Appendix Fig. 6 Detrending growth at individual level using cubic smoothing splines of 10  years for the three studied sites and a negative exponential method 1 3 86 Page 12 of 19 Annals of Forest Science (2021) 78: 86 Fig. 7 Detrending growth at individual level using the negative exponential method for the three studied sites 1 3 Annals of Forest Science (2021) 78: 86 Page 13 of 19 86 Fig. 8 Sample depth for the three stages of forest expansion and site. The dashed line indicates the 35 years of cambial age. The gray area indicates a sample depth of 5 trees 1 3 86 Page 14 of 19 Annals of Forest Science (2021) 78: 86 Fig. 9 Tree density variations (violin and boxplots) for each site (a). Percentage of trees per sex for each site (b); all sampled juniper trees were considered (n = 816). Significant differ ences between sexes at each site (b) are denoted by small letters 1 3 Annals of Forest Science (2021) 78: 86 Page 15 of 19 86 Fig. 10 Violins and boxplots displaying differences in repro- ductive trees (n = 451) among stages of forest expansion and at sites for the following vari- ables: tree age (a, f), number of stems per tree (b, g), quadratic diameter (QD, c, h), height (d, i), and crown diameter (e, j). The capital letters indicate significant differences among stages of forest expansion or at sites, while the lowercase letters indicate significant differences between sexes obtained in post- hoc tests from LMMs (Table 4) 1 3 86 Page 16 of 19 Annals of Forest Science (2021) 78: 86 Acknowledgements We are especially grateful for the help, advice, and support provided by David López-Quiroga, José Miguel Olano, Adrián Escudero, Pablo Álvarez, Esteban Manrique, Eduardo Serna and Miguel Díaz. We are also thankful to José Antonio Lozano, direc- tor of the Alto Tajo Natural Park, for facilitating the research at the park. Funding This study was funded by the grants SPONFOREST (Biodi- vERsA3-2015–58), PCIN-2016–055 (financed by the Spanish Research Agency (AEI) and the Spanish Ministry of Economy, Industry and Competitiveness (MINECO)), COMEDIAS (MINECO, CGL2017- 83170-R), and REMEDINAL TE (Ref. TE-CM. S2018/EMT-4338, 2019–2023-Comunidad de Madrid). Data availability The dataset used in the current study will be available on Figshare after a 1-month embargo period in this link: https:// doi. org/ 10. 6084/ m9. figsh are. 14806 683. v1. The provisional link to download the data during the embargo period is https:// figsh are. com/s/ d42a2 9f95d a711d 586a5. Declarations Consent for publication All authors gave their informed consent to this publication and its content. Conflict of interest The authors declare no competing interests. References Acuña-Míguez B, Valladares F, Martín-Forés I (2020) Both mature patches and expanding areas of Juniperus thurifera forests are vul- nerable to climate change but for different reasons. Forests 11:960. https:// doi. org/ 10. 3390/ f1109 0960 Alfaro-Sánchez R, Jump AS, Pino J et al (2019) Land use legacies drive higher growth, lower wood density and enhanced climatic sensitivity in recently established forests. Agric for Meteorol 276– 277:107630. https:// doi. org/ 10. 1016/j. agrfo rmet. 2019. 107630 Alfaro-Sánchez R, Valdés-Correcher E, Espelta JM, et al (2020) How do social status and tree architecture influence radial growth, wood density and drought response in spontaneously established oak forests? Annals of Forest Science 77. https://doi. or g/10. 1007/ s13595- 020- 00949-x Alfaro Sánchez R, Espelta, JM, Valladares F, Acuña-Míguez B, Martín- Forés I (2021) Disentangling the role of sex dimorphism and for- est structure as drivers of growth and wood density in expanding Juniperus thurifera L. woodlands. figshare. Dataset. 10.6084/ m9.figshare.14806683.v1 Anderegg WRL, Schwalm C, Biondi F et al (2015) Pervasive drought legacies in forest ecosystems and their implications for carbon cycle models. Science 349:528. https://do i. org/10 . 