This is the accepted version of the article: Mu, Zhaobin; Llusià Benet, Joan; Liu, Daijun; [et al.]. «Profile of foliar iso- prenoid emissions from Mediterranean dominant shrub and tree species under experimental nitrogen deposition». Atmospheric environment, Vol. 216 (Nov. 2019), art. 116951. DOI 10.1016/j.atmosenv.2019.116951 This version is avaible at https://ddd.uab.cat/record/218214 under the terms of the license 1 Profile of foliar isoprenoid emissions from Mediterranean dominant shrub and tree 2 species under experimental nitrogen deposition 3 Zhaobin Mua,b,*, Joan Llusiàa,b, Daijun Liua,b, Romà Ogayaa,b, Dolores Asensio a,b, Chao 4 Zhanga,b, Josep Peñuelasa,b 5 a 6 CSIC, Global Ecology Unit CREAF-CSIC-UAB, E08193 Bellaterra, Catalonia, Spain b 7 CREAF, E08193 Cerdanyola, Catalonia, Spain 8 9 *Correspondence 10 Zhaobin Mu 11 Email: zhaobin@creaf.uab.cat 12 13 ABSTRACT 14 Biogenic volatile organic compounds play important roles in atmospheric chemistry, and 15 their emissions can be greatly influenced by the variations in environmental conditions 16 and physiological activities caused by continuously increasing global nitrogen (N) 17 deposition. However, this influence is still poorly understood, especially in a natural 18 ecosystem. We conducted a one-year (2015-2016) experiment adding N deposition (60 19 kg N ha-1) with fertilization to a Mediterranean shrubland dominated by Erica multiflora 20 and a Mediterranean forest dominated by Quercus ilex and compared the seasonal and 21 daytime photosynthetic rates (A), stomatal conductances (gs) and rates of isoprenoid 22 emission with control (2015-2016) and pre-treatment (2014-2015) plots. N fertilization 23 increased A in warm weather as soil moisture increased, and assimilation became 24 saturated when the environment was sufficiently favorable, and excess soil N 25 significantly restrained A in cold weather. The plants were much more sensitive to soil 26 water availability than N content and terpene emissions increased synergistically due to 27 heat and drought stress in hot weather. N fertilization did not significantly affect isoprene 28 emission but significantly increased total terpene emissions and decreased the diversity 29 of terpenes. Our results suggest a successful acclimation of plants by emitting more 30 isoprenoids under environmental stress and that N deposition will further stimulate 31 emissions as the Mediterranean region becomes warmer and drier. The results highlight 32 the necessity for predicting the most realistic future of ecosystems under global 33 environmental change and for assessing the impacts of multiple factors acting in concert 34 on plant physiological and ecosystem functioning including biogenic VOC emissions. 35 36 Keywords Nitrogen deposition; Climate change; BVOC emissions; Isoprenoids; Erica 37 multiflora; Quercus ilex. 38 39 1. Introduction 40 Global environmental change (GEC) is accelerating and becoming more intense around 41 the world. The ecological impacts of the main drivers of GEC, such as climate change, 42 nitrogen (N) deposition, land-use changes and species invasions, are all expected to 43 become more conspicuous as human exploitation and pollution of the environment 44 continue to increase (Sala et al., 2000). Climate change has been the most widely and 45 thoroughly studied due to its influence. Other factors, however, are also receiving 46 increasing attention, especially N deposition. Global N deposition has increased in recent 47 decades and will likely double the existing levels to as much as 230 Mt N y-1 by 2050 48 (Galloway et al., 2004; Llusià et al., 2014; Yuan et al., 2017), especially in India, China 49 and Europe, which are also the major global manufacturers and emitters of reactive N 50 (Liu et al., 2011; Yuan et al., 2017). Higher N deposition from anthropogenic sources 51 such as fertilizers, combustion of fossil fuels and cattle residuals (Blanch et al., 2009; 52 Galloway et al., 2008; Peñuelas et al., 2013) has an important long-term impact on 53 ecosystem structure and function (Phoenix et al., 2012; Meunier et al., 2016), including 54 some potential threats to soil acidification that decreases the defensive capacity and 55 biodiversity of plants (Phoenix et al., 2012; Valliere and Allen, 2016; Zhang et al., 2017). 56 GEC research should therefore not ignore N deposition (Janssens et al., 2010). 57 More than 100 000 chemical products have been identified in plants (Dicke and Loreto, 58 2010), including many biogenic volatile organic compounds (BVOCs) (Loreto and 59 Schnitzler, 2010) that contribute about 90% of the global emission of volatile organic 60 compounds (VOCs) into the atmosphere (Guenther et al., 1995). BVOCs are a crucial 61 group of plant compounds due to their important role in the ecology of plants (Dicke and 62 Loreto, 2010; Niederbacher et al., 2015). Isoprenoids, which account for >80% of the 63 total emission of BVOCs (Guenther et al., 1995), are the most dominant components of 64 the biosphere-atmosphere exchange of BVOCs (Sharkey and Monson, 2017) because 65 they play important roles in plant responses to both abiotic and biotic stresses (Loreto and 66 Schnitzler, 2010; Holopainen and Gershenzon, 2010; Peñuelas and Llusià, 2003) and in 67 the chemistry of the atmosphere (Dicke and Loreto, 2010; Niederbacher et al., 2015). 68 Plants are essential components of complex communities that include organisms ranging 69 from microorganisms to mammals and have evolved intricate mechanisms for using 70 BVOCs to defend against enemies such as pathogens, parasitic plants and herbivores and 71 for interactions with other plants and beneficial organisms such as pollinators and 72 predators (Dicke and Loreto, 2010; Farré-Armengol et al., 2015). BVOCs are reactive 73 hydrocarbons that contribute to the production of tropospheric ozone in the presence of 74 NOx compounds and sunlight (Dicke and Loreto, 2010; Tiiva et al., 2017) and to the 75 formation and growth of aerosol particles in the atmosphere (Paasonen et al., 2013; Tiiva 76 et al., 2017). 77 Most BVOC emissions are associated with photosynthesis (Monson and Fall, 1989) 78 and account for a relevant amount of the carbon fixed by photosynthesis. Under stressed 79 conditions, a larger proportion of fixed carbon is often devoted to isoprenoid emissions 80 (Loreto et al., 2001; Vallat et al., 2005; Blanch et al., 2009), besides, the emission of some 81 terpenes may be limited by stomatal conductance (Niinemets et al., 2002; Harley et al., 82 2014) due to strong stomatal sensitivity (Niinemets et al., 2002). An abundant input of N 83 to the soil is usually beneficial for plants due to an increase in photosynthetic activity 84 (Häikiö et al., 2007; Handley and Grulke, 2008), and plants will allocate proportionately 85 more carbon toward growth and less toward carbon-based secondary compounds used for 86 defense (Loreto and Schnitzler, 2010; Peñuelas and Llusià, 2003). Low levels of N 87 addition, however, can increase terrestrial plant productivity (Vitousek and Howarth, 88 1991; Pivovaroff et al., 2016), and high levels exceeding critical loads may decrease 89 productivity and render plants more susceptible to environmental stressors (Cardoso- 90 Vilhena and Barnes, 2001; Yuan et al., 2017), such as excess of ozone, which induce 91 rapid stomatal closure (Kollist et al., 2007; Vahisalu et al., 2008; Li et al., 2017) and low 92 stomatal vitality (Paoletti and Grulke, 2010; Li et al., 2017). Immediate photosynthetic 93 responses during and after stress are primarily driven by modifications in stomatal 94 conductance (Kollist et al., 2007; Vahisalu et al., 2008), so limitations on CO2 uptake 95 may substantially decrease photosynthesis, which would also tend to affect BVOC 96 emissions. Even though the assimilative response to N deposition, is not uniform and 97 varies among species or even ecosystems (Bobbink et al., 2010; Alvarez-Clare et al., 2013) 98 because of differences in resource availability, acquisition and strategies of use (Vourlitis 99 and Pasquini, 2009; Pivovaroff et al., 2016), it may affect the emission of BVOCs by 100 directly affecting their biosynthesis or by affecting primary physiological metabolism 101 (Loreto and Schnitzler, 2010; Pivovaroff et al., 2016). 