Topographic Heterogeneity Drives the Functional Traits and Stoichiometry of Abies georgei var. smithii Bark in the Sygera Mountains, Southeast Tibet

Abstract

Bark is a multifunctional organ critical for tree survival, yet its functional plasticity in response to micro-environmental heterogeneity at alpine timberlines remains poorly understood. Here, we investigated the variations in bark physical traits (thickness, density), allometric scaling, and stoichiometric characteristics (C, N, P) of Abies georgei var. smithii (Viguie & Gaussen) W. C. Cheng & L. K. Fu on contrasting sunny and shady slopes in the Sygera Mountains, southeastern Tibetan Plateau. Despite the relative homogeneity of soil physicochemical properties between slope aspects, bark traits exhibited remarkable phenotypic plasticity. Trees on the shady slope possessed significantly thicker bark with higher nitrogen concentrations, adopting a “resource-acquisitive strategy”. Standardized Major Axis (SMA) regression indicated isometric scaling ($b\approx 1.03$) for trees on the shady slope, reflecting a sustained investment in bark thickness to provide thermal insulation against cold stress. Conversely, trees on the sunny slope exhibited negative allometry ($b \approx 0.87$), characterized by denser tissues and elevated C/N ratios. This shift represents a conservative strategy geared toward hydraulic safety and resistance to high radiation and evaporative loss. Crucially, our results show that bark traits are largely decoupled from soil nutrient gradients, being shaped instead by microclimate. The distinct trade-off—prioritizing insulation on shady slopes versus conservation on sunny slopes—underscores the importance of phenotypic plasticity for the persistence of timberline species in a changing climate.

1. Introduction

Plant functional traits serve as useful proxies for understanding how organisms obtain, make use of, and conserve resources using environmental filters [1]. While the “Leaf Economics Spectrum” (LES), which reveals a fundamental trade-off between resource acquisition and conservation, is extensively documented around the globe [2], barks as multifunctional organs constituting ~10%–20% of stem biomass are still less researched when compared with leaves, roots, and woods [3]. Bark provides not only protection but also facilitates movement of the photosynthates, mechanical support, pathogen defense, and water and nonstructural carbon [4]. Recently, the idea of a “Bark Economics Spectrum” has been proposed, suggesting that bark traits coordinate to mediate trade-offs between protection (thick, dense bark) and metabolic cost (thin, low-density bark) [5]. But we know little about how the same kind of bark can differ within the same species across heterogeneous micro-environments.

The Alpine timberline ecotone is one of the world’s most stressful boundaries on Earth. The alpine timberline ecotone is characterized by low temperatures, a short growing season, and strong solar radiation [6]. Under such extreme environments, the ability of phenotypic plasticity, a genotype can produce different phenotypes according to environmental conditions, which is very important to trees’ survival and persistence [7]. Topography creates distinct microclimates over short spatial scales. South-facing slopes generally receive higher solar irradiance and evaporative demand, resulting in larger daily temperature fluctuations, while north-facing slopes remain cooler and retain snowpack longer [8]. While previous studies have focused on how slope aspect affects leaf photosynthetic traits or xylogenesis [9], few have examined whether bark functional traits exhibit similar plastic responses to these microclimatic contrasts, or whether such plasticity is constrained by soil nutrient availability.

Furthermore, characterizing bark investment requires more than static morphological measurements; it necessitates an understanding of ontogenetic scaling and stoichiometric allocation. Allometric scaling relationships describe how trees prioritize resource allocation as they grow [10]. While early studies often assumed species-specific fixed scaling exponents, recent evidence suggests that plants may alter their allometric trajectories (i.e., slope b) to optimize fitness under stress (Optimal Partitioning Theory) [11]. Ecological stoichiometry (C/N/P ratios) provides a chemical method to evaluate the costs and benefits of building plant tissues. The Growth Rate Hypothesis (GRH) states that fast-growing tissues are usually rich in nitrogen and phosphorus, while stress-tolerant tissues are rich in carbon due to the accumulation of protective compounds [12]. Combining allometric scaling with stoichiometric analysis in bark ecology helps us understand how trees at the timberline balance between growing quickly and maintaining structure over time.