1126/ science . aab18 33 Anderson JT, Wagner MR, Rushworth CA et al (2014) The evolution of quantitative traits in complex environments. Heredity (edinb) 112:4–12. https:// doi. org/ 10. 1038/ hdy. 2013. 33 Arbuthnott D, Dutton EM, Agrawal AF, Rundle HD (2014) The ecol- ogy of sexual conflict: ecologically dependent parallel evolution of male harm and female resistance in Drosophila melanogaster. Ecol Lett 17:221–228. https:// doi. org/ 10. 1111/ ele. 12222 Barrett SCH, Yakimowski SB, Field DL, Pickup M (2010) Ecological genetics of sex ratios in plant populations. Philos Trans R Soc Lond B Biol Sci 365:2549–2557. https:// doi. org/ 10. 1098/ rs tb. 2010. 0002 1 3 Table 4 LMM coefficients f or the following dependent variables: tree age, number of stems per tree, quadratic mean diameter (QD), height, and crown diameter using only reproductive trees (n = 451). The variable QD is log-transformed. The marginal and conditional r is also provided (delta for number of stems). The levels sex: female, site: Maranchón, and gradient: long-existing are included in the intercept QD [log] Number of stems Crown diameter Tree age Height Predictors Estimates SE p Estimates SE p Estimates SE p Estimates SE p Estimates SE p (Intercept) 3.08 0.10 < 0.001 2.90 0.27 < 0.001 5.54 0.23 < 0.001 45.64 2.90 < 0.001 5.99 0.24 < 0.001 Gradient [transition] − 0.39 0.11 < 0.001 0.88 0.09 0.179 − 0.99 0.24 < 0.001 − 13.15 3.81 0.001 − 1.29 0.30 < 0.001 Gradient [expanding] − 0.53 0.11 < 0.001 0.71 0.08 0.002 − 1.38 0.26 < 0.001 − 18.34 3.99 < 0.001 − 1.50 0.32 < 0.001 Sex [male] − 0.22 0.05 < 0.001 0.85 0.06 0.012 − 0.51 0.15 0.001 − 2.83 1.17 0.015 − 0.42 0.13 0.001 Site [Ribarredonda] − 0.06 0.11 0.564 0.81 0.09 0.050 − 0.27 0.25 0.277 Site [Huertahernando] 0.18 0.11 0.101 1.04 0.10 0.696 0.84 0.24 < 0.001 Random effects σ 0.28 0.38 2.33 131.82 1.56 τ 0.02 0.01 0.07 34.92 0.20 00 Plot 2 2 Marginal r /conditional r 0.18/0.24 0.09/0.11 0.18/0.21 0.26/0.42 0.22/0.30 Annals of Forest Science (2021) 78: 86 Page 17 of 19 86 Barton K (2019) MuMIn: Multi-Model Inference Gimeno TE, Escudero A, Delgado A, Valladares F (2012b) Previous land use alters the effect of climate change and facilitation on Başnou C, Álvarez E, Bagaria G et al (2013) Spatial patterns of land expanding woodlands of Spanish Juniper. Ecosystems 15:564– use changes across a Mediterranean metropolitan landscape: 579. https:// doi. org/ 10. 1007/ s10021- 012- 9529-z implications for biodiversity management. Environ Manage Gimeno TE, Pías B, Martínez-Fernández J et al (2012c) The decreased 52:971–980. https:// doi. org/ 10. 1007/ s00267- 013- 0150-5 competition in expanding versus mature juniper woodlands is Blanco BC, Castro EB (1997) Los bosques ibéricos: una interpretación counteracted by adverse climatic effects on growth. Eur J Forest geobotánica. Planeta Res 131:977–987. https:// doi. org/ 10. 1007/ s10342- 011- 0569-2 Cook ER, Kairiukstis LA (eds) (1990) Methods of dendrochronology. Granda E, Rossatto DR, Camarero JJ et al (2014) Growth and carbon Springer Netherlands, Dordrecht isotopes of Mediterranean trees reveal contrasting responses to Cybis Elektronik (2018) CDendro and CooRecorder increased carbon dioxide and drought. Oecologia 174:307–317. Dawson TE, Bliss LC (1989) Patterns of water use and the tissue https:// doi. org/ 10. 1007/ s00442- 013- 2742-4 water relations in the dioecious shrub, Salix arctica: the physi- Greenwood S, Ruiz-Benito P, Martínez-Vilalta J et  al (2017) Tree ological basis for habitat partitioning between the sexes. Oecolo- mortality across biomes is promoted by drought intensity, lower gia 79:332–343. https:// doi. org/ 10. 1007/ BF003 84312 wood density and higher specific leaf area. Ecol Lett 20:539–553. De Schrijver A, De Frenne P, Ampoorter E et al (2011) Cumulative https:// doi. org/ 10. 1111/ ele. 12748 nitrogen input drives species loss in terrestrial ecosystems. Glob Guerrieri MR, Correia M, Martin-Forés I, et al (2021) Land-use lega- Ecol Biogeogr 20:803–816. https://doi. or g/10. 