Ambient levels and temporal trends of VOCs, including carbonyl compounds, and ozone at Cabañeros National Park border, Spain

Atmospheric Environment 85 (2014) 256e265

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

Ambient levels and temporal trends of VOCs, including carbonyl compounds, and ozone at Cabañeros National Park border, Spain Florentina Villanueva a, b, *, Araceli Tapia a, Alberto Notario c, José Albaladejo c, Ernesto Martínez a a

Laboratorio de Contaminación Atmosférica, Instituto de Investigación en Combustión y Contaminación Atmosférica, Universidad de Castilla-La Mancha, Camino de Moledores s/n, 13071 Ciudad Real, Spain Parque Científico y Tecnológico de Albacete, Paseo de la Innovación 1, 02006 Albacete, Spain c Departamento de Química Física, Facultad de Ciencias y Tecnologías Químicas, Universidad de Castilla la Mancha, Avenida Camilo José Cela s/n, 13071 Ciudad Real, Spain b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Atmospheric carbonyls, VOCs and ozone were monitored in gaseous atmospheric samples.
Monthly variations of carbonyl compounds were apparent with maximum values observed in July and August.
The levels of VOCs were very low ranged from not detected or below detection limit up to <0.54 mg m3.
Ozone shows a clear seasonal variation with the maximum monthly value registered in March.
Air mass back trajectories have been calculated.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 April 2013 Received in revised form 11 October 2013 Accepted 10 December 2013

Concentration levels of 15 carbonyls,17 VOCs and ozone were studied at Cabañeros National Park border, Spain, in an area mainly constituted by holm oaks (Quercus ilex) and cork oaks (Quercus suber), along with scrubland formations such as rock-rose and heather. The compounds were collected by means of diffusive samplers from AugusteNovember 2010 and FebruaryeAugust 2011. Carbonyl compounds, VOCs and O3 were analysed by HPLC with diode array UVeVis detector, GCeFID and by UVevisible spectrophotometry, respectively. The most abundant carbonyls were hexanal, acetoneeacrolein, formaldehyde and acetaldehyde. Seasonal variation was apparent with maximum values observed in summer months. Total carbonyl concentrations ranged from 2.8 to 19.7 mg m3. Most VOCs studied (using chemically desorbable cartridges) were either not detected or were below their detection limits, however, a parallel sampling using thermally desorbable cartridges, from May 22 to June 19, revealed the presence of much more VOCs, identified using GCeMS. O3 concentration ranged from 27.2 to 90.5 mg m3, reaching the maximum monthly mean concentration in March (84.4 mg m3). The analysis of back trajectories indicates the transport of polluted air masses from remote areas, mainly from the Mediterranean basin that should contribute to the high levels of ozone observed. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Air quality Aldehydes VOCs Ozone Passive samplers Cabañeros National Park

* Corresponding author. Laboratorio de Contaminación Atmosférica, Instituto de Investigación en Combustión y Contaminación Atmosférica, Universidad de Castilla-La Mancha, Camino de Moledores s/n, 13071 Ciudad Real, Spain. Tel.: þ34 926295300. E-mail addresses: [email protected], fl[email protected] (F. Villanueva), [email protected] (A. Tapia), [email protected] (A. Notario), Jose. [email protected] (J. Albaladejo), [email protected] (E. Martínez). 1352-2310/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2013.12.015

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

1. Introduction The atmospheric reactivity of volatile organic compounds (VOCs) may influence the concentration of tropospheric photochemical ozone, both in pollution episodes and in the background troposphere (Seinfeld and Pandis, 1998; Finlayson-Pitts and Pitts, 2000). Recent years have seen a growing appreciation of the importance of naturally generated VOCs in the atmosphere. In some areas with warm climates such as the studied in this work, VOCs from vegetation can play an equal or greater role than anthropogenic sources in contributing to low-level ozone formation (Figueruelo and Dávila, 2004). This is particularly important in rural areas of great ecologic interest (e. g. Cabañeros National Park studied in this paper) because elevated concentrations of surface ozone have been shown to damage vegetation (Bergweiler et al., 2008; Fishman et al., 2010). In order to assess this ozone damage to vegetation, the European Directive established two limits using the AOT40 parameter. The critical daytime AOT40 value for trees over six months is 20 000 mg m3 h (10 000 ppb h), calculated from April to September, whereas the AOT40 to protect vegetation is 6000 mg m3 h, calculated from May to July. Thus, emissions of VOCs in urban or rural areas and their impact on the environment are necessary to evaluate the exposure of the population and vegetation to gaseous pollutants better. Emission estimates of biogenic VOCs for regions in Europe are relatively uncertain and inventories are poor. Likewise, several measurements of VOCs in European rural and urban areas have been reported in the literature in recent years (Pilidis et al., 2005; Possanzini et al., 2007; Coll et al., 2010; Morknoy et al., 2011). However, only a few studies have been conducted in Spain (Parra et al., 2006; Gallego et al., 2008; Notario et al., 2013). In this sense, our group reported recently the first observations of VOCs concentrations, in the coastal, industrial area of Huelva near the Doñana National Park, South-west of the Iberian Peninsula (Villanueva et al., 2013). Acetone and formaldehyde were the most abundant carbonyls, followed by acetaldehyde and propanal. Maximum and minimum values for all these compounds in the period of measurement, and their relationship with meteorological parameters with influence of anthropogenic or/and biogenic emissions were also analysed. This work presents the first measurements and analysis related to VOCs, including aldehydes and aromatic hydrocarbons for a whole year, in the important ecologic area of Cabañeros National Park in central-southern Iberian Peninsula, in order to cover the existing lack of VOCs emissions in the studied zone and their relation to ozone formation. With respect to that, we recently analysed the levels and surface ozone temporal variations in the study carried out by our group in this rural area (Notario et al., 2012, 2013). The obtained results showed a great background average ozone concentration. We also observed that the European Directive limits to protect the human health and vegetation were widely exceeded; these facts pointed out an ozone problem in this rural region, where VOCs emissions probably play an important role. Our study must contribute to the understanding of the photochemical air pollution in the Western Mediterranean Basin and in the central-southern of the Iberian Peninsula that could be affected by high concentrations of photochemical pollutants. Finally, we intended to extend the database and the inventories of VOCs emissions in these regions of Europe. 2. Experimental section 2.1. Site description Field measurements were conducted on the southwestern border of the park at about 6 km east of the village of Horcajo de los

