Functional traits mediate interactions among organisms and the environment determining individuals’ fitness and have been central in the development of mechanistic models that project plant species responses to future climate (Myers-Smith et al. 2019, Heilmeier 2019). A key step in building a more mechanistic understating of how plants and ecosystems function is to define the interplay between above- and belowground trait variations that shapes the functional strategies within species, fostering knowledge on potential adjustments in physiology to changing climate (Meier and Leuschner 2008, Reich 2014, Anderegg 2015, Díaz et al. 2016). Among the well-studied functional trait dimensions, the Leaf Economics Spectrum (LES) has generated substantial attention. The LES describes functional trait variation that ranges from acquisitive to conservative leaf traits (Wright et al. 2004, Reich 2014). This axis not only dictates how plants efficiently acquire and process light but also underscores the tradeoff between rapid carbon acquisition and resource conservation (Wright et al. 2004, Reich 2014). In contrast, belowground traits are thought to be organized in a bidimensional root economics space (Bergmann et al. 2020, Carmona et al. 2021, Weigelt et al. 2021). One of these root axes resembles the acquisitive-conservative tradeoff seen in the LES, where fine root traits vary along a spectrum with high nitrogen concentration in roots at the acquisitive end, and roots with high root tissue density (RTD; root mass per unit volume) at the conservative end (Bergmann et al. 2020, Weigelt et al. 2021). The second root axis reflects a ‘collaboration axis’ that has been proposed to describe the degree to which roots associate with mycorrhizal fungi (Bergmann et al. 2020). One extreme of this collaboration axis is occupied by species that have long, thin roots (i.e., high specific root length, SRL; root length per unit root dry mass) able to efficiently explore the soil and acquire resources. The other extreme is occupied by species that have thick roots (high root diameter, RD) that cannot explore the soil efficiently by themselves; instead, plants “outsource” the acquisition of soil resources to mycorrhizal fungi, facilitating their colonization with increased root diameter.
Across species at global scales, these leaf and root traits have shown to form independent functional dimensions (Medeiros et al. 2017, Carmona et al. 2021, Weigelt et al. 2021), suggesting some degree of independence between above and belowground resource acquisition functions. However, we may expect different scenarios when trends are examined within species, as additional constraints may be at play and limit trait variability (Messier et al. 2010, 2017). For example, trait-trait relationships in the LES tend to weaken or even reverse when within species information is considered(Messier et al. 2010, 2017, Anderegg et al. 2018, Umaña and Swenson 2019) Examining whether functional leaf and root trait axes exhibit similar patterns intra- and interspecifically should provide valuable insights about the coordination in below- and aboveground tree functional strategies.
When examined at the within species level, above- and belowground traits may still exhibit decoupling due to constraints that tend to operate more strongly among traits within the same organ than across organs (Merila and Bjorklund 2004). This occurs because traits from the same organ tend to participate more cohesively in similar functions and/or share similar organ construction costs (Berg 1960, Poorter and Villar 1997). This shared functionality may result in stronger covariation among root traits than between root and aboveground traits, promoting correlation between dimensions that are independent at interspecific scales where these constraints are absent. However, this stronger association among traits from the same organ, may contribute to weaken the observed dimensionality of the root traits across species as root traits have shown strong phylogenetic conservatism, even more than leaf traits (Comas and Eissenstat 2009, Chen et al. 2013, Valverde-Barrantes et al. 2017). Furthermore, tree species tend to exhibit high specificity with mycorrhizal types (Smith and Read 2008) which can further limit independence in variation across root traits.
In addition, we may expect uncoupled variation between leaf and root traits due to the decoupled environmental variation between above- and belowground resources, that results in differential patterns of trait variation via plasticity or genetic variation (e.g., local adaptation)(Messier et al. 2010, 2017). For example, at local scales light heterogeneity is typically associated with tree fall gaps and may influence the distribution of leaf traits; under low light conditions trees display acquisitive leaves with high nutrient content and high specific leaf area (SLA; leaf area per unit leaf mass) to maximize carbon metabolism (Wright et al. 2006, Markesteijn et al. 2007, Petriţan et al. 2008, Barton 2023). Meanwhile, soil nutrients vary at a different spatial scale mainly associated to the distribution of microorganisms, vegetation type, climate, or to geological processes (Vitousek 2004, Fowler et al. 2013). This variation in soil nutrients has shown to have strong effects on diversity of belowground traits. When soil nutrient levels are high, plant roots tend to possess higher nutrient concentrations, while at lower soil nutrient levels, plants may have higher RTD to extend the lifespan of the root and conserve its resources, (Eissenstat et al. 2015, Cheng et al. 2016). The decoupled spatial scales at which variation in above- and belowground resources occur can thus result in independent variation in above- and belowground traits when examined within species (Asefa et al. 2022, Weemstra et al. 2022).