102 Environmental experiments have been widely used to predict potential physiological 103 and phenological profiles, with general considerations of realistic changed conditions and 104 their interactions to simulate future environmental scenarios (Beier et al., 2012; Leuzinger 105 et al., 2011; Ogaya et al., 2014). Increases in temperature caused by climate change and 106 increases in soil N content are thus two GEC factors that can interact. Climate change is 107 expected to increase the temperature by 1.5–3.7 °C and the frequency and intensity of 108 drought in many parts of the world by 2100, particularly during the summer and normally 109 drier months (IPCC, 2014; Tiiva et al., 2017). Rates of BVOC emission can increase 110 under moderate drought in conjunction with warmer temperatures caused by climate 111 change and help plants to resist stress (Breshears et al., 2005; Allen et al., 2015), but rates 112 can decrease under severe environmental conditions (Gershenzon et al., 1978; Llusià et 113 al., 2011; Llusià et al., 2013). Higher isoprenoid emissions can be expected in the warmer 114 and drier conditions projected by climatic and ecophysiological models for the coming 115 decades in the Mediterranean region (Peñuelas and Llusià, 2001; IPCC, 2014). The 116 important interaction between the emission of phytogenic VOCs and climate change has 117 also elicited great interest in detecting the effects of realistic combinations of other GEC 118 factors on the emission of BVOCs from vegetation. Current models predict that soil N 119 content, air temperature and soil moisture are expected to change concurrently, and 120 understanding their interactive influences on BVOC emissions is essential for predicting 121 future BVOC dynamics (Llusià et al., 2014; Zhang et al., 2017; Tiiva et al., 2017). The 122 triple threat of N deposition, drought and warming, representing the most realistic future, 123 on isoprenoid emissions have not been investigated but could provide valuable insights 124 for understanding the dynamics of isoprenoid emissions in the near future. These 125 changing GEC factors, in a long-term perspective, could also shift the composition and 126 structure of ecosystems (Valolahti et al., 2015), which could also influence emission 127 profiles. 128 The influence of N deposition on BVOC emissions is poorly understood. Most studies 129 of the effects on isoprenoids have been short-term or conducted in warm seasons or in 130 greenhouses (Blanch et al., 2007; Carriero et al., 2016) or open-top chambers (Llusià, et 131 al., 2014; Yuan et al., 2017), and the results have been inconsistent (Blanch et al., 2007; 132 Peñuelas and Staudt, 2010; Yuan et al., 2017) due to the variations in response to N 133 availability by different plant species. Yuan et al. (2017) reported that a moderate amount 134 of N fertilization (50 kg N ha-1) increased the emission of isoprene in Cathay poplar 135 (Populus cathayana), Llusià et al. (2014) reported that the emission of terpenes decreased 136 in two Mediterranean leguminous species (Ornithopus compressus and Trifolium 137 striatum) at similar N levels (40 kg N ha-1). Isoprene emission was also stimulated in 138 velvet bean (Mucuna sp.) (Harley et al., 1994), aspen and white oak (Litvak et al., 1996) 139 with increasing N availability, supporting the existence of links between foliar N status 140 and isoprene synthase activity (Litvak et al., 1996). Blanch et al. (2007) found that a high 141 amount of N fertilization (250 kg N ha-1) decreased the emission of terpenes in Pinus 142 halepensis but had no influence on Quercus ilex, and moderate drought increased the 143 terpene emissions of fertilized plants for both Mediterranean species in summer. 144 Kivimäenpää et al. (2016) observed that higher N availability increased the emission of 145 some minor terpene compounds in Scots pine and some major terpene compounds in 146 combination with warming in summer. Carriero et al. (2016) reported that individual 147 monoterpenes had a compound-specific response to N in silver birch (Betula pendula), 148 due to different pathways of biosynthetic formation of the emitted compounds 149 (Kesselmeier and Staudt, 1999; Niinemets et al., 2004), but N fertilization did not 150 significantly affect total terpene emissions. Ormeño et al. (2009) suggested that the 151 positive relationship between N fertilization and terpene emissions for two Mediterranean 152 species, Rosmarinus officinalis and Quercus coccifera, only occurred at optimal soil N 153 conditions. 154 Erica multiflora and Quercus ilex are two widespread species in western and central 155 Mediterranean shrublands and forests, respectively, and both are important isoprenoid 156 emitters, dominated by isoprene and terpenes, respectively (Kesselmeier and Staudt, 157 1999). Our study was relatively long-term and conducted in fields, which investigated 158 the seasonal and daytime variations of net photosynthetic rate (A), stomatal conductance 159 (gs) and rate of isoprenoid emissions in these two dominant species at two study sites, a 160 Garraf shrubland and a Prades forest (Llusià et al., 2011; Llusià et al., 2013; Ogaya et al., 161 2014). The aim of this study was to assess the realistic and systematic response of 162 isoprenoid emissions in these Mediterranean shrubland and forest ecosystems to 163 experimental N deposition and thus to improve the estimations of the emission dynamics 164 expected in the coming decades by models that take GEC into consideration. 165 166 2. Material and methods 167 2.1. Study sites and species description 168 The study was carried out in the Garraf and Prades Mountains in Catalonia, northeastern 169 Spain (Figure S1). The climate and vegetation at the two sites are typically Mediterranean. 170 Garraf Natural Park is a dry shrubland (Rosmarino-Ericion) south of Barcelona 171 (41°18′08″N, 1°49′05″E; 210 m a.s.l.). The vegetation has a coverage of 50–60% and a 172 maximum height of 70 cm. The Prades Mountains are in southwestern Catalonia 173 (41°20′42″N, 1°02′04″E; 970 m a.s.l.) and about 30 km from the Mediterranean Sea. The 174 Prades sampling site is a holm oak forest with tree heights between 1.5 and 10 m (Bolòs 175 and Vigo, 1990; Llusià et al., 2013; Ogaya et al., 2014). Both sites contain abundant 176 evergreen and deciduous species, and the dominant species are common throughout the 177 western Mediterranean Basin (Bolòs and Vigo, 1990; Llusià et al., 2013; Ogaya et al., 178 2014). We chose one dominant species at each site: E. multiflora in Garraf and Q. ilex in 179 Prades. 180 181 2.2. Experimental design 182 Six plots (5 × 4 m) were randomly established in Garraf in three replicate blocks, with 183 each block having one N and one control plot. Four plots (15 × 10 m) were established in 184 Prades at the same altitude along the slope, two as N and two as control plots. The N 185 treatments were fertilized with 60 kg N ha-1 using NH4NO3 (Fluka, Buchs, Switzerland), 186 and the control treatments were not fertilized. The N-treatment plots were fertilized 187 homogeneously three times each season (total of 15 kg N ha-1 season-1) in 2015. 188 Emissions were measured in the centers of the plots to reduce edge effects, with the outer 189 0.5 and 1 m serving as open buffer zones for the plots in Garraf and Prades, respectively. 190 Emissions were measured from spring 2014 to winter 2016 twice a day, morning (9:00– 191 12:00 solar time) and midday (13:00–16:00 solar time), on three consecutive days in the 192 middle of each season. The N deposition for the two dominant species had thus been 193 manipulated for one year, but we also obtained pre-treatment data for the previous year. 194 Emissions from sunlit and healthy E. multiflora needle clusters and Q. ilex leaves were 195 measured from three random plants in each plot. Air temperature was measured by an 196 automatic meteorological station, and soil moisture was measured by time-domain 197 reflectometry (Delta-T Devices Ltd, Cambridge, England), both about every 30 min on 198 the day of sampling. 199 200 2.3. Gas-exchange measurements and sampling of isoprenoid emissions 201 A and gs were measured and isoprenoid emissions were collected simultaneously using 202 a Licor-6400XT gas-exchange system (LI-COR, Lincoln, Nebraska USA). A and gs were 203 measured at a photosynthetic quantum flux density (PPFD) of 1000 ± 20 μmol m-2 s-1 204 under a controlled CO2 concentration of 400 ± 2 ppm. One E. multiflora needle cluster or 205 one Q. ilex leaf was enclosed in a clip-on gas-exchange cuvette with a surface area of 2 206 cm2. The emitted isoprenoids were pumped from the cuvette through a stainless-steel tube 207 (89 mm in length and 6.4 mm external diameter) manually filled with adsorbents (115 mg 208 of Tenax® TA and 230 mg of SulfiCarb®) separated by sorbent-retaining springs, fixed 209 using gauze-retaining springs and closed with air-tight caps (Markes International Inc. 210 Wilmington, USA). The flow across the tubes was measured with a Bios Defender 510 211 flow meter (Bios International Corporation, Butler, USA) and controlled at about 300 mL 212 min-1 with a metallic valve. The flow was generated using a Q-MAX air-sampling pump 213 (Supelco, Bellefonte, USA), and the sampling time was 10 min. The hydrophobic 214 properties of the activated adsorbents minimized any sample displacement by water, 215 without chemical transformation in the tube. Isoprenoid concentrations were determined 216 by reference to trapped standards (α-pinene, β-pinene, 3-carene, limonene, sabinene and 217 dodecane). The tubes were conditioned before isoprenoid sampling with a stream of 100 218 mL min-1 of purified helium at 350 °C for 35 min. The trapping and desorption 219 efficiencies of liquid and volatilized standards such as α-pinene, β-pinene or limonene 220 (the main terpenes accounting for about 65–90% of total emissions) were near 100%. 221 Blank samples of air without leaves in the cuvette were collected for 10 min immediately 222 before each measurement. The sampled parts of the leaves were cut to the perimeter 223 enclosed in the Licor-6400XT cuvette and stored in a portable cooler at 4 °C, taken to the 224 laboratory and oven-dried at 60 °C to constant weights. The metallic tubes (with trapped 225 BVOCs) were stored at 4 °C until analysis. 