1111/j. 1466- 8238. cies influence tree water-use efficiency and nitrogen availability 2011. 00652.x in recently established European forests. Functional Ecology. In DeSoto L, Olano JM, Rozas V (2016) Secondary growth and carbo- press hydrate storage patterns differ between sexes in Juniperus thu- Hacke UG, Sperry JS, Pockman WT et al (2001) Trends in wood den- rifera. Frontiers in Plant Science 7. https:// doi. org/ 10. 3389/ fpls. sity and structure are linked to prevention of xylem implosion 2016. 00723 by negative pressure. Oecologia 126:457–461. https://doi. or g/10. Escribano-Avila G, Sanz-Pérez V, Pías B et al (2012) Colonization of 1007/ s0044 20100 628 abandoned land by Juniperus thurifera is mediated by the inter- Hampe A, Alfaro-Sánchez R, Martín-Forés I (2020) Establishment of action of a diverse dispersal assemblage and environmental het- second-growth forests in human landscapes: ecological mecha- erogeneity. PLoS ONE 7:e46993. https:// doi. org/ 10. 1371/ journ nisms and genetic consequences. Ann For Sci 77:87, s13595–020– al. pone. 00469 93 00993–7. https:// doi. org/ 10. 1007/ s13595- 020- 00993-7 Espelta JM, Retana J, Habrouk A (2003) Resprouting patterns after fire Haylock MR, Hofstra N, Klein Tank AMG, et al (2008) A European and response to stool cleaning of two coexisting Mediterranean daily high-resolution gridded data set of surface temperature and oaks with contrasting leaf habits on two different sites. For Ecol precipitation for 1950–2006. Journal of Geophysical Research Manage 179:401–414. https:// doi. org/ 10. 1016/ S0378- 1127(02) 113. https:// doi. org/ 10. 1029/ 2008J D0102 01 00541-8 Holmes RL (1983) Computer-assisted quality control in tree-ring dat- Flinn KM, Marks PL (2007) Agricultural legacies in forest environ- ing and measurement. Tree-Ring Bulletin 43:69–78 ments: tree communities, soil properties, and light availability. Juvany M, Munné-Bosch S (2015) Sex-related differences in stress Ecol Appl 17:452–463. https:// doi. org/ 10. 1890/ 05- 1963 tolerance in dioecious plants: a critical appraisal in a physiologi- Forrester DI (2015) Transpiration and water-use efficiency in mixed- cal context. J Exp Bot 66:6083–6092. https://doi. or g/10. 1093/ jxb/ species forests versus monocultures: effects of tree size, stand erv343 density and season. Tree Physiol 35:289–304. https:// doi. org/ 10. Keenan RJ (2015) Climate change impacts and adaptation in forest 1093/ treep hys/ tpv011 management: a review. Ann for Sci 72:145–167. https:// doi. org/ Franceschini T, Longuetaud F, Bontemps J-D et al (2013) Effect of 10. 1007/ s13595- 014- 0446-5 ring width, cambial age, and climatic variables on the within-ring Kock N, Lynn G (2012) Lateral collinearity and misleading results in wood density profile of Norway spruce Picea abies (L.) Karst. variance-based SEM: an illustration and recommendations. JAIS Trees 27:913–925. https:// doi. org/ 10. 1007/ s00468- 013- 0844-6 13:546–580. https:// doi. org/ 10. 17705/ 1jais. 00302 Freschet GT, Östlund L, Kichenin E, Wardle DA (2014) Aboveground Kuznetsova A, Brockhoff PB, Christensen RHB (2017) lmerTest pack - and belowground legacies of native Sami land use on boreal for- age: tests in linear mixed effects models. Journal of Statistical est in northern Sweden 100 years after abandonment. Ecology Software 82:1–26. https:// doi. org/ 10. 18637/ jss. v082. i13 95:963–977. https:// doi. org/ 10. 1890/ 13- 0824.1 Lambers H, Chapin FS, Pons TL (2008) Plant physiological ecology, Gauquelin T, Bertaudiere V, Montes N et al (1999) Endangered stands 2nd ed. Springer, New York of thuriferous juniper in the western Mediterranean basin: eco- Lambin EF, Meyfroidt P (2011) Global land use change, economic logical status, conservation and management. Biodivers Conserv globalization, and the looming land scarcity. Proc Natl Acad Sci 8:1479–1498 108:3465–3472. https:// doi. org/ 10. 1073/ pnas. 11004 80108 Gauquelin T, Bertaudière-Montès A, Badri W, Montès N (2002) Sex Lenth RV (2016) Least-Squares Means: The R Package lsmeans. J Stat ratio and sexual dimorphism in mountain dioecious thuriferous Softw 69:1–33. https:// doi. org/ 10. 