A key question is whether soil factors or microclimate conditions have a stronger influence on bark functional traits. The “soil–plant continuum” concept suggests that the nutrient content in plant tissues closely matches the nutrient availability in the soil [13]. However, in the stressful environment of the alpine timberline, climate may have a greater impact than soil, leading to a separation between soil nutrients and plant nutrient traits [14]. It is important to determine whether bark traits simply reflect soil nutrient changes or are adjusted by the plant to cope with small-scale climate variations. This understanding will help predict how trees will respond to future environmental changes. Regions like the Tibetan Plateau are of particular concern because they are warming much faster than the global average [15].

Abies georgei var. smithii (Viguie & Gaussen) W. C. Cheng & L. K. Fu (taxonomically treated by some authors as a synonym of Abies forrestii var. smithii [16]) is a major treeline species on the southeastern Tibetan Plateau and is important for the stability of the alpine ecosystem. In this study, we examined the differences in bark functional traits (thickness, density), allometric scaling, and stoichiometry (C, N, P) of Abies georgei var. smithii on contrasting sunny and shady slopes in the Sygera Mountains. We proposed the following hypotheses:

(1)

Trees on the shady slope will develop thicker bark to improve insulation against low temperatures, while trees on the sunny slope will develop denser bark to reduce water loss and resist radiation stress.

(2)

The allometric scaling relationship for bark thickness is not fixed but flexible; specifically, we expect isometric scaling ($b\approx 1.0$) on shady slopes for thermal maintenance and allometric scaling ($b<1.0$) on sunny slopes for hydraulic safety.

(3)

Bark stoichiometric traits will change from a resource-acquisitive pattern (high nitrogen) on the shady slope to a conservative/defensive pattern (high C/N ratio) on the sunny slope, primarily due to microclimatic differences rather than soil nutrients.

2. Materials and Methods

2.1. Study Area and Sampling

Field sampling was conducted in August 2024. The study sites were established within the elevation range of 3600–4200 m on both sunny and shady slopes of the Sygera Mountains [17]. To minimize the confounding effects of micro-topography, sampling plots were selected in areas with consistent slope gradients slope variation < 5°. We established four standard plots per aspect, totaling eight plots across the study area along the altitudinal gradient. Within each plot, 27 healthy adult individuals of Abies georgei var. smithii (free of visible pests or diseases) were selected as standard sample trees (Figure 1; see also Figure S1). In total, 216 trees were sampled for this study (108 individuals per slope aspect $\times 2$ aspects). Bark samples were collected at breast height (1.3 m above ground) to minimize the confounding effects of root buttressing and basal swelling, ensuring comparability with standard functional trait protocols Samples were processed to determine total bark thickness (TBT), bark density, and stoichiometric traits. Concurrently, topsoil (0–20 cm) was collected within the canopy projection area of each sample tree, mixed thoroughly, and air-dried in the laboratory [18].

2.2. Measurements

Bark thickness was measured using a digital vernier caliper (0.01 mm precision), and bark density was determined using the water displacement method [19]. Total carbon (TC) and nitrogen (TN) contents of soil and bark samples were determined using an elemental analyzer (Vario EL III, Elementar, Langenselbold, Germany). Total phosphorus (TP) was determined by the molybdenum-antimony anti-colorimetric method after digestion with H₂SO₄-HClO₄ [20].

2.3. Statistical Analysis

All data processing, statistical analyses, and visualization were performed using Python 3.9 with Pandas (v. 1.3.5), SciPy (v. 1.7.3), Matplotlib (v. 3.5.1), and Seaborn (v. 0.11.2) libraries [21].

To avoid pseudoreplication arising from the nested sampling design (trees within plots), all statistical comparisons of bark traits and soil properties between slope aspects were performed using plot-level means (n = 8; 4 plots per aspect), rather than individual tree measurements. Data were checked for normality (Shapiro–Wilk test) and homogeneity of variance (Levene’s test) prior to analysis. Independent samples t-tests were used to compare differences between sunny and shady slopes. Although boxplots visualize the full variability of individual tree measurements (n = 216), the reported p-values are strictly based on the robust plot-level analysis.

To quantify the magnitude of trait differences between slope aspects, Cohen’s d effect size was calculated. Following Cohen (1988) [22], effect sizes were interpreted as negligible $\left(\right|d| < 0.2)$, small $(0.2 \le |d| < 0.5)$, medium $(0.5\le |d| < 0.8)$, and large $\left(\right|d| \ge 0.8)$.

