ORIGINAL RESEARCH published: 08 January 2019 doi: 10.3389/fpls.2018.01964 Early-Warning Signals of Individual Tree Mortality Based on Annual Radial Growth Maxime Cailleret 1,2* , Vasilis Dakos 3 , Steven Jansen 4 , Elisabeth M. R. Robert 5,6,7 , Tuomas Aakala 8 , Mariano M. Amoroso 9,10 , Joe A. Antos 11 , Christof Bigler 1 , Harald Bugmann 1 , Marco Caccianaga 12 , Jesus-Julio Camarero 13 , Paolo Cherubini 2 , Marie R. Coyea 14 , Katarina Čufar 15 , Adrian J. Das 16 , Hendrik Davi 17 , Guillermo Gea-Izquierdo 18 , Sten Gillner 19 , Laurel J. Haavik 20,21 , Henrik Hartmann 22 , Edited by: Ana-Maria Hereş 23,24 , Kevin R. Hultine 25 , Pavel Janda 26 , Jeffrey M. Kane 27 , Veronica De Micco, Viachelsav I. Kharuk 28,29 , Thomas Kitzberger 30,31 , Tamir Klein 32 , Tom Levanic 33 , University of Naples Federico II, Italy Juan-Carlos Linares 34 , Fabio Lombardi 35 , Harri Mäkinen 36 , Ilona Mészáros 37 , Reviewed by: Juha M. Metsaranta 38 , Walter Oberhuber 39 , Andreas Papadopoulos 40 , Louis S. Santiago, Any Mary Petritan 2,41 , Brigitte Rohner 2 , Gabriel Sangüesa-Barreda 42 , Jeremy M. Smith 43 , University of California, Riverside, Amanda B. Stan 44 , Dejan B. Stojanovic 45 , Maria-Laura Suarez 46 , Miroslav Svoboda 26 , United States Volodymyr Trotsiuk 2,26,47 , Ricardo Villalba 48 , Alana R. Westwood 49 , Peter H. Wyckoff 50 Minhui He, and Jordi Martínez-Vilalta 5,51 Northwest Institute of Eco-Environment and Resources 1 Department of Environmental Systems Science, Forest Ecology, Institute of Terrestrial Ecosystems, ETH Zürich, Zurich, (CAS), China Switzerland, 2 Swiss Federal Institute for Forest, Snow and Landscape Research – WSL, Birmensdorf, Switzerland, 3 CNRS, *Correspondence: IRD, EPHE, ISEM, Université de Montpellier, Montpellier, France, 4 Institute of Systematic Botany and Ecology, Ulm Maxime Cailleret University, Ulm, Germany, 5 CREAF, Cerdanyola del Vallès, Catalonia, Spain, 6 Ecology and Biodiversity, Vrije Universiteit maxime.cailleret@usys.ethz.ch Brussel, Brussels, Belgium, 7 Laboratory of Wood Biology and Xylarium, Royal Museum for Central Africa, Tervuren, Belgium, 8 Department of Forest Sciences, University of Helsinki, Helsinki, Finland, 9 Consejo Nacional de Investigaciones Científicas y Specialty section: Técnicas, CCT Patagonia Norte, Río Negro, Argentina, 10 Instituto de Investigaciones en Recursos Naturales, Agroecología y This article was submitted to Desarrollo Rural, Sede Andina, Universidad Nacional de Río Negro, Río Negro, Argentina, 11 Department of Biology, Functional Plant Ecology, University of Victoria, Victoria, BC, Canada, 12 Dipartimento di Bioscienze, Università degli Studi di Milano, Milan, Italy, 13 a section of the journal Instituto Pirenaico de Ecología (IPE-CSIC), Zaragoza, Spain, 14 Centre for Forest Research, Département des Sciences du Frontiers in Plant Science Bois et de la Forêt, Faculté de Foresterie, de Géographie et de Géomatique, Université Laval, Québec, QC, Canada, 15 Biotechnical Faculty, University of Ljubljana, Ljubljana, Slovenia, 16 United States Geological Survey, Western Ecological Received: 12 September 2018 Research Center, Sequoia and Kings Canyon Field Station, Three Rivers, CA, United States, 17 Ecologie des Forêts Accepted: 18 December 2018 Méditerranéennes (URFM), Institut National de la Recherche Agronomique, Avignon, France, 18 Centro de Investigación Published: 08 January 2019 Forestal (CIFOR), Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Madrid, Spain, 19 Institute of Forest Citation: Botany and Forest Zoology, TU Dresden, Dresden, Germany, 20 USDA Forest Service, Forest Health Protection, Saint Paul, Cailleret M, Dakos V, Jansen S, MN, United States, 21 Department of Entomology, University of Arkansas, Fayetteville, AR, United States, 22 Department Robert EMR, Aakala T, Amoroso MM, of Biogeochemical Processes, Max Planck Institute for Biogeochemistry, Jena, Germany, 23 Department of Forest Sciences, Antos JA, Bigler C, Bugmann H, Transilvania University of Brasov, Brasov, Romania, 24 BC3 – Basque Centre for Climate Change, Leioa, Spain, 25 Department " Caccianaga M, Camarero J-J, of Research, Conservation and Collections, Desert Botanical Garden, Phoenix, AZ, United States, 26 Faculty of Forestry Cherubini P, Coyea MR, Čufar K, and Wood Sciences, Czech University of Life Sciences, Prague, Czechia, 27 Department of Forestry and Wildland Resources, Das AJ, Davi H, Gea-Izquierdo G, Humboldt State University, Arcata, CA, United States, 28 Sukachev Institute of Forest, Siberian Division of the Russian Gillner S, Haavik LJ, Hartmann H, Academy of Sciences, Krasnoyarsk, Russia, 29 Siberian Federal University, Krasnoyarsk, Russia, 30 Department of Ecology, Hereş A-M, Hultine KR, Janda P, Universidad Nacional del Comahue, Río Negro, Argentina, 31 Instituto de Investigaciones en Biodiversidad y Medioambiente, Kane JM, Kharuk VI, Kitzberger T, Consejo Nacional de Investigaciones Científicas y Técnicas, Río Negro, Argentina, 32 Department of Plant and Environmental Klein T, Levanic T, Linares J-C, Sciences, Weizmann Institute of Science, Rehovot, Israel, 33 Department of Yield and Silviculture, Slovenian Forestry Institute, Lombardi F, Mäkinen H, Mészáros I, Ljubljana, Slovenia, 34 Department of Physical, Chemical and Natural Systems, Pablo de Olavide University, Seville, Spain, 35 Metsaranta JM, Oberhuber W, Department of Agricultural Science, Mediterranean University of Reggio Calabria, Reggio Calabria, Italy, 36 Natural Papadopoulos A, Petritan AM, Resources Institute Finland (Luke), Espoo, Finland, 37 Department of Botany, Faculty of Science and Technology, University Rohner B, Sangüesa-Barreda G, of Debrecen, Debrecen, Hungary, 38 Northern Forestry Centre, Canadian Forest Service, Natural Resources Canada, Smith JM, Stan AB, Stojanovic DB, Edmonton, AB, Canada, 39 Department of Botany, University of Innsbruck, Innsbruck, Austria, 40 Department of Forestry Suarez M-L, Svoboda M, Trotsiuk V, and Natural Environment Management, Technological Educational Institute of Stereas Elladas, Karpenisi, Greece, 41 National Villalba R, Westwood AR, Wyckoff PH Institute for Research and Development in Forestry “Marin Dracea”, Voluntari, Romania, 42 Departamento de Ciencias and Martínez-Vilalta J (2019) Agroforestales, EiFAB, iuFOR – University of Valladolid, Soria, Spain, 43 Department of Geography, University of Colorado, Early-Warning Signals of Individual Boulder, CO, United States, 44 Department of Geography, Planning and Recreation, Northern Arizona University, Flagstaff, Tree Mortality Based on Annual Radial AZ, United States, 45 Institute of Lowland Forestry and Environment, University of Novi Sad, Novi Sad, Serbia, 46 Grupo Growth. Front. Plant Sci. 9:1964. Ecología Forestal, CONICET – INTA, EEA Bariloche, Bariloche, Argentina, 47 Department of Environmental Systems Science, doi: 10.3389/fpls.2018.01964 Institute of Agricultural Sciences, ETH Zürich, Zurich, Switzerland, 48 Laboratorio de Dendrocronología e Historia Ambiental, Frontiers in Plant Science | www.frontiersin.org 1 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality Instituto Argentino de Nivología, Glaciología y Ciencias Ambientales, CCT CONICET Mendoza, Mendoza, Argentina, 49 Boreal Avian Modelling Project, Department of Renewable Resources, University of Alberta, Edmonton, AB, Canada, 50 Department of Biology, University of Minnesota, Morris, Morris, MN, United States, 51 Departament de Biologia Animal, de Biologia Vegetal i d’Ecologia, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain Tree mortality is a key driver of forest dynamics and its occurrence is projected to increase in the future due to climate change. Despite recent advances in our understanding of the physiological mechanisms leading to death, we still lack robust indicators of mortality risk that could be applied at the individual tree scale. Here, we build on a previous contribution exploring the differences in growth level between trees that died and survived a given mortality event to assess whether changes in temporal autocorrelation, variance, and synchrony in time-series of annual radial growth data can be used as early warning signals of mortality risk. Taking advantage of a unique global ring-width database of 3065 dead trees and 4389 living trees growing together at 198 sites (belonging to 36 gymnosperm and angiosperm species), we analyzed temporal changes in autocorrelation, variance, and synchrony before tree death (diachronic analysis), and also compared these metrics between trees that died and trees that survived a given mortality event (synchronic analysis). Changes in autocorrelation were a poor indicator of mortality risk. However, we found a gradual increase in inter- annual growth variability and a decrease in growth synchrony in the last ∼20 years before mortality of gymnosperms, irrespective of the cause of mortality. These changes could be associated with drought-induced alterations in carbon economy and allocation patterns. In angiosperms, we did not find any consistent changes in any metric. Such lack of any signal might be explained by the relatively high capacity of angiosperms to recover after a stress-induced growth decline. Our analysis provides a robust method for estimating early-warning signals of tree mortality based on annual growth data. In addition to the frequently reported decrease in growth rates, an increase in inter-annual growth variability and a decrease in growth synchrony may be powerful predictors of gymnosperm mortality risk, but not necessarily so for angiosperms. Keywords: tree mortality, ring-width, forest, growth, resilience indicators, drought, biotic agents, variance especially under progressively warmer and drier conditions (Pace INTRODUCTION et al., 2015; Trumbore et al., 2015). Episodes of tree mortality associated with drought and heat Evaluating individual tree mortality risk requires reliable stress have been reported in many forested biomes over the indicators that reveal temporal changes in tree vitality (Allen last decades (Allen et al., 2010; Hartmann et al., 2018), and et al., 2015; Hartmann et al., 2018). Such information can be are expected to increase under ongoing climate change in provided by physiological and anatomical data. Both abrupt and many regions (Allen et al., 2015). Forest dieback can induce long-term declines in hydraulic conductivity caused by drought- multiple changes in forest