257

Montes (39.2 N, 04.4 W, 617 m above sea level) in the region of Castilla La Mancha, in central-southern Spain (Fig. 1). The sampling point was located about 90 Km northwest of Ciudad Real, an urban area of around 72 000 inhabitants and about 116 km southwest of Toledo, an urban area of around 82 000 inhabitants. The climate of this region of Toledo Mountains has been classified as temperate Mediterranean climate with oceanic tendency in its western side. While the entire park is included in the mesoMediterranean floor. The most representative plant community, to the foothills of these mountains, is constituted by holm oak (Quercus ilex) and cork oak (Quercus suber), as the main species, along with other transition formations or scrubland formations such as rock-rose and heather that are accompanied by other aromatic plants such as Rosemary (Rosmarinus officinalis), the lavender (Lavandula stoechas) or Gorse (Genista hirsuta). The oaks, formation where dominates the holm oak, have a wide variety of associated plants and shrubs such as strawberry tree (Arbutus unedo), labiérnago (Phillyrea angustifolia), green olives (Pistacea terebinthus), honeysuckle (Lonicera implexa), common plants like peony (Paeonia broteroi) and thermophilic species as wild olive (Olea europaea) and mastic (Pistacia lentiscus). In order to determine the origin and pathway of the air masses affecting the results involved in this study, back trajectories were computed using the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model Version 4 developed by the NOAA’s Air Resources Laboratory (ARL) (Draxler et al., 2009). The GDAS input meteorological files have a spatial resolution of 1  1 and 24 levels of vertical resolution. The three-dimensional kinematic back trajectories were calculated using the vertical wind component provided by the meteorological model. Back trajectory duration of 48 h was considered enough to represent the synoptic air flows in this area. Two back trajectories were calculated every day (00:00 and 12:00 UTC). In order to understand the behaviour of the air masses circulating in the planetary boundary layer (PBL), these trajectories were calculated at a 100-m height. A cluster methodology was applied to form air mass groups using the tool incorporated in the HYSPLIT model. Air mass clusters for two periods 12e20 November 2010 and 20e27 March 2011 are shown in Fig. 2a and b, respectively. In November, air masses came clearly from the WeNW sector. Three have their origin in the Atlantic Ocean and the other two with higher occurrence frequency (30 and 32%) have their origin in the Iberian Peninsula. However, in March air masses came mainly from the Mediterranean basin. Another origin is found in air masses coming from the northern Iberian Peninsula. Therefore, the studied area may be affected by air masses loaded with air pollutants from Valencia metropolitan area (an eastern urban zone with a population greater than one million), Alicante and Cartagena (an industrial area) which may be injected into the Iberian Peninsula plateau thanks to the mesoscale processes developed in the coastal area (Millán et al., 2002). Although Madrid, in the North, is the largest urban area close to the measurement site, the air pollution generated in this area would reach the Cabañeros National Park with a lower frequency since the Toledo Mountains present an important orographic barrier. 2.2. Experimental design RadielloÒ passive samplers (Fondazione Salvatore Maugeri, Padova, Italy) were used for monitoring VOCs, carbonyl compounds and ozone and were placed at about 1.8 m above the ground level. VOCs were sampled by two configurations of the RadielloÒ diffusive sampler: the chemically desorbable sampler and thermally desorbable one. Each cartridge was exposed for 1 week for

258

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

Fig. 1. Location of Cabañeros National Park and the sampling point in the region of Castilla La Macha in central-southern Spain.

aldehydes, VOCs and ozone during the first and second campaign (AugusteNovember 2010 and FebruaryeAugust 2011) except for VOCs that was 2 weeks during the second campaign. The total number of samples was 46 for carbonyl compounds, 33 for VOCs and 39 for O3. In the case of VOCs sampled with thermally desorbable cartridges, each cartridge was exposed for 1 week from May 22 to June 19, 2011. Tubes were protected from bad weather conditions by a mountable polypropylene shelter. After exposure the cartridges were introduced in their sealed glass tubes and stored in the dark and refrigerated until the analysis. Field blanks were transported together with samplers to the sampling point.

