Volatile organic compounds (VOCs) produced by plants are typically lipophilic molecules with high vapor pressure. Hence, they can easily cross membranes and be released into the surrounding environment when diffusion barriers are lacking (Pichersky et al., 2006). As a result, copious quantities of constitutive and stress-induced plant volatiles are released into the atmosphere. Consequent to their release, VOCs act as critical mediators for within plant, between plants, plant-insect and plant-microbe communication (Holopainen, 2004; Dudareva et al., 2006; Heil, 2010; Ninkovic et al., 2020), which play a key role in shaping plant communities (Kegge and Pierik, 2010; Effah et al., 2019).

The composition of volatile bouquets is species-specific (Bracho-Nunez et al., 2011; Malik et al., 2018) yet influenced by biotic and abiotic environmental variables (Holopainen and Gershenzon, 2010; Clavijo McCormick, 2016; Copolovici and Niinemets, 2016). Herbivory has frequently been linked to increased emissions of terpenoids (Clavijo McCormick et al., 2019; Effah et al., 2020d) and lipoxygenase (LOX) products that account for over 50% of damage-induced volatiles (Holopainen, 2004). Damage-induced volatiles reduce herbivore loads and increase resistance to future attacks in natural systems (Kessler and Baldwin, 2001; Kessler et al., 2006; Hu et al., 2019). However, VOC emissions in response to herbivory can vary depending on the attacker’s density and identity, season, and the plant’s phenology (Clavijo McCormick et al., 2012, 2014; Effah et al., 2020d).

Belowground components like soil moisture and nutrients also have a strong influence on plant volatile emissions. However, the response of plants to water availability differs considerably between species. For instance, some plants release more VOCs in response to drought (Copolovici et al., 2014) while others significantly reduce their emissions (Lavoir et al., 2009), and outcomes are even more ambiguous for prolonged or different drought intensities (Wu et al., 2015; Pagadala Damodaram et al., 2021). Similarly, the impact of soil nutrients on VOC emissions depends on the nature of chemical compound but also plant species-specific and may even vary for different types of nutrients (Chen et al., 2008; Holopainen and Gershenzon, 2010; Effah et al., 2020a,c).

In contrast, studies investigating relationships between temperature and plant volatile emissions have shown more consistent results, with most demonstrating a strong positive relationship between high temperature (warming) and VOC emissions (Kramshøj et al., 2016; Lindwall et al., 2016; Tiiva et al., 2017; Effah et al., 2020c; Rinnan et al., 2020). This relationship can be explained by the direct effect of temperature on enzyme activities, biosynthetic pathways, stomatal conductance, and physicochemical properties of compounds; and indirectly through antagonistic effects on other variables like drought and plant phenology (Peñuelas and Staudt, 2010; Possell and Loreto, 2013).

In nature, environmental factors, including temperature, soil nutrients, and herbivore loads, fluctuate between seasons (Shiojiri and Karban, 2008; Omer et al., 2018; Effah et al., 2020c), which often correspond to changes in plants’ physiological status. Therefore, measuring VOCs can be a good indicator of plant responses to environmental changes associated with different seasons and provide valuable information about how plant communication changes over time. This is increasingly important since most of these environmental parameters are changing globally due to climate change (Collins et al., 2013; Pachauri et al., 2014). However, information about how these climate components influence plant volatile emissions, particularly for endemic plant species, is still limited. Such information is critical to predicting alterations in plant volatiles’ ecological roles, their interference with other atmospheric processes such as cloud formation and reacting with O₃, and will increase our understanding of the impact of climate change on plant communication networks.

Here, we explored seasonal VOC emission by the New Zealand native plant Dracophyllum subulatum Hook.f., (Ericaceae). This woody shrub is endemic to the North Island of New Zealand and an archetypal species of the frost flats found on the North Island Central Plateau that thrives in the cool climate and infertile volcanic soils (Smale, 1990; Smale et al., 2011). This evergreen plant typically grows up to 0.3–2.0 m tall and has paniculate inflorescence. The adult narrow grass-like leaves are 10–48 mm long (Figure 1) and have stomata only present on the abaxial surface. D. subulatum has 2–6 inflorescences, and its tiny reticulate seeds are dispersed by wind (Venter, 2009; de Lange, 2021).

