An increasingly scented world

We know that, at least in the short term, a rise in temperature exponentially increases the emission rates of most biogenic volatile organic compounds (BVOCs). It does so by enhancing the enzymatic activities of synthesis, by raising the BVOC vapour pressure, and by decreasing the resistance of the diffusion pathway (Tingey et al., 1991). BVOC emissions are thus expected to increase strongly with globally rising temperatures (IPCC, 2007). By applying the most frequently used algorithms of emission response to temperature (Guenther et al., 1995), it is easy to estimate that climate warming over the past 30 yr (IPCC, 2007) could have already increased BVOC global emissions by 10%. A further 2–3°C rise in the mean global temperature, which is predicted to occur early this century (IPCC, 2007), could increase BVOC global emissions by an additional 30–45%. Furthermore, global warming in boreal and temperate environments not only means warmer average and warmer winter temperatures, but also implies an extended plant activity season (Peñuelas & Filella, 2001), further increasing the total annual emissions. The biological and environmental effects of such increases in emissions can be substantial (Peñuelas & Llusià, 2003). However, many investigations are conducted in the laboratory, and we still do not know very much about the emissions of BVOCs in the field in response to warming. We know even less about medium- to long-term responses of BVOC emissions to warming, and about some regions of our planet, such as the Arctic, that are likely to experience the most pronounced effects of climatic warming. Tiiva et al. have started to fill this gap in our knowledge in this issue of New Phytologist (pp. 853–863). They have measured the emissions of isoprene, the most emitted reactive BVOC from vegetation, at a subarctic heath that they have experimentally subjected to a 3–4°C air temperature enhancement for 8 yr. Tiiva et al. have measured increases in emissions ranging between 56 and 83% depending on the year. Their results confirm the percentage increases in emissions expected using the standard algorithms of Guenther et al. (1995). 'Increasing production and emission of BVOCs may be largely beneficial for plants, which are likely to gain increased protection in the face of abiotic stressors ...' Warming does not only have direct effects on BVOC emissions. It has numerous indirect effects. One of these is linked to the consequent changes in land cover. In environments that have winters with freezing temperatures, an increase in minimum winter temperature of 5°C is expected to increase the number of species able to grow there by 7–20% (Niinemets & Peñuelas, 2008). In fact, as a consequence of the warming of the last few decades, migrations of the tree-line northward and upslope and increasing abundance of deciduous woody shrubs in Arctic vegetation communities have already occurred (e.g. Tape et al., 2006). These vegetation changes lead to increasing amounts of leaf litter on the ground (Cornelissen et al., 2007), and this, in turn, brings extra nutrients to the soil (Rinnan et al., 2008). In their study, Tiiva et al. also tested the effects of an addition of mountain birch (Betula pubescens ssp. czerepanovii) litter during the 7–8 yr of their experiment, thus simulating the warming-induced expansion of deciduous shrub species and migration of the tree-line, and therefore an increased availability of nutrients. They did not find any significant effect on BVOC emissions. The absence of response to the litter addition does not fit the hypothesis of increased emissions under increasing litter-fall which was based on observations of increased isoprene emission under nutrient fertilization (e.g. Harley et al., 1994), and on the expected increases in carbon fixation and in the activity of the enzymes responsible (Litvak et al., 1996). It might be that the expected positive response is still limited by temperature in this Arctic ecosystem, and that positive effects may thus only be found in more meridional latitudes which are not energy limited. However, there have been few studies in warmer regions, and only variable and complex species-specific and compound-specific responses have been reported in meridional latitudes (Blanch et al., 2007). The shifts in species dominance and the changes in land use and cover that are currently occurring, and that are projected for the next few decades, can also affect BVOC emissions dramatically because these emissions are species-specific, and many of the plant species migrating to northern latitudes and higher altitudes are strong emitters of BVOCs such as isoprene and monoterpenes. For instance, most broad-leaved species of Populus or Quercus and, essentially, all conifers are important emitters of volatile isoprenoids (Niinemets & Peñuelas, 2008 and references therein). Increases in emissions in response to changes in land cover are expected to occur not only in Arctic regions, but in many regions around the globe. For example, in some tropical areas the rainforest has been replaced by plantations; for instance, palm plantations in Malaysia and rubber tree plantations in southern China (Wang et al., 2007). Not only do these plantations emit up to 10 times more isoprenoids than natural forest, but some of their emitted compounds can respond more strongly to warming (S. Owen et al., unpublished; Wang et al., 2007). Other land cover changes might also greatly increase BVOC emissions, for example abandonment of agricultural land in temperate regions, and subsequent aforestation with evergreens such as Eucalyptus, Quercus or Pinus, which are strong emitters of BVOCs throughout the year. Warming, eutrophication and land cover changes are not the only global environmental changes that can potentially increase BVOC emissions (Fig. 1). Enhanced UV-B radiation may substantially increase emissions, as reported for the Arctic in another recent study by Tiiva et al. (2007). The rising atmospheric CO2 concentrations are likely to increase the productivity and standing biomass of plants, at least in the short term, and hence also facilitate further production and emission of BVOCs. However, the number of studies on CO2–BVOC interactions is still limited, and their results are sometimes contradictory. It is not clear whether or not elevated CO2 per se increases the release of BVOCs (Peñuelas & Llusià, 2003). In fact, recent work indicates that increasing CO2 concentration may uncouple isoprene emission from photosynthesis (the carbon source for BVOCs), and inhibit isoprene emission at the leaf level (Possell et al., 2005). Changes in water availability also affect BVOC emissions. The decrease of isoprenoid emission in response to severe