Linking seedling wood anatomical trade-offs with drought and seedling growth and survival in tropical dry forests

Abstract

Introduction

Water availability is a main factor driving functional and demographic variations across plant species and communities (Poorter & Markesteijn, 2008; Phillips et al., 2010; Comita & Engelbrecht, 2014) and is expected to become increasingly important as drought is predicted to intensify in many regions world-wide (Intergovernmental Panel on Climate Change, 2022). In this sense, understanding how plants persist under drought conditions is an important step toward predicting their response to future drier climates (Poorter & Markesteijn, 2008; Comita & Engelbrecht, 2014). Wood plays a central role in water transport (Carlquist, 2001; Baas et al., 2004) and, thus, in species ability to persist under drought conditions (Anderegg & Meinzer, 2015). In angiosperm wood, the vascular system responsible for transporting water consists of a network of interconnected vessels, and occasionally tracheids, all embedded in a matrix of fibers and parenchyma cells. Fibers primarily provide mechanical support and participate in storage in the case of living fibers, while living parenchyma cells are mainly involved in the storage of water and nonstructural carbohydrates (NSC; Carlquist, 2001). Although fibers and parenchyma cells may also participate in water transport, their roles in plant hydraulics are less recognized than that of vessels. Fibers, for example, can reinforce vessel walls and avoid vessel implosion under extreme negative pressures induced by drought (Jacobsen et al., 2005). In turn, parenchyma cells can favor stem capacitance by storing water, which buffers fluctuations in xylem water potentials, or they can participate in vessel refilling when they are in direct contact with vessels (Sauter et al., 1973; Morris et al., 2018a,b; Aritsara et al., 2021). However, the interplay among vessels, fibers and parenchyma cells under drought conditions remains elusive, as well as its effect in growth and survival.

The allocation of wood volume to vessels, fibers and parenchyma cells may lead to anatomical and functional trade-offs (Baas et al., 2004; Bittencourt et al., 2016; Pratt & Jacobsen, 2017). Two of these trade-offs may be particularly relevant to species performance under drought conditions. The first is between the wood fraction (i.e. % of wood cross-sectional area) allocated to either fibers or parenchyma cells (e.g. Ziemińska et al., 2015; Pratt & Jacobsen, 2017). Species with larger parenchyma fractions, and thus lower fiber fractions, generally store higher amounts of water or NSC (Plavcová & Jansen, 2015; Zhang et al., 2023), at the expense of mechanical support or vessel reinforcement due to a lower investment in fibers (Ziemińska et al., 2015). The second is between wood density and vessel lumen size (Ziemińska et al., 2015; Hietz et al., 2017). Plants with denser wood and narrower vessels tend to exhibit greater drought tolerance (e.g. Preston et al., 2006; Chave et al., 2009; Fajardo et al., 2022), although this implies a decreased in water transport efficiency, as xylem hydraulic conductivity is proportional to vessel lumen size (e.g. Tyree & Zimmermann, 2002; Baas et al., 2004; Hietz et al., 2017).

Given that NSC stored in parenchyma cells play a key role in maintaining hydraulic balance (O'Brien et al., 2014; Morris et al., 2016), one may expect larger parenchyma fractions, and thus lower fiber fractions, in drier environments (e.g. Morris et al., 2016). However, since parenchyma cells are alive and metabolically active (Spicer & Holbrook, 2007), having a higher proportion of parenchyma cells can result in increased maintenance costs, particularly in low-resource environments such as drier sites (Grime, 1977; Reich, 2014). Therefore, plants growing in drier sites may have stems with lower parenchyma and thus higher fiber fractions. Likewise, as narrow vessels are generally less vulnerable to drought-induced embolisms than wider ones (Hacke et al., 2017; Lens et al., 2022), it is likely that selection pressures favor the prevalence of plants with narrower vessels, and thus denser woods, in drier sites. Selection should also favor denser woods in drier sites because dense wood typically has more and thicker fibers (Ziemińska et al., 2015), which can reduce the risk of vessel implosion during severe droughts (Jacobsen et al., 2005). These expectations have been supported by a number of studies (e.g. Carlquist, 2001; Swenson & Enquist, 2007; Wheeler et al., 2007; Chave et al., 2009; Hacke et al., 2017), though are also exceptions (see ter Steege & Hammond, 2001; Muller-Landau, 2004).