Unfortunately, the occurrence of D. subulatum in this region has been reduced considerably by conversion to pasture, the planting of exotic invasive conifers and invasion by aggressive alien weeds such as Calluna vulgaris and Cytisus scoparius (Chapman and Bannister, 1990; Buddenhagen, 2000; Smale and Fitzgerald, 2004), which also impact the arthropod diversity associated with this plant (Effah et al., 2020b). Nevertheless, D. subulatum spp. are an important component of the unique New Zealand Flora, having associations with native fauna (Eagle, 2006). The plant retains its foliage throughout the year and has a broad flowering and fruiting period (de Lange, 2021), serving as a food source for various animals.

Previous studies on the ecology of D. subulatum demonstrate that the plant is resilient to fire and even suggest that occasional fires in the past bolstered its dominance by initiating new populations (Smale, 1990; Smale and Fitzgerald, 2004; Smale et al., 2011). Dracophyllum species are highly flammable and popularly referred to as “turpentine shrub” (Smale et al., 2011; Cui et al., 2020). Turpentine is composed of terpenes, with monoterpenes α- and β-pinenes being the main components (Mercier et al., 2009). The plant, therefore, likely produces other several secondary compounds that, along with terpenes, contribute to its flammability.

This study aimed to investigate the VOCs emitted by D. subulatum and explore the factors driving their emissions in nature. We measured the emissions of D. subulatum at different sites on the North Islands’ Central Plateau. VOCs were measured on four different occasions from December 2017 to September 2018 over several seasons. On each sampling occasion, we also collected information about biotic (herbivore damage) and abiotic factors (ambient temperature, soil moisture, and soil nutrients; N, P, K) and determined their relationship with the plant’s volatile emissions.

Our findings show that D. subulatum Hook. f. is rich in terpenoids, including α- and β-pinenes, which may contribute to the popular generic name of “turpentine shrub” and its high flammability (Smale et al., 2011). Information about volatile storage and secretory structures such as glandular trichomes and resin ducts in D. subulatum is scarce. However, the enormous amount of terpenes observed in our study and the plant’s high flammability, as reported earlier, are typical of terpene-storing species (Llusià et al., 2006; Pausas et al., 2016; Della Rocca et al., 2017). Our results also reveal a clear seasonal pattern in VOC emissions, with total emissions ranging from 22.27 to 545.75 ng gDW^–1 h^–1 during winter and summer seasons, respectively. The findings further demonstrate that several environmental variables, including ambient temperature, soil water content, P and K availability, differ between sampling occasions, corresponding with the observed variable emissions. We acknowledge that plant phenology can influence volatile emissions, too, and suggest future studies to explore the phenology of the plant in relation to its volatile emission. Unfortunately, D. subulatum’s phenology is poorly understood, but it is known to retain its foliage through the winter and have broad flowering and fruiting periods (de Lange, 2021). Therefore, we will focus on the effect of environmental variables.

Ambient temperature was significantly higher during the summer months and corresponded with lower soil moisture content (Figure 5). This was also true when comparison was made separately for night- and daytime temperatures (Supplementary Figure 2). Temperature had a consistently strong positive relationship with the production of monoterpenoids, sesquiterpenes, fatty acid derivatives, aldehydes, other esters, and alcohols, reflecting higher amounts of all chemical classes during summer. This is consistent with previous reports showing increased plant volatile emissions in response to elevated temperature (warming). Examples include strong increasing effects of warming on terpenoids and other VOCs emission from Arctic ecosystems (Rinnan et al., 2020), stimulation of monoterpenes by night-time warming in heath ecosystems (Tiiva et al., 2017), and increased emissions of major volatile classes by the L. scoparium in association with high temperatures (Effah et al., 2020c). Other authors (e.g., Gouinguené and Turlings, 2002; Kivimäenpää et al., 2016; Kramshøj et al., 2016; Lindwall et al., 2016) have also reported positive effects of higher temperatures on VOC emissions.

