Volume 36, Issue 11 p. 2806-2818
Open Access

Nonstructural carbohydrates predict survival in saplings of temperate trees under carbon stress

Frida I. Piper

Corresponding Author

Frida I. Piper

Instituto de Ciencias Biológicas (ICB), Universidad de Talca, Talca, Chile

Institute of Ecology and Biodiversity (IEB), Barrio Universitario S/N, Concepción, Chile


Frida I. Piper

Email: [email protected]

Search for more papers by this author
Paulo Moreno-Meynard

Paulo Moreno-Meynard

Centro de Investigación en Ecosistemas de la Patagonia (CIEP), Coyhaique, Chile

Search for more papers by this author
Alex Fajardo

Alex Fajardo

Instituto de Investigación Interdisciplinaria (I3), Vicerrectoría Académica, Universidad de Talca, Talca, Chile

Search for more papers by this author
First published: 10 August 2022
Citations: 3
Handling Editor Tamir Klein


  1. Nonstructural carbohydrates (NSCs) mediate plant survival when the plant's carbon (C) balance is negative, suggesting that NSCs could predict plant survival under C stress. To examine this possibility, we exposed saplings of six temperate tree species to diverse levels of C stress created by the combination of two light conditions (full light availability and deep shade) and two defoliation levels (severe defoliation and nondefoliation). We then measured survival, biomass and total NSCs and soluble sugars (SSs) concentrations in different organs of both dead and live saplings.
  2. We estimated mean NSCs and SSs contents and concentrations per sapling and fitted logistic generalized mixed-effects models to determine if NSCs and SSs predict survival. Using inverse prediction modelling, we also determined whether there is a common NSCs and SSs threshold across species at the time of sapling's death.
  3. Defoliation and shade reduced the mean sapling's NSCs and SSs contents, indicating C stress. Mean sapling NSCs and SSs contents and concentrations predicted survival and the robustness of the models improved with the inclusion of species. At death, saplings of the exotic deciduous tree species Acer pseudoplatanus exhibited significantly lower mean NSCs and SSs contents than saplings of the evergreen conifer species Podocarpus nubigenus and lower stem NSCs and SSs concentrations than the broadleaf evergreen species Drimys winteri.
  4. The energetic role that NSCs and SSs play in plants under C stress was evidenced by the capacity of these compounds to predict sapling survival under C stress. No common threshold of NSCs and SSs contents or concentrations for sapling survival amongst species was found, indicating that the level of these compounds may not be good proxies for interspecific comparisons of tolerance to C stress. Presumably, there are species-specific limits for the mobilization and use of NSCs and SSs in metabolism.
  5. Our results anticipate that the inclusion of NSCs and SSs in modelling will improve predictions regarding tree responses to ongoing climate change. Nonetheless, a better understanding of the many roles that carbohydrates play in plant survival under C stress is required to scale predictions up to the community level.

Read the free Plain Language Summary for this article on the Journal blog.


Nonstructural carbohydrates (NSCs) represent a principal form of carbon (C) storage in plants, particularly in angiosperm trees (Martínez-Vilalta et al., 2016). In most plant species, NSCs are largely made up of starch, a polysaccharide that is osmotically inert, in addition to low molecular weight sugars that are soluble in water (i.e. soluble sugars, SSs) and osmotically active (Chapin et al., 1990; Morgan, 1984). NSCs are a fundamental component of plant fitness because they mediate survival under stress (Piper & Paula, 2020; Signori-Müller et al., 2021) and fuel plant reproduction and growth (Löiez & Piper, 2021). These ecological and biological functions are promoted by several physiological roles that NSCs play in plants, which are critical for adequate plant functioning (Hartmann & Trumbore, 2016). The most recognized physiological role of NSCs is energetic; they fulfil the C demands required by growth and metabolism when current photoassimilation has been reduced by defoliation (Schönbeck et al., 2018), shade (Myers & Kitajima, 2007), reduced CO2 concentrations (Hartmann et al., 2013), or long-lasting droughts (Adams et al., 2013; Piper & Fajardo, 2016). The energetic role is also important for growth resumption in seasonal climates, especially in early spring, when metabolic demands peak after a long dormancy period (Furze et al., 2019; Löiez & Piper, 2021; Martínez-Vilalta et al., 2016; Reyes-Bahamonde et al., 2022). While the energetic role of NSCs is mainly related to starch remobilization, SSs play roles in osmoprotection and osmoregulation. In terms of osmoprotection, SSs can protect the cell structure and its organelles from the formation of ice crystals and scavenge reactive oxygen species under water deficit or freezing temperatures; furthermore, NSCs are also precursors of other molecules with osmoprotective activity (Slama et al., 2015). SSs are the main components responsible for the osmotic potential of cells because they are osmotically active and metabolically inert; they are, hence, responsible for maintaining the cell turgor (Aranda et al., 2020; Sapes et al., 2021; Turner, 2018). Under drought stress, many plants accumulate SSs, which decreases their osmotic potential and promotes water intake, ultimately preventing plasmolysis and tissue desiccation (Aranda et al., 2020). The maintenance of a pool of NSCs ensures the supply of SSs, but also of precursors of other osmotically active molecules, such as amino acids. Accordingly, experiments have shown NSCs depletion in plants to be associated with a reduction in the cell turgor, and also with an increased xylem vulnerability to embolism and a limited recovery capacity (i.e. vessels refilling) after drought (Ravi et al., 2021; Sapes et al., 2021; Sevanto et al., 2014; Tomasella et al., 2021). Due to the crucial roles that NSCs play in plant functioning under C stress, NSCs could predict plant survival under C stress.

