Temperate rain forest species partition fine-scale gradients in light availability based on their leaf mass per area (LMA)

Allometry of the quasi-pipe (qPipe) model for estimating tree leaf area and tree leaf mass applied to plant functional types

The allometry of the pipe model quantifies the approximate proportionality between the tree leaf amount and the stem cross-sectional area at the crown base (A_CB). It is useful for estimating and modeling carbon fixation abilities of trees but requires climbing the tree and is thus unsuitable for large-scale studies. Here, we adopted a previously proposed allometry (hereafter the quasi-pipe (qPipe) model allometry) formulating the relationship between the tree leaf amount and a surrogate of A_CB, A_{CB_Est}, calculated from tree dimensions measurable from the ground. Using published/unpublished data for 962 trees of 159 species collected between tropical rainforests and boreal forests, we established pipe and qPipe model allometries for evergreen-conifer, deciduous-conifer, evergreen-broadleaf, and deciduous-broadleaf plant functional types (PFTs). For the leaf area per tree (LA), allometric lines on a log-log plane were almost identical among the four PFTs in both models, with slopes of ~ 1. For the leaf mass per tree (LM), however, the allometric lines separated among the four PFTs in both models and had slopes greater than 1, indicating that the proportionality assumed in the pipe model held for LA but not LM. The applicability of the qPipe model in estimating the stand-scale leaf amount was further examined.

Phenology explains different storage remobilization in two congeneric temperate tree species with contrasting leaf habit

The dependence of trees on carbon and nutrient storage is critical to predicting the forest vulnerability under climate change, but whether evergreen and deciduous species differ in their use and allocation of stored resources during spring phenology is unclear. Using a high temporal resolution, we evaluated the role of spring phenology and shoot growth as determinants of the carbon and nutrient storage dynamics in contrasting leaf habits. We recorded the phenology and shoot elongation and determined the concentrations of total non-structural carbohydrates (NSCs), starch, soluble carbohydrates, nitrogen (N) and phosphorus (P) in buds, expanding shoots and previously formed shoots of two sympatric Nothofagus species with contrasting leaf habit. Species reached similar shoot lengths, though shoot expansion started 35 days earlier and lasted c. 40 days more in the deciduous species. Thus, although the deciduous species had a relatively constant shoot growth rate, the evergreen species experienced a conspicuous growth peak for c. 20 days. In the evergreen species, the greatest decreases in NSC concentrations of previously formed shoots and leaves coincided with the maximum shoot expansion rate and fruit filling, with minimums of 63 and 65% relative to values at bud dormancy, respectively. In contrast, minimum NSC concentrations of the previously formed shoots of the deciduous species were only 73% and occurred prior to the initiation of shoot expansion. Bud N and P concentrations increased during budbreak, whereas previously formed shoots generally did not decrease their nutrient concentrations. Late spring phenology and overlapping of phenophases contributed to the greater dependence on storage of proximal tissues in the studied evergreen compared with deciduous species, suggesting that phenology is a key determinant of the contrasting patterns of storage use in evergreen and deciduous species.

Leaf density and chemical composition explain variation in leaf mass area with spectral composition among 11 widespread forbs in a common garden

Leaf mass per area (LMA) is a key leaf functional trait correlated with plant strategies dictating morphology, physiology, and biochemistry. Although sunlight is generally accepted as a dominant factor driving LMA, the contribution of each spectral region of sunlight in shaping LMA is poorly understood. In the present study, we grew 11 widespread forb species in a common garden and dissected the traits underpinning differences in LMA, such as its morphological components (leaf density [LD] and leaf thickness [LT]), macroelement, and metabolite composition under five spectral-attenuation treatments: (1) transmitting c. 95% of the whole solar spectrum (> 280 nm), (2) attenuating ultraviolet-B radiation (UV-B), (3) attenuating both UV-A and UV-B radiation, (4) attenuating UV radiation and blue light, (5) attenuating UV radiation, blue, and green light. We found that LMA, LD, and chemical traits varied significantly across species depending on spectral treatments. LMA was significantly increased by UV-B radiation and green light, while LD was increased by UV-A but decreased by blue light. LMA positively correlated with LD across treatments but was only weakly related to LT, suggesting that LD was a better determinate of LMA for this specific treatment. Regarding leaf elemental and metabolite composition, carbon, nitrogen, and total phenolics were all positively correlated with LMA, whereas lignin, non-structural carbohydrates, and soluble sugars had negative relationships with LMA. These trends imply a tradeoff between biomass allocation to structural and metabolically functional components. In conclusion, sunlight can spectrally drive LMA mainly through modifying functional and structural support.