226 227 2.4. Isoprenoid analyses 228 The isoprenoids were analyzed using a GC-MS system (HP59822B, Hewlett Packard, 229 Palo Alto, USA) with an automatic sample processor (Combi PAL, FOCUS-ATAS GL 230 International BV 5500 AA Veldhoven, The Netherlands). The desorber was an OPTIC3 231 injector (ATAS GL International BV 5500 AA Veldhoven, The Netherlands), and 232 samples were applied to a 30 m × 0.25 mm × 0.25 µm film capillary column (HP-5, 233 Crosslinked 5% pH Me Silicone; Supelco, Bellefonte, USA). A detailed description of 234 the chromatographic method is provided by Mu et al. (2018). 235 The terpenes were identified by comparing their retention times with those of standards 236 from Fluka (Buchs, Switzerland), published spectra, GCD ChemStation G1074A HP and 237 the wiley7n mass-spectra library. The concentrations of common terpenes such as α- 238 pinene, 3-carene, β-pinene, myrcene, limonene, and sabinene were determined from 239 calibration curves every five analyses using four terpene concentrations (r2 > 0.99 for the 240 relationships between the signal and terpene concentrations). The most abundant terpenes 241 had very similar sensitivities, with differences <5% among the calibration factors. 242 243 2.5. Statistical analyses 244 Data variables were analyzed (ANOVA) using STATISTICA v.8.0 (StatSoft, Inc., 245 Tulsa, USA). Statistical differences between treatments were identified with a Student’s 246 t-test. Differences were considered significant at P < 0.05. The significance of the effects 247 of season, treatment and sampling time were determined by a repeated-measures ANOVA. 248 Regression analyses were conducted using Sigma Plot v. 14.0 for Windows (Systat 249 Software, Chicago, USA). The covariance of emissions of individual terpenes (dependent 250 variables, Y) with environmental conditions and physiological activities (independent 251 variables, X) was analyzed by partial least squares (PLS) regression using the plsdepot 252 package in R v. 3.3.3. The PLS model included two components and was validated by 253 cross-validation. Sigma Plot v. 14.0 and R v. 3.3.3 were also used to generate the figures. 254 255 3. Results 256 3.1. Seasonal and daytime variation of air temperature and soil moisture 257 Annual air temperature and soil moisture were similar between the control and N 258 treatments, with variation <10% for both sites. Mean air temperature on the sampling 259 dates in Garraf ranged between 10.1 ± 0.84 °C in winter mornings in 2015 and 34.4 ± 260 0.60 °C at summer middays in 2014. Soil moisture ranged between 5.5 ± 0.39% (v/v) in 261 summer middays in 2015 and 25.8 ± 0.47% (v/v) at winter middays in 2015 (Fig. S2). 262 Mean air temperature on the sampling dates in Prades ranged between 13.6 ± 1.11 °C in 263 winter mornings in 2016 and 32.7 ± 1.23 °C at summer middays in 2014. Soil moisture 264 ranged between 4.6 ± 0.30% (v/v) at summer middays in 2014 and 24.6 ± 1.37% (v/v) in 265 winter mornings in 2015 (Fig. S2). 266 267 3.2. Seasonal and daytime variation of A and gs 268 A for E. multiflora was always highest in spring or autumn and lowest in winter (Fig. 269 S3A). A was also notably high in mornings but decreased significantly at middays during 270 spring and summer. gs was always highest in autumn and lowest at winter. A and gs 271 differed significantly between the nitrogen and control treatments in winter, lower by 49.0% 272 (P < 0.05) in mornings for gs and by 40.5% for A (P < 0.05) and 40.8% for gs (P < 0.01) 273 at midday. gs was 41.0% (P < 0.05) lower in the N than the control treatments in summer 274 mornings. A increased sharply in spring relative to the pre-treatments after the first 275 addition of 15 kg N ha-1. A noticeably decreased and gs increased compared to autumn in 276 2014 after the third addition of 15 kg N ha-1. 277 A for Q. ilex was always highest in spring and lowest in summer (Fig. S3B). gs was 278 always highest in spring or winter and lowest in summer. A at winter midday was 279 significantly lower by 32.7% (P < 0.01) in the N than the control treatments. A in autumn 280 and winter was noticeably higher in the N treatments than the pre-treatments, and gs first 281 noticeably decreased in spring mornings and then increased steadily in the following 282 seasons, with higher values than the previous year after continual N fertilization. 283 284 3.3. Seasonal and daytime variation of isoprenoid emissions 285 Isoprenoid emission rates typically oscillated seasonally in both species, with maxima 286 in summer and minima in cold seasons (Figs. 1 and 2). Isoprenoid emissions were higher 287 for E. multiflora in 2015-2016 (Fig. 1), but Q. ilex emitted more isoprenoids in 2014- 288 2015 (Fig. 2). Isoprenoid emissions were always higher at midday than in the morning 289 for both species (Figs. 1 and 2). Both species emitted a variety of terpenes in spring and 290 summer, and most terpenes were detected at summer midday (Table S1). α-Pinene and 291 limonene were the two most abundant terpenes for both species and were detected at all 292 sampling times, with trends similar to those for total terpene emissions (Figs. 2 and S4). 293 Large amounts of tricyclene and β-caryophyllene for E. multiflora and β-pinene and β- 294 myrcene for Q. ilex were also detected in warm seasons (Fig. 3). 295 E. multiflora emitted both isoprene and terpenes, with isoprene the main compound 296 due to the large emissions in spring and summer (Fig. S5). Isoprene emissions ranged 297 between 0.03 ± 0.02 µg g-1 dw h-1 at autumn middays in 2015 and 7.74 ± 2.45 µg g-1 dw 298 h-1 at summer middays in 2015 (Fig. 1A). Isoprene emission rates did not differ 299 significantly between treatments but tended to increase after N fertilization in the same 300 plots, especially at summer middays. Total terpene emissions for E. multiflora ranged 301 between 0.10 ± 0.05 µg g-1 dw h-1 in winter mornings in 2015 and 3.99 ± 0.56 µg g-1 dw 302 h-1 at summer middays in 2015 (Fig. 1B). N fertilization increased total terpene emissions 303 in spring and summer relative to the pre-treatments and control treatments, especially at 304 summer midday, with a significant increase of 76.1% (P < 0.05). α-Pinene and limonene 305 were emitted mostly at summer middays in 2015, at rates of about 1.3 and 0.8 µg g-1 dw 306 h-1, respectively. 307 Q. ilex did not emit isoprene but was a large terpene emitter. The emission of total 308 terpenes ranged between 0.35 ± 0.08 µg g-1 dw h-1 at winter middays in 2016 and 23.9 ± 309 3.23 µg g-1 dw h-1 at summer middays in 2014 (Fig. 2). Q. ilex emitted fewer terpenes in 310 the fertilized year for the same plots, but total terpene emissions still increased 311 significantly relative to the control treatments at summer middays by 83.7% (P < 0.05), 312 coinciding with significantly higher limonene (P < 0.01). α-Pinene and limonene were 313 emitted mostly at summer middays in 2014, at rates of about 10 and 7 µg g-1 dw h-1, 314 respectively. 315 316 4. Discussion 317 4.1. Seasonal and daytime variations of A, gs and isoprenoid emissions with N 318 fertilization 319 Photosynthesis of most plants increases in warm weather if soil moisture and nutrients 320 are not limiting (Wan et al., 2009; Selsted et al., 2012). Plants in Mediterranean-type 321 climates have similar physiological trends, with A and gs highest in spring or autumn 322 when environmental conditions are favorable ( Llusià et al., 2013; Liu et al., 2016), with 323 A and gs lowest in winter for E. multiflora (Fig. S3A) and in summer for Q. ilex (Fig. 324 S3B) in our study. E. multiflora was particularly N responsive. Q. ilex, however, 325 displayed fewer physiological adjustments with N fertilization, suggesting poor N 326 acclimation (Pivovaroff et al., 2016). 327 The emission of isoprenoids differed between the species but followed a similar 328 seasonal pattern. The seasonal pattern agreed with previous results of isoprenoid 329 emissions in most Mediterranean species, with a maximum in summer and a minimum in 330 cold seasons (Llusià et al., 2011; Mu et al., 2018). Both species emitted most isoprenoids 331 at summer midday, coinciding with lower A, and the N treatments increased terpene 332 emissions significantly (Figs. 1, 2 and S3). Plants could temporarily decrease 333 photosynthetic activity under drought stress because of the increased resistance to CO2 in 334 both the stomata and mesophyll (Centritto et al., 2003; Mu et al., 2018), meanwhile, a 335 higher proportion of photosynthetically fixed carbon was used for increasing terpene 336 production to reduce the damage caused by oxidative stress at summer midday (Vallat et 337 al., 2005; Blanch et al., 2009), and plants increased terpene emissions with the higher N 338 deposition under the heat and drought stress, indicating a successful acclimation by 339 adjusting metabolism under environmental stress (Litvak et al., 1996; Loreto et al., 2001). 340 The N treatment favored terpene production (Blanch et al., 2009; Ormeño and Fernandez, 341 2012), especially under environmental stress. Higher N contents likely translate into 342 higher enzymatic activity and thus higher terpene production (Figs. 1B and 2) in these 343 two non-storing species, whose emission of terpenes depends on short-term production 344 (Litvak et al., 1996). 