Heliyon 10 (2024) e23854 Contents lists available at ScienceDirect Heliyon journal homepage: www.cell.com/heliyon Measuring BVOC emissions released by tomato plants grown in a soilless integrated rooftop greenhouse Gaia Stringari a, Joan Villanueva a, Elisa Appolloni b, Francesco Orsini b, Gara Villalba a, c, Xavier Gabarrell Durany a, c, * a Institut de Ciència i Tecnologia Ambientals ICTA-UAB (CEX2019-0940-M), Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain b Department of Agricultural and Food Sciences, University of Bologna Alma Mater Studiorum, Bologna, Italy c Department of Chemical, Biological and Environmental Engineering, Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain A R T I C L E I N F O A B S T R A C T Keywords: Urban design is currently promoting the inclusion of plants in buildings. However, plants emit Volatile sampling biogenic volatile organic compounds (BVOCs), which alone or in combination with other airborne Solanum lycopersicum L. molecules such as CO2, may result in a general increase in tropospheric pollution. Many studies i-RTG have documented the effects of biotic and abiotic factors on plant BVOC responses, but few have CO2 monitoring Indoor agriculture assessed the contribution of typical CO2 levels found in indoor work and meeting spaces. To Urban agriculture answer this question, we monitored CO2 and constitutive (MT-limonene) and induced (LOX-cis-3- hexenal) BVOC emissions of a fully developed tomato crop grown hydroponically inside an in­ tegrated rooftop greenhouse (i-RTG) in a Mediterranean climate. Two distinctive CO2 assays were performed at the level of the i-RTG by supplying or not CO2. The impact of CO2 on plant phys­ iological emittance was then assessed, and the resulting BVOC rates were compared with refer­ ence to EU-LCI values. MT-limonene was ubiquitous among the assays and the most abundant, while LOX-cis-3-hexenal was detected only under controlled CO2 management. The highest levels detected were below the indicated LCIs and were approximately tenfold lower than the corre­ sponding LCI for MT-limonene (50.88 vs. 5000 μg m− 3) and eightfold (6.63 μg m− 3) higher than the constitutive emission level for LOX-cis-3-hexenal. Over extended sampling (10 min) findings revealed a general emission decrease and significantly different CO2 concentration between the assays. Despite similar decreasing rates of predicted net photosynthesis (Pn) and stomatal conductance (gs) their correlation with decreasing CO2 under uncontrolled condition indirectly suggested a negative CO2 impact on plant emission activity. Conversely, increasing CO2 under the controlled assay showed a positive correlation with induced emissions but not with constitutive ones. Because of significantly higher levels of relative humidity registered under the uncontrolled condition, this factor was considered to affect more than CO2 the emission response and even its collection. This hypothesis was supported by literature findings and attributed to a common issue related with the sampling in static enclosure. Hence, we suggested a careful monitoring of the sampling conditions or further improvements to avoid bias and underestimation of actual emis­ sions. Based on the main outcomes, we observed no evidence of a hazardous effect of registered CO2 rates on the BVOC emissions of tomato plant. Furthermore, because of the low BVOC levels measured in the i-RTG, we assumed as safe the recirculation of this air along building’s indoor environments. * Corresponding author. Department of Chemical, Biological and Environmental Engineering, Universitat Autònoma de Barcelona (UAB), Campus UAB, 08193 Bellaterra, Barcelona, Spain. E-mail address: xavier.gabarrell@uab.cat (X. Gabarrell Durany). https://doi.org/10.1016/j.heliyon.2023.e23854 Received 31 July 2023; Received in revised form 1 December 2023; Accepted 14 December 2023 Available online 19 December 2023 2405-8440/© 2023 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). G. Stringari et al. Heliyon 10 (2024) e23854 1. Introduction Worldwide climate emergencies and intense gentrification are forcing the scientific community and citizens to enact timely actions for the containment of forecasted environmental and socioeconomical contingencies [1]. Among the effective strategies recom­ mended, the systematic expansion of green infrastructures in urban centres has been encouraged, particularly by European councils and local neighbourhoods [2,3], to solve issues related to food accessibility and social exclusion [4–6] as well as to tropospheric contamination [7–9]. The environmental performance of existing buildings can be improved by integrating plants inside and outside of them [10], not necessarily by means of high-cost technologies. Innovative systems, such as integrated rooftop greenhouses (i-RTGs) [11–13], have demonstrated the multiple benefits provided in terms of energy use efficiency [12], carbon emissions reduction [14], and energy and food production [15,16]. The benefits associated with plant removal capacity of air contaminants have been widely reviewed in the green retrofitting literature [7,17–20]. However, this capacity seemed to be overestimated with respect to the measured removal efficiency [21–23] due to species-specific plant mechanisms and to the same environment considerably affecting plant response in terms of indoor and outdoor air recovery [17,21,24]. In the wild, plants can release [25,26] tons of biogenic volatile organic compounds (BVOCs) annually into the atmosphere during photosynthetic activity [27–29], which have also been detected in urban areas [30,31] and domestic spaces [23,32,33]. The ultimate impact of BVOCs on indoor air quality is still uncertain [17,32,34], although because of the high concentration and chemical reactivity, potential harm to human health has been warned [35–37]. For some BVOCs, concern levels were assessed and included along with those of other volatiles in the EU-LCI tables for enclosed spaces [38,39]. Depending on indoor conditions, the composition and amount of BVOC emissions change and can increase considerably, as observed under elevated ozone [37,40] and temperature [25,41] levels. Emission blend mutations have been referred to as a substantial adaptation of plants to the presented environment [42,43] and were demonstrated to follow typical patterns. Without stimuli, either biotic or abiotic, plants emit constitutive BVOCs, mainly mono- and sesquiterpenes, regulated by the temperature–light gradient [25,44,45]. Under non­ physiological conditions, commonly during a stress event (e.g., drought and insect plague), plants emit predominantly volatile der­ ivates of the lipoxygenase pathway [25,46]. Among the stress factors affecting BVOC emissions, the implication of CO2 has also been studied [47,48] given the higher levels recorded currently than in the past 30 years and those foreseen [49]. Although there is still uncertainty about the effect of CO2 at both ambient and elevated concentrations, a trend toward reduced plant response under elevated CO2 has been documented [50–52], although it is connected to the length of exposure. More specifically, this has mainly been detected at short-term exposure. Longer term exposure was believed to result in plant adaptation with almost no consequences on total emission release [48,53]. Additionally, combining high CO2 with increasing temperature and other common stresses, such as drought, was shown in the short term to have a contrasting effect to the inhibiting effect of CO2, favouring plant emission activity [54–56]. In a recent study [57], different CO2 regimes were applied for a short time, individually or in combination with defined temperature rates, to the leafy species Artemisia annua L. The results demonstrated that a decreasing effect on total emissions was significant only when the plant was exposed to both elevated CO2 (800 ppm) and temperature (>30 ◦ C); this also confirmed the predominant effect of temperature on the induction of plant emissions. However, high variability and different outcomes were also noted depending on different factors, for instance, the sampling condition (field or laboratory study), the plant status and the species [51,58]. Among indoor crops, tomato (Solanum lycopersicum L.) is one of the top crops because of its remarkable nutritional and commercial aspects [59–61]. Regarding the volatile profile, tomato plants have been investigated under conventional and stressed conditions to determine BVOC variations between constitutive and induced emissions [62,63]. Among the wide range of BVOC classes, terpene components are usually detected in physiological emission blends [64,65]. Monoterpene and sesquiterpene release via plant photo­ synthesis