functions and dynamics (Franklin induced xylem embolism (Anderegg et al., 2013b; Adams et al., et al., 1987; Anderegg et al., 2013a, 2016b), including rapid shifts 2017; Choat et al., 2018) or changes in wood anatomical features in vegetation composition (Martínez-Vilalta and Lloret, 2016) (e.g., lower lumen area; Hereş et al., 2014; Pellizzari et al., or significant changes in terrestrial carbon sequestration with 2016) may indicate impending tree death. In association with resulting feedbacks to the climate system (e.g., Carvalhais et al., low whole-plant conductivity, reduced carbon assimilation and 2014). In addition to the direct loss of individuals, tree mortality depletion of stored carbohydrates may also occur due to the may also reduce forest regeneration capacity by decreasing the decline in stomatal conductance and leaf area, particularly for number of potential reproductive individuals, and by modifying gymnosperms (Galiano et al., 2011; Pangle et al., 2015; Adams the micro-environmental conditions and biotic interactions (e.g., et al., 2017). The determination of such mechanistic indicators is, Mueller et al., 2005; Royer et al., 2011). Being able to forecast however, costly, and temporally and spatially limited. Therefore, when and where tree mortality episodes are likely to occur is other approaches have been used to identify changes in tree thus a prerequisite for effective and adaptive forest management, health and mortality risk, such as temporal changes in crown Frontiers in Plant Science | www.frontiersin.org 2 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality defoliation (Dobbertin and Brang, 2001), or in radial growth can be difficult, as methodologies vary among studies, especially rates (e.g., Pedersen, 1998; Bigler and Bugmann, 2004; Dobbertin, for the quantification of the inter-annual variability in growth. 2005; Camarero et al., 2015; Hülsmann et al., 2018). Ring-width This aspect is fundamental, as opposite relationships could be (RW) data are especially suitable, as they provide retrospective obtained when using the standard deviation (SD) or the mean and long-term information about tree radial growth at an annual sensitivity (i.e., the mean relative change in RW between two resolution, and can be applied effectively at an affordable cost to consecutive rings; see Bunn et al., 2013) to characterize year-to- a large number of trees, sites, and species. year variability in RW series (Gillner et al., 2013; Macalady and A recent synthesis reported either abrupt or long-term Bugmann, 2014). Similarly, Camarero et al. (2015) did not find reduction in growth rates before death in most tree mortality any consistent change in growth synchrony between declining events recorded in dendrochronological studies worldwide and healthy trees among species. (Cailleret et al., 2017). However, this decrease in growth before Here, we tested whether EWS based on annual radial growth mortality was not ubiquitous, and its detection was subject to data can be used as universal indicators of tree mortality. important methodological constraints, especially related to the We used a unique, pan-continental database that contains sampling design (Cailleret et al., 2016). Therefore, additional paired growth time series for dead and surviving trees from metrics that go beyond changes in absolute growth rates are nearly 200 sites, including data for 13 angiosperm and 23 needed to identify individuals at high risk of mortality. Early- gymnosperm species. In particular, we measured temporal warning signals (EWS) have been proposed to characterize changes in tree growth variance, temporal autocorrelation, and (ecological) systems that are approaching a critical transition, synchrony (correlation among trees) after removing any effect i.e., a sudden and persistent shift in a system’s state (Scheffer driven by changes in absolute growth rates, which had been et al., 2009). EWS are caused by the gradual decrease in the studied in a previous publication (Cailleret et al., 2017). We recovery rate of a system after a perturbation – called “critical analyzed temporal changes in the properties of RW chronologies slowing down” (Wissel, 1984) – and have been identified prior of individual trees that died during a given stress event to population extinction in experiments under increasing levels (diachronic approach on dying trees), and compared the resulting of stress (e.g., Drake and Griffen, 2010; Dai et al., 2012; Veraart patterns to those from trees that survived this specific event et al., 2012). Tree death can be considered as system failure (synchronic approach). Contrary to standard tree growth analysis (Anderegg et al., 2012), and can be viewed as a critical transition that explores trends in RW chronologies, our approach here caused by the combined changes in the intensity, frequency and is to estimate changes in the dynamic properties of these duration of stress factors (Dakos et al., 2015), and high sensitivity time series (e.g., autocorrelation structure) that can be used of the tree to these specific stresses (Brandt et al., 2017). This as proxies of tree mortality risk. The methodology we develop would be somewhat analogous to recent applications of critical may assist in using such proxies for assessing individual tree transitions theory to human physiology, where health failures at resilience. the individual level can be anticipated with EWS (Olde Rikkert et al., 2016). In fact, the growth rate decline observed in most trees before mortality may be typical of such “critical slowing MATERIALS AND METHODS down” phenomenon, which can be captured by an increase in temporal autocorrelation and variance in time series of variables Tree-Ring Width Chronologies reflecting the functioning of the system (Scheffer et al., 2009; We used the pan-continental tree-ring width (mm) database Dakos et al., 2012b), and by a decrease in their synchrony with compiled by Cailleret et al. (2017), which includes 58 published the environment. These EWS would, respectively, reveal that the and unpublished datasets for which (i) both dying and surviving state of the system at any given moment becomes more and more trees growing together at the same site were cored, (ii) all like its recent past state, increasingly affected by shocks, and less individual chronologies had been successfully cross-dated, (iii) able to track the environmental fluctuations (Scheffer et al., 2009). mortality was proximally induced by stress (e.g., drought, Several studies have reported that RW time series of dying competition, and frost) and biotic agents in an endemic phase or declining individual trees tend to show increasing temporal (e.g., bark beetles, defoliator insects, fungi, acting as predisposing autocorrelation and variance over time or higher values than or contributing factor), and not by abrupt abiotic disturbances surviving individuals (e.g., Ogle et al., 2000; Suarez et al., 2004; such as windthrow, fire, or flooding, which may kill trees Millar et al., 2007; Kane and Kolb, 2014; Camarero et al., irrespective of their vitality and previous growth patterns (but 2015; see Supplementary Appendix A), especially in the case of see Nesmith et al., 2015). We grouped the datasets into four drought-induced mortality (McDowell et al., 2010; Heres et al., groups according to the main mortality sources determined by 2012; Gea-Izquierdo et al., 2014; Macalady and Bugmann, 2014). the authors of each study: (i) ‘drought’ corresponds to mortality However, it remains unclear whether rising growth variance caused by a single or several drought events without obvious and autocorrelation can be used as EWS for tree mortality. impact of biotic agents; (ii) ‘biotic’ includes sites in which First, other studies have reported opposite trends (e.g., Pedersen, mortality was induced primarily by biotic factors, including bark- 1998; Millar et al., 2012), or contrasting results depending on beetles, defoliator insects, and/or fungal infection; (iii) ‘drought the study species (Camarero et al., 2015), sites (Ogle et al., and biotic’ when the impact of biotic agents (including mistletoes 2000), and tree size (Herguido et al., 2016). Second, finding a and wood-borers) was associated with drought; (iv) and the common trend comparing results across different case studies group ‘others’ includes snow break, frost events, high competition Frontiers in Plant Science | www.frontiersin.org 3 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality intensity, and cases in which mortality were not evident or not SD and AR1 values are highly sensitive to the bandwidth of the specified. Gaussian kernel regression (see Supplementary Appendix D), The database analyzed here slightly differs from Cailleret et al. this one was fixed at 15 years rather than proportional to the (2017) as some sites for which we previously did not find any length of the time-series. Indeed, the latter approach would bias pair of dying/surviving tree with similar diameter at breast height the comparison among trees with different length of the time- (DBH) are considered in the present analysis, and as we excluded series (∼different ages). As we specifically focused on the end of trees with less than 20 measured rings (see below). A total of the RW time series, our analysis is prone to edge-effects that can 36 gymnosperm and angiosperm species were studied, with an emerge from Gaussian detrending (e.g., D’Arrigo et al., 2008; see overrepresentation of gymnosperms (64% of the species and 86% Supplementary Appendix E). Thus, the sensitivity of our results of the sites). Pinaceae was the most represented family, followed to the bandwidth length was also assessed (Supplementary by Fagaceae. Overall, the dataset analyzed in the main text Appendix D). (2) We used residuals (differences between the included 3065 dead trees and 4389 living trees growing at 198 sites original (raw) RW data and the smoothing spline from the mostly in boreal, temperate, and Mediterranean biomes of North Gaussian kernel regression) rather than ratios as done in America and Europe. More details on the sampling methods traditional dendrochronological studies. In this way, the output and on the assessments of the mortality sources, tree cambial chronology is centered on zero, is still heteroscedastic, and does age, DBH, and the year of death are available in Supplementary not include annual outliers