345 346 4.2. Comprehensive impacts of annual climate and N fertilization on isoprenoid emissions 347 N fertilization for E. multiflora increased A in spring 2015 when temperature was 348 suitable and water availability increased. The stimulation also indicated the importance 349 of even a small increase in soil moisture during spring in the warm plots, increasing N 350 availability in this nutrient-limited ecosystem (Figs. S1 and S2A) (Llusià et al., 2014; 351 Zhang et al., 2017; Tiiva et al., 2017). Abundant nutrition helps plants to recover from 352 slow physiological rhythms in winter and to return to growth quickly, especially at 353 midday. The higher temperature at spring midday in 2014, though, may have restrained 354 A, although not significantly, because the environmental conditions were sufficiently 355 suitable for assimilation, maintaining A at a high level for both treatments and saturating 356 N usage (Gundersen et al., 1998; Chen et al., 2016). N fertilization tended to have the 357 least influence in summer, supported by the similar environmental conditions between 358 the two years, in contrast to previous studies hypothesizing that drought-tolerant 359 evergreen shrubs are favored under a warming climate with increased CO2 or N levels 360 due to a higher A (Fineschi et al., 2013; Tiiva et al., 2017). In autumn and winter, 361 temperature variations may have been the main cause of the fluctuations in A between 362 two years, and excessive N began to negatively affect A and gs, especially at winter 363 midday. All fluctuations between the two years for Q. ilex were due to the variations in 364 environmental conditions (Figs. S1 and S2B), which was obvious for autumn and winter, 365 indicating that better water-heat interaction led to higher A in the fertilized year. A even 366 decreased significantly at winter midday for the N treatments, most likely due to the 367 decrease in both air temperature and soil moisture rather than to excess of available N. 368 Annual climate clearly played an important role due to the importance of some 369 environmental parameters (water and temperature) in setting the rates of isoprene and 370 terpene emissions (Jardine et al., 2014; Fernández-Martínez et al., 2018) under N-rich 371 conditions. For E. multiflora, more isoprenoids were emitted in spring and summer in 372 2015. Isoprenoid emissions from this temperate heath generally depend on both the 373 current environmental drivers and on the preceding season and weather (Niinemets et al., 374 2010; Llusià et al., 2013; Tiiva et al., 2017), which may suggest a high emission potential 375 for the ecosystem in the hot season if A was high in the previous growing season (Figs. 1 376 and S3A). Emissions then decreased to stable low levels. Increased A in spring stimulated 377 emissions in both spring and summer. Q. ilex emitted most terpenes in summer 2014 due 378 to a moderate drought, even though the N treatment significantly increased emissions at 379 summer midday in 2015. Q. ilex may thus be much more sensitive to water availability 380 than N content in hot seasons (Figs. 2 and S2). 381 Taking into account the results and conditions of this study and considering the climate 382 in combination with available N levels could help us to better understand the systematic 383 role of climate as a determinant of the dynamics of isoprenoid emissions (Fernández- 384 Martínez et al., 2018). Species without structures for storing terpenes are more likely to 385 have higher emission rates in N-rich conditions, supporting findings linking high foliar N 386 content to high rates of isoprenoid emission (Litvak et al., 1996; Possell et al., 2004; 387 Blanch et al., 2009; Fernández-Martínez et al., 2018). Our results thus suggest that N is 388 important for the emission of both isoprene and terpenes, and the variations in emission 389 may be associated with different strategies of N uptake and use (Litvak et al., 2002; 390 Fernández-Martínez et al., 2018), albeit the relationship between them was not very 391 strong, except in summer. This relationship may also be due to the mineralization of soil 392 N or because N absorption by plants is not likely to be limited by water under current 393 conditions (Gundersen et al., 1998; Wan et al., 2009; Chen et al., 2016). These favorable 394 environmental conditions (water and N availability) in the growing season enable high 395 rates of photosynthesis, which in turn have been linked to high rates of isoprenoid 396 emission (Monson et al., 1994; Litvak et al., 1996; Fernández-Martínez et al., 2018) in 397 summer. The positive correlation between soil N content and isoprenoid emission rates 398 may be due to both a direct effect of N on isoprenoid emission rates and an indirect effect 399 from the positive effect of N on photosynthesis (Monson et al., 1994; Fernández-Martínez 400 et al., 2018). 401 402 4.3. The influence of N deposition on the relationship between main physiological or 403 environmental parameters and isoprenoid emissions 404 Environmental conditions such as air temperature and soil moisture are the main factors 405 that determine BVOC emissions (Gershenzon et al., 1978; Breshears et al., 2005; Llusià, 406 et al., 2011; Allen et al., 2015). BVOC emissions, however, are also largely influenced 407 by A and gs, the two main physiological activities of plants (Eller et al., 2016). The PLS 408 regression analysis did not find a strong correlation between terpene emissions and 409 physiological activities but found a strong correlation with environmental conditions (Fig. 410 4). All terpene emissions by E. multiflora were correlated positively with air temperature 411 and negatively with soil moisture, except for β-myrcene, which had the opposite trend. β- 412 Pinene had the same relationship with β-myrcene, and 3-carene emission became strongly 413 correlated with physiological activities in the N plots. Trans-β-ocimene, sabinene and α- 414 caryophyllene emissions by Q. ilex were not obviously correlated with any parameter, 415 and other species followed a common trend with environmental conditions. α-Pinene and 416 limonene have recently been reported as the main emitted terpenes; α-pinene is sensitive 417 to temperature, and limonene responds more strongly to water deficits (Mu et al., 2018). 418 The correlation for β-caryophyllene with environmental conditions disappeared, and 419 sabinene emission was slightly positively correlated with gs in the N plots. N fertilization 420 generally increased the correlation between the most emitted compounds and the 421 environmental conditions but decreased it for the least emitted compounds (Fig. 4), which 422 may indicate that N fertilization increased the rates of terpene emission (Fig. 3) but 423 slightly decreased the diversity of terpenes in the warm seasons (Table S1). Q. ilex 424 emitted more terpenes in summer in the N plots, but mainly α-pinene, β-pinene and 425 limonene contributed to the increase, and almost all other terpene emissions decreased, 426 especially for β-myrcene and 3-carene (Fig. 3B) and very-least compounds, such as trans- 427 β-ocimene, β-caryophyllene and α-caryophyllene, tended to disappear (Table S1). Not all 428 terpene emissions, however, were higher in summer. This diversity of responses may 429 have been due to the environmental effects on the activities of synthases, the potential 430 protective roles of the various terpenes under environmental constraints (Blanch et al., 431 2009) or to the increases in foliar N content (Blanch et al., 2009). 432 433 We also analyzed the corresponding correlations that can be applied to the modelling 434 of BVOC emissions (Table S2). The emission rates of isoprene and total terpenes were 435 both correlated positively with air temperature and negatively with soil moisture for both 436 species (Fig. 5). N addition slightly decreased isoprene emission, especially at the high- 437 temperature and low-moisture site, and significantly increased total terpene emissions at 438 this same site. The emission rates of isoprenoids were correlated negatively with A and 439 gs for both species (Fig. S6), which supports the idea that BVOC emissions are mostly 440 influenced by environmental conditions (Gershenzon et al., 1978; Loreto and Schnitzler, 441 2010; Holopainen and Gershenzon, 2010; Llusià et al., 2011). Q. ilex emitted more 442 isoprenoids and A was lowest in summer than in the other seasons (Figs. 2, S3B), and 443 higher percentages of fixed carbon were devoted to isoprenoid emission in summer. The 444 relationships with physiological activities, however, became much weaker for the E. 445 multiflora in N plots, also indicating that this species was more sensitive to nutrient 446 availability than Q. ilex due to some fluctuations in A and gs (Figs. S3A and S6). 