Pearson correlation analysis was used to examine the coupling between soil and bark nutrient contents based on plot-level means. To assess nutrient limitation, a one-sample t-test was performed to compare bark N/P ratios against the critical threshold value of 14 [23].

The allometric relationship between bark thickness $(Y, TBT)$ and diameter at breast height $(X, DBH)$ was described using the power-law equation ($Y=\beta X^{\alpha }$). Data were log-transformed into a linear form for fitting:

$$\mathrm{l}\mathrm{n}\left(Y\right)=\mathrm{l}\mathrm{n}\left(\beta \right)+\alpha \mathrm{l}\mathrm{n}\left(X\right)$$

(1)

Since both the independent variable (DBH) and dependent variable (bark thickness) are subject to natural variation and measurement error, Ordinary Least Squares (OLS) regression is inappropriate as it assumes error only in Y. Therefore, Standardized Major Axis (SMA) regression was employed to correctly estimate the allometric exponent ($slope \alpha$) and intercept [24]. The heterogeneity of SMA slopes between sunny and shady aspects was tested using permutation tests (n = 2000) to determine if the allometric strategies differed significantly.

2.4. AI Tool Usage

During the preparation of this manuscript, the authors used ChatGPT 4.0 (OpenAI, San Francisco, CA, USA) for language polishing and proofreading. The authors have reviewed and edited the output and take full responsibility for the content of this publication.

3. Results

3.1. Variations in Soil Physicochemical Properties

The study plots were established across comparable altitudinal gradients on both slope aspects (Sunny: 3696–4200 m; Shady: 3680–4200 m;

Table 1. Physicochemical properties of surface soil (0–20 cm) on sunny and shady slopes of the Sygera Mountains. Values are mean ± SD. p-values indicate significance from independent samples t-tests.

Variable	Sunny Slope	Shady Slope	p-Value
Altitude Range (m)	3696–4200	3680–4200	-
Soil WC (%)	64.31 ± 21.13	54.41 ± 20.59	0.527
Soil TN (g/kg)	3.37 ± 0.62	2.94 ± 0.84	0.446
Soil TP (g/kg)	0.63 ± 0.09	0.54 ± 0.13	0.302
Soil N/P	5.34 ± 0.64	5.97 ± 3.32	0.733

). Independent samples t-tests based on plot-level means (n = 8) revealed no significant differences in surface soil (0–20 cm) physicochemical properties between the sunny and shady slopes, indicating a homogeneous edaphic background across the timberline ecotone (Table 1, Figure 2).

Sunny slope’s soil water content (SWC) was $(64.31\pm 21.13\%)$ numerically more but statistically equal to shady slope’s ($54.41 \pm 20.59\%; p = 0.527$; Figure 2a). And soil nutrients also did not show obvious topographic variation. The soil TN was higher on the sunny slope, $3.37 \pm 0.62 \mathrm{g}/\mathrm{k}\mathrm{g}$, compared to $2.94\pm 0.84 \mathrm{g}/\mathrm{k}\mathrm{g}$ on the shady slope (p = 0.446; Figure 2d). Soil total phosphorus (TP) followed a similar pattern, with values of $0.63\pm 0.09 \mathrm{g}/\mathrm{k}\mathrm{g}$ on the sunny slope and $0.54\pm 0.13 \mathrm{g}/\mathrm{k}\mathrm{g}$ on the shady slope ($p = 0.302$; Figure 2e). Furthermore, other measured soil properties, including pH and soil organic matter (SOM), showed no significant variation between slope aspects ($p > 0.05$; Figure 2b,c). Regarding stoichiometric traits, the soil N/P ratio was $5.34\pm 0.64$ on the sunny slope and $5.97\pm 3.32$ on the shady slope, with no significant difference observed ($p = 0.733$). These results confirm that soil water availability and nutrient status were relatively uniform across the contrasting slopes.

3.2. Variations in Bark Physical and Stoichiometric Traits

Bark functional traits exhibited significant phenotypic plasticity in response to slope aspect (

Table 2. Comparison of bark functional traits and stoichiometric characteristics between sunny and shady slopes. Effect size ($Cohen\prime s d$) indicates the magnitude of the difference.