As these two trade-offs can influence plant functioning, they are expected to affect growth and survival. A higher allocation to parenchyma cells might increase survival by enabling a higher storage of NSC (Plavcová & Jansen, 2015; Herrera-Ramírez et al., 2021), favoring recovery after herbivory (Myers & Kitajima, 2007), or by providing an active response against pathogens (Morris et al., 2016). It is less clear, however, how this trade-off might influence growth. One possibility is that species with higher parenchyma fractions, and lower fiber fractions, may incur in higher maintenance costs, which could ultimately limit growth (Chapin et al., 1990; Myers & Kitajima, 2007; Spicer & Holbrook, 2007). Alternatively, species with larger parenchyma fractions might grow faster given that wood with fewer fibers is in principle cheaper to build, and because these species likely have a higher number of parenchyma cells in direct contact with vessels (Morris et al., 2018a,b), which can favor growth by increasing hydraulic conductivity (Aritsara et al., 2021). Regarding the wood density–vessel size trade-off, it is well-established that low wood density and wider vessels promote faster growth by reducing construction costs and increasing xylem hydraulic conductivity, respectively (Chave et al., 2009; Hietz et al., 2017). Similarly, evidence indicates that species with high wood density and narrow vessels typically have lower mortality rates, as high wood density favors decay resistance and strength, and narrow conduits make species less vulnerable to drought-induced embolisms that may eventually cause hydraulic failure and tissue or plant death (Muller-Landau, 2004; King et al., 2006; Chave et al., 2009; Hietz et al., 2017; González-M et al., 2021; Hacke et al., 2022; Lens et al., 2022; Jacobsen & Pratt, 2023). However, the link between vessel lumen size and vulnerability to drought-induced embolisms can be weak (Zanne et al., 2010; Lens et al., 2022), and it ultimately depends on how vessel size is related to intervessel pit traits that are more mechanistically linked to embolism formation and spread (Choat et al., 2008; Lens et al., 2022).

Given that the effects of traits on plant functioning can vary according to resource availability (Grime, 1977; Reich, 2014), the links between trait trade-offs and demographic rates may be better understood by considering the abiotic environment (Laughlin et al., 2018; Yang et al., 2018; Iida & Swenson, 2020; Li et al., 2022). For example, as having light wood and wider vessels is a strategy that favors growth, it would represent a demographic advantage in sites with relatively higher water availability, but it can be disadvantageous in low-resource environments such as drier sites (Chave et al., 2009; Reich, 2014; Hietz et al., 2017). Likewise, since investing in larger amounts of parenchyma cells, and hence lower amounts of fibers, is a strategy that can reduce the risk of drought-induced mortality (e.g. Morris et al., 2016; Aritsara et al., 2021), it would be particularly advantageous in more drier sites, but it could be too costly in less dry sites where the benefits may not offset the costs of this allocation strategy (Reich, 2014).

Although previous research has examined the links of wood anatomical structure with environmental factors (Martínez-Cabrera et al., 2009; Fortunel et al., 2014; Lourenço et al., 2022; Zhang et al., 2023), and with demographic rates (Russo et al., 2010; Hietz et al., 2017; Aritsara et al., 2021), these studies have been mainly focused on adults. By contrast, our understanding of these links in the case of seedlings remains limited (but see Corcuera et al., 2006; Durante et al., 2011; Aref et al., 2013), despite the key role that the seedling stage plays in shaping species diversity and abundance at later stages due to its high mortality rates (e.g. Muller-Landau et al., 2002). As seedlings and adults experience contrasting local environments (e.g. Iida & Swenson, 2020), and given that wood anatomy and demographic rates can vary considerably during ontogeny (Osazuwa-Peters et al., 2017; Rungwattana & Hietz, 2018; Iida & Swenson, 2020), the trait–environment and trait–demography links reported previously for adults may not always hold true for seedlings. For instance, wood density seems to be a better predictor of growth rates for adults than for seedlings (Visser et al., 2016). Furthermore, recent studies have highlighted that vessel traits can explain mortality rates at the sapling, but not at the adult stage (Osazuwa-Peters et al., 2017; González-Melo et al., 2023). To our knowledge, our study is the first in evaluating seedling wood anatomical trade-offs and their relationships with drought conditions and demographic rates for species-rich communities in tropical regions.

In this study, we examined how wood anatomical trade-offs are related to drought conditions and 1-yr growth and survival in 65 species from four tropical dry forests in Colombia representing a steep gradient in rainfall conditions. In particular, we wanted to address the following questions: (1) How do the fiber vs parenchyma and wood density vs vessel size trade-offs change in response to drought? We anticipated that seedlings will allocate more wood volume to parenchyma cells and less to fibers as drought increases and that seedlings will have denser woods with narrow vessels in drier sites. (2) How are the fiber vs parenchyma and the wood density vs vessel size trade-offs related to growth and mortality, and to what extent are these relationships mediated by drought? We predicted that higher fractions of parenchyma, together with lower fractions of fibers, will be positively related to survival, but negatively to growth. We anticipated that low-density wood and wider vessels will favor growth, but increase mortality. We also predicted that the links between trait trade-off and demographic rates will be mediated by drought.