was reported under a positive temperature gradient sustained over time [64,66]. Despite the ubiquity of terpene [67], induced emissions from tomato may change substantially according to elicitation and intensity subordination [63]. Lipoxygenase derivates were the volatiles most documented under extreme temperature rates [62], ozone fumigation [68], insect wounding [69], and combined biotic and abiotic stresses [70,71]. To the best of our knowledge, potential variations in tomato emissions associated with different CO2 levels have not been reported, since related measurements were conducted at ambient CO2 levels only. Compared to most horticultural species, tomato plants account for a considerable emitting surface, and therefore, BVOC amounts eventually reach levels of attention over prolonged exposure [72]. Few studies have been conducted on the assessment of tomato emissions into the environment [66,73–75]. Among these, their association with CO2 was briefly mentioned [66]. Additionally, in an i-RTG on the ICTA building on the Universitat Autònoma de Barcelona campus, the emissions from a leafy plant (Phaseolus vulgaris L.) were previously investigated through a static chamber under ambient conditions by checking temperature, light and relative humidity parameters (with the exception of CO2), and emission rates below the advised EU-LCI [76] were found. Emissions from larger plants (e. g., tomato), which in southeast-oriented i-RTGs [77] are cultivated annually, have not yet been studied. Compared to a controlled greenhouse [66,70,73,74] and although frequently adjusted to crop requirements, climatic conditions in i-RTGs are much more variable due to the top location, which enhances exchanges with the surroundings. Additionally, according to the building’s recir­ culation system, air coming from the greenhouse floor could be efficiently exploited to warm up the lower space, increasing the living quality of indoor environments, such as offices [12]. Therefore, in this study, in addition to measuring tomato plant emissions, the contribution of CO2 was also analysed. This was performed under two sampling conditions in a static chamber, one at the CO2 con­ centration generated by the enclosure and one at a controlled concentration maintained at the ambient (i-RTG) levels. The research questions investigated were as follows: 2 G. Stringari et al. Heliyon 10 (2024) e23854 − Are the BVOC emissions of tomato plants grown under i-RTG conditions below the reference EU-LCI values to be safely recirculated to indoor spaces? − Does CO2 control significantly affect constitutive BVOC emissions during sampling? − If so, do emissions increase or decrease under uncontrolled CO2? 2. Materials and methods 2.1. Experimental conditions The study was conducted in two consecutive warm seasons (spring-summer), with one experiment conducted in 2021 and the following in 2022. Plant cultivation and experimental setup were carried out hydroponically in the abovementioned i-RTG (Fig. 1A.2). In both campaigns, Solanum lycopersicum cv. Arawak was grown in the greenhouse and intercropped with another cultivar, cv. Siranzo in 2021 and Rosa de Cadiz in 2022. Cultivars Siranzo and Arawak were chosen for the volatile profile investigation in 2021 and 2022, respectively. Despite slight morphological differences, the emission blend composition of the plants was assumed to be similar and consistent with previous studies (see Table S1 in the Supplementary Materials). In both cases, the hydroponic system consisted of 12 parallel rows of perlite substrate bags (40 L) where plants were distributed equally along the row (0.30 m) and between the rows (0.80 m) (Fig. 1A.1), providing a density of 3 plants per m2 and a total cultivated area equal to 41.8 m2 in 2021 and 27.8 m2 in 2022. In the two years, the crop cycle in the i-RTG had an equivalent duration of approximately 5 months (135 days). The duration of the cycle was defined as days after transplanting (DAT), which occurred in mid-March, and it finished at the end of July. Cultivation parameters were checked continuously and adjusted when necessary to maintain adequate and stable growing con­ ditions over the season. This included daily measurements of the pH and electrical conductivity (EC) of water tanks, dripped solution and leachates, as well as of environmental temperature, relative humidity, and radiation (CS215 system, Campbell Scientific and L202, Hukseflux). Ambient CO2 was periodically measured in the greenhouse space with a portable gas analyser (U-CO2-915, GLA132 Series, LGR) to provide updated ambient levels. The internal conditions of the i-RTG were managed by a remote station (Siemens Building Technologies Ltd) that regulated the opening of the surrounding façade to dissipate the passive heat collected from the building. Plant development was monitored weekly based on morphological traits by measuring stem height and leaf area, counting the blossoms, and weighing the biomass collected from the pruning and harvested fruits. Additionally, over the whole cultivation cycle of each season, the heating degree days (HDDs) inside the i-RTG were calculated by applying Equation [1] of Ref. [78]. The single HDD was derived from the average of the maximum and the minimum temperature (Tmax, Tmin) and the lowest temperature at which tomato can thrive (Tbase), which was fixed at 10 ◦ C. This value provided the minimum temperature required for the plant’s metabolic activity. This information was then used as a direct and indirect measurement of plant growth in the i-RTG to determine crop resemblance and diversity among the two cultivation seasons. (Tmax + Tmin ) HDD = − Tbase [1] 2 2.2. Volatile monitoring and CO2 management BVOC investigation was conducted on mature plants at first fruit ripening (82 DAT) over the whole productive stage. Emissions were concentrated in a static enclosure and trapped dynamically onto Anasorb tubes (CSC, 6 × 70 mm–100/50 mg, SKC) at 250 mL min− 1 constant flow using the sampling settings reported in a previous study for the determination of green bean emissions [76]. The aerial part of one tomato plant was gently enclosed in a suspended LDPE cylindrical chamber (1.16 m diameter × 2.23 m height) to avoid mechanical injury and minimize the load of the plastic wall around the plant (Fig. 2A.2). In our study, we investigated the emission response to CO2 variation because of the static enclosure. First, the inertia of the static enclosure was determined by measuring radiation (Rs), temperature (T), relative humidity (RH), and CO2 fluctuation in the empty chamber over different time intervals (5–60 min). Then, the variation in CO2 before and during plant enclosure was recorded every minute with the portable gas Fig. 1. Crop distribution in the i-RTG in the early growing stage (A.1) and a panoramic view from the building’s atrium of mature tomato plants in the i-RTG (A.2). Both pictures were taken in the 2021 season. 3 G. Stringari et al. Heliyon 10 (2024) e23854 analyser. After 5 min of closure, a progressive drop in CO2 relative to the starting concentration was observed. Therefore, short consecutive samplings of 5 (1.25 L) and 10 (2.50 L) minutes were carried out to contain alterations in constitutive plant emissions under uncontrolled CO2 (=unadministered CO2). Feedback on plant physiological performance was derived from 1-min records of Rs, T, RH, and CO2 (Fig. 2A.1). In addition, plant photosynthetic rate (Pn) and stomatal conductance (gs) were estimated based on the empirical formulas of Ref. [79,80], respectively, in absence of measuring tools. Missing input data for the calculation of gs were retrieved from comprehensive literature [81] based on C3 plants. The results obtained in 2021 were compared with an equivalent trial performed the following season (2022). Air samples were instead collected at CO2 levels equivalent to the ambient, achieved by controlled CO2 purge (≥99% purity) into the chamber throughout the length of sampling (=administered CO2). Sampling was usually carried out between 10:00 a.m. and 1:00 p.m. (CET) because most of the greenhouse tasks were performed in the morning, and this time was thus considered the longest period of exposure to plant emissions. Throughout the season, volatile collection was repeated to provide replicates for the statistical analysis. Blank samples from a pure air tank were generated as benchmarks for all samples collected. 