The role of NSCs as an indicator and predictor of C limitation in plants has been a subject of great debate in the last decade (Ryan, 2011; Sala et al., 2010; Weber et al., 2019; Wiley & Helliker, 2012). Arguments against the use of NSCs as a predictor of C limitation have been based on the transitory increases of NSCs under drought when C limitation was presumed. However, drought affects the sink strength more than the source strength (Palacio et al., 2014; Zweifel et al., 2021), and it may limit the access and remobilization of C stores (Millard et al., 2007; Sala et al., 2010). Additionally, plant mortality under drought might be the result of hydraulic failure preceding C starvation, implying that at the time of death trees may still have high NSCs levels (Rodríguez-Calcerrada et al., 2017). In contrast to drought, shade and defoliation are two factors that clearly affect the C sources more than the C sinks, causing a condition of C stress that induces a negative C balance in plants and the concomitant use of NSCs (Fischer et al., 2015; Piper & Fajardo, 2014; Piper & Fajardo, 2016). The combination of defoliation and shade has been unequivocally shown to reduce the NSCs levels and cause mortality (Canham et al., 1999; Myers & Kitajima, 2007; Weber et al., 2019), suggesting that under a severe C stress condition the NSCs are a precise indicator of C availability for metabolism. However, it remains unclear whether all NSCs are ultimately available. A global study identified a common minimal NSC concentration of 46% relative to the seasonal maximum across biomes, organs, and plant functional types, suggesting that there might be a threshold of NSCs for plant survival under C stress (Martínez-Vilalta et al., 2016). For example, species-specific differences in osmotic requirements (Bartlett et al., 2012) could, however, argue against such a common threshold.

Recent evidence shows that NSCs predict survival/mortality under C stress, supporting the existence of a critical threshold of NSCs required for survival (Barker Plotkin et al., 2021). Specifically, two winter deciduous Quercus species subjected to different defoliation history and contrasting light conditions exhibited a common mortality threshold of 1.5% dry weight of total NSCs (Barker Plotkin et al., 2021). Although defoliation generally reduces C stores, mortality responses in different species do not necessarily occur at similar NSC values. For example, complete defoliation reduced the NSC concentrations in stems and roots of both an evergreen and deciduous Nothofagus species at similar NSC levels; although the evergreen species showed full mortality while the deciduous species showed full survival (Piper & Fajardo, 2014). Such results could reflect interspecific differences in the minimal level of NSCs necessary to survive. Different minimal NSCs levels required for plant survival were suggested by a study where saplings of four temperate species were exposed to severe and prolonged darkness (Weber et al., 2018). This study found that in all evaluated species NSC concentrations at the time of death were below 1% in most tissues, with the exceptions of last year's needles in Pinus and Picea which retained significantly higher SS (1.6% and 2.4%) and starch (1.6% and 1.1%) concentrations (Weber et al., 2018). Different thresholds of NSCs at the time of sapling's death could be explained by different osmotic requirements. Considering that the constitutive osmotic potential predicts drought tolerance (Bartlett et al., 2012) and that freezing acts as a dehydrating factor on cells (Pearce, 2001), drought- and cold-adapted species are expected to have higher NSC requirements to meet osmotic demands than less adapted species (Bartlett et al., 2012; Sapes et al., 2021). In the same way, it has been hypothesized that higher requirements of NSCs for seedling survival could have been evolutionarily favoured under conditions of recurrent cavitation risk, driven by drought or low temperatures (Sala et al., 2012). High SS requirements could impose limits for metabolic use (Hartmann & Trumbore, 2016; Martínez-Vilalta et al., 2016), determining higher minimal NSCs levels for survival in drought- or cold-adapted species than those of more sensitive counterparts (Sapes et al., 2021). Differences in thresholds of NSCs for survival would indicate that at a given NSCs level, the leeway to support metabolic demands under C stress differs across species, implying that the level of NSCs is not a good predictor of tolerance to severe C limitation.

In the present study, we aimed to determine if the levels of NSCs and SSs under C stress ultimately predict sapling survival, and if so, whether the levels preceding C stress-induced mortality are similar across species. In other words, we asked whether there is a common threshold of NSCs amongst species at the time of saplings' death. To respond to these objectives, we experimentally exposed saplings of six temperate tree species to diverse levels of C stress created by the combination of two light conditions (open condition and severe shade) and two defoliation levels (severe defoliation and nondefoliation). After two growing seasons, we quantified saplings' survival, biomass, and NSCs and SSs concentrations per organ and for the whole-plant. We then estimated the NSCs and SSs contents per organ and for the whole-plant (i.e. the storage pool size), as these have been found to be a more precise indicator of the C budget than concentrations (Myers & Kitajima, 2007; Wiley et al., 2019). Our study included six temperate species that markedly differ in their successional status (early-, mid- and late- successional), and their cold- and drought-tolerance (Table 1). We anticipated that thresholds of NSCs for survival would differ across species, being higher in stress-tolerant species.