The Leaf Economics Spectrum Constrains Phenotypic Plasticity Across a Light Gradient

The leaf economics spectrum (LES) characterizes multivariate correlations that confine the global diversity of leaf functional traits onto a single axis of variation. Although LES is well established for traits of sun leaves, it is unclear how well LES characterizes the diversity of traits for shade leaves. Here, we evaluate LES using the sun and shade leaves of 75 woody species sampled at the extremes of a within-canopy light gradient in a subtropical forest. Shading significantly decreased the mean values of LMA and the rates of photosynthesis and dark respiration, but had no discernable effect on nitrogen and phosphorus content. Sun and shade leaves manifested the same relationships among N mass, P mass, A mass, and R mass (i.e., the slopes of log-log scaling relations of LES traits did not differ between sun and shade leaves). However, the difference between the normalization constants of shade and sun leaves was correlated with functional trait plasticity. Although the generality of this finding should be evaluated further using larger datasets comprising more phylogenetically diverse taxa and biomes, these findings support a unified LES across shade as well as sun leaves.

Leaf functional trait variation in a humid temperate forest, and relationships with juvenile tree light requirements

The species-rich arborescent assemblages of humid tropical forests encompass much of the known range of the leaf economics spectrum, often including >20-fold variation in leaf lifespan. This suite of traits underpins a life-history continuum from fast-growing pioneers to slow-growing shade-tolerant species. Less is known about the range of leaf traits in humid temperate forests, and there are conflicting reports about relationships of these traits with the light requirements of temperate evergreen angiosperms. Here I quantify the range of leaf functional traits in a New Zealand temperate evergreen forest, and relationships of these traits with light requirements of juvenile trees and shrubs. Foliage turnover of saplings of 19 evergreen angiosperms growing beneath gaps (12-29% canopy openness) and in understories (1.2-2.9%) was measured over 12 months. Dry mass per area (LMA), dry matter content, thickness, density and nitrogen content (N) of leaves were also measured. Species minimum light requirements were indexed as the 10th percentile of the distribution of saplings in relation to canopy openness. Interspecific variation of leaf lifespan was ∼6-fold in gaps (0.6 to 3.8 yrs), and ∼11-fold in the understorey (0.7 to 7.7 yrs). Six small tree and shrub species are effectively leaf-exchangers, with leaf lifespans of c.1 year in gaps-albeit usually longer in the shade. Interspecific variation in other leaf traits was 2.5 to 4-fold. Lifespans and LMA of both sun and shade leaves were negatively correlated with species light requirements i.e., positively correlated with shade tolerance. However, light environment (gap vs shade) explained about the same amount of variation in LMA as species' identity did. Species light requirements were not significantly correlated with leaf N, dry matter content, density or thickness-except for a marginally significant correlation with dry matter content of shade leaves. Species light requirements were thus less consistently related to leaf structural traits than appears to be the case in humid tropical forests. Whereas the wide interspecific variation in leaf economic traits of tropical rainforest species outweighs plastic response to light availability, temperate evergreen woody angiosperms appear to occupy a narrower range of the leaf economic spectrum. Standardization of the light environments in which LMA is measured is vital in comparative studies of humid temperate forest evergreens, because of countergradient responses of this trait to light, and because of the relative magnitudes of plastic and interspecific variation in LMA in these forests.

Intraspecific trait variation and the leaf economics spectrum across resource gradients and levels of organization

Understanding patterns of functional trait variation across environmental gradients offers an opportunity to increase inference in the mechanistic causes of plant community assembly. The leaf economics spectrum (LES) predicts global tradeoffs in leaf traits and trait-environment relationships, but few studies have examined whether these predictions hold across different levels of organization, particularly within species. Here, we asked (1) whether the main assumptions of the LES (expected trait relationships and shifts in trait values across resource gradients) hold at the intraspecific level, and (2) how within-species trait correlations scale up to interspecific or among-community levels. We worked with leaf traits of saplings of woody species growing across light and soil N and P availability gradients in temperate rainforests of southern Chile. We found that ITV accounted for a large proportion of community-level variation in leaf traits (e.g., LMA and leaf P) and played an important role in driving community-level shifts in leaf traits across environmental gradients. Additionally, intraspecific leaf trait relationships were generally consistent with interspecific and community-level trait relationships and with LES predictions-e.g., a strong negative intraspecific LMA-leaf N correlation-although, most trait relationships varied significantly among species, suggesting idiosyncrasies in the LES at the intraspecific level.