18637/ jss. v069. i01 juniper (Juniperus thurifera L., Cupressaceae). Bot J Linn Soc Linares J-C, Delgado-Huertas A, Julio Camarero J et al (2009) Com- 138:237–244. https:// doi. org/ 10. 1046/j. 1095- 8339. 2002. 13800 petition and drought limit the response of water-use efficiency 2237.x to rising atmospheric carbon dioxide in the Mediterranean fir Gazol A, Camarero JJ, Vicente-Serrano SM et  al (2018) Forest Abies pinsapo. Oecologia 161:611–624. https:// doi. org/ 10. 1007/ resilience to drought varies across biomes. Glob Change Biol s00442- 009- 1409-7 24:2143–2158. https:// doi. org/ 10. 1111/ gcb. 14082 Lüdecke D (2021) sjPlot: data visualization for statistics in social Gerstner K, Dormann CF, Stein A et al (2014) Effects of land use on science plant diversity - a global meta-analysis. J Appl Ecol 51:1690– Lüdecke D, Makowski D, Waggoner P, Patil I (2020) performance: 1700. https:// doi. org/ 10. 1111/ 1365- 2664. 12329 assessment of regression models performance Gimeno TE, Camarero JJ, Granda E et al (2012a) Enhanced growth Lundgren C (2004) Microfibril angle and density patterns of fertilized of Juniperus thurifera under a warmer climate is explained by a and irrigated Norway spruce. Silva Fenn 38. https:// doi. org/ 10. positive carbon gain under cold and drought. Tree Physiol 32:326– 14214/ sf. 439 336. https:// doi. org/ 10. 1093/ treep hys/ tps011 1 3 86 Page 18 of 19 Annals of Forest Science (2021) 78: 86 Martín-Forés I, Magro S, Bravo-Oviedo A, et al (2020) Spontaneous Rozas V, DeSoto L, Olano JM (2009) Sex-specific, age-dependent forest regrowth in South-West Europe: consequences for nature’s sensitivity of tree-ring growth to climate in the dioecious tree contributions to people. People and Nature. In press Juniperus thurifera. New Phytol 182:687–697. https://doi. or g/10. McKown AD, Klápště J, Guy RD, et al (2017) Sexual homomorphism 1111/j. 1469- 8137. 2009. 02770.x in dioecious trees: extensive tests fail to detect sexual dimor- Sánchez-Salguero R, Linares JC, Camarero JJ et al (2015) Disentan- phism in Populus. Scientific Reports 7. https:// doi. org/ 10. 1038/ gling the effects of competition and climate on individual tree s41598- 017- 01893-z growth: a retrospective and dynamic approach in Scots pine. For Montesinos D, De Luís M, Verdú M et al (2006) When, how and how Ecol Manage 358:12–25. https:// doi. org/ 10. 1016/j. foreco. 2015. much: gender-specific resource-use strategies in the dioecious tree 08. 034 Juniperus thurifera. Ann Bot 98:885–889. https://doi. or g/10. 1093/ Stewart JL, Salazar R (1992) A review of measurement options for aob/ mcl172 multipurpose trees. Agrofor Syst 19:173–183. https:// doi. org/ 10. Montesinos D, García-Fayos P, Verdú M (2012) Masting uncoupling: 1007/ BF001 38507 mast seeding does not follow all mast flowering episodes in a Tasser E, Walde J, Tappeiner U et al (2007) Land-use changes and dioecious juniper tree. Oikos 121:1725–1736. https:// doi. org/ 10. natural reforestation in the Eastern Central Alps. Agr Ecosyst 1111/j. 1600- 0706. 2011. 20399.x Environ 118:115–129. https://doi. or g/10. 1016/j. ag ee.2006. 05. 004 Nadal-Romero E, Otal-Laín I, Lasanta T et al (2018) Woody encroach- Thompson JD (2005) Plant Evolution in the Mediterranean. Oxford ment and soil carbon stocks in subalpine areas in the Central University Press, Oxford Spanish Pyrenees. Sci Total Environ 636:727–736. https:// doi. Tozawa M, Ueno N, Seiwa K (2009) Compensatory mechanisms for org/ 10. 1016/j. scito tenv. 2018. 04. 324 reproductive costs in the dioecious tree Salix integra. Botany Obeso JR (2002) The costs of reproduction in plants. New Phytol 87:315–323. https:// doi. org/ 10. 1139/ B08- 125 155:321–348. https://doi. or g/10. 1046/j. 1469- 8137. 2002. 