TABLE 1. Taxonomical and general ecological characteristics of the species included in this study
Study species Family Leaf habit Shade tolerance Leaf lifespan (years) Cold-tolerance Drought tolerance n for survival
Acer pseudoplatanus L. Sapindaceae Deciduous Avoider 0.5 Low Mid 34
Drimys winteri J.R. Forst. & G. Forst. Winteraceae Evergreen Mid 3.5 Low Low 39
Nothofagus betuloides (Mirb.) Oerst. Nothofagaceae Evergreen Intolerant 2.0 High Low 33
Nothofagus nitida (Phil.) Krasser Nothofagaceae Evergreen Mid 3.5 Mid Low 39
Nothofagus pumilio (Poepp. & Endl.) Krasser Nothofagaceae Deciduous Intolerant 0.5 High High 25
Podocarpus nubigenus Lindl. Podocarpaceae Evergreen Tolerant 7.3 Mid Mid 37
  • Note: Light requirements and leaf lifespan were based on literature (Grubb et al., 2013; Lusk, 2001; Piper et al., 2009; Piper et al., 2019; Vitasse, 2013).


2.1 Experimental design

We conducted an outdoor pot experiment over two growing seasons at the El Mallín nursery (Corporación Nacional Forestal, Conaf), located in Puerto Aysén, southern Chile (45°59′′S, 71°52′′W, 30 m above sea level [a.s.l.]). The climate in this area is temperate and rainy, with a mean annual precipitation of 2,034 mm that is distributed uniformly throughout the year, whereas the mean annual temperature is 8°C. The mean temperature and precipitation of the growing season (October–March) are 11°C and 347 mm, respectively (Puerto Aysén weather station, 32 m a.s.l.; Dirección General de Aguas, 2007–2017).

We studied two winter deciduous species and four evergreen species of contrasting successional status and light requirements. These tree species included five native and one exotic temperate tree species (Table 1). Nothofagus pumilio (Nothofagaceae) is a deciduous, early-successional tree species that dominates the temperate forest in the southern Andes of Chile and Argentina, where it is distributed along a wide elevational range up to the treeline, and from 35° to 55°S (Fajardo & Piper, 2017). Nothofagus dombeyi is an evergreen, early-successional tree species that is also dominant in southern Chilean temperate forests (Donoso 1993). Drimys winteri (Winteraceae) and Nothofagus nitida are two evergreen mid-successional tree species that are typical of the temperate rainforest of southern Chile (Donoso 1993). Finally, Podocarpus nubigenus (Podocarpaceae) is an evergreen conifer, late-successional species and one of the most shade-tolerant species of the temperate rainforest of southern Chile (Grubb et al., 2013). Acer pseudoplatanus (Sapindaceae) was also included in this experiment because it is an exotic tree species that occasionally invades the understory of the temperate rainforest (Fuentes et al., 2020); it is a fast-growing, winter deciduous species with resource-acquisitive leaves that can, nonetheless, thrive in the shade (Lechowicz, 1984) and become a late-successional tree species (Shouman et al., 2017; Vitasse et al., 2014). We consider A. pseudoplatanus to be a good counterexample in our experiment because it is both deciduous and shade-tolerant (typical of flora in the Northern Hemisphere), a combination that is nonexistent in the temperate forests of southern Chile.

In October 2013 (early spring), two-year-old individuals of each species were transplanted to 2 L pots completely filled with a homogenized mixture of commercial substrate (20% organic matter content, 50 carbon: nitrogen [C/N] ratio and a pH of 6). The substrate of each pot was c. 2.0 Kg, a sufficient amount of substrate for root exploration in these overall slow-growing species. However, for Acer pseudoplatanus we used 4 L pots, as this species is notably fast-growing when compared to the rest of the study species. All species' saplings were randomly assigned to four blocks with two light treatments (severe shade and open conditions). The open light condition was represented by full light availability, whereas the severe shade condition was achieved by placing the pots on the ground beneath the sparse shade of Alnus glutinosa trees, and by completely covering the lateral sides and the top with frames of a black Raschel mesh with 80% shading. Each block contained at least 2 individual saplings per species and light condition. The photosynthetic photon flux density (PPFD, mean ± SE, n = 4 blocks) was 888 ± 169 and 85 ± 199 μmol m−2 s−1 for full light and severe shade conditions, respectively. Saplings were regularly watered at field capacity during the growing seasons. By late November 2014, budbreak was completed and all individuals showed new leaves. In order to provoke further C limitation in saplings, by late spring (December 2014), a complete manual defoliation was applied to half of the saplings of each light condition. By late March 2015 (early autumn), all saplings were assessed for survival and subsequently harvested. Saplings were considered dead when they had shed all their leaves or when their leaves were all brown, in addition, the stem was dead (i.e. dry and brown). A total of 128 saplings, including 75 live and 53 dead, were harvested for biomass and nonstructural carbohydrate (NSC) determination.

2.2 Biomass determination

In late April 2015, all of the pots were transported from the nursery to the laboratory (Centro de Investigación en Ecosistemas de la Patagonia, Coyhaique, Chile), where each sapling was separated into roots, stems (including branches and twigs), and leaves. Most defoliated saplings re-foliated after artificial defoliation, except for those of N. pumilio and N. betuloides under shade (Figure S1). The leaf biomass in defoliated saplings under shade was generally very low and insufficient to perform analyses of NSCs in all species except N. nitida (Figure S1). Roots were thoroughly washed with tap water, gently brushed and put into labelled paper bags along with the rest of the saplings' organs. Samples were then heated in a microwave for three 20-s cycles at 900 W in order to stop the enzymatic activity of the living tissues (Popp et al., 1996). After this, samples were dried in a forced-air stove (Memmert, Schwabach, Germany) at 70°C for 72 hr; afterwards, dry tissues were weighed on a scale with a precision of 0.0001 g. Tissue samples were then ground into a fine powder with the help of a ball mill (Retsch®MM200) and finally stored at 4°C until the corresponding chemical analyses were carried out.