2.3. Determination of selected BVOCs in the sampled emissions Only two elective volatiles were analysed in the total emissions sampled that represented specific BVOC classes of constitutive and induced tomato plant emissions (see Table S1 in the Supplementary Materials): limonene for the monoterpenes (MT) and cis-3-hexenal for the lipoxygenase derivatives (LOX). Quantitation of BVOCs was performed following the NIOSH protocol [82] BVOCs trapped in Anasorb tubes were extracted with 1 mL of liquid carbon disulfide (CS2 for spectroscopy, ≥99.9%, Thermo Scientific) collected in 1.5-mL vials. Samples from 2021 were analysed using a Thermo GC‒MS system (GC: TRACE 1300/1310, MS: ISQ7000, Thermo Scientific, USA) using the working conditions described elsewhere [76]. Samples obtained in 2022 were analysed using an Agilent GC‒MS system (Agilent 7890A GC coupled to a 5975C MS). Calibration with external standard solutions ensured data comparability between both instruments. The instrumental conditions were maintained constant among instruments. Extract volumes of 2 μL were injected in splitless mode (1 min, 240 ◦ C), and BVOCs were separated using a TG-WAXMS capillary column (TraceGOLD™, 30 m × 0.32 mm × 1 μm, Thermo Scientific, USA) with a constant helium flow (3 mL min− 1). The GC oven programming consisted of an initial temperature step at 35 ◦ C (held for 2 min), followed by a 10 ◦ C min− 1 gradient up to 215 ◦ C (maintained for 4 min). The MS transfer line temperature was set at 250 ◦ C, and the ion source was heated at 150 ◦ C with 70 eV as the ionization potential. Data acquisition was performed in SIM mode to optimize instrumental sensitivity and reproducibility. Data processing and BVOC quantitation were performed with either MassHunter Workstation 10.1 (Agilent GC/MS data) or Chromeleon Software (Thermo GC/MS data). Compound identification and quantitation were performed by comparison of peak retention times and peak areas with external standard solutions as described elsewhere [76]. In this study, the instrumental lower limit of quantitation (LLOQ) corresponds to the concentration value that can be quantified with a relative standard deviation (RSD%) ≤ 21% upon consecutive analysis (See Table S2 in the Supplementary Materials). The instrumental dynamic range was typically between 10 and 100 ng mL− 1. The calculation of the BVOC emission rate (ERBVOC) was based on the canonical Equation [2] for the determination of the emission rate of a volatile. This is the difference between the products of the volatile emitted by the plant (BVOCp, μg m− 3) in the sampled volume (Vc, m3) and its amount found in a blank sample (either in the empty chamber or in scrubbed air, BVOCc) divided by the product between the sampling time (h, hr) and either the fresh weight (gfw, g) of the leaves or of the total biomass (stem + leaves + fruits) enclosed, or the projected leaf area (m2). In the results section, this is referred to as the weight of fresh leaves. Fig. 2. Simplified representation of the dynamic (active) emission sampling with static enclosure, with or without CO2 supply (A.1). Picture taken during the setup of the sampling chamber on a tomato plant (A.2). 4 G. Stringari et al. Heliyon 10 (2024) e23854 ( ) BVOCp × Vc − (BVOCc × Vc ) ERBVOC = [2] gfw × h 2.4. Statistical analysis and variable impact calculation Individual BVOC emissions were compared between the sampling intervals (5 and 10 min) within the same CO2 conditions (administered or unadministered) and between the CO2 conditions within the sampling intervals using the Wilcox test (paired or unpaired). Dependencies and independencies between the variation rates of monitored and predicted variables (Rs, T, RH, and CO2 concentration) were also compared using linear regression models and ANOVA tests (type II). The relationships between the sampled BVOC emissions at the 5- and 10-min intervals of each CO2 assay and the considered variables were preliminarily assessed through principal component analysis (PCA). Finally, the variation rates of both emissions and variables were log-transformed before their computation in paired plot matrices for the determination of the correlation coefficient (R2). Differences were assigned as significant at a probability value p ≤ 0.05, and correlations were considered meaningful at a R2 coefficient comprised between ±0.50 and ±1 and significant at a probability value p ≤ 0.05. All tests were performed with R software (version 4.2.2). 3. Results 3.1. Crop heat accumulation and growth during the two cultivation seasons The sum of the heating degree days (HDDs) calculated from the DAT in the i-RTG until the last sampling day for each of the two experimental seasons (see Fig. S1 in Supplementary Materials) was significantly higher (p = 0.001) in 2021 (3039.82) than in 2022 Fig. 3. Average ± SEM (n = 9, 2021; n = 5, 2022) ERs of the monoterpene (limonene) and lipoxygenase derivative (cis-3-hexenal) detected in the static enclosure containing one mature plant at 5- and 10-min sampling intervals under different CO2 assays. Different lowercase letters indicate significantly different emissions (p ≤ 0.05); average ERs of the single volatile are compared between the CO2 assays within the same sampling interval (A.1) or between the sampling intervals within the same CO2 assay (A.2). 5 G. Stringari et al. Heliyon 10 (2024) e23854 (2323.41). In detail, comparing the maximum and minimum T rates of the two periods, there was a significant difference (p = 0.001) between the Tmax average of 2022 (37.12 ◦ C) and 2021 (32.81 ◦ C). In fact, BVOC collection in 2022 covered approximately one week less than in 2021, and it was concluded 21 days before the last day of sampling in 2021. Considering plant morphological development over the growing cycles (see Table S3 in the Supplementary Materials), the mean total fresh biomass (stem + leaves + fruits) measured Fig. 4. Mean ± SD of CO2 levels interpolated with the monitored environmental variables (T: ‘A’ and ‘B’; RH: ‘C’ and ‘D’) and predicted physi­ ological ones (Pn: ‘E’ and ‘F’; gs: ‘G’ and ‘H’) under administered (left) and unadministered (right) CO2 condition along plant enclosure sampling. Red bars delimit the principal sampling intervals considered (5- and 10- minutes). 6 G. Stringari et al. Heliyon 10 (2024) e23854 in 2021 was significantly higher (4.23 kg, p = 0.007) than that measured in 2022 (3.08 kg). However, comparing individual organ weights, the overall difference was provided only by the mean weight of fruits (p = 0.002), rather than that of leaves or stem. 3.2. BVOC emission rates and amounts in the i-RTG with respect to indoor LCI values Average emission rates (ERs) for the two volatiles at both the 5- and 10-min sampling intervals under the two different CO2 assays are reported in Fig. 3 and summarized in Table S6 of the Supplementary Materials. During the two experimental seasons, MT-limonene was usually detected under both the administered and unadministered CO2 assays, while LOX-cis-3-hexenal was detected only under administered CO2 (Fig. 3A.1). The emissions of both volatiles were higher in the first 5 min than in the following 10. Between short and long sampling intervals there was no significant difference in the levels of limonene (p > 0.1) in either of the two CO2 assays, whereas a significant difference was observed in those of cis-3-hexenal (p = 0.002) (Fig. 3A.2). Overall, Limonene emissions were substantially higher but not significantly different under the administered assay (Fig. 3A.1). Moreover, under the same condition, MT emissions were on average the most abundant in both the 5- and 10-min samples, accounting for 65.2% and 72.8%, respectively, compared to LOX emissions, which accounted for 34.8% and 27.2%, respectively. Likewise, the average minimum and maximum percentages collected during 5 and 10 min were 44.4–78.9% and 50–87.2% for limonene and 21–55.6% and 12.8–50% for cis-3-hexenal, respectively. Considering the calculated ERs (Table S6), for the constitutive MT, the highest amount was reported under unadministered CO2 (1752.19 ng g− 1 h− 1), while for the induced LOX, it was reported under the administered assay (181.8 ng g− 1 h− 1). 3.3. Enclosure performance prior to emission sampling The tested inertia of the empty chamber provided no significant variation in the measured Rs, T, RH, and CO2 concentration during the whole enclosure time (up to 60 min). Radiation inside and outside the chamber did not vary significantly. Over the whole cycle, Rs recorded at the crop level (see Table S4.1 in the Supplementary Materials) was approximately 29% (2021) and 48% (2022) of the total Rs measured 2 m above the canopy, which was far below the standard illumination supplied in other greenhouses [57,74,83]. However, according to previous assessments performed on this i-RTG [84], such rates were still consistent with those found in common greenhouses and would ensure plant photosynthetic activity. Therefore, the effect of radiation on CO2 assimilation was not contemplated henceforth. 