when RW is close to zero, which often Appendix B and in Cailleret et al. (2017). occurs in dying trees. In contrast, most dendrochronological studies using RW data calculate ratios to get series that are Growth Metrics centered on one and are assumed to be homoscedastic (see Following Dakos et al. (2012a) and Camarero et al. (2015), Cook and Peters, 1997; Büntgen et al., 2005; Frank et al., 2006; we estimated levels and trends of Standard Deviation (SD) Supplementary Figure C2). To detect short-term (∼decadal) but and first-order autocorrelation (AR1) in detrended RW time still robust changes in growth metrics, SD, AR1 and COR were series of individual trees (Figure 1). Contrary to most calculated within a 20-year moving time-window (hereafter SD20 , dendrochronological studies, where AR1 is calculated using raw AR120 , and COR20 ). Trees with fewer than 20 rings were thus RW time series (e.g., Martín-Benito et al., 2008; Esper et al., discarded from this analysis. Other lengths of the moving time- 2015; Hartl-Meier et al., 2015), chronologies were detrended window were tested and showed similar results (Supplementary to correct for decadal to centennial trends, including decadal Appendix F). decreases in growth rates that are commonly observed prior to mortality (Cailleret et al., 2017). Such negative growth trends Detecting Trends in Growth Metrics would automatically lead to increasing trends in AR1 before tree Before Tree Mortality death (Figure 2B and Supplementary Appendix C), irrespective Our dataset allowed us to follow two approaches for estimating of the potential intrinsic change in the AR1 properties related to EWS that helped us to increase the robustness of our conclusions changes in tree vitality. In addition, we calculated the Pearson and to assess potential methodological biases. The first approach correlation (COR) coefficient between individual time series and was based on the analysis of the temporal changes in growth the site chronology (Figure 1). In contrast to the study by patterns of dying trees (diachronic approach), and the second Camarero et al. (2015), where COR coefficients corresponded on the comparison between dying and surviving individuals to the correlations between separated mean chronologies of coexisting at the same site (synchronic approach). ‘declining’ and ‘non-declining’ trees, we analyzed COR values between each individual detrended time series of dying trees Temporal Change in Growth Metrics of Dying Trees and the corresponding site- and species-specific chronology For each of the 3065 dying trees, we calculated SD20 , AR120 , and (including both dying and surviving trees), to reduce potential COR20 until the last year with complete ring formation, i.e., the biases at sites where few living trees had been sampled. Site year before tree death. We determined whether absolute values chronologies were derived using the bi-weight robust mean in SD20 , AR120 , and COR20 calculated during the last 20 years of the individual residual chronologies (Figure 1) to reduce preceding mortality (SD20f , AR120f , and COR20f for final values) the importance of outliers. This is particularly important when were significantly different than those during any other previous sample size is low, which is the case for some of our sites 20-year period. (Supplementary Appendix B). As SD20 calculated on the detrended chronology was still As we aimed at analyzing temporal changes in growth SD and positively related to mean growth rate calculated over the same AR1, and at comparing them among trees with different ages, period (meanRW20 ; see Supplementary Appendix C), we did sizes, or growth rates, two precautionary measures were taken to not directly analyze this metric, but instead we analyzed the detrend the RW data. (1) Most tree-ring- based studies remove residuals of a linear mixed-effect model (LMM) fitted to the size-effects on the RW data while keeping climate-induced overall dataset with meanRW20 as a fixed explanatory variable. decadal to centennial changes in growth rates using negative The same approach was used for AR120 and COR20 to center exponential curves or using the Regional Curve Standardization them on zero, which allows for an easier comparison among trees, method (e.g., Peters et al., 2015; Büntgen et al., 2017). In contrast, species, and periods with different mean growth rates. This is we used smoothing splines which are more flexible and more especially important as our sampling is not equal in terms of adapted to remove decadal trends (Cook and Peters, 1997). As mean tree age per species, which could lead to problems when Frontiers in Plant Science | www.frontiersin.org 4 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality FIGURE 1 | Example of early-warning signals of tree mortality based on ring-width (RW) data from two Abies alba trees from Mont Ventoux, France (Cailleret et al., 2014). The Standard Deviation (SD), first-order autocorrelation (AR1) and Pearson correlation coefficients (COR) were calculated on the original (Left) and detrended (Right) RW data using 20-year moving time windows. Frontiers in Plant Science | www.frontiersin.org 5 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality averaging results to analyze the overall temporal dynamics in in DBH (diffD−S DBHf ) as fixed effects. Random effects were growth metrics. Bootstrap resampling procedures were then used estimated for the intercept with species crossed with site as to test if the LMM residuals for SD20f , AR120f , and COR20f grouping factor. Direct age effects were not considered here significantly differed from zero (500 re-samplings). assuming that senescence only marginally affects tree function SD20 and meanRW20 were log-transformed unlike AR120 and (Mencuccini and Munné-Bosch, 2017). LMMs were finally COR20 values because their distributions were normal. As each used to predict diffD−S SD20f , diffD−S AR120f , and diffD−S COR20f tree species may have different SD and AR1 values for a similar growth rate (e.g., higher AR1 values are expected for evergreen species; Anderegg et al., 2015b), and COR values may depend on the number of trees used to derive the reference chronology, random effects were estimated for the intercept and the slope with species crossed with site as a grouping factor. Differences in Growth Metrics Between Conspecific Dying and Surviving Trees Although RW data were detrended using Gaussian filtering before calculating SD20 , AR120 , and COR20 , temporal changes in these metrics could be affected by site-specific decadal-scale changes in environmental conditions (e.g., change in climatic conditions or in canopy dynamics; Brienen et al., 2006; Carrer and Urbinati, 2006; Esper et al., 2015), regardless of individual intrinsic changes in tree vitality. Thus, to account for this possibility, we compared SD20f , AR120f , and COR20f between conspecific dying and surviving trees for each mortality event, i.e., for each combination of species, site, and mortality year (see Cailleret et al., 2017). For each dying tree, two approaches were followed for selecting comparable conspecific surviving trees from the same site: we only considered trees (i) with a similar DBH at the given mortality year (difference in final DBH between dying and surviving trees diffD−S DBHf ≤ 2.5 cm), or (ii) with a similar mean RW during the 20-year period before the mortality year (diffD−S meanRW20f ≤ 5%). In cases where none of the surviving trees fulfilled this condition, the corresponding dying tree was discarded. Following these two approaches, we considered 2887 (94.2% of the dying trees) and 2093 (68.3%) pairs of trees, respectively. On the one hand, comparing trees with similar DBH removes both geometric and structural (∼size) effects (see Bowman et al., 2013). For instance, large and dominant trees tend to show more plastic growth than small and suppressed ones (Martín-Benito et al., 2008; Mérian and Lebourgeois, 2011). On the other hand, comparing trees with similar mean RW removes mathematical effects related to changes in growth rate (see Supplementary Appendix C), and allows us to detect the presence of growth-based EWS in case of unchanging growth level before tree death (relative to the surviving trees). Thus, these two sampling approaches may individually bias the results, but they are complementary and should be considered together. On both datasets, we analyzed if the differences in SD20f , AR120f , and COR20f between conspecific dying and surviving trees (diffD−S SD20f , diffD−S AR120f , and diffD−S COR20f ) were significantly different from zero for all species groups and FIGURE 2 | Temporal change in SD20 (A), AR120 (B), and COR20 (C) before mortality sources using LMMs and bootstrapping methods. For death averaged for all dying trees and calculated on the original and each of these response variables, we fitted a LMM considering detrended RW data. We also show the temporal change in the residuals of the the species group and mortality source as interactive fixed linear mixed-effects models fitted to these metrics (right y-axes). Shaded areas represent the 95% confidence intervals of the means. Note that COR20 effects. As size or geometric effects could remain, we also values were not calculated on not-detrended RW data. included the difference in final mean RW (diffD−S RW20f ) and Frontiers in Plant Science | www.frontiersin.org 6 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality values in the theoretical situation in which dying trees have The first-order autocorrelation increased on average before similar meanRW20f and DBHf as surviving ones. tree death both in detrended RW chronologies (AR120 ) and in the residuals of the LMMs (resAR120 ) (Figure 2B). In fact, Sampling Scheme the residual AR1 (after removing both growth level and trend To account for the heterogeneity in the number of dying effects, Supplementary Appendix C) was higher than zero trees per site and per species in the dataset, we used two in the final 20-year period preceding tree death (resAR120f ; resampling procedures (Cailleret et al., 2017). First, we randomly Figure 3B). However, this was mostly true for gymnosperms sampled with replacement the same number of dying trees (except when mortality was caused by both drought and biotic (diachronic approach) or the same number of dying-surviving agents in samples including equal number of dying trees