2.3 Determination of NSCs

Nonstructural carbohydrates concentrations were determined for each organ as the sum of the three most abundant low-molecular weight SSs (glucose, fructose and sucrose) and starch. NSCs concentrations were analysed following the procedure of Hoch et al. (2002) with some modifications (Piper & Reyes, 2020). About 13 mg of dried sample powder was extracted with 1.6 ml of distilled water at 100°C for 60 min. An aliquot of the extract was used to determine low-molecular weight soluble sugars after enzymatic conversion (invertase and phosphoglucose isomerase from Saccharomyces cerevisiae, Sigma Aldrich I4504 and P5381, respectively) of sucrose and fructose to glucose. The concentration of free glucose was determined photometrically after the enzymatic conversion of glucose to gluconate-6-phosphate (Glucose Assay Reagent, G3293 Sigma Aldrich) in a 96-well multiplate reader. Following the overnight degradation of starch to glucose using a purified fungal amylase (“amiloglucosydase” from Aspergillus niger, Sigma Aldrich 10,115) at 45 °C, NSCs were determined in a separate analysis. The starch concentration was calculated as NSCs minus the sum of free sugars. SSs, starch and NSC concentrations are presented as percent of dry matter. Organ-specific NSCs concentrations were scaled up to the whole sapling using the weighted mean concentrations across all sapling organs, which consider the organ-specific NSCs concentration (% dry matter) and the organ-specific biomass (Hoch et al., 2002; Weber et al., 2018). Additionally, organ-specific NSCs and SSs contents (i.e. pool size, or mass, in g) were calculated per sapling as the product of NSCs or SSs concentration and the biomass of each organ. Organ-specific contents were then added to obtain whole-plant NSCs and SS contents. Finally, the NSCs and SS fraction represented by each organ was estimated for each sapling as the ratio between the organ-specific content and the whole-plant content.

2.4 Statistical analyses

All statistical analyses were conducted in R version 4.2.0 (R Development Core Team, 2022). First, to determine if the light and defoliation treatments provoked C stress across species, we fitted linear mixed-effects models (LMMs) using the R/NLME package (Pinheiro et al., 2021). In particular, we evaluated treatment and species effects on the NSCs and SSs contents and concentrations of each organ and of the whole-sapling, as well as for organ-specific biomasses. The fixed effects of the models were light condition, defoliation level, species, and all of their interactions. The random effect included the blocks. Sapling survival was further analysed using generalized linear mixed-effects models (GLMMs), with a binomial family and a logit link function because the sapling was considered either live or dead, with random effects for blocks and fixed effects for light condition, defoliation level, species, and the interaction amongst them. GLMMs were also used to analyse NSCs fractions because the response variable (e.g. NSCs fraction) is bounded (by 1 above and 0 below). To assess significance of terms for these GLMMs, we used the Akaike information criterion (AIC) corrected for small samples and likelihood-ratio tests (LRTs). This was carried out by comparing the fit of the model with the fixed factor and a model with the random terms only. If the comparison resulted in a p-value <0.05, the model with the fixed factor was deemed significant and the levels have significant differences from one another. When assessing whether the interaction term was significant, the model with the interaction term should show a better fit (p < 0.05) than the model with the additive terms, which was diagnosed using LRTs. NSCs and SS contents and concentrations were compared between live and dead saplings using LMMs, where the block was the random factor and the survival, the species, and the interaction between them were the fixed factors.

To determine whether the levels of NSCs and SSs predict sapling survival across species, we fitted mixed-effects binary logistic regression models, that is, logistic GLMM using the R/LME4 package v1.1–27 (Bates et al., 2015), considering the NSCs or SSs contents or concentrations for each organ or for the whole-sapling, with species as fixed factors, the block as a random factor, and the survival response as the binary response variable (Barker Plotkin et al., 2021). To determine if the means were significantly different from one another, we performed posthoc comparisons with a Tukey's honest significance test, and a α of 0.05. For all models, we assessed model fit by examining standardized residuals using the R/DHARMA package (Hartig, 2020). Finally, to determine if the NSCs and SSs levels predicting survival under C stress were similar across species, we estimated the contents and concentrations and their respective confidence intervals for 1%, 25%, 50%, 75% and 100% survival by inverse prediction, using the R/GGEFFECTS package (Lüdecke, 2018). For NSCs, the model structure was as follows:
Survival ~ 1 + species + NSCs + 1 block ,
where Survival is the response variable expressed as either 1, live sapling, or 0, dead sapling; species and NSCs (or SSs) correspond to the fixed factors, whereas block is the random effect. These analyses were performed at both the organ- and whole-sapling level.