3.4. Comparison of environmental and physiological trends across different CO2 assays Across the CO2 assays, between 5- and 10-min sampling intervals no significant differences (p > 0.1) were highlighted in the mean values (Table S4) of both the monitored and predicted variables. However, between the assays significantly higher levels of RH and gs under the unadministered condition were observed at both 5- (+16%, p = 0.04) and 10- (+18%, p = 0.02) minutes sampling for RH, at 10 min only (+124 mmol m− 2 s− 1, p = 0.02) for gs. Fig. 4 reports variables performance in relation to the CO2 tendency during the enclosing time. According to the mean values registered, with respect to the i-RTG concentration by the end of the first and the second sampling term CO2 decreased up to 14 and 30 ppm under the unadministered condition and increased up to 8 and 11 ppm under the administered one. However, according to the variation rates, a general CO2 drop was always observed within the first 5 min of enclosure across both the assays. Although the decrease was higher under unadministered (− 57.01 ppm) than under unadministered (− 17.60 ppm) CO2 it was not statistically significant (p = 0.24). However, on the overall 10 min CO2 decrease under the first condition (− 69.04 ppm) was then significantly higher (p = 0.042) than under the second one (− 11.22 ppm) where partial recovery of CO2 (+6.38 ppm) occurred. Considering T and RH, throughout the sampling a progressive increase was registered under both CO2 assays (Fig. 4A and B; C and D), with overall higher RH increase (±18%) than T one (±1.50 ◦ C). Despite slightly higher rates observed under the unadministered (0.40–0.80 ◦ C; 9.70–11.57%) than under the administered condition (0.23–0.41 ◦ C; 2.15–3.64%) no significant differences (p > 0.05) were individuated. Consistent with CO2 trend, also predicted Pn and gs were found to decrease along the sampling (Fig. 4E and F; G and H). Within the individual assay substantial Pn and gs decrease was higher under administered CO2 (− 21.53–22.91 μmol CO2 m− 2 s− 1 and − 413–442.72 mmol m− 2 s− 1) rather than the unadministered condition (− 6.40–10.38 μmol CO2 m− 2 s− 1 and − 64.39–161.41 mmol m− 2 s− 1). However, their variation rates of between the assays were not significantly different. 3.5. Implication of the CO2 assay on plant emissions and the enclosing conditions Exploratory principal component analysis (PCA) scores and data distribution (see Fig. S2 and Table S5 in the Supplementary Materials) revealed that emissions collected under administered CO2 were primarily related to environmental factors, especially those sampled within the 5-min interval and mostly for the LOX volatile. Temperature and CO2 emerged as the main relevant variables, while Pn, gs, and RH in this order showed less influence. Conversely, the data suggested little to no relationship between environmental and physiological factors and MT emissions. Correlations between the five variables monitored and predicted are shown in Fig. S3 of the Supplementary Materials. Between constitutive and induced emissions there were no relationships (R2 ≤ 0.40) under any of the CO2 assays. Between the emissions and the five variables, there were correlations only under administered CO2. There were negative correlations between MT emission and increasing T (R2 = − 0.60) and RH (R2 = − 0.50), respectively in the first and in the second sampling interval. There was a positive correlation between LOX emission and decreasing CO2 (R2 = 0.60) in the second interval only. Among the variables several 7 G. Stringari et al. Heliyon 10 (2024) e23854 correlations were additionally individuated (Fig. S3). Within the 5 and the total 10 min of sampling T and RH were significantly positively correlated (R2 = 0.67 and 0.73, p ≤ 0.01) and mutually explained their increasing rate (p = 0.03 and p = 0.01) under unadministered CO2, whereas this was not found under the administered condition. The variation of CO2 appeared to be uncorrelated with T and RH in any of the two cases. However, significant positive correlations with CO2 were found under unadministered CO2 for the whole enclosure time with Pn (R2 = 0.78 and 0.73, p ≤ 0.01) and gs (R2 = 0.58 and 0.68, p ≤ 0.05 and 0.01). Decreasing CO2 levels also explained significantly (p ≤ 0.01 and 0.05) and partially significantly (p ≤ 0.03 and p = 0.07) both Pn and gs variations, respectively. These correlations were however not highlighted under administered CO2, where instead both Pn and gs were correlated with T in the first sampling term and with RH in both the terms. The relationship was negative (R2 = − 0.70) with the first but positive (R2 = 0.5 and 0.90, p ≤ 0.01) with the second. In contrast to CO2 in the unadministered assay, in this case none of the variables explained (p ≥ 0.05) the variation rates of the physiological variables. 4. Discussion 4.1. Effects of CO2 conditions on plant emissions during the two sampling intervals The aim of this study was to assess tomato plant emissions in an i-RTG and compare them to existing LCI values and levels found in previous assessments to evaluate the potential harm to human health in the i-RTG and in the offices where this air could be recir­ culated. Short sampling intervals of 5 and 10 min were used to investigate the impact of CO2, comparing two sampling conditions: controlled (=administered) and uncontrolled (=unadministered) environmental CO2 levels. The study also measured HDDs as a function of plant development during each separate season, which significantly differed only in terms of average Tmax that was higher during the experiment performed with the administration of CO2. However, the weight of traditional emitting biomass (leaves and stem), excepting the fruits, remained similar throughout the cycles. These preliminary outcomes suggested that investigated crops were rather similar, therefore emission variations were researched in the corresponding sampling conditions. Similar MT-limonene emissions were found across the different CO2 assays. However, administration of CO2 affected the detection of LOX-cis-3-hexenal volatiles, with significant differences observed between the two intervals. Overall, average emissions were higher in 5-min samples and under administered CO2. The most abundant and ubiquitous volatile was MT-limonene. Nevertheless, the highest emission recorded in the i-RTG (50.88 μg m− 3) over the 2021 season was approximately tenfold lower than the corresponding EU-LCI (5000 μg m− 3). Likewise, the highest LOX emission was at least eightfold lower (6.63 μg m− 3) than that of the MT. Comparing these levels with resembling studies on tomato emission assessment in greenhouse environments [66,73], there were some consistencies. In our i-RTG limonene and cis-3-hexenal levels were comprised between 0.01–9.13 and 0.27–1.65 ppb, respectively, which fell within the intervals found by Ref. [73] under normal or stressed conditions (0.035–40.4 ppb for MTs and 0–20 ppb for LOX volatiles). Under similar experimental conditions (open bottom chamber and similar inner CO2 trends), Ref. [66] retrieved for terpenes and LOX de­ rivatives emission indices (EIs) at steady and stressed condition. Higher EI (>3) was reported for terpenes than for LOX compounds in the first condition, then much lower EI (<2) for the first ones was measured than the second ones (EI > 6) during the stress occurrence. In contrast to these results equivalent limonene and cis-3-hexenal EIs measured under our CO2 assays were substantially similar (<3). Typically, emissions of MTs are observed to increase in response to a quasilinear temperature gradient [46,85] within a tolerant range for the plants before experiencing structural damages which may affect the normal emission activity. However, studies have reported emission disruptions due to occasional or prolonged temperature increases [86]. With respect to CO2, the relatively few studies conducted on the emission responses at determined CO2 concentrations provided conflicting or not fully consistent results [48]. Physiological MT emission is generally slowed by increasing CO2, as terpene production implies the activation of the same enzymatic system involved in the transportation of O2 during photosynthesis [42,46,86–88]. However, it was generally observed that supplying CO2 up to normal ambient concentrations (~400 ppm) did not have a relevant impact on MTs and overall emission feedback unless temperature rose without exceeding damage thresholds (over 35 ◦ C and up to 47 ◦ C) [54–57]. In contrast, applying high CO2 concentrations (>800 ppm) at high T would decrease overall emissions [56,57,86]. In our case, under administered CO2 flush within the first 5 min, the relationship between the positive T gradient and limonene