pairs (synchronic approach) for each of the 36 species. Second, per species; Supplementary Appendix G), and such level of a similar approach was followed to provide the same weight positive resAR120 values was not exclusive to the end of a in the calibration dataset for each of the 198 sites. With both approaches, each species or each site contributes equally to the results, which minimizes the bias related to under-sampling or over-sampling of specific sites or species (Supplementary Appendix G). Theoretical Expectations Finally, to detect which combinations of temporal trends in SD and AR1 can be expected when growth rates gradually decrease (commonly reported for dying trees), we generated theoretical RW time series based on simple growth models that included (i) an autocorrelation component, (ii) a long-term change in the mean, and (iii) some noise reflecting the environmental stochasticity (Supplementary Appendix E). The calculation of moving SD20 , AR120 , and COR20 values, and LMM analyses were performed using the packages earlywarnings (Dakos et al., 2012a), lme4 (Bates et al., 2014), and lmerTest (Kuznetsova et al., 2017) of the open-source software R (R Core Team, 2017). RESULTS Temporal Changes in Growth Metrics of Dying Trees SD20 calculated on detrended RW data started decreasing around 30 years before tree death (Figure 2A). This trend in SD20 was related to the general reduction in mean RW, as both variables are highly correlated (Supplementary Appendix C). After removing the effect of the mean RW using a LMM, SD residuals revealed an increase in inter-annual variability of RW before trees died (Figure 2A). The variability calculated for the 20-year period before mortality (resSD20f ) was generally higher than during the rest of the lives of dying trees (Figure 3). For gymnosperms, this pattern was significant irrespective of the mortality cause and of the method used to account for the heterogeneity in sample properties (Figure 3A and Supplementary Appendix G). In addition, the increase in variability was even stronger in the last 10-year period before mortality (Supplementary Appendix F). Results were less clear for angiosperms. Although variability was generally significantly FIGURE 3 | Variation in the residuals of SD (A), AR1 (B), and COR (C) higher at the end of an angiosperm’s life, this pattern was not calculated over the last 20-year period of the detrended ring-width time series present for all sources of mortality (e.g., when mortality was preceding tree death (resSD20f , resAR120f , and resCOR20f ) among mortality sources and species groups. Error bars depict 95% confidence intervals of caused by both drought and biotic agents, Figure 3A), and the mean residuals, which were determined from 500 bootstrap resamplings resSD20 did not monotonically increase toward the end of a tree’s of the original dataset. life (Supplementary Figure G1B). Frontiers in Plant Science | www.frontiersin.org 7 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality gymnosperm’s life (Supplementary Figure G1C). Thus, the period before mortality. These trends were more robust for high AR1 values calculated during the 20-year period before gymnosperms than for angiosperms, irrespective of the main gymnosperm mortality should not be interpreted as an exclusive cause of mortality. However, this result only partly conforms response indicative of impending tree death. In the case of to the patterns that are expected to characterize systems prior angiosperms, no significant or monotonic change in resAR120 to transitions due to critical slowing down (Scheffer et al., was observed consistently before mortality (Figure 3B and 2009; Dakos et al., 2012b), as no consistent changes in growth Supplementary Figure G1D). autocorrelation was detected for either taxonomic group. On average, Pearson correlations calculated between individual RW time series of dying trees and site chronologies decreased gradually ca. 30 years before death (Figure 2C). Mechanisms Underlying the Differences However, residual correlation values (resCOR20 ; after correcting Between Angiosperms and for mean RW, Supplementary Appendix C) were not Gymnosperms consistently below zero or lower than any previous period The increase in growth variance (for a given growth level) of across mortality sources, species groups, or sampling strategies dying gymnosperms may indicate an increase in susceptibility (Figure 3C and Supplementary Appendix G). The only to external influences such as climatic factors or pathogen exceptions were mortality caused by both drought and biotic diseases (e.g., Csank et al., 2016; Timofeeva et al., 2017). In agents for angiosperms and mortality caused by other factors in addition, their growth seems to be less coupled to high-frequency gymnosperms (Figure 3C and Supplementary Figure G2). climate fluctuations than surviving gymnosperms, as revealed by the decrease in growth synchrony with the overall site Differences in Temporal Changes of chronology (Fritts, 1976; Boden et al., 2014). Both changes may be Growth Metrics Between Conspecific associated with small-scale differences in atmospheric conditions Dying and Surviving Trees and in water availability that may become more important under stress, and with alterations in carbon allocation patterns, Dying trees generally showed higher variability in growth in the which may reflect the higher sensitivity of gymnosperms’ carbon last 20 years of their lives compared to surviving trees. Estimated economy to stress events (Adams et al., 2017). Some studies have differences in variance between dying and surviving trees (diffD−S shown stronger stomatal control and reduced non-structural SD) based on LMMs adjusted for growth rate (meanRW20f ) carbohydrate (NSC) concentrations in tissues of dying conifers, and size effects (DBHf ) were significantly higher than zero in relative to coexisting surviving individuals (Galiano et al., 2011; most cases for both angiosperms and gymnosperms and across Timofeeva et al., 2017). For instance, Pinus sylvestris saplings mortality drivers, except when trees were killed by biotic agents survived experimental drought longer when keeping assimilation (Figures 4A,B). This result was generally robust to different rates relatively high, even at the expense of higher water loss sampling schemes (unbalanced original dataset in Figure 4 (Garcia-Forner et al., 2016). Associated changes in xylogenesis vs. equal weight among species or sites in Supplementary phenology are also likely to be important. Compared to healthy Appendix G). Dying gymnosperms showed more consistent trees, defoliated pines showed a delay in the onset and reduction effects, although the magnitude of the SD difference between in the duration of cambial activity (Guada et al., 2016). Such dying and surviving trees was generally higher for angiosperms physiological responses could explain the observed higher growth (Figures 4A,B). variability in dying trees that goes along with a different Contrary to variance, autocorrelation did not significantly synchrony relative to surviving individuals. differ between dying and surviving trees. In specific In contrast, no consistent increase in growth variance was cases, differences were significantly higher than zero (e.g., observed for angiosperms. This is in line with reported small gymnosperms for drought-induced mortality and pairing by and short-term reductions in tree growth before angiosperm meanRW20f ), but this was never consistent across mortality death (Cailleret et al., 2017). Several reasons may explain drivers or sampling schemes (Figures 4C,D and Supplementary the lack of growth-based signals in angiosperms, including Appendix G). greater functional diversity (Augusto et al., 2014), species- Finally, we found predominantly lower COR20f for dying trees dependent responses to tree size compared to gymnosperms than surviving ones (Figures 4E,F). This pattern was largely (Steppe et al., 2011), the relatively loose coupling between consistent and of similar magnitude for every mortality source hydraulic failure and carbon depletion during drought (Adams for gymnosperms, but it was less clear for angiosperms, as et al., 2017), and their high recovery rates once favorable some differences in correlation (e.g., when biotic agents were environmental conditions prevail after drought (Augusto et al., the main mortality source) strongly depended on the sampling 2014; Anderegg et al., 2015b; Yin and Bauerle, 2017). Compared strategy, i.e., on the species and sites considered (Supplementary with gymnosperms, angiosperms generally have a higher capacity Appendix G). to (i) store NSC in their wood parenchyma (Plavcová et al., 2016), (ii) rebuild NSC pools owing to their higher stomatal DISCUSSION conductance (Lin et al., 2015) and growth efficiency, and (iii) replace conducting area via new xylem growth (Brodribb et al., We found a gradual increase in inter-annual growth variability 2010), resprouting (Zeppel et al., 2015), and potentially by and a decrease in growth synchrony during the ∼20-year refilling embolized xylem conduits (Johnson et al., 2012). In Frontiers in Plant Science | www.frontiersin.org 8 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality FIGURE 4 | Mean difference in SD20f (A,B), AR120f (C,D), and COR20f (E,F) values between dying and surviving trees predicted by the linear mixed-effects models (LMMs) fitted to the original dataset, fixing diffD−S RW20f and diffD−S DBHf at zero. Positive values mean that dying trees showed higher SD20f , AR120f , or COR20f compared to conspecific surviving trees. Standardization was based on similar meanRW20f (Left) and similar DBHf (Right). Error bars depict 95% confidence intervals of the predicted mean differences, which were determined from 500 bootstrap resamplings. Estimates of the LMMs are available in Supplementary Table H1. addition, all gymnosperms studied are evergreen species, whereas The lack of change in AR1 for both taxonomic groups may most analyzed angiosperms are deciduous (except Nothofagus be explained by antagonistic effects of the stress-induced changes betuloides, Nothofagus dombeyi, and Tamarix chinensis) which in key components of growth autocorrelation. On the one hand, may make them less dependent on previous-year leaf area and the growth dependency on NSC reserves may induce lagged growth efficiency. The relatively low number of angiosperm responses (‘growth memory’; Schulman, 1956; Esper et al., 2015; species included in our study, together with the higher variation Timofeeva et al., 2017; von Arx et al., 2017). On the other hand, in leaf and growth strategies (e.g., diffuse- vs. ring-porous reductions in hydraulic conductivity through xylem embolism species) and in recovery performance across species relative to and lower production of new functional xylem (Brodribb et al., gymnosperms (Cailleret et al., 2017; Yin and Bauerle, 2017) 2010), as well as reductions in overall crown area, or in leaf may have also contributed to the lack of consistent increases in size, number and longevity (Bréda et al., 2006; Girard et al., variance before tree mortality. 