Trait	Sunny Slope (Mean ± SD)	Shady Slope (Mean ± SD)	p-Value	Cohen’s d
Total Bark Thickness (mm)	15.83 ± 9.02	21.19 ± 14.76	0.002	−0.44
Relative Bark Thickness	0.39 ± 0.14	0.49 ± 0.21	<0.001	−0.60
Bark Density (g/cm3)	0.39 ± 0.08	0.34 ± 0.08	<0.001	0.59
Bark Water Content (%)	0.96 ± 0.03	0.97 ± 0.03	0.079	−0.24
Bark Total C (g/kg)	505.95 ± 15.57	505.61 ± 28.13	0.912	0.02
Bark Total N (g/kg)	2.93 ± 1.99	3.98 ± 0.81	<0.001	−0.69
Bark Total P (g/kg)	0.57 ± 0.20	0.47 ± 0.19	<0.001	0.52
Bark C/N Ratio	194.82 ± 50.99	132.15 ± 26.55	<0.001	1.54
Bark N/P Ratio	5.50 ± 2.93	9.45 ± 3.25	<0.001	−1.28

, Figure 3). Trees on the shady slope invested more in bark thickness compared to those on the sunny slope. Specifically, as shown in Figure 3a, the total bark thickness (TBT) on the shady slope was $21.19\pm 14.76$ mm, significantly higher than the $15.83\pm 9.02$ mm on the sunny slope ($p = 0.002$). The substantial standard deviation observed ($CV \approx 70\%$ on the shady slope) reflects high intraspecific variability likely driven by micro-site heterogeneity. Despite this variability, the plot-level analysis confirmed a consistent trend. Relative bark thickness (RBT) followed a similar pattern ($p < 0.001$), with a medium effect size ($Cohen\prime s d = -0.60$).

In contrast, bark density showed an opposite pattern (Figure 3b). Trees on the sunny slope possessed significantly denser bark (${0.39 \pm 0.08 \mathrm{g}/\mathrm{c}\mathrm{m}}^3$) compared to those on the shady slope (${0.34 \pm 0.08\mathrm{g}/\mathrm{c}\mathrm{m}}^3; p < 0.001$), with a medium effect size ($Cohen\prime s d = 0.59$). Bark water content (WC) showed no significant difference between aspects ($p > 0.05$; Figure 3d), suggesting that tissue hydration levels are maintained relatively stable despite contrasting evaporative demands. Regarding stoichiometric traits (Table 2, Figure 4, slope aspect exerted a profound influence on bark nutrient composition. While bark total carbon (C) content remained statistically indistinguishable across slopes ($p = 0.912$), confirming a stable carbon allocation to structural tissues, bark nitrogen (N) and phosphorus (P) contents differed significantly. The shady slope was characterized by significantly higher bark N content (3.98 ± 0.81 g/kg) compared to the sunny slope ($2.93 \pm 1.99 \mathrm{g}/\mathrm{k}\mathrm{g}; p < 0.001$). Conversely, bark P content was higher on the sunny slope ($0.57 \pm 0.20 \mathrm{g}/\mathrm{k}\mathrm{g}$) than on the shady slope ($0.47 \pm 0.19 \mathrm{g}/\mathrm{k}\mathrm{g};p < 0.001$).

Most notably, the stoichiometric ratios exhibited large effect sizes. The bark C/N ratio on the sunny slope reached $194.82 \pm 50.99$, which was substantially higher than that on the shady slope ($132.15 \pm 26.55; p < 0.001$), with a very large effect size ($\mathrm{C}\mathrm{o}\mathrm{h}\mathrm{e}\mathrm{n}\prime \mathrm{s} d = 1.54$). Similarly, the bark N/P ratio differed sharply, being significantly lower on the sunny slope ($5.50 \pm 2.93$) compared to the shady slope ($9.45 \pm 3.25; p < 0.001, Cohen\prime s d =-1.28$). Despite these variations, the mean N/P ratios on both slopes were well below 14 (Figure 4a), indicating widespread N-limitation for bark growth in this region.