3.1 Species, light and defoliation effects on NSCs and SSs

Shade significantly reduced the NSCs contents of each organ and the whole-sapling, as well as the NSCs concentrations of leaves and stems (Table 2, Table S1; Figure 1). Similarly, defoliation significantly reduced the NSCs contents of each organ and the whole-sapling, in addition to the NSCs concentrations of stems, roots and whole-saplings (Table 2, Table S1; Figure 1, Figures S2–S4). The decrease of NSCs contents and concentrations driven by defoliation was similar in the two light conditions (i.e. no significant interaction was observed between light and defoliation, Table 2, Table S1). As a result, the most severe reduction in NSCs was caused by the combination of shade and defoliation (Figure 1, Figure S2). Whole-sapling NSC contents and concentrations were also different amongst species, an effect largely driven by A. pseudoplatanus in the open condition (Table 2, Table S2; Figure 1, Figure S2), for which NSC contents were c. 10-fold higher than those of the other species. All species responded similarly to defoliation in terms of NSC contents and concentrations (i.e. a nonsignificant interaction was found between species and defoliation, Table 2, Table S1). However, species responded differently to light in terms of NSC contents, with A. pseudoplatanus as the only species showing a significant decrease in NSC content in response to shade (Table 2, Figure 1). Although the interaction amongst light, defoliation, and species was not significant (Table 2), in P. nubigenus and D. winteri whole-sapling NSC contents and concentrations were more affected by defoliation than shade (Figure 1, Figure S2). Conversely, in A. pseudoplatanus, the NSC contents and concentrations were more severely reduced by shade than defoliation (Figure 1, Figure S2). These contrasting responses to defoliation were associated with interspecific differences in the fraction of NSCs represented by each organ; while the greatest fraction of NSCs in P. nubigenus and D. winteri was found in leaves, the greatest fraction of NSCs in A. pseudoplatanus was found in stems and roots (i.e. the species factor significantly affected the fraction of NSCs in leaves and stems, Table S2, Figure S5).

TABLE 2. Results of the generalized (GLMM) and linear (LMM) mixed-effects models testing the effects of the light environment, species, defoliation and their interactions, on the survival and the whole-sapling nonstructural carbohydrate (NSCs) and soluble sugars (SSs) contents following defoliation in saplings of five species of the temperate forest of southern Chile and one exotic temperate species, Acer pseudoplatanus, under two experimental light conditions
Model effect χ 2 AIC R 2 p-value
Light 9.389 280.85 0.067 0.002
Defoliation 55.603 234.63 0.284 <0.001
Species 14.425 283.81 0.110 0.013
Light * defoliation 1.062 220.15 0.430 0.303
Light * species 24.001 261.43 0.797 <0.001
Defoliation * species 20.786 215.63 0.927 <0.001
Light * defoliation * species 46.421 194.80 0.985 <0.001
F Log-like R 2 p-value
−226.837 0.369
Light 8.526 0.004
Defoliation 44.86 <0.001
Species 5.883 0.017
Light * defoliation 2.105 0.084
Light * species 7.732 <0.001
Defoliation * species 0.37 0.611
Light * defoliation * species 0.067 0.800
−255.116 0.293
Light 0.894 0.346
Defoliation 40.339 <0.001
Species 2.294 0.133
Light * defoliation 0.948 0.332
Light * species 4.17 0.043
Defoliation * species 0.377 0.54
Light * defoliation * species 0.208 0.649
Details are in the caption following the image
Mean contents of nonstructural carbohydrate (NSC) and soluble sugar (SS) in saplings of six temperate tree species (Table 1) exposed to open and shade conditions and subjected to two levels of defoliation. Data are missing for defoliated saplings of Nothofagus betuloides in shade due to insufficient plant material to analyse NSCs. Notice the different y-axis scales for each light condition.

Variations in SSs amongst species and treatments mirrored those found for NSCs (Figure 1, Figure S2), although the effects of species and light on the whole-sapling SSs contents were not significant (Table 2, Table S1). Defoliation significantly reduced whole-sapling, root, and stem SSs contents and concentrations, but increased leaf SSs concentrations (Table 2, Table S1; Figure 1, Figures S2, S6 and S7). Species differed in their whole-sapling SSs concentrations but not contents (Table 2, Table S1) and responded differently to light (but not to defoliation) in terms of their SSs contents and concentrations (i.e. significant species * light interaction, Table 2, Table S1).

3.2 Species, light and defoliation effects on survival

As expected, sapling survival was affected by the level of defoliation and the light conditions, varying significantly across species (Table 2). Mortality was significantly higher in defoliated than undefoliated saplings, and in shade than in open conditions (Table 2, Figure 2). The effects of light varied across species (i.e. significant interaction light*species) in accordance with the differences in shade tolerance (Table 2, Figure 2). Although defoliation caused high mortality in all species under shade, deciduous species exhibited a significantly higher defoliation tolerance than evergreen species under open conditions (Figure 2). Additionally, not all of the species were affected by shade in the absence of defoliation. For example, control saplings of D. winteri and P. nubigenus had no mortality regardless of the light condition (Figure 2).

Details are in the caption following the image
Saplings' survival proportions of five tree species belonging to the temperate forest of southern Chile and one exotic temperate species (Acer pseudoplatanus) after 17 months of experimental exposition to two light conditions (open canopy and shade), and following two defoliation levels (control and defoliated) that were applied at the beginning of the second growing season, that is, 3 months before harvest.