Although emissions typically correlate positively with temperature, there may be a threshold at which emissions will decline. For instance, Gouinguené and Turlings (2002) found that total VOCs emissions of Zea mays decreased when temperature was increased from 27 to 37°C. Analogously, heat stress caused massive emissions from Nicotiana tabacum at temperatures 52–54°C, but a decline in emissions at 55°C (Turan et al., 2019). Temperature, therefore, can have varied effects on VOC emissions, as summarized below.

Elevated temperature stimulates both gross and net plant primary productivity, which coincides with augmented isoprene emissions as simulated by the photosynthetic-based isoprene scheme (Yue et al., 2015). This implies that temperature can regulate the availability of substrates required for VOCs synthesis. Photosynthetic products like pyruvate and glyceraldehyde-3-phosphate supply energy for the formation of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) from which isoprene and terpenes originate via the 2-C-methyl-D-erythritol 4-phosphate (MEP) and mevalonate (MVA) pathways (Dudareva et al., 2006). This makes photosynthetic products the main constraints for the production of volatile compounds via these pathways, and several studies have found a positive correlation between gross photosynthetic capacity and isoprene emission (Kuhn et al., 2004; Possell et al., 2004; Lahr et al., 2015).

High temperatures also directly affect biosynthetic enzymes and the expression of their associated VOCs regulatory genes, thus impacting emissions. For example, the isoprene synthase (IspS) activity responded strongly to temperature, with exponentially increasing activity between 15 and 40°C. IspS activity reached an optimum at 50°C and correlated strongly with VOC emission from Quercus robur (Lehning et al., 1999). Similarly, heat stress strongly induced IspS activity in Populus alba (Sasaki et al., 2005) and transgenic Arabidopsis thaliana plants (Sasaki et al., 2007), corresponding with higher emission levels. However, VOCs regulation by enzyme activities and related gene expressions is a complicated mechanism (Ibrahim et al., 2010) and could be species and compound dependant. For example, increasing night-time temperature led to variable expressions of the genes 1-deoxy-D-xylulose 5-phosphate reductoisomerase (DXR), 1-deoxy-D-xylulose 5-phosphate synthase (DXS), IPP, and hydroxymethylglutaryl CoA reductase in Betula pendula and Populus tremula, which coincided with the differences in volatile emissions (Ibrahim et al., 2010). Both day and night-time temperatures were significantly different between the sampling occasions in our study (Supplementary Figure 2) and could relate to the observed emissions.

Elevated temperature is linked to increased stomatal conductance (Urban et al., 2017) and can directly affect the diffusion rate of volatile compounds at a given time. Stomatal opening is expected to be accompanied by large bursts of emissions, mainly from storage pools (Niinemets et al., 2002; Hüve et al., 2007). However, the intensity of stomatal control on emissions is contingent on the pool size of volatiles stored in lipid and water phases (Harley, 2013), with noticeable control on more soluble compounds like alcohols and carboxylic acids compared to highly volatile compounds (Niinemets et al., 2002, 2004; Niinemets and Reichstein, 2003).

Temperature can also indirectly affect volatile emissions through its influence on different biotic and abiotic variables. For example, elevated temperature can amplify the severity of drought by enhancing soil moisture evaporation. In our study, higher temperatures in summer corresponded to relatively lower soil water content (Figure 5) and may account for the observed higher emissions in summer, as shown in earlier reports (Copolovici et al., 2014; Campbell et al., 2018). We indeed observed a strong negative relationship between soil moisture content and the emissions of aldehydes and other esters and a moderate negative relationship with sesquiterpenes (Figure 6). Another potential indirect effect of temperature on the variable emissions observed in this study could be the interaction between temperature and plant phenology. D. subulatum’s phenology would have been changing between the four sampling occasions. For example, flowering occurred from November to March. This covers the summer months in our study and may impact the composition of volatile blends measured during this period. Although we did not directly test this in the present study, higher temperatures have been shown to increase the floral volatiles from some plants, for instance (Hu et al., 2013; Farré-Armengol et al., 2014).