447 448 5. Conclusions 449 This study investigated the seasonal and daytime variations in the response of 450 isoprenoid emissions to experimental N deposition in dominant species at Mediterranean 451 ecosystems. In the context of GEC, the combination of environmental factors, such as air 452 temperature, soil moisture and N deposition, mimicked future conditions in this temperate 453 ecosystem. The complex effects among warming, drought and N deposition can decrease 454 the regularity of single-factor effects on isoprenoid emissions, even though the responses 455 varied strongly between seasons, and the emission profile was mainly dominated by 456 synergetic increases in summer. Our results indicate a successful acclimation of plants by 457 increasing isoprenoid emissions under environmental stress, which are expected to be 458 increased by climate change in the Mediterranean region (Litvak et al., 1996; Loreto et 459 al., 2001). N deposition will also further stimulate these emission trends in the warmer 460 and drier conditions projected by climatic and ecophysiological models for the coming 461 decades (Peñuelas and Llusià, 2001; Penuelas and Staudt, 2010; IPCC, 2014). The 462 changes in isoprenoid emissions in this region, however, also thus depend on the species, 463 and therefore on the changes in land covers, for example from forests to shrublands with 464 different emission capacities. Further long-term and quantitative research on the detailed 465 emission mechanisms with multiple factors acting in concert is still warranted. 466 467 Declare on interest 468 All authors declare that there are no conflicts of interest relevant to this work. 469 470 Acknowledgements 471 This work was financially supported by the Spanish Government project CGL2016- 472 79835-P, the Catalan government project SGR2017-1005 and the ERC Synergy project 473 SyG-2013-610028 IMBALANCE-P. We are also grateful for the financial support from 474 the China Scholarship Council. 475 476 Appendix A. Supplementary data 477 Supplementary data to this article can be found online. 478 479 References 480 Allen, C.D., Breshears, D.D., McDowell, N.G., 2015. On underestimation of global 481 vulnerability to tree mortality and forest die-off from hotter drought in the 482 Anthropocene. Ecosphere 6, 1–55. 483 Alvarez-Clare, S., Mack, M.C., Brooks, M., 2013. A direct test of nitrogen and 484 phosphorus limitation to net primary productivity in a lowland tropical wet forest. 485 Ecology 94, 1540–1551. 486 Beier, C., Beierkuhnlein, C., Wohlgemuth, T., Peñuelas, J., Emmett, B., Körner, C., de 487 Boeck, H., Christensen, J.H., Leuzinger, S., Janssens I.A., Hansen, K., 2012. 488 Precipitation manipulation experiments – challenges and recommendations for the 489 future. Ecol. Lett. 15, 899–911. 490 Blanch, J., Peñuelas, J., Llusià, J., 2007. Sensivity of terpene emissions to drought and 491 fertilization in terpene-storing Pinus halepensis and non-storing Quercus ilex. 492 Physiol. Plant. 131, 211–225. 493 Blanch, J., Peñuelas, J., Sardans, J., Llusià, J., 2009. Drought, warming and soil 494 fertilization effects on leaf volatile terpene concentrations in Pinus halepensis and 495 Quercus ilex. Acta Physiolo. Plant. 31, 207–218. 496 Bobbink, R., Hicks, K., Galloway, J., Spranger, T., Alkemade, R., Ashmore, M., 497 Bustamante, M., Cinderby, C., Davidson, E., Dentener, F., Emmett, B., Erisman, 498 J.W., Fenn, M., Gilliam, F., Nordin, A., Pardo, L., de Vries, W., 2010. Global 499 assessment of nitrogen deposition effects on terrestrial plant diversity: a synthesis. 500 Ecol. Appl. 20, 30–59. 501 Bolòs, O., Vigo, J., 1990. Flora dels Països Catalans. Barcelona, Spain: Barcino Press. 502 Breshears, D.D, Cobb, N.S., Rich, P.M., Price, K.P., Allen , C.D., Balice, R.G., Romme, 503 W.H., Kastens J.H., Floyd, M.L., Belnap J., Anderson, J.J., Myers, O.B., Meyer, 504 C.W., 2005. Regional vegetation die-off in response to global-change-type drought. 505 Proc. Natl. Acad. Sci. U.S.A. 102, 15144–15148. 506 Cardoso-Vilhena, J., Barnes, J., 2001. Does nitrogen supply affect the response of wheat 507 (Triticum aestivum cv. Hanno) to the combination of elevated CO2 and O3? J. Exp. 508 Bot. 52, 1901–1911. 509 Carriero, G., Brunetti, C., Fares, S., Hayes, F., Hoshika, Y., Mills, G., Tattini, M., Paoletti, 510 E., 2016. BVOC responses to realistic nitrogen fertilization and ozone exposure in 511 silver birch. Environ. Pollut. 213, 988–995. 512 Centritto, M., Loreto, F., Chartzoulakis, K., 2003. The use of low [CO2] to estimate 513 diffusional and non-diffusional limitations of photosynthetic capacity of salt- 514 stressed olive saplings. Plant, Cell Environ. 26, 585–594. 515 Chen, H., Gurmesa, G.A., Zhang, W., Zhu, X., Zheng, M., Mao, Q., Zhang, T., Mo, J., 516 2016. Nitrogen saturation in humid tropical forests after 6 years of nitrogen and 517 phosphorus addition: hypothesis testing. Funct. Ecol. 30, 305–313. 518 Dicke, M., Loreto, F., 2010. Induced plant volatiles: from genes to climate change. Trends 519 Plant Sci. 15, 115–117. 520 Eller, A.S.D., Young, L.L., Trowbridge, A.M., Monson, R.K., 2016. Differential controls 521 by climate and physiology over the emission rates of biogenic volatile organic 522 compounds from mature trees in a semi-arid pine forest. Oecologia 180, 345–358. 523 Farré-Armengol, G., Filella, I., Llusia, J., Peñuelas, J., 2015. Relationships among floral 524 VOC emissions, floral rewards and visits of pollinators in five plant species of a 525 Mediterranean shrubland. Plant Ecol. Evol. 148, 90-99. 526 Fernández-Martínez, M., Llusià, J., Filella, I., Niinemets, Ü., Arneth, A., Wright, I.J., 527 Loreto, F., Peñuelas, J., 2018. Nutrient-rich plants emit a less intense blend of 528 volatile isoprenoids. New Phytol. 220, 773–784. 529 Fineschi, S., Loreto, F., Staudt, M., Peñuelas, J., 2013. Diversification of volatile 530 isoprenoid emissions from trees: evolutionary and ecological perspectives. In: 531 Niinemets, Ü., Monson R. K., eds. Biology, Controls and Models of Tree Volatile 532 Organic Compound Emissions. Tree Physiol 5, 1-20. 533 Galloway, J.N., Dentener, F.J., Capone, D.G., Boyer, E.W., Howarth, R.W., Seitzinger, 534 S.P., Asner, G.P., Cleveland, C.C., Green, P.A., Holland, E.A., Karl, D.M., Michaels, 535 A.F., Porter, J.H., Townsend A.R., Vörösmarty, C.J., 2004. Nitrogen cycles: past, 536 present, and future. Biogeochemistry 70, 153–226. 537 Galloway, J.N., Townsend, A.R., Erisman, J.W., Bekunda, M., Cai, Z., Freney, J.R., 538 Martinelli, L.A., Seitzinger S.P., Sutton, M.A., 2008. Transformation of the nitrogen 539 cycle: recent trends, questions and potential solutions. Science, 320, 889–892. 540 Gershenzon, J., Lincoln, D.E., Langenheim, J.H., 1978. The effect of moisture stress on 541 monoterpenoid yield and composition in Satureja douglasii. Biochem. Syst. Ecol. 6, 542 33–43. 543 Guenther, A., Hewitt, C.N., Erickson, D., Fall, R., Geron, C., Graedel, T., Harley, P., 544 Klinger, L., Lerdau, M., McKay, W.A., Piere, T., Scholes, B., Steinbrecher, R., 545 Tallamraju, R., Taylor, J., Zimmerman, P., 1995. A global model of natural volatile 546 organic compound emissions. J. Geophys. Res. 100, 8873–8892. 547 Gundersen, P., Emmett, B.A., Kjonaas, O.J., Koopmans, C.J., Tietema, A., 1998. Impact 548 of nitrogen deposition on nitrogen cycling in forests: a synthesis of NITREX data. 549 Forest Ecol. Manag. 101, 37–55. 550 Häikiö, E., Freiwald, V., Silfver, T., Beuker, E., Holopainen, T., Oksanen, E., 2007. 551 Impacts of elevated ozone and nitrogen on growth and photosynthesis of European 552 aspen (Populus tremula) and hybrid aspen (P. tremula × Populus tremuloides) 553 clones. Can. J. Forest Res. 37, 2326–2336. 554 Handley, T., Grulke, N.E., 2008. Interactive effects of O3 exposure on California black 555 oak (Quercus kelloggii Newb.) seedlings with and without N amendment. Environ. 556 Pollut. 156, 53–60. 557 Harley, P., Eller, A., Guenther, A., Monson, R.K., 2014. Observations and models of 558 emissions of volatile terpenoid compounds from needles of ponderosa pine trees 559 growing in situ : control by light, temperature and stomatal conductance. Oecologia 560 176, 35–55. 561 Harley, P.C., Litvak, M.E., Sharkey, T.D., Monson, R.K., 1994. Isoprene emission from 562 velvet bean leaves-interactions among nitrogen availability, growth photon flux 563 density, and leaf development. Plant Physiol. 105, 279–285. 564 Holopainen, J. K., Gershenzon, J., 2010. Multiple stress factors and the emission of plant 565 VOCs. Trends Plant Sci. 15, 176–184. 566 IPCC, 2014. Climate Change 2014: Impacts, adaptation, and vulnerability. Part A: Global 567 and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment 568 Report of the Intergovernmental Panel on Climate Change. New York, USA: 569 Cambridge University Press. 570 Janssens, I.A., Dieleman,W., Luyssaert, S., Subke, J.A., Reichstein, M., Ceulemans, R., 571 Ciais, P., Dolman, A.J., Grace, J., Matteucci, G., Papale, D., Piao, S.L., Schulze, 572 E.D., Tang, J., Law, B.E., 2010. Reduction of forest soil respiration in response to 573 nitrogen deposition. Nat. Geosci. 3, 315–322. 574 Jardine, K., Chambers, J., Alves, E.G., Teixeira, A., Garcia, S., Holm, J., Higuchi, N., 575 Manzi, A., Abrell, L., Fuentes, J.D., Nielsen, L.K., Torn, M.S., Vickers C. E., 2014. 576 Dynamic balancing of isoprene carbon sources reflects photosynthetic and 577 photorespiratory responses to temperature stress. Plant Physiol. 166, 2051–2064. 578 Kesselmeier, J., Staudt, M., 1999. Biogenic Volatile Organic Compunds (VOC): An 579 Overview on Emission, Physiology and Ecology. J. Atmos. Chem. 33, 23–88. 