3.3 Whole-sapling NSCs and SSs in live and dead saplings

Across defoliation and light treatments, surviving saplings had significantly higher NSCs and SSs contents and concentrations than dead saplings (Figure 3, Figure S8). Differences in whole-sapling NSCs and SSs contents and concentrations between live and dead saplings were species-specific, with some species showing similar values between dead and live sapling (i.e. significant interaction between species and survival, Figure 3, Figure S8). Notably, species with higher NSC contents or concentrations in live saplings were not necessarily more tolerant to C stress. For example, A. pseudoplatanus and D. winteri exhibited the lowest mortality under shade and defoliation, but they had highly contrasting NSCs and SSs contents (Figures 2 and 3). Similarly, mean whole-sapling NSC concentrations of live saplings were significantly higher in N. pumilio than in D. winteri; however, survival under C stress was significantly lower in the former (Figure 2, Figure S8).

Details are in the caption following the image
Mean sapling content of nonstructural carbohydrates (NSCs) and soluble sugars (SSs) in five tree species belonging to the temperate forest of southern Chile and one exotic temperate species (Acer pseudoplatanus) after 17 months of experimental exposition to two light conditions (open canopy and under shade), and following two defoliation levels (control and defoliated) that were applied at the beginning of the second growing season, that is, 3 months before harvest. Insets show F-ratios and statistical significance for the effects of survival, species, and the interaction between them. The lack of asterisks indicates a nonsignificant effect of the factor, while *, and ** indicate significant effects at p < 0.05 and p < 0.01, respectively. The asterisk between bars stands for the only significant difference in NSCs between live and dead saplings within species. Data correspond to both live and dead saplings.

3.4 Predictors of sapling survival

Logistic GLMMs rendered highly significant and positive effects of the mean whole-sapling and organ-specific NSCs and SSs contents and concentrations on sapling survival (Table 3, Tables S2–S4). Models indicated that the greater the NSCs and SSs contents or concentrations, the lower the mortality under C stress. The statistical robustness of the models was generally higher when they included the factor “species” (Table S5). In all of these models species was also a significant predictor of survival, indicating that the survival probability differed depending on the species. The multiple comparisons of the average survival of species showed significantly higher survival in A. pseudoplatanus than in P. nubigenus for models of the mean whole-sapling, stem, and root NSCs and SSs contents (Table 3, Table S3). Minimal thresholds of NSCs and SSs contents for sapling survival also showed significant interspecific differences between A. pseudoplatanus and P. nubigenus (Figures 4 and 5). For example, for 75% of survival, mean whole-sapling NSCs contents were 0.007 g in A. pseudoplatanus and 0.033 g in P. nubigenus, while mean whole-sapling SSs contents were 0.005 and 0.023 in A. pseudoplatanus and P. nubigenus, respectively (Figures 4 and 5). Similarly, at 0.02 g of whole-sapling NSCs contents, survival was c. 94% in A. pseudoplatanus, but only c. 29% in P. nubigenus, (Figure 4). In the models of mean whole-sapling and stem NSCs concentrations, and also in the model of stem SSs concentrations, survival was significantly higher in A. pseudoplatanus than D. winteri (i.e, Tables S2 and S4). Accordingly, minimal thresholds of NSCs and SS concentrations for sapling survival were significantly lower in A. pseudoplatanus than D. winteri (Figures S9–S11). For example, mean whole-sapling NSCs and SS concentrations for 50% survival were 0.30% and 0.25% for A. pseudoplatanus, but 2.21% and 1.40% for D. winteri (Figures S9 and S10).

TABLE 3. Probabilistic statistics, log-likelihood (logLik), and fixed-effects results from the logistic generalized linear mixed-effects (GLMM) models predicting sapling survival under carbon stress caused by shade and defoliation in six temperate tree species (Table 1) as a function of the species (SP) and the whole-sapling content (in g) of nonstructural carbohydrates (NSCs) and soluble sugars (SSs). DrWi: Drimys winteri, NoBe: Nothofagus betuloides, NoNi: N. nitida, NoPu: N. pumilio, PoNu: Podocarpus nubigenus. AIC stands for Akaike Information Criterion index; BIC stands for Bayesian Information Criterion index; LL and UL stand for the parameters lower and upper confidence intervals. Asteriks stand for species that showed statistical significance
Survival AIC BIC logLik Parameter Estimate Std. Error LL UL Pr(>|z|)
~ 1 + SP + NSCg + (1|block) 95.4 118.2 −39.7 (Intercept) 13.8436 2.9549 8.052054 19.63513496 2.80E-06 ***
speciesDrWi −2.3754 1.0047 −4.344537 −0.40623291 0.018062 *
speciesNoBe −1.7982 1.1747 −4.100663 0.50426856 0.125835
speciesNoNi −1.3852 0.8904 −3.130336 0.36002199 0.119789
speciesNoPu −3.6764 1.9023 −7.405022 0.05213187 0.053286
speciesPoNu −4.1723 1.2448 −6.611998 −1.73257513 0.000803 ***
NSCgr 159.3981 34.1633 92.438095 226.3581372 3.07E-06 ***
~ 1 + SP + SSg + (1|block) 110.8 133.6 −47.4 (Intercept) 7.0749 1.4177 4.296261 9.8534884 6.02E-07 ***
speciesDrWi −1.6436 0.9045 −3.416547 0.1292521 0.069201
speciesNoBe −1.21 1.0644 −3.296185 0.8761883 0.255618
speciesNoNi −1.4279 0.881 −3.154727 0.2988504 0.105063
speciesNoPu −1.6861 1.1946 −4.027469 0.6553182 0.158119
speciesPoNu −3.4584 1.0456 −5.507786 −1.4089696 0.000941 ***
SSgr 194.5696 37.084 121.884918 267.2543198 1.55E-07 ***
Details are in the caption following the image
Mean marginal predicted probabilities of survival under carbon stress caused by shade and defoliation in six temperate tree species (Table 1) as a function of whole-sapling nonstructural carbohydrates (NSCs) contents, estimated by inverse prediction. The shaded area around the line indicates the 95% confidence intervals.
Details are in the caption following the image
Mean marginal predicted probabilities of survival under carbon stress caused by shade and defoliation in six temperate tree species (Table 1) as a function of mean sapling soluble sugars (SSs) contents, estimated by inverse prediction. The shaded area around the line indicates the 95% confidence intervals.