emissions was found to be unexpectedly negative, and no relationship at all was found with increasing RH or with (decreasing) CO2 levels. Conversely, despite CO2 decrease, ERs were still slightly higher within 5 min than after CO2 stabilization achieved after until the end of sampling. In this second interval, MT were instead negatively correlated with the increasing RH rather than with T and still uncorrelated with CO2. Under unadministered CO2, no relationship was found with the constitutive emissions, but negative correlation between T and RH was maintained throughout the sampling, as well as the positive correlation between overall decreasing CO2, Pn and gs. . Based on these outcomes, it could be rationally assumed that T and RH had a relevant implication on the MT emission response when no administration is applied. The increase of RH has been traditionally associated with a common issue individuated during volatile sampling in static enclosures. As reported by experimental tests [89–91], within a relatively short timeframe (3 min), despite preconditioning and thorough sampling, humidity levels can increase quickly due to instant plant transpiration substantially affecting both the plant’s regular emittance and the goodness of the emission sampled [89–91]. This condition would be confirmed in our study by the tight correlations found between physiological Pn and gs and the CO2 variation that even explained the two variables. On the other hand, under the administered condition the highest T and RH registered (Tmax = 32 ◦ C, RHmax 60%) were much below extreme levels reported, plus no correlation was found between CO2 rates and the physiological response (Pn and gs decrease). This suggested that supplying CO2 up to fixed environmental levels perhaps mitigates the quenching action of RH on plant emission release and sampling bias. However, beyond 5 min, sampling constitutive emissions can be tricky making the sample less reliable due to the loss of stable conditions inside the static enclosure. 8 G. Stringari et al. Heliyon 10 (2024) e23854 LOX emissions have been extensively investigated in field studies, and they have been found to be rapidly released following an abiotic or biotic stress event, usually associated with mechanical injury of the cellular structure. Instead, MT emissions are mostly subordinated to T under a characteristic slow and extended release [43,88,92–95]. Currently, the relationship between LOX emissions and the variation of CO2 is unclear, mainly due to the few investigations on this topic. So far, there was no evidence of a correlation between high CO2 levels (720 μmol mol− 1) and these volatiles in comparison with environmental CO2 concentration did not proved. Furthermore, even the combination of a mechanical stressor (insect chewing) with each of these two conditions (high and ambient CO2 levels), missed this correlation [93,96]. It was also unclear the positive feedback of combined increasing T (below stress thresholds) and ambient CO2 levels, though hypothesis was scarcely supported due to the lower correlation between LOX volatiles and T compared to the MTs [67,92]. There were exceptions, such as in the case of tomato, where Ref. [62] found direct correlation with the increase of LOX emission and T under normal CO2 concentration starting from chilling and heating degrees (− 7 ◦ C and 49 ◦ C) because of a breakage in the cell membrane. However, further field studies [67,71] found out that stress-induced LOX emissions were quite common and independent of temperature increases. In the case study, under administered CO2, the significantly higher rates of LOX emissions collected in the short sampling interval were strongly correlated with the initial CO2 drop and still correlated when restoration of pre-sampling levels was achieved (10 min). Conversely, it should be noted that without CO2 supply, emissions were not detected. It is important to consider that the occurrence or absence of LOX compounds is likely due to a stress event, which may include a combination of environmental factors such as tem­ perature, humidity, and CO2 concentration. In the case of tomato plants, the connection between these volatiles and the type of elicitor [67,69,71,83] and the role of the LOX kinetic response in the expression and the total amount of volatile released has been demon­ strated [97]. Therefore based on the reviewed literature, the inverse correlation between progressive CO2 decrease plus significantly higher RH rates and LOX emissions could be hypothesized even if it was not determined. Conversely, MTs were generally found to be the most abundant and far ubiquitous under multiple stress conditions [64,67,73,83]. According to these findings and given the much higher gradient of RH (up to 71.9% at min 5 and up to 79.4% at min 10) under uncontrolled rather than controlled CO2 assays, humidity could again be considered the determining factor in the fulfilment of LOX-cis-3-hexenal collection. Eventually, RH may also be determined as the driving factor in the modulation of the stress response in plants. 5. Conclusions In this study, it was observed that BVOC emissions generally decreased over extended sampling under both unadministered and administered CO2. Similar variations of Pn and gs observed between the two assays suggested a similar physiological performance under the different CO2 conditions. However, in the unadministered assay the progressive decrease of CO2 appeared to be tightly connected with the decrease of both the physiological variables. Even if no relationships were found between constitutive nor induced BVOCs and the variation of these variables (Pn, gs, and CO2) lower emissions collected under this condition suggested, indirectly, the impact of low CO2 concentration on overall plant emission activity. On the other hand, in the administered assay positive correlations individuated among increasing CO2 and LOX emissions also suggested the implication of this variable on the release of induced emissions. Conversely, the absence of correlations with MTs made difficult to assess the actual impact of CO2 on constitutive emissions. Furthermore, it has been observed a significant implication of the increase of relative humidity on both MT and LOX emissions under unadministered condition that was likely to be connected to the sampling set up, as demonstrated in the literature. Instead, this effect was not fully individuated under administered condition where indeed lower RH levels were observed and that were positively correlated with plant physiological response. It could be reasonably assumed but not confirmed, then, that under administered condition CO2 supply partially contributed on plant BVOCs emission, whereas under unadministered condition its contribution remained unclear and was rather prevailed by relative humidity factor. To fill this gap, altogether proper adjustments of the sampling setting and accurate monitoring of all the variables involved may improve the performance of static sampling and reduce technical issues responsible of errors and incorrect BVOCs estimation. Overall, monoterpene and lipoxygenase derivatives emissions released by tomato plants in an i-RTG at environmental conditions (temperature, relative humidity, radiation, and CO2) are expected to be safe over longer periods of exposure. However, it should be noted that further research would be needed to confirm measured BVOC levels since for many of them referencing EU-LCI values are currently still missing (e.g., LOX derivatives). In addition, based on the configuration of the building, designed air recirculation in the closed spaces of the lower floors for energy and quality purposes is likely to be implemented in the future without bringing about dangerous conditions for human health. Data availability statement Data are included in the article and in the supplementary material referenced. Additional data will be made available on request. CRediT authorship contribution statement Gaia Stringari: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptuali­ zation. Joan Villanueva: Writing – review & editing, Validation, Supervision, Resources, Methodology, Data curation. Elisa Appolloni: Writing – review & editing, Investigation. Francesco Orsini: Writing – review & editing, Validation, Supervision, Re­ sources, Project administration, Methodology, Funding acquisition, Conceptualization. Gara Villalba: Writing – review & editing, 9 G. Stringari et al. Heliyon 10 (2024) e23854 Supervision, Resources, Conceptualization. Xavier Gabarrell Durany: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing in­ terests:Xavier Gabarrell reports financial support was provided by Spain Ministry of Science and Innovation. Francesco Orsini