580 Kivimäenpää, M., Ghimire, R.P., Sutinen, S., Häikiö, E., Kasurinen, A., Holopainen, T., 581 Holopainen, J.K., 2016. Increases in volatile organic compound emissions of Scots 582 pine in response to elevated ozone and warming are modified by herbivory and soil 583 nitrogen availability. Eur. J. Forest Res. 135, 343–360. 584 Kollist, T., Moldau, H., Rasulov, B., Oja, V., Rämma, H., Hüve, K., Jaspers, P., 585 Kangasjärvi J., Kollist, H., 2007. A novel device detects a rapid ozone-induced 586 transient stomatal closure in intact Arabidopsis and its absence in abi2 mutant. 587 Physiol. Plantarum, 129, 796–803. 588 Leuzinger, S., Luo, Y., Beier, C., Dieleman, W., Vicca, S., Körner, C., 2011. Do global 589 change experiments overestimate impacts on terrestrial ecosystems? Trends Ecol. 590 Evol. 26, 236–241. 591 Li, S., Harley, P.C., Niinemets, Ü., 2017. Ozone-induced foliar damage and release of 592 stress volatiles is highly dependent on stomatal openness and priming by low-level 593 ozone exposure in Phaseolus vulgaris. Plant Cell Environ. 40, 1984–2003. 594 Litvak, M.E., Constable, J.V., Monson, R.K., 2002. Supply and demand processes as 595 controls over needle monoterpene synthesis and concentration in Douglas fir 596 [Pseudotsuga menziesii (Mirb.) Franco]. Oecologia 132, 382–391. 597 Litvak, M. E., Loreto, F., Harley, P.C., Sharkey, T. D., Monson, R. K., 1996. The response 598 of isoprene emission rate and photosynthetic rate to photon flux and nitrogen supply 599 in aspen and white oak trees. Plant Cell Environ. 19, 549–559. 600 Liu, D., Llusià, J., Ogaya, R., Estiarte, M., Llorens, L., Yang, X., Peñuelas, J., 2016. 601 Physiological adjustments of a Mediterranean shrub to long-term experimental 602 warming and drought treatments. Plant Sci. 252, 53–61. 603 Liu, X., Duan, L., Mo, J., Du, E., Shen, J., Lu, X., Zhang, Y., Zhou, X., He, C., Zhang, 604 F., 2011. Nitrogen deposition and its ecological impact in China: an overview. 605 Environ. Pollut. 159, 2251–2264. 606 Llusià, J., Bermejo-Bermejo, V., Calvete-Sogo, H., Peñuelas, J., 2014. Decreased rates 607 of terpene emissions in Ornithopus compressus L. and Trifolium striatum L. by 608 ozone exposure and nitrogen fertilization. Environ. Pollut. 194, 69–77. 609 Llusià, J., Peñuelas, J., Alessio, G.A., Ogaya, R., 2011. Species-specific, seasonal, inter- 610 annual, and historically-accumulated changes in foliar terpene emission rates in 611 Phillyrea latifolia and Quercus ilex submitted to rain exclusion in the Prades 612 Mountains (Catalonia). Russ. J. Plant Physiol. 58, 126–132. 613 Llusià, J., Peñuelas, J., Guenther, A., Rapparini, F., 2013. Seasonal variations in terpene 614 emission factors of dominant species in four ecosystems in NE Spain. Atmos. 615 Environ. 70, 149–158. 616 Loreto, F., Fischbach, R.J., Schnitzler, J.P., Ciccioli, P., Brancaleoni, E., Calfapietra, C., 617 Seufert, G., 2001. Monoterpene emission and monoterpene synthase activities in the 618 Mediterranean evergreen oak Quercus ilex L. grown at elevated CO2 concentrations. 619 Global Change Biol. 7, 709–717. 620 Loreto, F., Schnitzler, J.P., 2010. Abiotic stresses and induced BVOCs. Trends Plant Sci. 621 15, 154–166. 622 Meunier, C.L., Gundale, M.J., Sánchez, I.S., Liess, A., 2016. Impact of nitrogen 623 deposition on forest and lake food webs in nitrogen-limited environments. Global 624 Change Biol. 22, 164–179. 625 Monson, R.K., Fall, R., 1989. Isoprene emission from aspen leaves : Influence of 626 environment and relation to photosynthesis and photorespiration. Plant Physiol. 90, 627 267–274. 628 Monson, R.K., Harley, P.C., Litvak, M.E., Wildermuth, M., Guenther, A.B., Zimmerman, 629 P.R., Fall, R., 1994. Environmental and developmental controls over the seasonal 630 pattern of isoprene emission from aspen leaves. Oecologia 99, 260–270. 631 Mu, Z., Llusià, J., Liu, D., Ogaya, R., Asensio, D., Zhang, C., Peñuelas, J., 2018. Seasonal 632 and diurnal variations of plant isoprenoid emissions from two dominant species in 633 Mediterranean shrubland and forest submitted to experimental drought. Atmos. 634 Environ. 191, 105–115. 635 Niederbacher, B., Winkler, J.B., Schnitzler, J.P., 2015. Volatile organic compounds as 636 non-invasive markers for plant phenotyping. J. Exp. Bot. 66, 5403–5416. 637 Niinemets, Ü., Loreto, F., Reichstein, M., 2004. Physiological and physicochemical 638 controls on foliar volatile organic compound emissions. Trends Plant Sci. 9, 180– 639 186. 640 Niinemets, Ü., Monson, R.K., Arneth, A., Ciccioli, P., Kesselmeier, J., Kuhn, U., Noe, 641 S.M., Penuelas, J., Staudt, M., 2010. The leaf-level emission factor of volatile 642 isoprenoids: caveats, model algorithms, response shapes and scaling. 643 Biogeosciences 7, 1809–1832. 644 Niinemets, Ü., Reichstein, M., Staudt, M., Seufert, G., Tenhunen, J.D., 2002. Stomatal 645 constraints may affect emission of oxygenated monoterpenoids from the foliage of 646 Pinus pinea. Plant Physiol. 130, 1371–1385. 647 Ogaya, R., Llusià, J., Barbeta, A., Asensio, D., Liu, D., Alessio, G.A., Peñuelas, J., 2014. 648 Foliar CO2 in a holm oak forest subjected to 15 years of climate change simulation. 649 Plant Sci. 226, 101–107. 650 Ormeño, E., Fernandez, C., 2012. Effect of soil nutrient on production and diversity of 651 volatile terpenoids from plants. Curr. Bioact. Compd. 8, 71–79. 652 Ormeño, E., Olivier, R., Mévy, J.P., Baldy, V., Fernandez, C., 2009. Compost may affect 653 volatile and semi-volatile plant emissions through nitrogen supply and chlorophyll 654 fluorescence. Chemosphere 77, 94–104. 655 Paasonen, P., Asmi, A., Petäjä, T., Kajos, M.K., Äijälä, M., Junninen, H., Holst, T., 656 Abbatt, J.P.D., Arneth, A., Birmili, W., van der Gon, H.D., Hamed, A., Hoffer, A., 657 Laakso, L., Laaksonen, A., Leaitch , W.R., Plass-Dülmer, C., Pryor, S.C., Räisänen, 658 P., Swietlicki, E., Wiedensohler, A., Worsnop, D.R., Kerminen, V., Kulmala, M., 659 2013. Warming-induced increase in aerosol number concentration likely tomoderate 660 climate change. Nat. Geosci. 6, 438–442. 661 Paoletti E. Grulke N.E., 2010. Ozone exposure and stomatal sluggishness in different 662 plant physiognomic classes. Environ. Pollut. 158, 2664–2671. 663 Peñuelas, J., Llusià, J., 2001. The complexity of factors driving volatile organic 664 compound emissions. Biol. Plantarum, 44, 481–487. 665 Peñuelas, J., Llusià, J., 2003. BVOCs: Plant defense against climate warming? Trends 666 Plant Sci. 8, 105–109. 667 Peñuelas, J., Poulter, B., Sardans, J., Ciais, P., Van Der Velde, M., Bopp, L., Boucher, O., 668 Godderis, Y., Hinsinger, P., Llusia, J., Nardin, E., Vicca, S., Obersteiner, M., 669 Janssens, I.A., 2013. Human-induced nitrogen-phosphorus imbalances alter natural 670 and managed ecosystems across the globe. Nat. Commun. 4, 2934. 671 Peñuelas, J., Staudt, M., 2010. BVOCs and global change. Trends Plant Sci. 3, 115–184. 672 Phoenix, G.K., Emmett, B.A., Britton, A.J., Caporn, S.J.M., Dise, N.B., Helliwell, R., 673 Jones, L., Leake, J.R., Leith, I.D., Sheppard, L.J., Sowerby, A., Pilkington, M.G., 674 Rowe, E.C., Ashmore, M.R., Power, S.A., 2012. Impacts of atmospheric nitrogen 675 deposition: responses of multiple plant and soil parameters across contrasting 676 ecosystems in long-term field experiments. Global Change Biol. 18, 1197–1215. 677 Pivovaroff, A.L., Santiago, L.S., Vourlitis, G.L., Grantz, D.A., Allen, M.F., 2016. Plant 678 hydraulic responses to long-term dry season nitrogen deposition alter drought 679 tolerance in a Mediterranean-type ecosystem. Oecologia 181, 115–184. 680 Possell, M., Heath, J., Hewitt, C.N., Ayres, E., Kerstiens, G., 2004. Interactive effects of 681 elevated CO2 and soil fertility on isoprene emissions from Quercus robur. Global 682 Change Biol. 10, 1835–1843. 683 Sala, O.E., Chapin III, F.S., Armesto, J.J., Berlow, E., Bloomfield J., Dirzo, R., Huber- 684 sanwald, E., Huenneke, L.F., Jackson, R.B., Kinzig, A., Leemans, R., Lodge, D.M., 685 Mooney, H.A., Oesterheld, M., Poff, N.L., Sykes, M.T.,Walker, B.H.,Walker, M., 686 Wall, D.H., 2000. Global Biodiversity Scenarios for the Year 2100. Science 287, 687 1770–1774. 688 Selsted, M.B., van der Linden, L., Ibrom, A., Michelsen, A., Larsen, K.S., Pedersen, J.K., 689 Mikkelsen, T.N., Pilegaard, K, Beier, C., Ambus, P., 2012. Soil respiration is 690 stimulated by elevated CO2 and reduced by summer drought: Three years of 691 measurements in a multifactor ecosystem manipulation experiment in a temperate 692 heathland (CLIMAITE). Global Change Biol. 18, 1216–1230. 693 Sharkey, T.D., Monson, R.K., 2017. Isoprene research – 60 years later, the biology is still 694 enigmatic. Plant Cell Environ. 40, 1671–1678. 695 Tiiva, P., Tang, J., Michelsen, A., Rinnan, R., 2017. Monoterpene emissions in response 696 to long-term night-time warming, elevated CO2 and extended summer drought in a 697 temperate heath ecosystem. Sci. Total Environ. 580, 1056–1067. 698 Valolahti, H., Kivimäenpää, M., Faubert, P., Michelsen, A., Rinnan, R., 2015. Climate 699 change-induced vegetation change as a driver of increased subarctic biogenic 700 volatile organic compound emissions. Global Change Biol. 21, 3478–3488. 701 Vallat, A., Gu, H., Dorn, S., 2005. How rainfall, relative humidity and temperature 702 influence volatile emissions from apple trees in situ. Phytochemistry 66, 1540–1550. 