2012; Jump et al., 2017), may reduce the importance of lag effects. Frontiers in Plant Science | www.frontiersin.org 9 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality Finally, species-specific changes in water and carbon economy, Appendix C), differences are to be preferred over ratios (see also during and after high stress levels (Galiano et al., 2017), can Scheffer et al., 2009; Dakos et al., 2012a). explain the lack of a consistent change in AR1 preceding tree Second, it is always advisable to combine both diachronic and death. For instance, after intense drought, carbon assimilates may synchronic approaches to control for potential biases that are be invested into storage and restoration of root functions rather typical of field data; i.e., to focus on the temporal change in than into stem growth (Palacio et al., 2012; Hagedorn et al., 2016; growth metrics of dying trees before they actually die, and on Martínez-Vilalta et al., 2016), and the allocation priority level the comparison between coexisting trees that died and survived varies among species (Galiano et al., 2017). a specific mortality event (see also Gessler et al., 2018). Still, the synchronic approach is prone to artifacts, due to the fact that the Methodological Considerations group of ‘surviving’ trees at a given mortality event, which are Our results did not agree with some previous studies that used as a control, may include trees that died shortly after the showed that declining/dying trees had higher radial growth stress event. On the other hand, using the diachronic approach variance, autocorrelation, and synchrony than healthy/surviving only is not sufficient to disentangle changes in growth patterns ones, or showed an increase of these growth metrics before that are caused by variations in tree functions or in environmental death (e.g., Sánchez-Salguero et al., 2010; Amoroso et al., 2012; conditions (e.g., mortality of neighbors). For instance, first-order Camarero et al., 2015; Cailleret et al., 2016). They also indicate temporal autocorrelation calculated for the 20-year period before that the contrasting results obtained among previous studies the death of gymnosperms (AR120f ) was generally higher than (Supplementary Appendix A) may be due to methodological average AR120 (Figure 3B), which could indicate that high AR1 choices. In addition to the prescriptions that are inherent to is associated with impending tree death. However, it cannot be the characteristics of our database, e.g., regarding the inequality used as a predictive tool, as high AR1 values were also observed in sample sizes among sites and species (Supplementary during other periods of the trees’ lives, and because conspecific Appendix G), or the potential biases related to the assessment trees that survived the mortality event showed similar AR120f of the year of tree death (see Bigler and Rigling, 2013) or to values (Figures 4C,D). the measurement of narrow rings, there are three particularly Third, the unexpected lack of significant and meaningful important elements to consider, which we discuss in the following differences in growth-based EWS among the mortality groups paragraphs. considered here (see Cailleret et al., 2017) highlights the need First, if one aims at understanding the ecological mechanisms for a more precise determination of the mortality source(s) behind changes in the variance (quantified here with SD) and in the field. It is now well accepted that tree mortality is a autocorrelation of ring-width chronologies, the effects of tree phenomenon induced by multiple biotic and abiotic drivers with size, growth level, and growth trend should be removed or strong interdependencies (Manion, 1991; Anderegg et al., 2015a), accounted for. All these growth-related metrics are highly inter- and rarely occurs because of one single factor. Trees in the correlated (Supplementary Appendix C), which can lead to a ‘drought’ category might actually belong in ‘drought-biotic,’ and misinterpretation of the results. For instance, the decrease in trees in the ‘others’ category might belong in the ‘biotic agents’ SD20 calculated on raw RW data before tree death was caused by category (Das et al., 2016). In addition to information on climate, the gradual decrease in RW increment, and thus did not indicate soil, and stand characteristics, detailed pathological data would an intrinsic decrease in growth sensitivity to inter-annual changes be highly needed as biotic factors are involved in many individual in environmental conditions (Figure 2A). Four procedures can mortality reports (Das et al., 2016). be used to account for these effects: (i) detrending the RW time series to remove part of the low- and medium-frequency fluctuations, (ii) removing the remaining effects of growth rate Application of Early-Warning Signals of on the composite SD, AR1 and COR individual time series Tree Mortality Based on Radial Growth using mixed-effects models, (iii) comparing dying and surviving Our results expand previous assessments of the association trees with similar size or growth rate, and (iv) including the between tree radial growth and mortality risk based on the remaining differences in size and growth rate between dying direct effects of (absolute) growth rates (cf. Cailleret et al., and surviving trees of a given pair as an additional explanatory 2017) by focusing on subtler properties of the growth time variable in the statistical models. As in all dendrochronological series. Overall, we found that an increase in inter-annual growth analyses, the detrending method should be carefully selected (e.g., variability and a low growth synchrony could be used as EWS Esper et al., 2015). For instance, the bandwidth of the kernel of gymnosperm mortality. Because these results were clear even regression smoother should be constant among trees and should after accounting for any indirect effect driven by changing growth have an adequate length to capture enough medium-frequency levels, high growth variability and low synchrony could be used (∼decadal-scale) variability (Supplementary Appendix D) while as independent diagnostics to identify gymnosperm trees or minimizing end-effect biases (Supplementary Appendix E). populations at high risk of mortality. However, these trends Also, and in contrast to classical dendroclimatic studies that aim were much less consistent for angiosperms, and we did not find at getting homoscedastic growth time series by calculating ratios significant changes in autocorrelation prior to mortality. Hence, (Cook and Peters, 1997; Frank et al., 2006), the heteroscedasticity our results do not support the idea that critical slowing down of growth residuals needs to be retained. As using one or the indicators in radial growth data can be used as universal early other approach may lead to opposite trends (Supplementary warnings for tree mortality. Frontiers in Plant Science | www.frontiersin.org 10 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality There are many reasons why early-warning indicators based scales (e.g., Verbesselt et al., 2016; Rogers et al., 2018). Based on a on radial growth metrics may not be accurate indicators of rich pan-continental ring-width database of dying and surviving stress-induced tree mortality. First, although we did not detect trees, and by combining diachronic and synchronic approaches, any consistent difference in growth metrics between mortality our results highlight that in addition to the analysis of the multi- sources, some types of mortality stress may be too abrupt to be annual growth rates and trends (Cailleret et al., 2017), the inter- reflected in gradual changes in tree-ring width, and can occur annual variability of the growth time series can be used to assess without previous warning. For example, fungal diseases, bark- mortality risk, particularly for gymnosperm species. beetle outbreaks, or intense droughts can kill trees irrespective of their vitality, or at least, irrespective of their previous radial growth (Cherubini et al., 2002; Raffa et al., 2008; Sangüesa- AUTHOR CONTRIBUTIONS Barreda et al., 2015; Cailleret et al., 2017). Second, for a similar stress event, there is a large variety in the type and timing MC, VD, and JM-V conceived the ideas and designed the of responses among and within species (Jump et al., 2017) methodology. MC, TA, MA, JA, CB, HB, J-JC, PC, MRC, KČ, that may confound detection of common changes in growth AD, HD, GG-I, SG, LH, HH, A-MH, KH, PJ, JK, VK, TKi, TKl, sensitivity. Third, annual radial growth may not be the most TL, J-CL, FL, HM, IM, JM, WO, AP, AMP, BR, GS-B, JS, AS, appropriate variable to derive such early warnings, as it is not DS, M-LS, MS, VT, RV, AW, PW, and JM-V collected the tree- only dependent on tree carbon and water status, but also on the ring data. MC, SJ, ER, and JM-V compiled and cleaned the environmental influences on sink activity (Körner, 2015). Other ring-width database. MC analyzed the data and led the writing xylem-based physiological, anatomical, hydraulic, and isotopic of the manuscript with inputs from VD and JM-V. All authors properties that can be measured in tree rings may provide contributed critically to the drafts and gave final approval for complementary information on tree mortality probability (e.g., publication. Hereş et al., 2014; Anderegg et al., 2016a; Csank et al., 2016; Pellizzari et al., 2016; Timofeeva et al., 2017; Gessler et al., 2018). Fourth, despite recent developments (Gea-Izquierdo et al., ACKNOWLEDGMENTS 2015; Schiestl-Aalto et al., 2015; Guillemot et al., 2017), we lack This study generated from the COST Action STReESS (FP1106) mechanistic models of