3.3. Allometric Scaling of Bark Thickness

Standardized Major Axis (SMA) regression revealed distinct allometric scaling relationships between bark thickness (TBT) and diameter at breast height (DBH) across slope aspects (

Table 3. Allometric scaling parameters of bark thickness vs. DBH estimated by Standardized Major Axis (SMA) regression. The model is $ln\left(TBT\right)=a+b·ln\left(DBH\right)$.

Slope Aspect	Intercept (SMA a)	Slope (SMA b)	R2	p-Value (Correlation)	n
Sunny	−0.54	0.87	0.77	<0.001	108
Shady	−0.89	1.03	0.77	<0.001	108

, Figure 5). Both slopes showed strong positive correlations ($R^2 = 0.77,p <0.001$). However, the scaling strategies differed significantly. Permutation tests confirmed significant heterogeneity in SMA slopes between the two aspects ($p = 0.004$; Figure 5).

Trees on the shady slope exhibited isometric scaling (b = 1.03), indicating that bark thickening keeps pace with radial stem growth. In contrast, trees on the sunny slope displayed negative allometry (b = 0.87), meaning that investment in bark thickness proportionally decreases as trees grow larger. This divergence highlights a trade-off where sunny slope trees prioritize other functions over thickness accumulation during ontogeny.

3.4. Correlations Between Soil and Bark Traits

Pearson correlation analysis was conducted to explore the potential coupling between edaphic drivers and bark functional traits (Figure 4b and Figure 6).

Regarding the nutrient coupling within the “soil–plant” continuum, the relationship between soil TN and bark TN exhibited a distinct “decoupling” pattern shown in Figure 4b. Statistical analysis indicated a non-significant correlation between soil nitrogen availability and bark nitrogen accumulation $(r = 0.01, p = 0.979)$. This suggests that, in the studied timberline ecotone, bark nutrient status is relatively independent of local soil nutrient supply, implying strong physiological regulation by the plant rather than passive uptake from the soil.

The correlation heatmap based on plot-level means (Figure 6) further revealed the complex interplay among multiple variables. A strong trade-off was observed between bark physical properties: bark density was strongly negatively correlated with bark thickness ($r = -0.75$), confirming the mechanical and physiological trade-off between tissue density (“quality”) and construction thickness (“quantity”).

In terms of environmental drivers, bark thickness (TBT) showed a negative correlation with soil total nitrogen (Soil N)$(r = -0.63)$ and a weak negative correlation with soil water content (Soil WC) $(r = -0.33)$. This negative association indicates that trees investing in thicker bark (typical of shady slopes) tend to occur in microsites that do not necessarily possess higher soil resource levels, further supporting the finding that microclimatic factors, rather than soil resources, are the primary drivers of bark thickening. Additionally, elevation showed weak correlations with most bark traits (e.g., $r = 0.12$ for TBT; $r = 0.37$ for Bark N), suggesting that within the studied altitudinal range, slope aspect acts as the dominant determinant of bark phenotypic plasticity.

4. Discussion

4.1. Trade-Offs in Bark Physical Architecture: Insulation vs. Hydraulic Safety

Our results demonstrated that Abies georgei var. smithii exhibits remarkable phenotypic plasticity in bark functional traits across contrasting slope aspects, despite the relative homogeneity of soil physicochemical properties (Table 1). This supports the hypothesis that at the timberline ecotone, microclimatic factors rather than soil nutrient pools act as the primary drivers shaping plant functional traits [7]. Interestingly, we observed slightly higher (though statistically non-significant) soil water content on the sunny slope compared to the shady slope (Table 1). This pattern likely reflects the homogenizing effect of monsoon rainfall during the growing season (August sampling), which may temporarily override the differences in evaporation demand between slopes [25,26]. Additionally, local variations in soil texture and organic matter capacity to retain water might clearly offset the higher evaporation demand typically associated with south-facing aspects.

This reflects a functional trade-off in terms of what portion of it is bark and how closely it is packed. Shady slope trees invested more carbon in making thicker barks (Figure 3a) with isometric scaling (Figure 5). High altitude thick bark is thought of mainly as insulation against thermal damage from the vascular cambium freezing (or re-freezing) at lethal temperatures frequently and thermal protection [27,28]. The shady slopes of the Sygera Mountains are experiencing lower average temperatures and longer-lasting snow cover. Therefore, increasing bark thickness becomes a critical “thermal buffer” to ensure cambial survival [29].