reports financial support was provided by H2020 Food Security Sustainable Agriculture and Forestry Marine Maritime and Inland Water Research and the Bioeconomy. Gaia Stringari reports was provided by Government of Catalonia Agency for Administration of Uni­ versity and Research Grants. Gara Villalba reports financial support was provided by European Research Council. Joan Villanueva reports financial support was provided by European Research Council. Acknowledgements This study was funded by the European Union under the European Horizon 2020 Research and Innovation Program under grant agreement n◦ 862663 (Food Systems in European Cities, FoodE); Ministerio de Ciencia e Innovación under grant n◦ PID2021- 126845OB-C21 (MOVE4EDU, MCIN/AEI/10.13039/501100011033/FEDER) and Maria de Maeztu award (CEX2019-000940-M); Ministerio de Ciencia e Innovación under grant n◦ TED2021-130047B-C21 (BINAFET, MCIN/AEI/10.13039/501100011033/) and the European Union under the grant NextGenerationEU/PRTR; Catalan Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR) under grant agreements n◦ 2020FI_B 01004, 2020 PANDE 00021 and 2021SGR00734 Sostenipra; European Research Council (ERC) Consolidator project, Integrated System Analysis of Urban Vegetation and Agriculture (818002-URBAG), and PALADYN Project (#834934). The publication reflects the author’s views. The Research Executive Agency (REA) is not liable for any use that may be made of the information contained therein. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.heliyon.2023.e23854. References [1] IPCC, Synthesis Report of the Ipcc Sixth Assessment Report (AR6), Panmao Zhai, Interlaken, Switzerland, 2023. https://report.ipcc.ch/ar6syr/pdf/IPCC_AR6_ SYR_LongerReport.pdf. (Accessed 10 April 2023). [2] L.C. Hui, C.Y. Jim, Y. Tian, Public views on green roofs and green walls in two major Asian cities and implications for promotion policy, Urban For. Urban Green. 70 (2022), 127546, https://doi.org/10.1016/j.ufug.2022.127546. [3] European Commission, Communication from the Commission to the European Parliament, The European Council, The Council, The European economic and social committee and the committee of the regions - the European Green Deal, Brussels, 2019. http://eur-lex.europa.eu/resource.html?uri=cellar:208111e4- 414e-4da5-94c1-852f1c74f351.0004.02/DOC_1&format=PDF. (Accessed 12 April 2022). [4] D. Gould, T. Caplow, 8 - building-integrated agriculture: a new approach to food production, in: F. Zeman (Ed.), Metropolitan Sustainability, Woodhead Publishing, 2012, pp. 147–170, https://doi.org/10.1533/9780857096463.2.147. [5] D. Buehler, R. Junge, Global trends and current status of commercial urban rooftop farming, Sustainability 8 (2016), https://doi.org/10.3390/su8111108. [6] J. Yudelson, Green Building Trends: Europe, Island Press, 2012. [7] F. Torpy, M. Zavattaro, Bench-study of Green-Wall Plants for Indoor Air Pollution Reduction, 2018. [8] P. Ramasubramanian, O. Starry, T. Rosenstiel, E.T. Gall, Pilot study on the impact of green roofs on ozone levels near building ventilation air supply, Build. Environ. 151 (2019) 43–53, https://doi.org/10.1016/j.buildenv.2019.01.023. [9] Y. Shao, J. Li, Z. Zhou, Z. Hu, F. Zhang, Y. Cui, H. Chen, The effects of vertical farming on indoor carbon dioxide concentration and fresh air energy consumption in office buildings, Build. Environ. 195 (2021), https://doi.org/10.1016/j.buildenv.2021.107766. [10] O. Pons, A. Nadal, E. Sanyé-Mengual, P. Llorach-Massana, E. Cuerva, D. Sanjuan-Delmàs, P. Muñoz, J. Oliver-Solà, C. Planas, M.R. Rovira, Roofs of the future: rooftop greenhouses to improve buildings metabolism, Procedia Eng. 123 (2015) 441–448, https://doi.org/10.1016/j.proeng.2015.10.084. [11] A. Nadal, P. Llorach-Massana, E. Cuerva, E. López-Capel, J.I. Montero, A. Josa, J. Rieradevall, M. Royapoor, Building-integrated rooftop greenhouses: an energy and environmental assessment in the mediterranean context, Appl. Energy 187 (2017) 338–351, https://doi.org/10.1016/j.apenergy.2016.11.051. [12] J. Muñoz-Liesa, S. Toboso-Chavero, A. Mendoza Beltran, E. Cuerva, E. Gallo, S. Gassó-Domingo, A. Josa, Building-integrated agriculture: are we shifting environmental impacts? An environmental assessment and structural improvement of urban greenhouses, Resour. Conserv. Recycl. 169 (2021), 105526, https://doi.org/10.1016/J.RESCONREC.2021.105526. [13] A.B. Milford, S. Kårstad, M. Verheul, Exploring the Opportunities for Building a Rooftop Greenhouse Case Study from Bergen, NIBIO Rapport, Norway, 2019. [14] J. Lampinen, O. García-Antúnez, A.S. Olafsson, K.C. Kavanagh, N.M. Gulsrud, C.M. Raymond, Envisioning carbon-smart and just urban green infrastructure, Urban For. Urban Green. 75 (2022), 127682, https://doi.org/10.1016/j.ufug.2022.127682. [15] F. Orsini, G. Pennisi, F. Zulfiqar, G. Gianquinto, Sustainable use of resources in plant factories with artificial lighting (PFALs), Eur. J. Hortic. Sci. 85 (2020) 297–309, https://doi.org/10.17660/eJHS.2020/85.5.1. [16] P. Zambrano, J. Muñoz Liesa, A. Josa, J. Rieradevall, R. Alamús, S. Gassó, X. Gabarrell Durany, Assessment of the food-water-energy nexus suitability of rooftops. Methodological remote sensing approach in a urban Mediterranean area, Sustain. Cities Soc. 75 (2021), 103287, https://doi.org/10.1016/j. scs.2021.103287. [17] M. Tomson, P. Kumar, Y. Barwise, P. Perez, H. Forehead, K. French, L. Morawska, J.F. Watts, Green infrastructure for air quality improvement in street canyons, Environ. Int. 146 (2021), 106288, https://doi.org/10.1016/j.envint.2020.106288. [18] T. Ysebaert, K. Koch, R. Samson, S. Denys, Green walls for mitigating urban particulate matter pollution—a review, Urban For. Urban Green. 59 (2021), 127014, https://doi.org/10.1016/j.ufug.2021.127014. [19] K. Gunawardena, K. Steemers, Living walls in indoor environments, Build. Environ. 148 (2019) 478–487, https://doi.org/10.1016/j.buildenv.2018.11.014. 10 G. Stringari et al. Heliyon 10 (2024) e23854 [20] S. Sharma, A. Bakht, M. Jahanzaib, H. Lee, D. Park, Evaluation of the effectiveness of common indoor plants in improving the indoor air quality of studio apartments, Atmosphere (Basel) 13 (2022), https://doi.org/10.3390/atmos13111863. [21] B.E. Cummings, M.S. Waring, Potted plants do not improve indoor air quality: a review and analysis of reported VOC removal efficiencies, J. Expo. Sci. Environ. Epidemiol. 30 (2020) 253–261, https://doi.org/10.1038/s41370-019-0175-9. [22] H.H. Kim, J.Y. Lee, J.Y. Yang, K.J. Kim, Y.J. Lee, D.C. Shin, Y.W. Lim, Evaluation of indoor air quality and health related parameters in office buildings with or without indoor plants, J. Jpn. Soc. Hortic. Sci. 80 (2011) 96–102, https://doi.org/10.2503/jjshs1.80.96. [23] M. Zorić, M. Simić, S. Orlović, E. Mladenović, Z. Babić, Indoor ecosystem services: impacts of plants on air quality, Contemp. Agric. 68 (2019) 12–16, https:// doi.org/10.2478/contagri-2019-0003. [24] F. Brilli, S. Fares, A. Ghirardo, P. de Visser, V. Calatayud, A. Muñoz, I. Annesi-Maesano, F. Sebastiani, A. Alivernini, V. Varriale, F. Menghini, Plants for sustainable improvement of indoor air quality, Trends Plant Sci. 23 (2018) 507–512, https://doi.org/10.1016/j.tplants.2018.03.004. [25] F. Loreto, C. Barta, F. Brilli, I. Nogues, On the induction of volatile organic compound emissions by plants as consequence of wounding or fluctuations of light and temperature, Plant Cell Environ. 29 (2006) 1820–1828, https://doi.org/10.1111/j.1365-3040.2006.01561.x. [26] C. Arena, T. Tsonev, D. Doneva, V. De Micco, M. Michelozzi, C. Brunetti, M. Centritto, S. Fineschi, V. Velikova, F. Loreto, The effect of light quality on growth, photosynthesis, leaf anatomy and volatile isoprenoids of a monoterpene-emitting herbaceous species (Solanum lycopersicum L.) and an isoprene-emitting tree (Platanus orientalis L.), Environ. Exp. Bot. 130 (2016) 122–132, https://doi.org/10.1016/j.envexpbot.2016.05.014. [27] J. Drewer, D. Howard, N. McNamara, Greenhouse Gas (GHG) and Biogenic Volatile Organic Compound (bVOC) Fluxes Associated with Land-Use Change to Bioenergy Crops, Elsevier Inc., 2018, https://doi.org/10.1016/B978-0-08-101036-5.00006-9. [28] J. Laothawornkitkul, J.E. Taylor, N.D. Paul, C.N. Hewitt, Biogenic volatile organic compounds in the Earth system: Tansley review, New Phytol. 183 (2009) 27–51, https://doi.org/10.1111/j.1469-8137.2009.02859.x. [29] X. Lun, Y. Lin, F. Chai, C. Fan, H. Li, J. Liu, Reviews of emission of biogenic volatile organic compounds (BVOCs) in Asia, J. Environ. Sci. 95 (2020) 266–277, https://doi.org/10.1016/j.jes.2020.04.043. [30] G. Churkina, F. Kuik, B. Bonn, A. Lauer, R. Grote, K. Tomiak, T. Butler, Effect of VOC emissions from vegetation on air quality in Berlin during a heatwave, Environ. Sci. Technol. 