Our study reveals that NSCs and SSs levels play a significant role in the survival of tree saplings under conditions of C stress driven by shade and defoliation. For all evaluated species, we found that sapling survival under C stress was significantly and consistently predicted by NSCs and SSs contents and concentrations; the higher the NSCs and SSs, the higher the probability of sapling survival. This implies that NSCs are a robust indicator of C stress. The positive relationship found between NSCs and SSs levels and tree sapling survival under C stress demonstrates the ecological importance of the energetic role of NSCs (Chapin et al., 1990; Hoch, 2015). According to this role, NSCs act as a buffer against C stress during periods of negative C balance (i.e. a carbon safeguard, Sapes et al., 2021), permitting a longer period of survival (Piper & Paula, 2020). Thus, the greater survival proportion exhibited by the two winter deciduous species included in this study (A. pseudoplatanus and N. pumilio) in response to defoliation under open conditions can be explained by their greater NSCs and SS concentrations under this condition. Likewise, the high survival of nondefoliated saplings of D. winteri, P. nubigenus and A. pseudoplatanus exposed to shade can be explained by the high NSCs and SSs contents exhibited by these species under this condition. The identification of NSCs and SSs levels as predictors of sapling survival under C stress supports the need to consider these physiological parameters in modelling efforts aimed at predicting tree and forest responses to climate change (Hartmann et al., 2018; Tague et al., 2013).

In addition to NSCs and SSs, species also appeared as a significant factor predicting sapling survival under C stress; that is, not all species responded similarly to shade and defoliation in terms of survival. Provided the interspecific differences in geographic distribution (five native, one exotic), phylogenies (one conifer, five broadleaved), leaf habits (two deciduous, four evergreens), and successional status (two late-, two mid-successional, and two pioneer species), this is not a surprising result. The significant effect of species as a predictor of survival indicates that mortality under C stress cannot solely be explained by the dynamics of NSCs and SSs; that is, species exhibiting the lowest NSCs and SS levels were not necessarily more vulnerable than those with the highest NSC and SS levels. Indeed, P. nubigenus had the highest mean whole-sapling NSCs and SSs contents amongst the native species and, at the same time, it exhibited one of the highest mortalities in response to defoliation. The high mortality of P. nubigenus in response to defoliation in both open and shade conditions is clearly not explained by a reduction in NSCs and SSs levels, as other species achieved similar contents and concentrations but exhibited a lower mortality (e.g. N. nitida). The high NSCs concentrations at the time of death in P. nubigenus, along with its high mortality following defoliation, could be explained by a greater reliance on C reserves other than NSCs for remobilization under C stress, for example, lipids, which were not accounted for in our study and appear to be more abundant in conifers than angiosperms (Hoch et al., 2003). This potential explanation is supported by an experiment showing that saplings of Pinus sylvestris, a conifer species similar to P. nubigenus, switched the substrate for respiration from carbohydrates to lipids over 18 days subjected to shade (Fischer et al., 2015). Additionally, the significant effect of species as a predictor of survival can be explained by interspecific differences in the minimal levels of NSCs and SSs required to survive.

The thresholds of NSCs and SSs contents for survival under C stress were different amongst species, particularly between A. pseudoplatanus and P. nubigenus, as indicated by the lack of overlap in the confidence intervals estimated by the inverse prediction (Figures 4 and 5). Specifically, our results show that a similar NSCs content confers significantly higher survival against C stress in A. pseudoplatanus than in P. nubigenus. Likewise, thresholds of NSCs and SSs concentrations for survival were significantly higher in A. pseudoplatanus than in D. winteri. The consequence of this result is that species with greater NSCs contents or concentrations may not necessarily account for a greater availability of NSCs to meet metabolic demands under C stress relative to species with lower NSCs contents or concentrations. Since NSCs and SSs predicted survival across species, but the NSCs and SSs thresholds for plant survival differed amongst some species, the robustness of NSCs and SSs levels to predict survival/mortality seems to be higher within species than between species. Although important, this conclusion must be considered in the context of the singular assessment of NSCs of our study since the NSC contents and concentrations vary throughout the season (Martínez-Vilalta et al., 2016); thus, it remains possible that the minimal NSC level required for survival varies as well.