703 Vahisalu T., Kollist H., Wang Y., Nishimura N., Chan W., Valerio G., Lamminmäki, A., 704 Brosché, M., Moldau H., Desikan R., Schroeder, J.I., Kangasjärvi J., 2008. SLAC1 705 is required for plant guard cell S-type anion channel function in stomatal signalling. 706 Nature 452, 487–491. 707 Valliere, J.M., Allen, E.B., 2016. Interactive effects of nitrogen deposition and drought- 708 stress on plant-soil feedbacks of Artemisia californica seedlings. Plant Soil 403, 709 277–290. 710 Vitousek, P.M., Howarth, R.W., 1991. Nitrogen limitation on land and in the sea: how 711 can it occur? Biogeochemistry 13, 87–115. 712 Vourlitis, G.L., Pasquini, S.C., 2009. Plant and soil N response of Southern Californian 713 semi-arid shrublands after 1 year of experimental N deposition. Ecosystems 10, 263– 714 279 715 Wan, S., Xia, J., Liu, W., Niu, S., 2009. Experimental dry-season N deposition alters 716 species composition in southern Californian mediterranean-type shrublands. 717 Ecology 90, 2183–2189. 718 Yuan, X., Shang, B., Xu, Y., Xin, Y., Tian, Y., Feng, Z., Paoletti, E., 2017. No significant 719 interactions between nitrogen stimulation and ozone inhibition of isoprene emission 720 in Cathay poplar. Sci. Total Environ. 601–602, 222–229. 721 Zhang, R., Wu, J., Li, Q., Hänninen, H., Peng, C., Yao, H., Song, X., Ying, Y., 2017. 722 Nitrogen deposition enhances photosynthesis in Moso bamboo but increases 723 susceptibility to other stress factors. Front. Plant Sci. 8, 1–11. 724 725 726 727 728 729 730 731 732 733 A) Morning Midday Season P < 0.001 Season P < 0.001 10 Control Nitrogen Rate of isoprene emissions 8 ( µg g dw h ) 6 -1 -1 4 2 0 N N ng r r r r r r r r pri me mn nte rin g me mn nte pri ng me mn nte rin g me mn nte S um utu Wi Sp um utu Wi S um utu Wi Sp um utu Wi 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 20 2 20 20 2 20 20 2 20 20 2 Season 734 735 B) Morning Midday 5 Control Season P < 0.001 Season P < 0.001 Nitrogen Rate of total terpene emissions 4 ( µg g dw h ) 3 -1 -1 2 * 1 0 N N r r r r g r r g r r rin g me mn nte rin g me mn nte rin me mn nte rin me mn nte Sp um utu Wi Sp um utu Wi Sp um utu Wi Sp um utu Wi 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 20 2 20 20 2 20 20 2 20 20 2 Season 736 737 Fig. 1. Seasonal variation of the rates of emission of isoprene (A) and total terpenes (B) for Erica multiflora. ‘N’ 738 indicates the start of the fertilization treatment. Error bars indicate standard errors of the means (n = 6). Significant 739 differences between treatments identified by Student’s t-tests are indicated by asterisks (*, P < 0.05). The effects of 740 season, treatment and sampling time are depicted in the panels when significant. 741 742 743 744 Morning Midday 30 Season P < 0.01 Season P < 0.001 Time P < 0.05 Control Nitrogen 25 Rate of total terpene emissions 20 ( µg g dw h ) -1 15 -1 10 5 * 0 N N 12 Season P < 0.001 Season P < 0.01 Rate of a-pinene emissions 9 ( µg g dw h ) -1 -1 6 3 0 N N Season × treatment P < 0.05 Season P < 0.05 10 Rate of limonene emissions 8 ( µg g dw h ) 6 -1 -1 4 2 ** 0 g r mn r N ing r mn r g r mn r N ing r mn r rin me nte r me nte rin me nte r me nte Sp um utu Wi Sp um utu Wi Sp um utu Wi Sp um utu Wi 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 20 2 20 20 2 20 20 2 20 20 2 Season 745 746 Fig. 2. Seasonal variation of the rates of emission of total terpenes, α-pinene and limonene for Quercus ilex. ‘N’ 747 indicates the start of the fertilization treatment. Error bars indicate standard errors of the means (n = 6). Significant 748 differences between treatments identified by Student’s t-tests are indicated by asterisks (*, P < 0.05; **, P < 0.01). 749 The effects of season, treatment and sampling time are depicted in the panels when significant. 750 751 752 753 754 755 756 A) 757 758 759 760 761 762 763 764 765 766 767 768 B) 769 770 Fig. 3. Distribution of seasonal terpene emissions for Erica multiflora (A) and Quercus ilex (B) in the morning and at 771 midday for the fertilized year. 772 773 774 775 776 777 778 779 780 781 782 Erica multiflora 783 A) Control B) Nitrogen 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 Quercus ilex 802 C) Control D) Nitrogen 803 804 Fig. 4. Partial least squares (PLS) regression between main environmental or physiological parameters and terpene 805 emissions for Erica multiflora (A for control treatments, B for nitrogen treatments) and Quercus ilex (C for control 806 treatments, D for nitrogen treatments) in the fertilized year. Blue represents environmental or physiological parameters 807 (independent variables, X), and yellow represents emission rates of individual terpenes (dependent variables, Y). Tair, 808 air temperature; SM, soil moisture; A, net photosynthetic rate; gs, stomatal conductance. Individual terpenes: α-pinene, 809 tricyclene, β-pinene, β-myrcene, 3-carene, limonene, sabinene, trans-β-ocimene, α-terpinolene, β-caryophyllene, α- 810 caryophyllene. 811 812 813 814 815 816 817 818 819 820 821 822 823 824 A) Control : r = 0.55 P < 0.0001 Rate of isoprene emissions for E. multiflora Control : r = 0.51 P < 0.001 Control Nitrogen : r = 0.43 P < 0.01 Nitrogen : r = 0.41 P < 0.01 Nitrogen 15 Correlation curve for control Correlation curve for nitrogen 10 ( µg g dw h ) -1 -1 5 0 8 Rate of total terpene emissions for E. multiflora Control : r = 0.53 P < 0.0001 Control : r = 0.55 P < 0.0001 Nitrogen : r = 0.51 P < 0.001 Nitrogen : r = 0.48 P < 0.001 6 ( µg g dw h ) -1 4 -1 2 0 10 15 20 25 30 35 0 5 10 15 20 Air temperature (°C) Soil moisture (%) 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 B) Control : r = 0.62 P < 0.0001 Control : r = 0.48 P < 0.001 Control Rate of total terpene emissions for Q. ilex Nitrogen : r = 0.66 P < 0.0001 Nitrogen : r = 0.39 P < 0.01 30 Nitrogen Correlation curve for control Correlation curve for nitrogen ( µg g dw h ) 20 -1 -1 10 0 10 20 30 0 5 10 15 20 Air temperature (°C) Soil moisture (%) 842 843 844 Fig. 5. Relationships for the rate of isoprenoid emissions with main environmental conditions (air temperature and soil 845 moisture) for Erica multiflora (A) and Quercus ilex (B) in the fertilized year. 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 Supporting Information 864 865 866 Profile of foliar isoprenoid emissions from Mediterranean dominant 867 shrub and tree species under experimental nitrogen deposition 868 Zhaobin Mua,b,*, Joan Llusiàa,b, Daijun Liua,b, Romà Ogayaa,b, Dolores Asensio a,b, Chao 869 Zhanga,b, Josep Peñuelasa,b 870 a 871 CSIC, Global Ecology Unit CREAF-CSIC-UAB, E08193 Bellaterra, Catalonia, Spain b 872 CREAF, E08193 Cerdanyola, Catalonia, Spain 873 874 *Correspondence 875 Zhaobin Mu 876 Email: zhaobin@creaf.uab.cat 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 Tables 892 Table S1. Distribution of seasonal terpene emissions (µg g-1 dw h-1) for Erica multiflora 893 (A and B) and Quercus ilex (C and D) in the morning and at midday for the fertilized year. 894 ND, not detected. 895 896 A) Morning Spring Summer Autumn Winter Control Nitrogen Control Nitrogen Control Nitrogen Control Nitrogen α-Pinene 0.81 ± 0.26 0.48 ± 0.14 0.86 ± 0.31 0.94 ± 0.26 0.17 ± 0.02 0.13 ± 0.06 0.14 ± 0.05 0.06 ± 0.03 Tricyclene ND 0.16 ± 0.09 0.04 ± 0.02 0.13 ± 0.04 0.02 ± 0.01 0.02 ± 0.01 0.01 ND β-Pinene ND ND 0.14 ± 0.05 0.22 ± 0.08 0.04 ± 0.02 <0.01 ND ND β-Myrcene ND ND ND ND 0.01 <0.01 ND ND 3-Carene 0.01 ND 0.18 ± 0.05 0.31 ± 0.11 0.02 ± 0.01 ND ND ND Limonene 0.18 ± 0.03 0.21 ± 0.04 0.58 ± 0.18 0.71 ± 0.30 0.02 ± 0.01 0.07 ± 0.01 0.1 ± 0.04 0.07 ± 0.02 Sabinene ND ND 0.08 ± 0.05 <0.01 <0.01 ND <0.01 <0.01 Trans-β-ocimene 0.02 ± 0.01 ND ND 0.01 0.04 ± 0.02 0.04 ± 0.02 ND ND α-Terpinolene ND ND ND ND ND ND ND ND β-Caryophyllene 0.01 ND 0.14 ± 0.08 0.19 ± 0.10 0.05 ± 0.03 <0.01 ND ND 897 898 899 B) Midday Spring Summer Autumn Winter Control Nitrogen Control Nitrogen Control Nitrogen Control Nitrogen α-Pinene 0.50 ± 0.09 0.71 ± 0.19 0.65 ± 0.14 1.31 ± 0.27 0.25 ± 0.02 0.33 ± 0.14 0.23 ± 0.09 0.16 ± 0.04 Tricyclene 0.10 ± 0.03 0.14 ± 0.06 0.28 ± 0.06 0.68 ± 0.23 0.04 ± 0.02 0.02 ± 0.01 0.02 ± 0.01 ND β-Pinene 0.17 ± 0.06 0.25 ± 0.13 0.07 ± 0.01 0.01 0.01 0.01 ND ND β-Myrcene ND ND ND ND 0.01 ND ND ND 3-Carene 0.10 ± 0.04 0.07 ± 0.04 0.06 ± 0.03 0.03 ± 0.01 <0.01 0.01 ND ND Limonene 0.27 ± 0.11 0.60 ± 0.20 0.49 ± 0.16 0.82 ± 0.21 0.09 ± 0.05 0.15 ± 0.09 0.09 ± 0.01 0.04 ± 0.01 Sabinene 0.10 ± 0.06 ND 0.01 0.05± 0.02 ND 0.01 0.02 ± 0.01 ND Trans-β-ocimene ND ND 0.32 ± 0.09 0.30± 0.18 ND 0.01 ND ND α-Terpinolene ND ND <0.01 0.03 ± 0.01 ND ND ND ND β-Caryophyllene 0.20 ± 0.05 0.19 ± 0.10 0.38± 0.10 1.01± 0.40 0.06 ± 0.04 ND ND ND 900 901 902 903 904 905 906 907 C) Morning Spring Summer Autumn Winter Control Nitrogen Control Nitrogen Control Nitrogen Control Nitrogen α-Pinene 0.93 ± 0.42 0.78 ± 0.38 1.87 ± 0.49 2.57 ± 0.26 0.54 ± 0.22 0.44 ± 0.24 0.03 ± 0.02 0.10 ± 0.07 Tricyclene 0.03 ± 0.02 0.01 0.11 ± 0.06 0.06 ± 0.02 0.01 0.04 ± 0.01 ND 0.06 ± 0.03 β-Pinene 0.43 ± 0.17 0.14 ± 0.02 1.24 ± 0.25 1.86 ± 0.21 0.32 ± 0.07 0.16 0.10 ± 0.05 0.10 ± 0.06 β-Myrcene 0.42 ± 0.24 ND 1.44 ± 0.58 0.15 ± 0.02 ND ND ND ND 3-Carene 0.21 ± 0.03 0.15 ± 0.04 0.04 ± 0.02 ND 0.29 ± 0.10 0.17 ± 0.10 ND <0.01 