Other environmental factors like soil nutrient availability also affect plant volatile emissions. We found differences in K and P between seasons, with higher levels measured in ES (Figure 5), which coincide with the higher volatile emissions at this sampling time. We also found that increased K levels positively relate to the emissions of fatty acid derivatives, aldehydes, and other esters but negatively correlated with monoterpenoids and sesquiterpenes. On the other hand, high P levels were associated with increased emissions of monoterpenoids, other esters and sesquiterpenes (Figure 6). The mechanisms behind the relationships between soil nutrients and VOC emissions are poorly documented. Phosphorus, however, is crucial in terpenoid production because it is a significant component of ATP and NADPH required for terpenoid synthesis and an essential constituent of terpenoid precursors (IPP and DMAPP) (Ormeno and Fernandez, 2012). In comparison, nitrogen stimulates electron transport rate and photosynthesis, which supplies energy and carbon substrate for isoprenoid production (Ormeno and Fernandez, 2012). Like our results, other studies have reported a positive relationship between nutrient availability and VOC emissions. These include increased total emissions from Z. mays under high fertilization (Gouinguené and Turlings, 2002), enhancement of α-thujene, acetic acid and LOX compounds with increasing N availability from Brassica napus (Veromann et al., 2013) and induced higher emissions by infested conventional B. napus grown at a high-soil nutrient level (Ibrahim et al., 2008). Lower emission in nutrient-depleted soils is likely due to the lack of substrates required to synthesize compounds, as previously suggested by Gouinguené and Turlings (2002).

This study is a contribution to our understanding of VOC emissions of native plant species in the wild and the environmental factors impacting them, but other aspects require further research. Future studies should investigate the associations between plant phenology and its VOC emissions in nature. We also recommend more studies to explore the potential links between increased photosynthesis in warmer summers and VOC emissions. The potential effects of global warming on plant biochemical and physiological processes and their ecological impacts also deserve attention. Lastly, we acknowledge the possibility of our results being influenced by microbes and unidentified belowground processes and encourage future research to investigate the effect of such organisms and processes on VOC emissions over several seasons.

For the first time, we characterized the VOCs emissions of New Zealand’s endemic D. subulatum and showed that this plant is a prolific emitter of terpenoids. Among the measured parameters, D. subulatum’s emissions showed a clear seasonal pattern with higher emissions during summer, mainly related to higher temperatures. Although other environmental variables such as soil moisture content and nutrients were associated with changes in emissions, the relationships were compound-dependent.

Climate change is expected to impact soil chemistry, air temperature, and plant development. Our results support the claim that the current warming trend is likely to increase plant volatile emissions globally and possibly affect their ecological roles (Peñuelas and Staudt, 2010; Rinnan et al., 2020). For instance, changes in emission in response to climate change can shape competitive outcomes for plants (Wang et al., 2010). Modification in emissions can also influence the communication between plants, herbivores and natural enemies of herbivores (Clavijo McCormick, 2016), which could be significant in shaping plant communities (Kegge and Pierik, 2010; Effah et al., 2019). Furthermore, volatile compounds including isoprene, monoterpenes, and green leaf volatiles play critical roles in improving thermotolerance, resistance to cold-stress, and serve as antioxidants (Holopainen, 2004; Zuo et al., 2017; Cofer et al., 2018). Therefore, further investigation is needed to establish whether altering emissions by climate change could enhance or impair VOCs’ protective functions.

Many volatile compounds also interfere with some atmospheric processes, reacting with O₃, nitrogen oxides and hydroxyl radicals, and are linked to cloud formation (Blande et al., 2014; Zhao et al., 2017). The products of these reactions can be redeposited on plants and could affect plants communication with their environment and atmospheric quality. A recent study showed that α-pinene’s oxidation products are deposited on plants and re-emitted, affecting plants acceptability to herbivores (Mofikoya et al., 2020). This research area requires further investigation to increase our knowledge of the impact of climate-induced emissions on plant-plant, plant-insect, and plant-microbe interactions, especially in vulnerable and threatened native ecosystems.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

EE and ACM conceived the project, collected the data, and led the writing of the manuscript. DB, PP, MP, and JH contributed to the final design and data collection. EE analyzed and interpreted the data with the advice of ACM. All authors were involved in editing and provided critical comments on the manuscript, and approved it for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.