cambial activity based on sink demand, financially supported by the EU Framework Programme for carbon uptake and reserves and water relations, which can go Research and Innovation Horizon 2020. We would like to beyond simplistic formulations to produce clear expectations thank Don Falk (University of Arizona) and two reviewers for of ring-width dynamics before mortality (cf. Supplementary their valuable comments, all the colleagues for their help while Appendix E). Finally, depending on which state variable(s) compiling the database, and Louise Filion, Michael Dorman, are affected by the environmental ‘noise’ and by the change and Demetrios Sarris for sharing their datasets. MC was funded in tree vitality, the temporal trends in AR1 and in SD by the Swiss National Science Foundation (project number prior to the transition can vary (Dakos et al., 2012b). For 140968). ER was funded by the Research Foundation – Flanders instance, the simple autoregressive models we developed to (FWO, Belgium) and got support from the EU Horizon 2020 simulate decreasing growth rate over time, highlighted that all Programme through a Marie Skłodowska-Curie IF Fellowship combinations of SD and AR1 trends can theoretically occur (No. 659191). KČ was funded by the Slovenian Research Agency (Supplementary Appendix E). Considering that climate modifies (ARRS) Program P4-0015. IM was funded by National Research, tree growth based on multiple direct and indirect pathways Development and Innovation Office, project number NKFI- (e.g., via changes in cambial activity and in the water and SNN-125652. AMP was funded by the Ministry of Research carbon economy), the relationship between climate variability and Innovation, CNCS – UEFISCDI, project number PN-III- and growth autocorrelation and variance is not straightforward. P1-1.1-TE-2016-1508, within PNCDI III (BIOCARB). GS-B Similarly, the SD metric integrates both tree resistance and was supported by a Juan de la Cierva-Formación grant from recovery to specific events that could be independently analyzed MINECO (FJCI 2016-30121). DS was funded by the project III (Lloret et al., 2011; Dakos et al., 2015). 43007 financed by the Ministry of Education and Science of the Climate change is predicted to modify mean temperature and Republic of Serbia. AW was funded by Canada’s Natural Sciences precipitation, but also to increase the inter-annual variability and Engineering Research Council and Manitoba Sustainable and persistence of climatic fluctuations (Fischer et al., 2013; Development. JM-V benefited from an ICREA Academia Award. Lenton et al., 2017), and to modify the population dynamics of Any use of trade, firm, or product names is for descriptive biotic agents (Allen et al., 2015). Several physiological thresholds purposes only and does not imply endorsement by the United can be exceeded during extreme biotic or abiotic conditions States Government. (e.g., during drought; Adams et al., 2017), which may ultimately lead to individual tree mortality, and potentially to widespread forest decline in many regions (Lloret et al., 2012; Reyer et al., SUPPLEMENTARY MATERIAL 2013; Allen et al., 2015). However, we still lack a general set of mechanistic and empirical EWS of tree mortality at the individual The Supplementary Material for this article can be found online scale (Gessler et al., 2018) that could be used to complement the at: https://www.frontiersin.org/articles/10.3389/fpls.2018.01964/ signals used for detecting dieback at the forest stand or landscape full#supplementary-material Frontiers in Plant Science | www.frontiersin.org 11 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality REFERENCES Brienen, R. J., Zuidema, P. A., and During, H. J. (2006). Autocorrelated growth of tropical forest trees: unraveling patterns and quantifying consequences. For. Adams, H. D., Zeppel, M. J., Anderegg, W. R., Hartmann, H., Landhäusser, Ecol. Manage. 237, 179–190. doi: 10.1016/j.foreco.2006.09.042 S. M., Tissue, D. T., et al. (2017). A multi-species synthesis of physiological Brodribb, T. J., Bowman, D. J., Nichols, S., Delzon, S., and Burlett, R. (2010). Xylem mechanisms in drought-induced tree mortality. Nat. Ecol. Evol. 1, 1285–1291. function and growth rate interact to determine recovery rates after exposure to doi: 10.1038/s41559-017-0248-x extreme water deficit. New Phytol. 188, 533–542. doi: 10.1111/j.1469-8137.2010. Allen, C. D., Breshears, D. D., and McDowell, N. G. (2015). On underestimation 03393.x of global vulnerability to tree mortality and forest die-off from hotter Bunn, A. G., Jansma, E., Korpela, M., Westfall, R. D., and Baldwin, J. (2013). drought in the Anthropocene. Ecosphere 6, 1–55. doi: 10.1890/ES15- Using simulations and data to evaluate mean sensitivity as a useful statistic 00203.1 in dendrochronology. Dendrochronologia 31, 250–254. doi: 10.1016/j.dendro. Allen, C. D., Macalady, A. K., Chenchouni, H., Bachelet, D., McDowell, N., 2013.01.004 Vennetier, M., et al. (2010). A global overview of drought and heat-induced tree Büntgen, U., Esper, J., Frank, D. C., Treydte, K., Schmidhalter, M., Nicolussi, K., mortality reveals emerging climate change risks for forests. For. Ecol. Manage. et al. (2005). “The effect of power transformation on RCS–evidence from 259, 660–684. doi: 10.1016/j.foreco.2009.09.001 three millennial length alpine chronologies,” in Proceedings of the International Amoroso, M. M., Daniels, L. D., and Larson, B. C. (2012). Temporal patterns of conference Tree rings in archaeology, climatology and ecology, TRACE, radial growth in declining Austrocedrus chilensis forests in Northern Patagonia: Birmensdorf, 141–149. the use of tree-rings as an indicator of forest decline. For. Ecol. Manage. 265, Büntgen, U., Krusic, P. J., Verstege, A., Sangüesa-Barreda, G., Wagner, S., 62–70. doi: 10.1016/j.foreco.2011.10.021 Camarero, J. J., et al. (2017). New tree-ring evidence from the Pyrenees reveals Anderegg, W. R., Berry, J. A., and Field, C. B. (2012). Linking definitions, Western Mediterranean climate variability since medieval times. J. Climate 30, mechanisms, and modeling of drought-induced tree death. Trends Plant Sci. 5295–5318. doi: 10.1175/JCLI-D-16-0526.1 17, 693–700. doi: 10.1016/j.tplants.2012.09.006 Cailleret, M., Bigler, C., Bugmann, H., Camarero, J. J., Cufar, K., Davi, H., et al. Anderegg, W. R., Hicke, J. A., Fisher, R. A., Allen, C. D., Aukema, J., Bentz, B., (2016). Towards a common methodology for developing logistic tree mortality et al. (2015a). Tree mortality from drought, insects, and their interactions in a models based on ring-width data. Ecol. Appl. 26, 1827–1841. doi: 10.1890/15- changing climate. New Phytol. 208, 674–683. doi: 10.1111/nph.13477 1402.1 Anderegg, W. R., Kane, J. M., and Anderegg, L. D. (2013a). Consequences of Cailleret, M., Jansen, S., Robert, E. M., DeSoto, L., Aakala, T., Antos, J. A., et al. widespread tree mortality triggered by drought and temperature stress. Nat. (2017). A synthesis of radial growth patterns preceding tree mortality. Glob. Clim. Change 3, 30–36. doi: 10.1038/nclimate1635 Change Biol. 23, 1675–1690. doi: 10.1111/gcb.13535 Anderegg, W. R., Klein, T., Bartlett, M., Sack, L., Pellegrini, A. F., Choat, B., Cailleret, M., Nourtier, M., Amm, A., Durand-Gillmann, M., and Davi, H. (2014). et al. (2016a). Meta-analysis reveals that hydraulic traits explain cross-species Drought-induced decline and mortality of silver fir differ among three sites in patterns of drought-induced tree mortality across the globe. Proc. Natl. Acad. Southern France. Ann. For. Sci. 71, 643–657. doi: 10.1007/s13595-013-0265-0 Sci. U.S.A. 113, 5024–5029. doi: 10.1073/pnas.1525678113 Camarero, J. J., Gazol, A., Sangüesa-Barreda, G., Oliva, J., and Vicente-Serrano, Anderegg, W. R., Martinez-Vilalta, J., Cailleret, M., Camarero, J. J., Ewers, B. E., S. M. (2015). To die or not to die: early warnings of tree dieback in response to Galbraith, D., et al. (2016b). When a tree dies in the forest: scaling climate- a severe drought. J. Ecol. 103, 44–57. doi: 10.1111/1365-2745.12295 driven tree mortality to ecosystem water and carbon fluxes. Ecosystems 19, Carrer, M., and Urbinati, C. (2006). Long-term change in the sensitivity of tree- 1133–1147. doi: 10.1007/s10021-016-9982-1 ring growth to climate forcing in Larix decidua. New Phytol. 170, 861–872. Anderegg, W. R., Plavcová, L., Anderegg, L. D., Hacke, U. G., Berry, J. A., doi: 10.1111/j.1469-8137.2006.01703.x and Field, C. B. (2013b). Drought’s legacy: multiyear hydraulic deterioration Carvalhais, N., Forkel, M., Khomik, M., Bellarby, J., Jung, M., Migliavacca, M., et al. underlies widespread aspen forest die-off and portends increased future risk. (2014). Global covariation of carbon turnover times with climate in terrestrial Glob. Change Biol. 19, 1188–1196. doi: 10.1111/gcb.12100 ecosystems. Nature 514, 213–217. doi: 10.1038/nature13731 Anderegg, W. R., Schwalm, C., Biondi, F., Camarero, J. J., Koch, G., Litvak, M., et al. Cherubini, P., Fontana, G., Rigling, D., Dobbertin, M., Brang, P., and Innes, J. L. (2015b). Pervasive drought legacies in forest ecosystems and their implications (2002). Tree-life history prior to death: two fungal root pathogens affect tree- for carbon cycle models. Science 349, 528–532. doi: 10.1126/science.aab1833 ring growth differently. J. Ecol. 90, 839–850. doi: 10.1046/j.1365-2745.2002. Augusto, L., Davies, T. J., Delzon, S., and Schrijver, A. (2014). The enigma of 00715.x the rise of angiosperms: Can we untie the knot? Ecol. Lett. 17, 1326–1338. Choat, B., Brodribb, T. J., Brodersen, C. R., Duursma, R. A., López, R., and Medlyn, doi: 10.1111/ele.12323 B. E. (2018). Triggers of tree mortality under drought. Nature 558, 531–539. Bates, D., Mächler, M., Bolker, B., and Walker, S. (2014). Fitting linear mixed- doi: 10.1038/s41586-018-0240-x effects models using lme4. arXiv preprint arXiv: 1406.5823. Available at: https: Cook, E. R., and Peters, K. (1997). Calculating unbiased tree-ring indices for the //arxiv.org/abs/1406.5823. study of climatic and environmental change. Holocene 7, 361–370. doi: 10.1177/ Bigler, C., and Bugmann, H. (2004). Predicting the time of tree death using 095968369700700314 dendrochronological data. Ecol. Appl. 14, 902–914. doi: 10.1890/03-5011 Csank, A. Z., Miller, A. E., Sherriff, R. L., Berg, E. E., and Welker, J. M. (2016). Tree- Bigler, C., and Rigling, A. (2013). Precision and accuracy of tree-ring-based death ring isotopes reveal drought sensitivity in trees killed by spruce beetle outbreaks dates of mountain pines in the Swiss National Park. Trees 27, 1703–1712. in south-central Alaska. Ecol. Appl. 26, 2001–2020. doi: 10.1002/eap.1365 doi: 10.1007/s00468-013-0917-6 Dai, L., Vorselen, D., Korolev, K. S., and Gore, J. (2012). Generic indicators for loss Boden, S., Kahle, H. P., von Wilpert, K., and Spiecker, H. (2014). Resilience of of resilience before a tipping point leading to population collapse. Science 336, Norway spruce (Picea abies (L.) Karst) growth to changing climatic conditions 1175–1177. doi: 10.1126/science.1219805 in Southwest Germany. For. Ecol. Manage. 