droughts, possibly through effects on protein levels or substrate supply (Fortunati et al., 2008), might largely offset the predicted impact of rising temperatures on the emission of isoprenoids in arid and semiarid terrestrial ecosystems suffering more frequent severe droughts. Scheme of the possible effects of global environmental changes on plant volatile organic compound emission rates and the still unknown biological and environmental consequences. Thus, there is still a lack of precise and complete data addressing the question of what the combined effects of these components of global change, and many others that have not been considered here, such as changing irradiance, stillness and air pollution, will be on BVOC emissions. For example, we still wonder whether or not BVOC emissions will acclimate to long-term warming. Moreover, the complex interaction of each one of these global change drivers with other biotic and abiotic factors introduces a great deal of variability into the responses to global environmental changes. However, our current knowledge seems to indicate that the most likely overall response will be an increase in BVOC emissions. At this point, the reader may ask the question: so what? Why do we bother about these changes in the amounts of emitted BVOCs? We bother because the effects, both biological and environmental, can be far-reaching and substantial. Increasing production and emission of BVOCs may be largely beneficial for plants, which are likely to gain increased protection in the face of abiotic stressors such as the high temperatures themselves, air pollution, high irradiance, oxidative stress, or mechanical wounding. Will the degree of protection increase in proportion to the stress? This is a challenging issue to address. The increased emissions will also affect plant communication and relationships with other organisms. Plants with increased emissions may have enhanced deterrence against pathogens or herbivores, enhanced allelopathic effects against neighbors, enhanced attraction of both pollinators and herbivore predators and parasitoids, or enhanced antimicrobial defense. If plants become more scented, we will experience a dramatically changed world in which olfactory cues are much more important as ecological and evolutionary factors. But will all these expected enhancements in defenses actually occur, or will the organisms receiving the enhanced BVOC messages from plants be puzzled by the altered emissions? Whatever the direction of the responses, the consequences for the structure and functioning of life on our planet may be very significant. The changes resulting from increases in BVOC emissions will not only be biological. Increased emissions will also affect the atmosphere biogeochemically and biophysically. The loss of carbon as isoprene to the atmosphere in the study by Tiiva et al. was in general less than 0.1% of the net ecosystem carbon assimilation at the heathland study site. However, the exponential response of BVOC emissions to temperature translates into a 3- to 6-fold increase for a 10°C rise in temperature (Q10 value), whereas the Q10 of typical biochemical reactions such as photosynthesis is only 2–3 (Niinemets, 2004). Therefore, during the periods of the highest isoprene emissions in Tiiva et al.'s study, the carbon loss reached 1% of the mean net ecosystem carbon assimilation rate. This amount of carbon loss is at the same level as previously measured, but BVOC fluxes can account for 5–10% or even more of total net carbon exchange, especially under stressed conditions (Peñuelas & Llusià, 2003). Therefore, these BVOC emissions may represent a significant plant carbon loss on an ecosystem basis and on a global basis. The global average BVOC emission for vegetated surfaces is 0.7 g C m−2 yr−1 but could exceed 100 g m−2 yr−1 in some tropical locations (Guenther, 2002). BVOC emissions may become an even more significant component in local and regional carbon budgets as they increase in response to global changes. BVOCs influence the oxidizing potential of the troposphere by affecting the concentration of the main atmospheric oxidant, the hydroxyl radical. Thus, increased BVOC emissions will affect crucial features of atmospheric chemistry such as ozone dynamics, aerosol formation, carbon monoxide production and methane oxidation (Peñuelas & Llusià, 2003, and references therein) The effects will thus be multiple; for example, currently, there is only limited formation of spring or winter smog in northern and high-altitude habitats, but conditions might become conducive for smog production if warming continues. Furthermore, these increases in the emissions of BVOCs might make a major contribution (via positive or negative feedback) to the complex processes associated with global warming itself. Until recently it was thought that the short lifetime of BVOCs would preclude them from having any significant direct influence on climate. However, there is emerging evidence that this influence might be significant at different spatial scales, from local to regional and global, through aerosol formation and direct and indirect greenhouse effects. BVOCs generate large quantities of organic aerosols (Claeys et al., 2004) that could affect climate significantly by forming cloud condensation nuclei. As a result, there should be a net cooling of the Earth's surface during the day because of radiation interception. Apart from the direct local BVOC greenhouse effect, which has detectable effects only when canopy-scale BVOC emissions are high, an additional global indirect greenhouse effect must also be considered because BVOCs increase ozone production and the atmospheric lifetime of methane, and hence enhance the greenhouse effect of both these gases. Whether the increased BVOC emissions will cool or warm the environment will depend on the relative weights of the negative (increased albedo) and positive (increased greenhouse action) feedbacks (Peñuelas & Llusià, 2003). The study of Tiiva et al. thus shows increased BVOC emissions with warming in field conditions and, moreover, it reminds us of the many unanswered questions regarding the relationship between global change and BVOCs. We still do not know how much BVOC emissions will increase in response to the different global change drivers, but we know that we must focus not only on the substantial effects of climate warming on BVOC emissions, but also on the effects of the other global change drivers, especially increasing changes in land cover. The expected increases in BVOC emissions are likely to have far-reaching biological and environmental effects, warranting interactive interdisciplinary research by biologists, physicists and chemists at foliar, ecosystem, regional and global scales.