Limonene 1.12 ± 0.39 0.8 ± 0.28 0.88 ± 0.37 1.33 ± 0.76 0.95 ± 0.49 0.38 ± 0.28 0.41 ± 0.10 0.05 ± 0.02 Sabinene ND ND 0.02 ± 0.01 0.04 ± 0.02 0.08 ± 0.05 0.01 0.02 ± 0.01 0.14 ± 0.04 Trans-β-ocimene 0.03 ± 0.02 0.04 ± 0.02 0.01 ND 0.02 ± 0.01 0.01 0.06 ± 0.03 0.07 ± 0.04 β-Caryophyllene ND ND ND ND ND ND ND ND α-Caryophyllene ND ND ND ND ND ND ND ND 908 909 910 D) Midday Spring Summer Autumn Winter Control Nitrogen Control Nitrogen Control Nitrogen Control Nitrogen α-Pinene 2.04 ± 0.57 1.02 ± 0.14 2.65 ± 0.61 6.18 ± 2.12 0.43 ± 0.12 1.06 ± 0.38 0.12 ± 0.08 0.02 ± 0.01 Tricyclene 0.12 ± 0.04 0.01 0.35 ± 0.08 0.50 ± 0.18 ND 0.05 ± 0.03 ND 0.06 ± 0.04 β-Pinene 0.23 ± 0.06 0.29 ± 0.04 1.72 ± 0.26 3.72 ± 1.20 0.16 ± 0.08 0.34 ± 0.17 0.08 ± 0.04 0.06 ± 0.04 β-Myrcene 0.19 ± 0.12 0.38 ± 0.12 2.21 ± 1.19 0.66 ± 0.30 ND ND ND ND 3-Carene 0.37 ± 0.07 0.44 ± 0.03 0.43 ± 0.13 0.20 ± 0.08 ND 0.14 ± 0.08 0.02 ± 0.01 0.09 ± 0.06 Limonene 1.18 ± 0.55 0.32 ± 0.18 1.60 ± 0.33 5.45 ± 0.81 0.80 ± 0.21 1.10 ± 0.43 0.24 ± 0.03 0.15 ± 0.02 Sabinene ND ND 0.06 ± 0.04 0.04 ± 0.02 0.08 ± 0.05 0.13 ± 0.03 0.03 ± 0.02 0.01 Trans-β-ocimene 0.01 ND 0.07 ± 0.05 ND 0.04 ± 0.02 0.01 0.02 ± 0.01 0.08 ± 0.03 β-Caryophyllene 0.05 ± 0.02 0.07 ± 0.03 0.24 ± 0.15 ND ND ND ND ND α-Caryophyllene ND ND ND ND 0.28 ± 0.15 0.18 ± 0.08 ND 0.01 911 912 913 914 915 916 917 918 Table S2. Relationships for the rate of isoprenoid emissions with main environmental 919 conditions (air temperature and soil moisture) and physiological activities (net 920 photosynthetic rate and stomatal conductance) for Erica multiflora (A for isoprene and B 921 for total terpenes) and Quercus ilex (C for total terpenes) in the fertilized year (SE, 922 Standard Error). 923 A) Control Nitrogen Linear Exponential Linear Exponential R 0.5540 0.453 0.4299 0.3969 Air temperature P <0.0001 0.0018 0.0045 0.0093 SE 2.3937 2.5634 2.3847 2.4243 Correspondent equation y = -5.745 + 0.342x y = -2.504 + 0.183x R 0.4682 0.5086 0.3899 0.4059 Soil moisture P 0.0012 0.0004 0.0107 0.0076 SE 2.5406 2.4756 2.4323 2.4139 Correspondent equation y = 9.360*0.873 ˄x y = 7.686*0.874 ˄x R 0.4151 0.3937 0.0711 0.0636 Net photosynthetic rate P 0.0046 0.0074 0.6545 0.6893 SE 2.6159 2.6431 2.6346 2.636 Correspondent equation y = 5.780 - 0.650x y = 2.590 - 0.261x R 0.4893 0.5177 0.3110 0.2993 Stomatal conductance P 0.0006 0.0003 0.0450 0.0541 SE 2.5076 2.46 2.5103 2.5202 y = 7.718*8.277E-010 Correspondent equation y = 3.532 - 22.24x ˄x 924 925 926 927 928 929 930 931 932 933 934 935 936 937 B) Control Nitrogen Linear Exponential Linear Exponential R 0.5260 0.4861 0.4887 0.514 Air temperature P <0.0001 0.0001 0.0006 0.0003 SE 0.8429 0.8661 1.2814 1.2599 Correspondent equation y = -1.155 + 0.090x y = 0.117*1.097 ˄x R 0.5199 0.5496 0.4272 0.4832 Soil moisture P <0.0001 <0.0001 0.0031 0.0007 SE 0.8466 0.828 1.3279 1.2859 Correspondent equation y = 3.423*0.892 ˄x y = 5.669*0.858 ˄x R 0.0712 0.0704 0.2844 0.2674 Net photosynthetic rate P 0.6021 0.6062 0.0527 0.0692 SE 0.9886 0.9887 1.4024 1.4096 Correspondent equation y = 0.835 - 0.037x y = 0.107 + 0.226x R 0.1552 0.1537 0.0808 0.0766 Stomatal conductance P 0.2769 0.2816 0.6066 0.6252 SE 1.0017 1.0019 1.4908 1.4913 Correspondent equation y = 1.419 - 5.228x y = 1.065 - 3.286x 938 939 C) Control Nitrogen Linear Exponential Linear Exponential R 0.5373 0.6177 0.6304 0.6572 Air temperature P 0.0001 <0.0001 <0.0001 <0.0001 SE 3.2277 3.0096 5.1507 5.0009 Correspondent equation y = 0.165*1.135 ˄x y = 0.317*1.125 ˄x R 0.4611 0.4831 0.3836 0.392 Soil moisture P 0.0013 0.0007 0.0114 0.0089 SE 3.3960 3.3508 6.1275 6.1042 Correspondent equation y = 19.18*0.832 ˄x y = 45.42*0.773 ˄x R 0.4064 0.5558 0.5205 0.531 Net photosynthetic rate P 0.0051 <0.0001 0.0008 0.0006 SE 3.4968 3.1816 5.6657 5.6225 Correspondent equation y = 15.45*0.753 ˄x y = 17.44*0.769 ˄x R 0.3069 0.4334 0.4407 0.5144 Stomatal conductance P 0.0380 0.0026 0.0056 0.0013 SE 3.6424 3.4489 5.9563 5.6901 y = 13.47*6.120E-011 y = 17.36*6.628E-011 Correspondent equation ˄x ˄x 940 941 942 943 Figures 944 945 946 947 948 949 950 951 GAR 952 953 PRA 954 955 956 957 958 959 960 961 962 963 964 965 966 PRA GAR 967 968 969 Fig. S1. Location and landform of experimental sites. GAR, Garraf; PRA, Prades. 970 971 972 973 974 975 976 Garraf (Erica multiflora) Prades (Quercus ilex) 40 Season P < 0.001 Season P < 0.001 Time P < 0.01 Time P < 0.01 30 Air temperature (°C) 20 10 N N 0 30 Season P < 0.001 Season P < 0.001 Morning control Morning nitrogen Midday control Midday nitrogen Soil moisture (%) 20 10 N N 0 r r er r r r r r rin g me mn nte rin g mn nte rin g me mn nte rin g me mn nte Sp um utu Wi Sp mm utu Wi Sp um utu Wi Sp um utu Wi 14 4S 4A 15 15 5 Su 5A 16 14 4S 4A 15 15 5S 5A 16 20 1 20 1 20 2 0 1 20 1 20 20 20 1 01 20 2 0 20 1 01 20 20 20 2 2 Season 977 978 979 Fig. S2. Seasonal morning and midday variation of mean air temperature and soil 980 moisture in Garraf and Prades. ‘N’ indicates the start of the fertilization treatment. Error 981 bars indicate standard errors of the means (n = 6). The significance of the effects of season 982 and sampling time (repeated-measures ANOVA) is depicted in the panels. 983 984 985 986 987 988 989 990 A) Erica multiflora Morning Midday Control Season P < 0.01 Season P < 0.05 10 Nitrogen Season × treatment P < 0.05 Treatment P < 0.05 Net photosynthetic rate (µmol m s ) -2 -1 8 6 4 2 N N * 0 0.18 Season P < 0.01 Season P < 0.001 Time P < 0.05 Season × treatment P < 0.01 Stomatal conductance (mol m s ) -2 -1 0.15 0.12 0.09 0.06 0.03 * N * ** N 0.00 g r r er r r r er r rin me mn nte rin g mn nte rin g me mn nte rin g mn nte Sp um utu Wi Sp mm utu Wi Sp um utu Wi Sp mm utu Wi 14 4S 4A 15 15 5 Su 5A 16 14 4S 4A 15 15 5 Su 5A 16 20 20 1 20 1 20 2 0 20 1 20 1 20 20 20 1 20 1 20 2 0 20 1 20 1 20 Season 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 B) Quercus ilex 1006 Morning Midday 14 Season P < 0.001 Season P < 0.001 Time P < 0.01 Control Net photosynthetic rate (µmol m s ) -2 -1 Nitrogen 12 10 8 6 N 4 ** N 2 0 0.18 Season P < 0.001 Season P < 0.001 Time P < 0.001 Stomatal conductance (mol m s ) -2 -1 0.15 0.12 0.09 0.06 N 0.03 N 0.00 rin g er n er ng me r n er rin g er n er ng me r n er Sp mm tum int pri tum int Sp mm tum int pri tum int Su Au 5W 5S um Au 6W Su Au 5W 5S um Au 6W 14 14 14 01 01 5S 15 01 14 14 14 01 01 5S 15 01 20 20 20 2 2 2 01 20 2 20 20 20 2 2 2 01 20 2 Season 1007 1008 1009 Fig. S3. Seasonal net photosynthetic rates and stomatal conductances for Erica multiflora 1010 (A) and Quercus ilex (B) in the morning and at midday. ‘N’ indicates the start of the 1011 fertilization treatment. Error bars indicate standard errors of the means (n = 6). Significant 1012 differences between treatments identified by Student’s t-tests are indicated by asterisks 1013 (*, P < 0.05; **, P < 0.01). The significance of the effects of season, treatment and 1014 sampling time (repeated-measures ANOVA) is depicted in the panels. 1015 1016 1017 1018 Season P < 0.01 Season P < 0.01 1.5 Control Time × treatment P < 0.01 Nitrogen Rate of a-pinene emissions 1.2 ( µg g dw h ) -1 0.9 -1 0.6 0.3 0.0 N N Season P < 0.01 Season P < 0.05 1.0 Rate of limonene emissions 0.8 ( µg g dw h ) -1 0.6 -1 0.4 0.2 0.0 N N r r r r ng r r r r pri ng me mn nte rin g me mn nte pri me mn nte rin g me mn nte S um utu Wi Sp um utu Wi S um utu Wi Sp um utu Wi 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 14 14S 1 4A 015 2 015 15S 1 5A 016 20 20 2 20 20 2 20 20 2 20 20 2 1019 Season 1020 1021 Fig. S4. Seasonal variation of the rates of emission of α-pinene and limonene for Erica 1022 multiflora. ‘N’ indicates the start of the fertilization treatment. Error bars indicate 1023 standard errors of the means (n = 6). The effects of season, treatment and sampling time 1024 are depicted in the panels when significant. 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 Fig. S5. Distribution of seasonal isoprenoid emissions for Erica multiflora in the morning 1039 and at midday for the fertilized year. 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 A) Control : r = 0.42 P < 0.01 Control : r = 0.52 P < 0.001 Rate of isoprene emissions for E. multiflora Nitrogen : ns Nitrogen : r = 0.31 P < 0.1 Control 15 Nitrogen Correlation curve for control Correlation curve for nitrogen 10 ( µg g dw h ) -1 -1 5 0 8 Rate of total terpene emissions for E. multiflora Control : ns Control : ns Nitrogen : ns Nitrogen : ns 6 ( µg g dw h ) -1 4 -1 2 0 0 3 6 9 0.00 0.03 0.06 0.09 0.12 0.15 Net photosynthetic rate (µmol m-2 s-1) -2 -1 Stomatal conductance (mol m s ) 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 B) Control : r = 0.43 P < 0.01 Control Rate of total terpene emissions for Q. ilex 30 Nitrogen : r = 0.51 P < 0.01 Nitrogen Correlation curve for control Correlation curve for nitrogen Control : r = 0.56 P < 0.0001 20 Nitrogen : r = 0.53 P < 0.001 ( µg g dw h ) -1 -1 10 0 0 5 10 15 0.00 0.05 0.10 0.15 0.20 0.25 -2 -1 -2 -1 Net photosynthetic rate (µmol m s ) Stomatal conductance (mol m s ) 1065 1066 Fig. S6. Relationships for the rate of isoprenoid emissions with main physiological 1067 activities (net photosynthetic rate and stomatal conductance) for Erica multiflora (A) and 1068 Quercus ilex (B) in the fertilized year. ns, not significant. 1069 1070