315, 12–21. doi: 10.1016/j.foreco. Dakos, V., Carpenter, S. R., Brock, W. A., Ellison, A. M., Guttal, V., Ives, A. R., 2013.12.015 et al. (2012a). Methods for detecting early warnings of critical transitions in Bowman, D. M., Brienen, R. J., Gloor, E., Phillips, O. L., and Prior, L. D. (2013). time series illustrated using simulated ecological data. PLoS One 7:e41010. Detecting trends in tree growth: not so simple. Trends Plant Sci. 18, 11–17. doi: 10.1371/journal.pone.0041010 doi: 10.1016/j.tplants.2012.08.005 Dakos, V., Carpenter, S. R., van Nes, E. H., and Scheffer, M. (2015). Resilience Brandt, L. A., Butler, P. R., Handler, S. D., Janowiak, M. K., Shannon, P. D., and indicators: prospects and limitations for early warnings of regime shifts. Philos. Swanston, C. W. (2017). Integrating science and management to assess forest Trans. R. Soc. B 370:20130263. doi: 10.1098/rstb.2013.0263 ecosystem vulnerability to climate change. J. For. 115, 212–221. doi: 10.5849/ Dakos, V., Van Nes, E. H., D’Odorico, P., and Scheffer, M. (2012b). Robustness of jof.15-147 variance and autocorrelation as indicators of critical slowing down. Ecology 93, Bréda, N., Huc, R., Granier, A., and Dreyer, E. (2006). Temperate forest trees and 264–271. doi: 10.1890/11-0889.1 stands under severe drought: a review of ecophysiological responses, adaptation D’Arrigo, R., Wilson, R., Liepert, B., and Cherubini, P. (2008). On the ‘divergence processes and long-term consequences. Ann. For. Sci. 63, 625–644. doi: 10.1051/ problem’ in northern forests: a review of the tree-ring evidence and possible forest:2006042 causes. Glob. Planet. Change 60, 289–305. doi: 10.1016/j.gloplacha.2007.03.004 Frontiers in Plant Science | www.frontiersin.org 12 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality Das, A. J., Stephenson, N. L., and Davis, K. P. (2016). Why do trees die? drought-induced tree and forest mortality. New Phytol. 218, 15–28. doi: 10. Characterizing the drivers of background tree mortality. Ecology 97, 2616–2627. 1111/nph.15048 doi: 10.1002/ecy.1497 Hereş, A. M., Camarero, J. J., López, B. C., and Martínez-Vilalta, J. (2014). Dobbertin, M. (2005). Tree growth as an indicator of tree vitality and of tree Declining hydraulic performances and low carbon investments in tree rings reaction to environmental stress: a review. Eur. J. For. Res. 124, 319–333. predate Scots pine drought-induced mortality. Trees 28, 1737–1750. doi: 10. doi: 10.1007/s10342-005-0085-3 1007/s00468-014-1081-3 Dobbertin, M., and Brang, P. (2001). Crown defoliation improves tree mortality Heres, A. M., Martínez-Vilalta, J., and López, B. C. (2012). Growth patterns in models. For. Ecol. Manage. 141, 271–284. doi: 10.1016/S0378-1127(00)00335-2 relation to drought-induced mortality at two Scots pine (Pinus sylvestris L.) Drake, J. M., and Griffen, B. D. (2010). Early warning signals of extinction in sites in NE Iberian Peninsula. Trees 26, 621–630. doi: 10.1007/s00468-011- deteriorating environments. Nature 467, 456–459. doi: 10.1038/nature09389 0628-9 Esper, J., Schneider, L., Smerdon, J. E., Schöne, B. R., and Büntgen, U. (2015). Herguido, E., Granda, E., Benavides, R., García-Cervigón, A. I., Camarero, J. J., and Signals and memory in tree-ring width and density data. Dendrochronologia Valladares, F. (2016). Contrasting growth and mortality responses to climate 35, 62–70. doi: 10.1016/j.dendro.2015.07.001 warming of two pine species in a continental Mediterranean ecosystem. For. Fischer, E. M., Beyerle, U., and Knutti, R. (2013). Robust spatially aggregated Ecol. Manage. 363, 149–158. doi: 10.1016/j.foreco.2015.12.038 projections of climate extremes. Nat. Clim. Change 3, 1033–1038. doi: 10.1038/ Hülsmann, L., Bugmann, H., Cailleret, M., and Brang, P. (2018). How to kill a tree: nclimate2051 empirical mortality models for 18 species and their performance in a dynamic Frank, D., Esper, J., and Cook, E. R. (2006). “On variance adjustments in tree- forest model. Ecol. Appl. 28, 522–540. doi: 10.1002/eap.1668 ring chronology development,” in Tree Rings in Archaeology, Climatology and Johnson, D. M., McCulloh, K. A., Woodruff, D. R., and Meinzer, F. C. (2012). Ecology, TRACE, Vol. 4, eds I. Heinrich, H. Gärtner, M. Monbaron, and Hydraulic safety margins and embolism reversal in stems and leaves: Why are G. Schleser, 56–66. conifers and angiosperms so different? Plant Sci. 195, 48–53. doi: 10.1016/j. Franklin, J. F., Shugart, H. H., and Harmon, M. E. (1987). Tree death as an plantsci.2012.06.010 ecological process. BioScience 37, 550–556. doi: 10.2307/1310665 Jump, A. S., Ruiz-Benito, P., Greenwood, S., Allen, C. D., Kitzberger, T., Fritts, H. C. (1976). Tree Rings and Climate. London: Academic Press . Fensham, R., et al. (2017). Structural overshoot of tree growth with climate Galiano, L., Martínez-Vilalta, J., and Lloret, F. (2011). Carbon reserves and canopy variability and the global spectrum of drought-induced forest dieback. Glob. defoliation determine the recovery of Scots pine 4 yr after a drought episode. Change Biol. 23, 3742–3757. doi: 10.1111/gcb.13636 New Phytol. 190, 750–759. doi: 10.1111/j.1469-8137.2010.03628.x Kane, J. M., and Kolb, T. E. (2014). Short-and long-term growth characteristics Galiano, L., Timofeeva, G., Saurer, M., Siegwolf, R., Martínez-Vilalta, J., associated with tree mortality in southwestern mixed-conifer forests. Can. J. Hommel, R., et al. (2017). The fate of recently fixed carbon after drought For. Res. 44, 1227–1235. doi: 10.1139/cjfr-2014-0186 release: towards unravelling C storage regulation in Tilia platyphyllos and Pinus Körner, C. (2015). Paradigm shift in plant growth control. Curr. Opin. Plant Biol. sylvestris. Plant Cell Environ. 40, 1711–1724. doi: 10.1111/pce.12972 25, 107–114. doi: 10.1016/j.pbi.2015.05.003 Garcia-Forner, N., Sala, A., Biel, C., Savé, R., and Martínez-Vilalta, J. (2016). Kuznetsova, A., Brockhoff, P. B., and Christensen, R. H. B. (2017). LmerTest Individual traits as determinants of time to death under extreme drought in package: tests in linear mixed effects models. J. Stat. Softw. 82, 1–26. doi: 10. Pinus sylvestris L. Tree Physiol. 36, 1196–1209. doi: 10.1093/treephys/tpw040 18637/jss.v082.i13 Gea-Izquierdo, G., Guibal, F., Joffre, R., Ourcival, J. M., Simioni, G., and Guiot, J. Lenton, T. M., Dakos, V., Bathiany, S., and Scheffer, M. (2017). Observed trends (2015). Modelling the climatic drivers determining photosynthesis and carbon in the magnitude and persistence of monthly temperature variability. Sci. Rep. allocation in evergreen Mediterranean forests using multiproxy long time series. 7:5940. doi: 10.1038/s41598-017-06382-x Biogeosciences 12, 3695–3712. doi: 10.5194/bg-12-3695-2015 Lin, Y. S., Medlyn, B. E., Duursma, R. A., Prentice, I. C., Wang, H., Baig, S., et al. Gea-Izquierdo, G., Viguera, B., Cabrera, M., and Cañellas, I. (2014). Drought (2015). Optimal stomatal behavior around the world. Nat. Clim. Change 5, induced decline could portend widespread pine mortality at the xeric ecotone 459–464. doi: 10.1038/nclimate2550 in managed mediterranean pine-oak woodlands. For. Ecol. Manage. 320, 70–82. Lloret, F., Escudero, A., Iriondo, J. M., Martínez-Vilalta, J., and Valladares, F. doi: 10.1016/j.foreco.2014.02.025 (2012). Extreme climatic events and vegetation: the role of stabilizing processes. Gessler, A., Cailleret, M., Joseph, J., Schönbeck, L., Schaub, M., Lehmann, M., et al. Glob. Change Biol. 18, 797–805. doi: 10.1111/j.1365-2486.2011.02624.x (2018). Drought induced tree mortality–a tree-ring isotope based conceptual Lloret, F., Keeling, E. G., and Sala, A. (2011). Components of tree resilience: effects model to assess mechanisms and predispositions. New Phytol. 219, 485–490. of successive low-growth episodes in old ponderosa pine forests. Oikos 120, doi: 10.1111/nph.15154 1909–1920. doi: 10.1111/j.1600-0706.2011.19372.x Gillner, S., Rüger, N., Roloff, A., and Berger, U. (2013). Low relative growth Macalady, A. K., and Bugmann, H. (2014). Growth-mortality relationships in rates predict future mortality of common beech (Fagus sylvatica L.). For. Ecol. piñon pine (Pinus edulis) during severe droughts of the past century: shifting Manage. 302, 372–378. doi: 10.1016/j.foreco.2013.03.032 processes in space and time. PLoS One 9:e92770. doi: 10.1371/journal.pone. Girard, F., Vennetier, M., Guibal, F., Corona, C., Ouarmim, S., and Herrero, A. 0092770 (2012). Pinus halepensis Mill. crown development and fruiting declined with Manion, P. D. (1991). Tree Disease Concepts. Englewood Cliffs, NJ: Prentice-Hall. repeated drought in Mediterranean France. Eur. J. For. Res. 131, 919–931. Martín-Benito, D., Cherubini, P., Del Río, M., and Cañellas, I. (2008). Growth doi: 10.1007/s10342-011-0565-6 response to climate and drought in Pinus nigra Arn. trees of different crown Guada, G., Camarero, J. J., Sánchez-Salguero, R., and Cerrillo, R. M. N. (2016). classes. Trees 22, 363–373. doi: 10.1007/s00468-007-0191-6 Limited growth recovery after drought-induced forest dieback in very defoliated Martínez-Vilalta, J., and Lloret, F. (2016). Drought-induced vegetation trees of two pine species. Front. Plant Sci. 7:418. doi: 10.3389/fpls.2016.00418 shifts in terrestrial ecosystems: the key role of regeneration dynamics. Guillemot, J., Francois, C., Hmimina, G., Dufrêne, E., Martin-StPaul, N. K., Glob. Planet. Change 144, 94–108. doi: 10.1016/j.gloplacha.2016. Soudani, K., et al. (2017). Environmental control of carbon allocation matters 07.009 for modelling forest growth. New Phytol. 214, 180–193. doi: 10.3389/fpls.2016. Martínez-Vilalta, J., Sala, A., Asensio, D., Galiano, L., Hoch, G., Palacio, S., et al. 00418 (2016). Dynamics of non-structural carbohydrates in terrestrial plants: a global Hagedorn, F., Joseph, J., Peter, M., Luster, J., Pritsch, K., Geppert, U., et al. (2016). synthesis. Ecol. Monogr. 