Due to this coordination when examining the relationships between traits and environmental (e.g., light availability or soil nutrient content), it is important to account for the potential coordination (and lack of it) among different traits that participate in similar functions. However, most work relating traits to environment has done so in a univariate framework (e.g., Freschet et al. 2013; de la Riva et al. 2018; Defrenne et al. 2019), in which each trait-environment relationship is examined independently. This approach ignores the co-variation between traits which is widely accepted exist and is fundamental for determining functionality as multiple traits tend to work together to achieve specific functions (e.g., LES and photosynthesis)(Wright et al. 2004, Armbruster et al. 2014). This integrated functionality is important across organizational levels but especially at the intraspecific scale because, as mentioned above, traits have the potential to be more constrained within species. To counter this issue, we develop a novel modeling approach that accounts for co-variation (or lack thereof) between traits in the hopes of better linking traits and the environment.
Here we aim to gain understanding of the coordination and intraspecific variation of leaf and root traits as well as the extent to which trait variation is explained by light and soil nutrient heterogeneity for four widespread and commonly co-occurring woody species from forests in the northeastern United States. We measured six leaf and root functional traits of 131 tree seedlings across eight sites in temperate, deciduous, broadleaf forests in the northeastern USA, distributed in two climatic regions: north and south. We focus on seedlings because the seedling stage is essential to the recruitment process in forests (Harper 1977, Green et al. 2014). The selected species represent different ecological strategies with respect to shade and drought tolerance and form associations with different mycorrhizal types (Table 1). We address the following two questions:
Table 1
– Summary of the life history of the four species used in our study. Mycorrhizal type refers to the primary type of mycorrhizal association, either arbuscular mycorrhizal (AM) or ectomycorrhizal (EcM). Shade and drought tolerance are scored from 0 (no tolerance) to 5 (high tolerance) and include standard deviation. All tolerance data presented is retrieved from Niinemets and Valladares (2006).
Species | Common Name | Mycorrhizal Type | Shade Tolerance | Drought Tolerance |
Acer rubrum | Red Maple | AM | 3.44 ± 0.23 | 1.84 ± 0.16 |
Acer saccharum | Sugar Maple | AM | 4.76 ± 0.11 | 2.25 ± 0.25 |
Prunus serotina | Black Cherry | AM | 2.46 ± 0.34 | 3.02 ± 0.02 |
Quercus rubra | Northern Red Oak | EcM | 2.75 ± 0.18 | 2.88 ± 0.12 |
-
Do the three independent above- and belowground trait axes found globally across species exist within species? Owing to the substantial differences in the variation of light and soil nutrients both within and between sites we expect to find weak relationships between comparable traits in roots and leaves (e.g., root N and leaf N), resulting in independent above- and belowground resource-acquisition axes. However, we also expect that there will be strong relationships between traits within individual organs, as there will be additional constraints that ensure optimal organ-level functionality. Root traits in particular have shown stronger phylogenetic constraints than leaf traits and high specificity in the type of mycorrhizal associations that developed, which may lead to stronger coordination between root traits than between leaf traits within species, resulting in only a single axis belowground.
-
How does intraspecific variation in traits relate to smaller scale (i.e. individual and site level) variation in light and soil nutrients, versus variation at larger scales (i.e., climatic differences between northern or southern sites)? Furthermore, is one scale more closely associated with trait variation than the other? We expect that aboveground trait variation will be more strongly associated with variation in light availability, while belowground trait variation will be more closely associated with soil nutrients. Specifically, belowground we expect variation in traits to be associated with variation in N, as N tends to be the most limiting nutrient in these systems (Vitousek 2004, Du et al. 2020). Finally, we predict that seedlings in northern sites that experience a shorter growing season will be selected to maximize growth during that short window and so express a more acquisitive strategy than those in the southern sites with more mild climates (Bonito et al. 2011). Overall, we expect seedlings to be more associated with smaller scale variations in abiotic variables because smaller variations in the environment can have large impacts on seedlings that lack the same storage capacities and occupy substantially smaller areas than adult trees.