In contrast, trees on the sunny slope possessed significantly denser bark (Figure 3b). High tissue density is a hallmark of the ‘conservative strategy,’ which provides mechanical strength and resistance to physical damage, albeit at a higher construction cost per unit volume [30]. The sunny slope experiences higher solar radiation and evaporative demand, subjecting trees to physiological drought stress even in the absence of severe soil water deficits. Although we found no significant difference in bark water content between slopes ($p > 0.05$), the higher density likely serves as a compensatory mechanism. Denser bark tissues reduce hydraulic conductance, thereby creating an effective barrier against desiccation [31]. This shift from an insulation-oriented strategy (thicker bark on shady slopes) to a conservation-oriented strategy (denser bark on sunny slopes) reflects a phenotypic adaptation to microclimatic stress.

It is also important to consider physical weathering as a driver of thinner bark on sunny slopes. The higher diurnal temperature fluctuations and intense UV radiation on south-facing slopes may accelerate the exfoliation and weathering of the outer rhytidome, contributing to the observed negative allometry ($b = 0.87$) where thickness accumulation lags behind stem growth [32]. Thus, the phenotypic plasticity observed here reflects a trade-off: prioritizing thermal insulation on the cold, shady slope versus maximizing hydraulic safety and density on the exposed, sunny slope.

4.2. Stoichiometric Plasticity and Resource Allocation

The divergence from stoichiometry in this study shows that soil nutrients do not match up with plant nutrients. Soil N availability was statistically similar regardless of the slopes, but bark N was significantly higher on the shady slopes, and the C/N was remarkably higher on the sunny slopes ($Cohen\prime s d = 1.54$, Table 2). GRH claims that an organism’s N content is positively associated with its growth rate because ribosomes needed for protein synthesis are N-rich [33]. Bark N and isometric growth pattern are higher on the shady slope, indicating that the individuals in this habitat would adopt an “acquire-resource” strategy to maximize their growth during the short growing period [34].

On the contrary, the extremely high C/N ratio (>190) and extremely low N/P ratio on the sunny slope reflect a strong limitation by N accumulation compared to C. In accordance with the Carbon-Nutrient Balance Hypothesis (CNBH), plants growing under nutrient-limited or stressed conditions accumulate excess carbon into defensive compounds (phenolics, lignin) as opposed to making new biomass [35]. Rather than “trying” to survive, these trees exhibit stress-tolerance traits, allocating more carbon to construct dense, lignified structures (high C cost) while conserving limiting nutrients, such as nitrogen, against oxidative stress or high pest populations [36].

4.3. Divergent Allometric Scaling and Ontogenetic Shifts in Bark Allocation

According to our SMA analysis, the allometric scaling of bark thickness is not a fixed species-constant but a plastic trait. It is affected by local environmental conditions. Slope b, the scaling exponent, is considered an important parameter for relative growth rates among organs [37]. The different values of the scaling exponents in sunny areas (b = 0.87) compared to shady areas (b = 1.03) imply that during ontogeny, Abies georgei var. smithii uses different resource allocation strategies to deal with contrasting environments.

Those on the shady slope showed isometric scaling (b ≈ 1.0) and, therefore, the bark becomes thicker by a fixed amount in relation to the stem diameter. It matches the “Hydraulic Safety-Thermal Insulation Hypothesis” in cold areas [38]. As taller trees grow and their stems increase, their total heat loss increases. Maintaining a relatively constant bark thickness is energetically costly. However, it is necessary to maintain the thermal inertia of the vascular cambium in the understory, which is shaded and snow-covered and therefore always cooler [39]. The better soil moisture and nutrient conditions on the shady slope (Table 1, though not statistically significant, had a trend towards more resources), which would presumably reduce the carbon cost of maintaining the thick bark structure, thus being able to make an investment in the “luxury” of structural defense for the rest of the tree’s life [40].