86, 495–516. doi: 10.1002/ecm.1231 Recovery of trees from drought depends on belowground sink control. Nat. McDowell, N., Allen, C. D., and Marshall, L. (2010). Growth, carbon-isotope Plants 2:16111. doi: 10.1038/nplants.2016.111 discrimination, and drought-associated mortality across a Pinus ponderosa Hartl-Meier, C., Zang, C., Büntgen, U., Esper, J., Rothe, A., Göttlein, A., et al. elevational transect. Glob. Change Biol. 16, 399–415. doi: 10.1111/j.1365-2486. (2015). Uniform climate sensitivity in tree-ring stable isotopes across species 2009.01994.x and sites in a mid-latitude temperate forest. Tree Physiol. 35, 4–15. doi: 10.1093/ Mencuccini, M., and Munné-Bosch, S. (2017). “Physiological and biochemical treephys/tpu096 processes related to ageing and senescence in plants,” in The Evolution of Hartmann, H., Moura, C. F., Anderegg, W. R., Ruehr, N. K., Salmon, Y., Allen, Senescence in the Tree of Life, eds R. P. Shefferson, O. R. Jones, and R. C. D., et al. (2018). Research frontiers for improving our understanding of Salguero-Gómez (Cambridge: Cambridge University Press), 257–283. Frontiers in Plant Science | www.frontiersin.org 13 January 2019 | Volume 9 | Article 1964 Cailleret et al. Early-Warning Signals of Tree Mortality Mérian, P., and Lebourgeois, F. (2011). Size-mediated climate–growth increases understorey solar input regionally: primary and secondary ecological relationships in temperate forests: a multi-species analysis. For. Ecol. Manage. implications. J. Ecol. 99, 714–723. doi: 10.1111/j.1365-2745.2011.01804.x 261, 1382–1391. doi: 10.1016/j.foreco.2011.01.019 Sánchez-Salguero, R., Navarro, R. M., Camarero, J. J., and Fernández-Cancio, Á Millar, C. I., Westfall, R. D., and Delany, D. L. (2007). Response of high-elevation (2010). Drought-induced growth decline of Aleppo and maritime pine forests limber pine (Pinus flexilis) to multiyear droughts and 20th-century warming, in south-eastern Spain. For. Syst. 19, 458–470. doi: 10.5424/fs/2010193-9131 Sierra Nevada, California, USA. Can. J. For. Res. 37, 2508–2520. doi: 10.1139/ Sangüesa-Barreda, G., Linares, J. C., and Camarero, J. J. (2015). Reduced X07-097 growth sensitivity to climate in bark-beetle infested Aleppo pines: connecting Millar, C. I., Westfall, R. D., Delany, D. L., Bokach, M. J., Flint, A. L., and Flint, climatic and biotic drivers of forest dieback. For. Ecol. Manage. 357, 126–137. L. E. (2012). Forest mortality in high-elevation whitebark pine (Pinus albicaulis) doi: 10.1016/j.foreco.2015.08.017 forests of eastern California, USA; influence of environmental context, bark Scheffer, M., Bascompte, J., Brock, W. A., Brovkin, V., Carpenter, S. R., Dakos, V., beetles, climatic water deficit, and warming. Can. J. For. Res. 42, 749–765. et al. (2009). Early warning signals for critical transitions. Nature 461, 53–59. doi: 10.1139/x2012-031 doi: 10.1038/nature08227 Mueller, R. C., Scudder, C. M., Porter, M. E., Talbot Trotter, R., Gehring, C. A., Schiestl-Aalto, P., Kulmala, L., Mäkinen, H., Nikinmaa, E., and Mäkelä, A. and Whitham, T. G. (2005). Differential tree mortality in response to severe (2015). CASSIA–a dynamic model for predicting intra-annual sink demand drought: evidence for long-term vegetation shifts. J. Ecol. 93, 1085–1093. and interannual growth variation in Scots pine. New Phytol. 206, 647–659. doi: 10.1111/j.1365-2745.2005.01042.x doi: 10.1111/nph.13275 Nesmith, J. C., Das, A. J., O’Hara, K. L., and van Mantgem, P. J. (2015). The Schulman, E. (1956). Dendroclimatic Changes in Semiarid America. Tucson: influence of prefire tree growth and crown condition on postfire mortality of University of Arizona Press. sugar pine following prescribed fire in Sequoia National Park. Can. J. For. Res. Steppe, K., Niinemets, Ü, and Teskey, R. O. (2011). “Tree size- and age-related 45, 910–919. doi: 10.1139/cjfr-2014-0449 changes in leaf physiology and their influence on carbon gain,” in Size- and Ogle, K., Whitham, T. G., and Cobb, N. S. (2000). Tree-ring variation in pinyon Age-Related Changes in Tree Structure and Function, eds F. C. Meinzer, B. predicts likelihood of death following severe drought. Ecology 81, 3237–3243. Lachenbruch, and T. E. Dawson (Dodrecht: Springer), 235–254. doi: 10.1007/ doi: 10.1890/0012-9658(2000)081[3237:TRVIPP]2.0.CO;2 978-94-007-1242-3 Olde Rikkert, M. G., Dakos, V., Buchman, T. G., de Boer, R., Glass, L., Cramer, Suarez, M. L., Ghermandi, L., and Kitzberger, T. (2004). Factors predisposing A. O., et al. (2016). Slowing down of recovery as generic risk marker for episodic drought-induced tree mortality in Nothofagus–site, climatic sensitivity acute severity transitions in chronic diseases. Crit. Care Med. 44, 601–606. and growth trends. J. Ecol. 92, 954–966. doi: 10.1111/j.1365-2745.2004.00941.x doi: 10.1097/CCM.0000000000001564 Timofeeva, G., Treydte, K., Bugmann, H., Rigling, A., Schaub, M., Siegwolf, R., et al. Pace, M. L., Carpenter, S. R., and Cole, J. J. (2015). With and without warning: (2017). Long-term effects of drought on tree-ring growth and carbon isotope managing ecosystems in a changing world. Front. Ecol. Environ. 13, 460–467. variability in Scots pine in a dry environment. Tree Physiol. 37, 1028–1041. doi: 10.1890/150003 doi: 10.1093/treephys/tpx041 Palacio, S., Hernández, R., Maestro-Martínez, M., and Camarero, J. J. (2012). Fast Trumbore, S., Brando, P., and Hartmann, H. (2015). Forest health and global replenishment of initial carbon stores after defoliation by the pine processionary change. Science 349, 814–818. doi: 10.1126/science.aac6759 moth and its relationship to the re-growth ability of trees. Trees 26, 1627–1640. Veraart, A. J., Faassen, E. J., Dakos, V., van Nes, E. H., Lürling, M., and Scheffer, M. doi: 10.1007/s00468-012-0739-y (2012). Recovery rates reflect distance to a tipping point in a living system. Pangle, R. E., Limousin, J. M., Plaut, J. A., Yepez, E. A., Hudson, P. J., Boutz, Nature 481, 357–359. doi: 10.1038/nature10723 A. L., et al. (2015). Prolonged experimental drought reduces plant hydraulic Verbesselt, J., Umlauf, N., Hirota, M., Holmgren, M., Van Nes, E. H., Herold, M., conductance and transpiration and increases mortality in a piñon–juniper et al. (2016). Remotely sensed resilience of tropical forests. Nat. Clim. Change 6, woodland. Ecol. Evol. 5, 1618–1638. doi: 10.1002/ece3.1422 1028–1031. doi: 10.1038/nclimate3108 Pedersen, B. S. (1998). The role of stress in the mortality of midwestern oaks von Arx, G., Arzac, A., Fonti, P., Frank, D., Zweifel, R., Rigling, A., et al. (2017). as indicated by growth prior to death. Ecology 79, 79–93. doi: 10.1890/0012- Responses of sapwood ray parenchyma and non-structural carbohydrates of 9658(1998)079[0079:TROSIT]2.0.CO;2 Pinus sylvestris to drought and long-term irrigation. Funct. Ecol. 31, 1371–1382. Pellizzari, E., Camarero, J. J., Gazol, A., Sangüesa-Barreda, G., and doi: 10.1111/1365-2435.12860 Carrer, M. (2016). Wood anatomy and carbon-isotope discrimination Wissel, C. (1984). A universal law of the characteristic return time near thresholds. support long-term hydraulic deterioration as a major cause of drought- Oecologia 65, 101–107. doi: 10.1007/BF00384470 induced dieback. Glob. Change Biol. 22, 2125–2137. doi: 10.1111/gcb. Yin, J., and Bauerle, T. L. (2017). A global analysis of plant recovery performance 13227 from water stress. Oikos 126, 1377–1388. doi: 10.1111/oik.04534 Peters, R. L., Groenendijk, P., Vlam, M., and Zuidema, P. A. (2015). Detecting Zeppel, M. J., Harrison, S. P., Adams, H. D., Kelley, D. I., Li, G., Tissue, D. T., long-term growth trends using tree rings: a critical evaluation of methods. Glob. et al. (2015). Drought and resprouting plants. New Phytol. 206, 583–589. Change Biol. 21, 2040–2054. doi: 10.1111/gcb.12826 doi: 10.1111/nph.13205 Plavcová, L., Hoch, G., Morris, H., Ghiasi, S., and Jansen, S. (2016). The amount of parenchyma and living fibers affects storage of nonstructural carbohydrates Conflict of Interest Statement: The authors declare that the research was in young stems and roots of temperate trees. Am. J. Bot. 103, 603–612. conducted in the absence of any commercial or financial relationships that could doi: 10.3732/ajb.1500489 be construed as a potential conflict of interest. R Core Team (2017). R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing. The handling Editor declared a past co-authorship with the authors JC, PC, and Raffa, K. F., Aukema, B. H., Bentz, B. J., Carroll, A. L., Hicke, J. A., Turner, M. G., KČ. et al. (2008). Cross-scale drivers of natural disturbances prone to anthropogenic amplification: the dynamics of bark beetle eruptions. BioScience 58, 501–517. Copyright © 2019 Cailleret, Dakos, Jansen, Robert, Aakala, Amoroso, Antos, doi: 10.1641/B580607 Bigler, Bugmann, Caccianaga, Camarero, Cherubini, Coyea, Čufar, Das, Davi, Reyer, C. P., Leuzinger, S., Rammig, A., Wolf, A., Bartholomeus, R. P., Bonfante, A., Gea-Izquierdo, Gillner, Haavik, Hartmann, Hereş, Hultine, Janda, Kane, Kharuk, et al. (2013). A plant’s perspective of extremes: terrestrial plant responses to Kitzberger, Klein, Levanic, Linares, Lombardi, Mäkinen, Mészáros, Metsaranta, changing climatic variability. Glob. Change Biol. 19, 75–89. doi: 10.1111/gcb. Oberhuber, Papadopoulos, Petritan, Rohner, Sangüesa-Barreda, Smith, Stan, 12023 Stojanovic, Suarez, Svoboda, Trotsiuk, Villalba, Westwood, Wyckoff and Martínez- Rogers, B. M., Solvik, K., Hogg, E. H., Ju, J., Masek, J. G., Michaelian, M., et al. Vilalta. This is an open-access article distributed under the terms of the Creative (2018). Detecting early warning signals of tree mortality in boreal North Commons Attribution License (CC BY). The use, distribution or reproduction in America using multiscale satellite data. Glob. Change Biol. 24, 2284–2304. other forums is permitted, provided the original author(s) and the copyright owner(s) doi: 10.1111/gcb.14107 are credited and that the original publication in this journal is cited, in accordance Royer, P. D., Cobb, N. S., Clifford, M. J., Huang, C. Y., Breshears, D. D., Adams, with accepted academic practice. No use, distribution or reproduction is permitted H. D., et al. (2011). Extreme climatic event-triggered overstorey vegetation loss which does not comply with these terms. Frontiers in Plant Science | www.frontiersin.org 14 January 2019 | Volume 9 | Article 1964