51 (2017) 6120–6130, https://doi.org/10.1021/acs.est.6b06514. [31] H. Zhang, Y. Zhang, Z. Huang, W.J.F. Acton, Z. Wang, E. Nemitz, B. Langford, N. Mullinger, B. Davison, Z. Shi, D. Liu, W. Song, W. Yang, J. Zeng, Z. Wu, P. Fu, Q. Zhang, X. Wang, Vertical profiles of biogenic volatile organic compounds as observed online at a tower in Beijing, J. Environ. Sci. 95 (2020) 33–42, https:// doi.org/10.1016/j.jes.2020.03.032. [32] D.S. Yang, K.C. Son, S.J. Kays, Volatile organic compounds emanating from indoor ornamental plants, Hortscience 44 (2009) 396–400, https://doi.org/ 10.21273/hortsci.44.2.396. [33] A. Tiwary, A. Namdeo, J. Fuentes, A. Dore, X.M. Hu, M. Bell, Systems scale assessment of the sustainability implications of emerging green initiatives, Environ. Pollut. 183 (2013) 213–223, https://doi.org/10.1016/j.envpol.2013.03.049. [34] A. Tiwari, P. Kumar, R. Baldauf, K.M. Zhang, F. Pilla, S. Di Sabatino, E. Brattich, B. Pulvirenti, Considerations for evaluating green infrastructure impacts in microscale and macroscale air pollution dispersion models, Sci. Total Environ. 672 (2019) 410–426, https://doi.org/10.1016/j.scitotenv.2019.03.350. [35] X. Song, H. Li, C. Li, J. Xu, D. Hu, Effects of VOCs from leaves of Acer truncatum Bunge and Cedrus deodara on human physiology and psychology, Urban For. Urban Green. 19 (2016) 29–34, https://doi.org/10.1016/j.ufug.2016.06.021. [36] M.E. Maffei, J. Gertsch, G. Appendino, Plant volatiles: production, function and pharmacology, Nat. Prod. Rep. 28 (2011) 1359–1380, https://doi.org/10.1039/ c1np00021g. [37] European Collaborative Action (ECA), Urban Air, Indoor Environment and Human Exposure, Report No.26, Impact of Ozone-Initiated Terpene Chemistry on Indoor Air Quality and Human Health, 2008. Luxembourg, https://op.europa.eu/en/publication-detail/-/publication/f3703ff2-812f-4107-9694- ec52747db2f7/language-en/format-PDF/source-199594952. (Accessed 11 April 2022). [38] European Collaborative Action (ECA), Urban Air, Indoor Environment and Human Exposure, Report No.29, Harmonisation Framework for Health Based Evaluation of Indoor Emissions from Construction Products in the European Union Using the EU-LCI Concept, Office for Official Publications of the European Communities, Luxembourg, 2013. https://publications.europa.eu/en/publication-detail/-/publication/d3d78842-bc95-4984-a2fe-2317731324bd. (Accessed 12 April 2022). [39] European Union, Agreed EU-LCI Values (December 2022), Luxemburg, 2022. [40] S. Li, P.C. Harley, Ü. Niinemets, Ozone-induced foliar damage and release of stress volatiles is highly dependent on stomatal openness and priming by low-level ozone exposure in Phaseolus vulgaris, Plant Cell Environ. 40 (2017) 1984–2003, https://doi.org/10.1111/pce.13003. [41] C. Zhou, Y. Zhan, S. Chen, M. Xia, C. Ronda, M. Sun, H. Chen, X. Shen, Combined effects of temperature and humidity on indoor VOCs pollution: intercity comparison, Build. Environ. 121 (2017) 26–34, https://doi.org/10.1016/j.buildenv.2017.04.013. [42] Ü. Niinemets, F. Loreto, M. Reichstein, Physiological and physicochemical controls on foliar volatile organic compound emissions, Trends Plant Sci. 9 (2004) 180–186, https://doi.org/10.1016/j.tplants.2004.02.006. [43] C. Calfapietra, E. Pallozzi, I. Lusini, V. Velikova, R.K. Monson, Ü. Niinemets, Biology, Controls and Models of Tree Volatile Organic Compound Emissions, 2013, https://doi.org/10.1007/978-94-007-6606-8. [44] J.-C. Kim, K.-J. Kim, D.-S. Kim, J.-S. Han, Seasonal variations of monoterpene emissions from coniferous trees of different ages in Korea, Chemosphere 59 (2005) 1685–1696, https://doi.org/10.1016/j.chemosphere.2004.10.048. [45] S. Juuti, J. Arey, R. Atkinson, Monoterpene emission rate measurements from a Monterey pine, J. Geophys. Res. 95 (1990) 7515, https://doi.org/10.1029/ jd095id06p07515. [46] F. Loreto, J.P. Schnitzler, Abiotic stresses and induced BVOCs, Trends Plant Sci. 15 (2010) 154–166, https://doi.org/10.1016/j.tplants.2009.12.006. [47] M.J. Wilkinson, R.K. Monson, N. Trahan, S. Lee, E. Brown, R.B. Jackson, H.W. Polley, P.A. Fay, R. Fall, Leaf isoprene emission rate as a function of atmospheric CO2 concentration, Global Change Biol. 15 (2009) 1189–1200, https://doi.org/10.1111/j.1365-2486.2008.01803.x. [48] C. Calfapietra, E. Pallozzi, I. Lusini, V. Velikova, Modification of BVOC Emissions by Changes in Atmospheric [CO2] and Air Pollution, 2013, pp. 253–284, https://doi.org/10.1007/978-94-007-6606-8_10. [49] C. Parmesan, M.D. Morecroft, R. Trisurat, Y. Adrian, G.Z. Anshari, A. Arneth, Q. Gao, P. Gonzalez, R. Harris, J. Price, N. Stevens, G.H. Talukdar, Terrestrial and freshwater ecosystems and their services, in: Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, UK and New York, NY, USA, 2022, https://doi.org/10.1017/ 9781009325844.004. [50] A. Jamloki, M. Bhattacharyya, M.C. Nautiyal, B. Patni, Elucidating the relevance of high temperature and elevated CO2 in plant secondary metabolites (PSMs) production, Heliyon 7 (2021), 1–13, https://doi.org/10.1016/j.heliyon.2021.e07709. [51] L. Copolovici, A.C. Popitanu, D.M. Copolovici, Volatile organic compound emission and residual substances from plants in light of the globally increasing CO2 level, Curr. Opin. Environ. Sci. Health 19 (2021), https://doi.org/10.1016/j.coesh.2020.10.004. [52] M. Kulmala, T. Nieminen, R. Chellapermal, R. Makkonen, J. Bäck, V.-M. Kerminen, Climate feedbacks linking the increasing atmospheric CO2 concentration, BVOC emissions, aerosols and clouds in forest ecosystems, in: Ü. Niinemets, R.K. Monson (Eds.), Biology, Controls and Models of Tree Volatile Organic Compound Emissions, Springer Netherlands, Dordrecht, 2013, pp. 489–508, https://doi.org/10.1007/978-94-007-6606-8_17. [53] M. Staudt, X. Morin, I. Chuine, Contrasting direct and indirect effects of warming and drought on isoprenoid emissions from Mediterranean oaks, Reg. Environ. Change 17 (2017) 2121–2133, https://doi.org/10.1007/s10113-016-1056-6. [54] P. Tiiva, J. Tang, A. Michelsen, R. Rinnan, Monoterpene emissions in response to long-term night-time warming, elevated CO2 and extended summer drought in a temperate heath ecosystem, Sci. Total Environ. 580 (2017) 1056–1067, https://doi.org/10.1016/j.scitotenv.2016.12.060. [55] V. Velikova, T. Tsonev, C. Barta, M. Centritto, D. Koleva, M. Stefanova, M. Busheva, F. Loreto, BVOC emissions, photosynthetic characteristics and changes in chloroplast ultrastructure of Platanus orientalis L. exposed to elevated CO 2 and high temperature, Environ. Pollut. 157 (2009) 2629–2637, https://doi.org/ 10.1016/j.envpol.2009.05.007. 11 G. Stringari et al. Heliyon 10 (2024) e23854 [56] T. Räisänen, A. Ryyppö, S. Kellomäki, Effects of elevated CO2 and temperature on monoterpene emission of Scots pine (Pinus sylvestris L.), Atmos. Environ. 42 (2008) 4160–4171, https://doi.org/10.1016/j.atmosenv.2008.01.023. [57] J. Daussy, M. Staudt, Do future climate conditions change volatile organic compound emissions from Artemisia annua? Elevated CO2 and temperature modulate actual VOC emission rate but not its emission capacity, Atmos. Environ. X 7 (2020), 100082, https://doi.org/10.1016/j.aeaoa.2020.100082. [58] S. Hantson, W. Knorr, G. Schurgers, T.A.M. Pugh, A. Arneth, Global isoprene and monoterpene emissions under changing climate, vegetation, CO2 and land use, Atmos. Environ. 155 (2017) 35–45, https://doi.org/10.1016/j.atmosenv.2017.02.010. [59] S. Cruz, E. van Santen, C. Gómez, Evaluation of compact tomato cultivars for container gardening indoors and under sunlight, Horticulturae 8 (2022), https:// doi.org/10.3390/horticulturae8040294. [60] M.L. Richardson, C.G. Arlotta, Producing cherry tomatoes in urban agriculture, Horticulturae 8 (2022), https://doi.org/10.3390/horticulturae8040274. [61] A. Peña, M.R. Rovira-Val, J.M.F. Mendoza, Life cycle cost analysis of tomato production in innovative urban agriculture systems, J. Clean. Prod. 367 (2022), https://doi.org/10.1016/j.jclepro.2022.133037. [62] L. Copolovici, A. Kännaste, L. Pazouki, Ü. Niinemets, Emissions of green leaf volatiles and terpenoids from Solanum lycopersicum are quantitatively related to the severity of cold and heat shock treatments, J. Plant Physiol. 169 (2012) 664–672, https://doi.org/10.1016/j.jplph.2011.12.019. [63] R.M.C. Jansen, J. Wildt, I.F. Kappers, H.J. Bouwmeester, J.W. Hofstee, E.J. van Henten, Detection of diseased plants by analysis of volatile organic compound emission, Annu. Rev. Phytopathol. 49 (2011) 157–174, https://doi.org/10.1146/annurev-phyto-072910-095227. [64] V. Falara, T.A. Akhtar, T.T.H. Nguyen, E.A. Spyropoulou, P.M. Bleeker, I. Schauvinhold, Y. Matsuba, M.E. Bonini, A.L. Schilmiller, R.L. Last, R.C. Schuurink, E. Pichersky, The tomato terpene synthase gene family, Plant Physiol. 