Negative allometry (b < 1.0) from the sunny slope is the “diminishing return” investment. As trees mature, DBH increases, and the proportion of allocation to bark thickness increases as well. The Optimal Partitioning Theory (OPT) can explain this phenomenon: plants allocate resources to the organ that acquires the most limiting resource [41]. A sunny slope, with higher radiation and higher evaporation demand, is most likely limited by water and metabolism for large trees. Therefore, big ones might favor investing more resources in conducting tissues (xylem) or taking up a lot of ground with their root systems rather than just making a thick bark [42]. Additionally, higher bark densities and higher C/N ratios of the sunny slopes (Table 2) indicate a change from “more” (the total thickness) to “better” (higher density and chemical defense). As trees grow larger, it is no longer the most carbon-efficient way to get more bark simply to have thicker bark, but instead to increase the tissue that provides mechanical and drought-resistant support and has less volume increase [43]. This ontogenetic shift minimizes construction costs under environmental stress.

4.4. Decoupling of Soil and Bark Traits: Ecological Implications

A key finding of this study is the lack of significant correlation (decoupling) between soil nutrient availability and bark traits (Figure 4b). We find that Abies georgei var. smithii exhibits considerable phenotypic plasticity in its bark traits and can thus dissociate its allocation strategy from soil nutrient heterogeneity. Several factors may explain this independence. First, internal nutrient retranslocation within the tree often masks a direct relationship with soil pools.

Second, and critically, our soil sampling was limited to the surface layer (0–20 cm). Adult Abies trees typically possess deep root systems that access nutrients and water from deeper soil horizons, which were not analyzed in this study [44]. Consequently, the surface soil properties might not fully represent the actual resource availability experienced by the trees. Finally, the influence of microclimate appears to override edaphic factors. This confirms that for high-altitude timberlines, topographic modulation of climate is a more potent driver of phenotypic plasticity than soil nutrient gradients.

This insight is crucial for understanding (or predicting) how treelines will respond in the future. The “conservative strategy” on the sunny slope, with denser bark, negative allometry, and a high C/N ratio, might be a pre-adaptation for the warmer, drier climate of the Tibetan Plateau [45]. And because global warming worsens drought stress and vapor pressure demand on the sunny slope, the capacity to go from a “resource-acquisitive” (shady slope) to a “stress-tolerant” (sunny slope) phenotypic strategy will be a key determinant of the persistence and stability of alpine timberline populations [46].

And this also challenges the old idea that bark ecology was totally based on the fire regime. In the fire-independent ecosystems of the Sygera mountains, we showed that bark traits are actively coordinated with micro-environmental drivers (temperature and water availability) through finely tuned trade-offs between thickness (insulation) and density (hydraulic safety) [47]. Thus, it is proven that bark is a multidimensional function organ with strong integration with the Whole-Plant Economics Spectrum [36]. And we also take it further, and broaden the scope of the spectrum concept to include bark as well as leaves and roots. As resource-acquisitive leaves (high specific leaf area, high N) are associated with fast-growing roots, likely, the “acquisitive” bark phenotype (high N, isometric) on shady slopes helps facilitate rapid nutrient acquisition and phloem loading for growth under temperature-limited conditions [48]. But the more “conservative” bark on sunny slopes is like a plant that prioritizes having leaves that are dense and full of defenses. We argue that future vegetation models and carbon accounting should account for the phenotypic plasticity of bark traits—specifically, stoichiometric plasticity (C/N/P)—to improve the accuracy of ecosystem productivity estimates in alpine regions.

5. Conclusions

This study provides the first comprehensive evidence that topographic heterogeneity drives divergent phenotypic strategies in the bark of Abies georgei var. smithii at the alpine timberline. Our results demonstrate that bark functional traits are largely decoupled from surface soil nutrient gradients and are instead shaped by the distinct microclimates of sunny and shady slopes. Trees on shady slopes exhibit isometric scaling and invest in thicker, nitrogen-rich bark, a strategy likely geared towards thermal insulation and rapid growth during short vegetative seasons. Conversely, trees on sunny slopes adopt a conservative strategy characterized by negative allometry, higher tissue density, and elevated C/N ratios, prioritizing hydraulic safety and resistance to physical weathering over thickness accumulation.

These findings offer valuable insights into how timberline species might adjust their allocation priorities under changing environmental conditions. The observed shift from “resource-acquisitive” to “stress-tolerant” bark phenotypes suggests that phenotypic plasticity plays a critical role in species persistence. While our study does not provide a quantitative predictive model for climate change responses, it highlights that future vegetation modeling should account for the multidimensional plasticity of bark—beyond simple allometry—to better understand forest resilience in a warmer, more variable climate.