1. Introduction
Isoprenoids are volatile organic compounds (VOCs) that are emitted to the atmosphere mostly by plants. They have diverse functional roles at multiple scales, from cellular protection and defense at the foliar level, through chemical signaling within and among plants, up to the regulation of large-scale biogeochemical processes, such as the effect on atmospheric chemical composition and contribution to aerosol formation (Laothawornkitkul, Taylor, Paul & Hewitt 2009). Volatile isoprenoids are represented by isoprene (C5H8), monoterpenes (C10H16), and sesquiterpenes (C15H24), and their largest source is from tropical trees that contribute ca. 80% of global emissions (Guenther et al. 2012). With half of the world’s tropical forests, Amazonia is recognized as the most important global source of isoprenoids to the atmosphere (Sindelarova et al. 2014). Since the early 1980s, multiple investigations have studied canopy flux, canopy concentrations, and to a lesser degree, leaf-level emissions of isoprenoids. These studies have reported meaningful insights into the emission drivers and how these compounds are involved in subsequent atmospheric processes (Yáñez‐Serrano et al. 2020). These combined findings contributed to developing and optimizing an isoprenoid emission model (Guenther et al. 2012). Despite this effort, emission estimates from tropical vegetation still carry a high uncertainty due to a poor understanding of the biological controls that determine thecapacity of emission and its plasticity in response to ecological and environmental conditions, as well as of the environmental controls that determine rates of emissions (Alves et al. 2018).
The constitutive emission capacity of isoprenoids is determined by the emission at leaf standard conditions (1000 𝜇mol m-2s-1photosynthetically active radiation, 30 ˚C), and this is a physiological property that may vary plastically (across individuals) under different ecological and physiological conditions. Actual emission rates are a function of emission capacity and variation in light and temperature (Niinemets et al. 2011). This implies that, for modeling isoprenoid emissions, first it is necessary to quantify emission capacities across species, and their plasticity across individuals, to then quantify the emission variation driven by environmental factors such as light and temperature (Duhl, Helmig & Guenther 2008; Niinemets et al. 2011; Guenther et al.2012). For several practical reasons, such as the remoteness and high plant species diversity of Amazonia, most studies so far have only investigated isoprenoid emission rates at the ecosystem level and how they vary with environmental factors. The majority of these studies measured canopy concentration or flux of isoprenoids, focusing on isoprene, known to be the strongest emitted compound (Eerdekens et al. 2009). Only a few studies measured monoterpenes, and very few studies quantified sesquiterpenes (Yáñez‐Serrano et al. 2020).
A first attempt to address isoprene emission capacities at the leaf level and the upscaling to ecosystems was made by Harley et al.(2004). Knowing that not all plant species emit isoprene (Monson, Jones, Rosenstiel & Schnitzler 2013), the aforementioned study aimed at quantifying the isoprene emission capacity for multiple plant species from different Amazonian regions, and a method was created to impute the isoprene trait to other non-measured trees by using species identification and phylogenetic proximity; then, results were used to upscale the isoprene emission capacity to the ecosystem level based on the trees fraction of isoprene emitters. Subsequent work has further expanded the number of species measured (Jardine et al. 2020; Taylor et al. 2021). Recent work has derived more mechanistic approaches to scaling isoprene emission across the landscape by determining how the fraction of emitters relates to mean climate conditions (Taylor et al. 2018) due to differential performance between isoprene-emitting and non-emitting species (Taylor, Smith, Slot & Feeley 2019). These studies were important as they identified the emission capacity of isoprene, and in a few cases also of monoterpenes, and certainly contributed to the overall body of work from nearly 30 years of research on modeling isoprenoid emission in Amazonia.
Yet, it is also known that isoprenoid composition is conserved within plant species, but that these compounds’ emission capacity may vary significantly within species and individuals with photosynthetic capacity, carbon and nutrient investment tradeoffs, habitat, and the environment (Harrison et al. 2013). This variability in emission capacities is an essential factor to explain why we observe seasonal variation in isoprenoid emission (see a synthesis of studies in Yáñez‐Serrano et al. 2020), which does not entirely follow the seasonal variation in solar radiation and temperature in central Amazonia (Alves et al. 2016, 2018).
Seasonal factors such as leaf demography and phenology are important drivers of variability in leaf emission capacities and the composition of emitted isoprenoids. During early leaf development, young leaves synthesize less isoprene and more monoterpenes and sesquiterpenes (Gershenzon & Croteau 1991; Kuhn et al. 2004b), and the opposite occurs with leaf maturation (Alves, Harley, Gonçalves, Silva & Jardine 2014). This shift in emission composition results from physiological and ecological factors, which cannot be explained by atmospheric observations and direct abiotic effects alone. Variation in leaf physiology with ontogeny scales up through leaf age distributions — with a higher proportion of young leaves during the dry season (Lopeset al. 2016; Wu et al. 2016 ) — to influence seasonal variation in isoprenoid emission capacities and total ecosystem emissions (Alves, Harley, Gonçalves, Silva & Jardine 2014; Alveset al. 2016, 2018).
Faced with these reports, we can infer that most of the studies in Amazonia were focused on emission rates — canopy-level sensitivity concerning environmental factors and landscape-level sensitivity to leaf quantity and emitter fraction. Relatively little work has focused on mechanisms of variation in emission capacity,either within or between species, and these have focused exclusively on leaf age, and primarily on isoprene (Alves et al. 2014, 2016, 2018). There is still a lack of understanding of the different physiological roles of the other light-dependent isoprenoids, and, by measuring all of them across habitats and seasons within species, we can begin to infer conditions under which one or the other compound is favored and how this can be more accurately scaled to the ecosystem. Therefore, the determination of intraspecific variation in emission capacities of different isoprenoids is a critical knowledge gap that this study intends to address.
In this study, we present a uniquely comprehensive set of leaf-level isoprenoid emission measurements to characterize variation in emission capacities and chemical compositions within species, across habitats, and across seasons. We performed our measurements on trees of three hyperdominant species from central Amazonia — Protium hebetatum, Eschweilera grandiflora, and Eschweilera coriacea (ter Steege et al. 2013) — distributed along a topographic and edaphic environmental gradient at the Amazon Tall Tower Observatory (ATTO) site, during the wet and the dry seasons. Simultaneous measurement of photosynthesis and emissions allowed us to assess shifting isoprenoid investments in the context of the leaf carbon balance and the inferred availability of photosynthetic substrates for isoprenoid synthesis.