157 (2011) 770–789, https://doi.org/10.1104/pp.111.179648. [65] A.L. Schilmiller, I. Schauvinhold, M. Larson, R. Xu, A.L. Charbonneau, A. Schmidt, C. Wilkerson, R.L. Last, E. Pichersky, Monoterpenes in the Glandular Trichomes of Tomato Are Synthesized from a Neryl Diphosphate Precursor rather than Geranyl Diphosphate, 2009. www.pnas.org/cgi/content/full/. [66] K. Takayama, R.M.C. Jansen, E.J. van Henten, F.W.A. Verstappen, H.J. Bouwmeester, H. Nishina, Emission index for evaluation of volatile organic compounds emitted from tomato plants in greenhouses, Biosyst. Eng. 113 (2012) 220–228, https://doi.org/10.1016/j.biosystemseng.2012.08.004. [67] L. Pazouki, A. Kanagendran, S. Li, A. Kännaste, H. Rajabi Memari, R. Bichele, Ü. Niinemets, Mono- and sesquiterpene release from tomato (Solanum lycopersicum) leaves upon mild and severe heat stress and through recovery: from gene expression to emission responses, Environ. Exp. Bot. 132 (2016) 1–15, https://doi.org/10.1016/j.envexpbot.2016.08.003. [68] J. Peñuelas, J. Llusiaá, B.S. Gimeno, Effects of Ozone Concentrations on Biogenic Volatile Organic Compounds Emission in the Mediterranean Region, 1999. [69] M.A. Farag, P.W. Paré, C6-green leaf volatiles trigger local and systemic VOC emissions in tomato, Phytochemistry 61 (2002) 545–554, https://doi.org/ 10.1016/S0031-9422(02)00240-6. [70] S. Catola, M. Centritto, P. Cascone, A. Ranieri, F. Loreto, L. Calamai, R. Balestrini, E. Guerrieri, Effects of single or combined water deficit and aphid attack on tomato volatile organic compound (VOC) emission and plant-plant communication, Environ. Exp. Bot. 153 (2018) 54–62, https://doi.org/10.1016/j. envexpbot.2018.05.001. [71] D.C. Degenhardt, S. Refi-Hind, J.W. Stratmann, D.E. Lincoln, Systemin and jasmonic acid regulate constitutive and herbivore-induced systemic volatile emissions in tomato, Solanum lycopersicum, Phytochemistry 71 (2010) 2024–2037, https://doi.org/10.1016/j.phytochem.2010.09.010. [72] European Collaborative Action (ECA), Urban Air, Indoor Environment and Human Exposure, Report No.30, Framework for Health-Based Ventilation Guidelines in Europe, 2020, https://doi.org/10.2788/17476. Luxembourg. [73] R.M.C. Jansen, J.W. Hofstee, J. Wildt, F.W.A. Verstappen, H.J. Bouwmeester, M.A. Posthumus, E.J. Van Henten, Health monitoring of plants by their emitted volatiles: trichome damage and cell membrane damage are detectable at greenhouse scale, Ann. Appl. Biol. 154 (2009) 441–452, https://doi.org/10.1111/ j.1744-7348.2008.00311.x. [74] R.M.C. Jansen, J.W. Hofstee, J. Wildt, B.H.E. Vanthoor, F.W.A. Verstappen, K. Takayama, H.J. Bouwmeester, E.J. van Henten, Health monitoring of plants by their emitted volatiles: a model to predict the effect of Botrytis cinerea on the concentration of volatiles in a large-scale greenhouse, Biosyst. Eng. 106 (2010) 37–47, https://doi.org/10.1016/j.biosystemseng.2010.01.009. [75] B. Fabbri, M. Valt, C. Parretta, S. Gherardi, A. Gaiardo, C. Malagù, F. Mantovani, V. Strati, V. Guidi, Correlation of gaseous emissions to water stress in tomato and maize crops: from field to laboratory and back, Sensor. Actuator. B Chem. 303 (2020), https://doi.org/10.1016/j.snb.2019.127227. [76] G. Stringari, J. Villanueva, A. Rosell-Melé, N. Moraleda-Cibrián, F. Orsini, G. Villalba, X. Gabarrell, Assessment of greenhouse emissions of the green bean through the static enclosure technique, Sci. Total Environ. 874 (2023), 162319, https://doi.org/10.1016/j.scitotenv.2023.162319. [77] F. Parada, X. Gabarrell, M. Rufí-Salís, V. Arcas-Pilz, P. Muñoz, G. Villalba, Optimizing irrigation in urban agriculture for tomato crops in rooftop greenhouses, Sci. Total Environ. 794 (2021), 148689, https://doi.org/10.1016/J.SCITOTENV.2021.148689. [78] C.E. Aguilar-Rodríguez, J. Flores -Velázquez, F. Rojano-Aguilar, W. Ojeda-Bustamante, M. Iñiguez-Covarrubias, Tomato (Solanum lycopersicum L.) crop cycle estimation in greenhouse, based on degree day heat (GDC) simulated in CFD, Tecnologia y Ciencias Del Agua 11 (2020) 27–57, https://doi.org/10.24850/j- tyca-2020-04-02. [79] J. Yin, X. Liu, Y. Miao, Y. Gao, R. Qiu, M. Zhang, H. Li, M. Li, Measurement and prediction of tomato canopy apparent photosynthetic rate, Int. J. Agric. Biol. Eng. 12 (2019) 156–161, https://doi.org/10.25165/j.ijabe.20191205.4982. [80] J.T. Ball, I.E. Woodrow, J.A. Berry, A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions, in: J. Biggins (Ed.), Progress in Photosynthesis Research: Volume 4 Proceedings of the VIIth International Congress on Photosynthesis Providence, Rhode Island, USA, August 10–15, 1986, Springer Netherlands, Dordrecht, 1987, pp. 221–224, https://doi.org/10.1007/978-94-017-0519-6_48. [81] G.L. Miner, W.L. Bauerle, D.D. Baldocchi, Estimating the sensitivity of stomatal conductance to photosynthesis: a review, Plant Cell Environ. 40 7 (2017) 1214–1238. https://api.semanticscholar.org/CorpusID:25675348. [82] NIOSH, Volatile Organic Compounds (Screening) METHOD 2549, fourth ed., 1995, pp. 1–8. Measurement. [83] T. Kasal-Slavik, J. Eschweiler, E. Kleist, R. Mumm, H.E. Goldbach, A. Schouten, J. Wildt, Early biotic stress detection in tomato (Solanum lycopersicum) by BVOC emissions, Phytochemistry 144 (2017) 180–188, https://doi.org/10.1016/j.phytochem.2017.09.006. [84] J.I. Montero, E. Baeza, E. Heuvelink, J. Rieradevall, P. Muñoz, M. Ercilla, C. Stanghellini, Productivity of a building-integrated roof top greenhouse in a Mediterranean climate, Agric. Syst. 158 (2017) 14–22, https://doi.org/10.1016/J.AGSY.2017.08.002. [85] Ü. Niinemets, M. Reichstein, Controls on the emission of plant volatiles through stomata: differential sensitivity of emission rates to stomatal closure explained, J. Geophys. Res. Atmos. 108 (2003) 1–17, https://doi.org/10.1029/2002jd002620. [86] K.G.S. Dani, I.M. Jamie, I.C. Prentice, B.J. Atwell, Evolution of isoprene emission capacity in plants, Trends Plant Sci. 19 (2014) 439–446, https://doi.org/ 10.1016/j.tplants.2014.01.009. [87] J. Peñuelas, J. Llusià, Effects of carbon dioxide, water supply, and seasonality on terpene content and emission by Rosmarinus officinalis, J. Chem. Ecol. 23 (1997) 979–993, https://doi.org/10.1023/B:JOEC.0000006383.29650.d7. [88] Ü. Niinemets, A. Kännaste, L. Copolovici, Quantitative patterns between plant volatile emissions induced by biotic stresses and the degree of damage, Front. Plant Sci. 4 (2013) 1–15, https://doi.org/10.3389/fpls.2013.00262. [89] O. Pérez-Priego, A. López-Ballesteros, E.P. Sánchez-Cañete, P. Serrano-Ortiz, L. Kutzbach, F. Domingo, W. Eugster, A.S. Kowalski, Analysing uncertainties in the calculation of fluxes using whole-plant chambers: random and systematic errors, Plant Soil 393 (2015) 229–244, https://doi.org/10.1007/s11104-015-2481-x. [90] L. Li, A.B. Guenther, S. Xie, D. Gu, R. Seco, S. Nagalingam, D. Yan, Evaluation of semi-static enclosure technique for rapid surveys of biogenic volatile organic compounds (BVOCs) emission measurements, Atmos. Environ. 212 (2019) 1–5, https://doi.org/10.1016/j.atmosenv.2019.05.029. [91] Ü. Niinemets, U. Kuhn, P.C. Harley, M. Staudt, A. Arneth, A. Cescatti, P. Ciccioli, L. Copolovici, C. Geron, A. Guenther, J. Kesselmeier, M.T. Lerdau, R.K. Monson, J. Peñuelas, Estimations of isoprenoid emission capacity from enclosure studies: measurements, data processing, quality and standardized measurement protocols, Biogeosciences 8 (2011) 2209–2246, https://doi.org/10.5194/bg-8-2209-2011. [92] M.E. Maffei, Sites of synthesis, biochemistry and functional role of plant volatiles, South Afr. J. Bot. 76 (2010) 612–631, https://doi.org/10.1016/j. sajb.2010.03.003. [93] A. Herrmann, The Chemistry and Biology of Volatiles, 2010, https://doi.org/10.1002/9780470669532. 12 G. Stringari et al. Heliyon 10 (2024) e23854 [94] L. Copolovici, A. Kännaste, T. Remmel, Ü. Niinemets, Volatile organic compound emissions from Alnus glutinosa under interacting drought and herbivory stresses, Environ. Exp. Bot. 100 (2014) 55–63, https://doi.org/10.1016/j.envexpbot.2013.12.011. [95] A. Mozaffar, N. Schoon, A. Bachy, A. Digrado, B. Heinesch, M. Aubinet, M.L. Fauconnier, P. Delaplace, P. du Jardin, C. Amelynck, Biogenic volatile organic compound emissions from senescent maize leaves and a comparison with other leaf developmental stages, Atmos. Environ. 176 (2018) 71–81, https://doi.org/ 10.1016/j.atmosenv.2017.12.020. [96] T. Vuorinen, G.V.P. Reddy, A.M. Nerg, J.K. Holopainen, Monoterpene and herbivore-induced emissions from cabbage plants grown at elevated atmospheric CO2 concentration, Atmos. Environ. 38 (2004) 675–682, https://doi.org/10.1016/j.atmosenv.2003.10.029. [97] A.C. Heiden, K. Kobel, C. Langebartels, G. Schuh-Thomas, J. Wildt, Emissions of Oxygenated Volatile Organic Compounds from Plants Part I: Emissions from Lipoxygenase Activity, 2003. 13