Higher thresholds of NSCs and SSs for survival in the two evergreen species (P. nubigenus and D. winteri) relative to A. pseudoplatanus might be explained by differences amongst these species in the importance of leaves for storage. Podocarpus nubigenus and Drimys winteri showed the highest leaf NSCs concentrations and leaf NSCs fraction (Figures S3–S5). As a result, defoliation led to a substantial reduction in the NSCs pool of these species (Figure 1, Figure S5). This was not the case in A. pseudoplatanus. Accordingly, the survival difference amongst these three species was caused by defoliation, not by shade (Figure 2). Leaf NSCs are the main source of NSCs remobilization in evergreen conifers (Hoch et al., 2003), as in some evergreen broadleaf species (Martínez-Vilalta et al., 2016). Thus, we suggest that the higher thresholds of NSCs for survival of the two evergreen species relative to A. pseudoplatanus were driven by the elimination (through defoliation) of leaves, the main source of NSCs remobilization. Although our research design did not allow us to determine whether the thresholds of NSCs for survival differed between shade and defoliation treatments, our results could imply that interspecific differences in thresholds of NSCs for survival could be stress-type-dependent. In accordance, interspecific differences in the thresholds of NSCs contents and concentrations for survival under C stress could be explained by interspecific differences in the capacity to remobilize NSCs from woody organs (Millard et al., 2007).

Our results showed that D. winteri died with greater stem NSCs concentrations than A. pseudoplatanus, and that P. nubigenus died with greater stem and root NSCs contents than A. pseudoplatanus. We suggest that these results are related to the ecological status of the species. Both P. nubigenus and D. winteri are evergreen species of the cold-temperate rainforest, typically slow-growing and shade-tolerant species, as such, they are expected to have a low tissue turnover and a high tissue longevity (Lusk, 2002). Indeed, leaf longevity is 7.5 years in P. nubigenus (Lusk, 2001) and 3.5 years in D. winteri (Lusk & Contreras, 1999). In contrast, A. pseudoplatanus is a fast-growing deciduous that avoids shade by flushing earlier than conspecific adults and gaining sufficient C to grow in the understory once the canopy closes, after which it overwinters (Vitasse, 2013). Acer pseudoplatanus is also an invasive species (Fuentes et al., 2014), which exhibits faster growth in invasive than in native ranges (Shouman et al., 2017). It has been shown that both growth and storage are under selection and compete with one another (Blumstein et al., 2022). Allocation to storage has also sometimes proven to have priority over allocation to growth (Reyes-Bahamonde et al., 2021). The advantages of NSCs for plant survival under stress conditions might be related to the presence of starch as a safeguard (metabolic role), but also to osmoregulation (Martínez-Vilalta et al., 2016), provided that both NSCs and SSs promote the maintenance of turgor and hydraulic conductivity (Sapes et al., 2021; Sevanto et al., 2014; Tomasella et al., 2021). To be durable, tissues of slow-growing, stress-adapted species may need to be able to face environmental fluctuations in water availability and vapour pressure that could cause dehydration. If so, the NSCs fraction allocated to osmoregulation should be proportionally greater in stress tolerant than in fast-growing species. This could account for the lower access to pools of NSCs for remobilization in woody organs of D. winteri and P. nubigenus in comparison to A. pseudoplatanus. On the contrary, winter deciduous species often shed their leaves when exposed to drought (Marchin et al., 2010), a response that has been specifically found in A. pseudoplatanus (Piper & Fajardo, 2016). On the other hand, the aggressive invasive character of some deciduous temperate species, like A. pseudoplatanus, has been linked to the capacity to store and remobilize starch in roots to grow opportunistically (Hinman & Fridley, 2018). Our results suggest that a lower threshold of NSCs for survival contributes to the fast growth of A. pseudoplatanus, along with its high storage capacity, which determines a greater leeway for the use of NSCs in growth and metabolism.


This study shows that NSCs and SSs levels predict sapling survival under C stress, thus demonstrating the role of these compounds as energetic buffers. The fact that the thresholds of NSCs and SSs for survival were ultimately species-specific indicates a lack of a common threshold for sapling survival amongst species. This result calls into question the use of the NSCs level as a proxy of C availability to tolerate C stress in interspecific comparisons. Species accounting for higher NSCs contents or concentrations will not unequivocally exhibit higher survival under conditions of C stress than species with lower NSCs levels. We suggest that interspecific differences in the thresholds of NSCs and SSs for survival are related to plant functions that are critical for the plant survival, and that require NSCs and SSs as such (i.e. without being consumed), like osmoregulation and osmoprotection (Martínez-Vilalta et al., 2016; Sapes et al., 2021). These functions can therefore be expected to limit the C availability for metabolism. Considering the great plant diversity in the capacity of osmoregulation, the demand of NSCs and SSs for these functions is likely species-specific. Finally, the fact that high NSC and SS contents and concentrations were associated with significantly higher survival implies that NSCs could be a trait under natural selection (Blumstein et al., 2022).


Frida I. Piper conceived the ideas and designed methodology; Frida I. Piper and Alex Fajardo set the experiment and collected the data; Alex Fajardo and Paulo Moreno-Meynard analysed the data; Frida I. Piper led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication.


The authors are indebted to Jonathan Riquelme for field and laboratory assistance. P.M.M. thanks the financial support of Programa Regional ANID R17A10002, and R20F0002.


    The authors have no conflicts of interest to declare.


    Data available from the Dryad Digital Repository, https://doi.org/10.5061/dryad.zw3r228bd, (Piper et al., 2022).