00477.x Valdés-Correcher E, van Halder I, Barbaro L et al (2019) Insect her- Olano JM, García-Cervigón AI, Arzac A, Rozas V (2015) Intra-annual bivory and avian insectivory in novel native oak forests: divergent effects of stand size and connectivity. For Ecol Manage 445:146– wood density fluctuations and tree-ring width patterns are sex- 153. https:// doi. org/ 10. 1016/j. foreco. 2019. 05. 018 and site-dependent in the dioecious conifer Juniperus thurifera L. Vasiliauskas SA, Aarssen LW (1992) Sex ratio and neighbor effects in Trees 29:1341–1353. https://doi. or g/10. 1007/ s00468- 015- 1212-5 monospecific stands of Juniperus Virginiana. Ecology 73:622– Olano JM, Rozas V, Bartolomé D, Sanz D (2008) Effects of changes in 632. https:// doi. org/ 10. 2307/ 19407 68 traditional management on height and radial growth patterns in a Vicente-Serrano SM, Beguería S, López-Moreno JI (2010) A multi- Juniperus thurifera L. woodland. For Ecol Manage 255:506–512. scalar drought index sensitive to global warming: the standard- https:// doi. org/ 10. 1016/j. foreco. 2007. 09. 015 ized precipitation evapotranspiration index. J Clim 23:1696–1718. Olano JM, Zavala MA, Rozas V (2012) Disruption of Juniperus thu- https:// doi. org/ 10. 1175/ 2009J CLI29 09.1 rifera woodland structure in its northwestern geographical range: Vilà-Cabrera A, Espelta JM, Vayreda J, Pino J (2017) “New Forests” potential drivers and limiting factors. Eur J Forest Res 131:563– from the twentieth century are a relevant contribution for c storage 570. https:// doi. org/ 10. 1007/ s10342- 011- 0531-3 in the Iberian peninsula. Ecosystems 20:130–143. https://d oi.o rg/ Ortiz PL, Arista M, Talavera S (2002) Sex ratio and reproductive effort 10. 1007/ s10021- 016- 0019-6 in the dioecious Juniperus communis subsp. alpina (Suter) Celak. Villellas J, Martín-Forés I, Mariette S, et al (2020) Functional distance (Cupressaceae) along an altitudinal gradient. Ann Bot 89:205– is driven more strongly by environmental factors than by genetic 211. https:// doi. org/ 10. 1093/ aob/ mcf028 relatedness in Juniperus thurifera L. expanding forest stands. Ann Palmero-Iniesta M, Espelta JM, Gordillo J, Pino J (2020) Changes For Sci 77:66. https:// doi. org/ 10. 1007/ s13595- 020- 00973-x in forest landscape patterns resulting from recent afforestation Williamson GB, Wiemann MC (2010) Measuring wood specific grav - in Europe (1990–2012): defragmentation of pre-existing forest ity…Correctly. Am J Bot 97:519–524. https:// doi. org/ 10. 3732/ versus new patch proliferation. Ann For Sci ajb. 09002 43 Pretzsch H, Biber P, Schütze G et al (2018) Wood density reduced while wood volume growth accelerated in Central European for- ests since 1870. For Ecol Manage 429:589–616. https:// doi. org/ Publisher's note Springer Nature remains neutral with regard to 10. 1016/j. foreco. 2018. 07. 045 jurisdictional claims in published maps and institutional affiliations. 1 3 Annals of Forest Science (2021) 78: 86 Page 19 of 19 86 Authors and Affiliations 1 2 3 3 4 Raquel Alfaro‑Sánchez  · Josep Maria Espelta  · Fernando Valladares  · Belén Acuña‑Míguez  · Irene Martín‑Forés * Raquel Alfaro-Sánchez Department of Biology, Wilfrid Laurier University, 75 r.alfarosanchez@gmail.com University Avenue W, Waterloo, ON N2L 3C5, Canada Josep Maria Espelta Centre de Recerca Ecològica I Aplicacions Forestals, josep.espelta@uab.cat CREAF, Bellaterra (Cerdanyola de Vallès), 08193 Catalonia, Spain Fernando Valladares valladares@ccma.csic.es Departament of Biogeography and Global Change, National Museum of Natural Sciences, Spanish Council for Scientific Belén Acuña-Míguez Research, CSIC, C/Serrano, 115dpdo, 28006 Madrid, Spain belacumig@gmail.com School of Biological Sciences, The University of Adelaide, Irene Martín-Forés Adelaide, South Australia 5005, Australia imfores@pdi.ucm.es 1 3

Journal

Annals of Forest Science – Springer Journals

Published: Dec 1, 2021

Keywords: Dioecy; Juniperus thurifera L.; Tree rings; Forest expansion; Spontaneous tree establishment; Abandoned fields; Mediterranean region

There are no references for this article.