2.3. Sampling and analytical methods The analysis and the analytical system for carbonyl compounds and VOCs were described in detail elsewhere (Villanueva et al., 2013). RadielloÒ passive samplers for carbonyls consist of a stainless steel cartridge filled with 2,4-dinitrophenylhydrazine coated FlorisilÒ. In this study, the LC column was SupelcosilÔ LC-18 250  4.6 mm  5 mm and samples of 20 ml of solution were injected and eluted as follows: 0e7 min, 60% acetonitrile (HPLC grade) and 40% water (from a Milli-Q system); 7e20 min, a gradient up to 100% acetonitrile. The flow was 1 ml min1. Identification and quantification of carbonyls were based on their

Fig. 2. Cluster mean of hourly back trajectories for 12e20 November 2010 (a) and 20e27 March 2011 (b).

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

retention times and ultraviolet (UV) spectra and peak areas, respectively. A series of standards (TO11/IP-6A Aldehyde/Ketone-DNPH Mix, Supelco, Bellefonte, USA) containing formaldehyde, acetaldehyde, acrolein, acetone, propanal, crotonaldehyde, butanal, benzaldehyde, isopentanal, pentanal, o-tolualdehyde, m-tolualdehyde, ptolualdehyde, hexanal and 2,5-dimethylbenzaldehyde in acetonitrile, were used to obtain a five-point calibration curve for each compound in concentration ranges similar to the tested samples (0.2e4 mg ml1). There were very good linear relationships between concentration and instrumental response for all carbonyls measured (R2 > 0.99). A control point with a calibration standard was run daily before or after the samples analysis. In the case of VOCs, the chemically desorbable sampler contains a stainless steel net cylinder packed with activated charcoal. Pentane, hexane, heptane, octane, nonane, decane, undecane, methylcyclohexane, benzene, toluene, o-xylene, m-xylene, pxylene, ethylbenzene, 1,2,4-trimethylbenzene, styrene, a-pinene, limonene, 1-butanol (analytical standards) and the solvent carbon disulphide (Purity 99.9%), were obtained from Aldrich (Steimheim, Germany). The calibration was performed by the phase equilibrium technique, adding to new, unexposed cartridges accurately measured 2 ml aliquots of a series of calibration solutions, ranging from 0.17 to 8.2 mg ml1 of each compound prepared by serial dilution. 100 ml of internal standard solution was added to each cartridge containing the standard solution. Five replicates of every calibration point were carried out. Good linear relationships were obtained for each hydrocarbon (R2 > 0.99). The thermally desorbable sampler is composed by a cylindrical stainless steel net containing Carbograph 4. Before sampling, the cartridges were conditioned during 8 h in the thermal desorber (Turbomatrix 100, Perkin Elmer, Norwalk, CT, USA) with helium at 300  C and analysed to verify the blank levels. For the analysis, the absorbing cartridges were fitted into empty stainless steel standard thermal desorption tubes and introduced in the automated thermal desorber. The VOCs were thermally desorbed for 15 min at 300  C with a flow of 35.6 ml min1 of ultrapure helium passing through and carrying the desorbed VOCs to a pre-concentration trap at 20  C (cold trap was filled with Tenax TA). Then, the analytes were desorbed from the trap by rapid heating to 290  C into the injector of the gas chromatograph coupled to the mass spectrometer (Shimadzu 17A QP5050). Analytes were separated onto the same capillary column and the same

259

analytical conditions as those used in the GCeFID. Identification of VOCs was performed in scan mode using electron impact ionization (70 eV). This sampling was used only with qualitative purposes as it will be explained below. The identification of the peaks obtained in the chromatogram was carried out taking into account the retention times and by comparing the mass spectrum measured to reference spectra registered in mass spectral libraries (NIST, Wiley and an own Laboratory Library). For ozone, the radial diffusive cartridge is filled with silica gel coated with 1,2-di(4-pyridyl)ethylene. All details of analytical procedure are described in the previous work (Martin et al., 2010). 2.4. Quality assurance Blanks samples, method detection limits and reproducibility were measured for quality assurance. Two or three field blanks were transported along with each sampler lot to the sampling site, stored in the laboratory during the exposure period and later used to check the possible contamination during the transportation and storage of samples. For each campaign a set of 3e5 unexposed samplers (laboratory and field blanks) were analysed to determine the blank value. Method detection limits were defined as three times the standard deviation of the blanks or, for compounds with zero blank, three times the standard deviation of low concentration analytical standards. The coefficients of variation in the reproducibility test ranged from 0.5% (acetaldehyde) to 4.5% (m/p-tolualdehyde) for carbonyl compounds except for acetone that was 10%, from 0.005% (benzene) to 0.11% (heptane) for VOCs and 0.001% for ozone. Method detection limits calculated for a sampling period of 7 or 14 days depending on the compound are presented in Table 1. 3. Results and discussion 3.1. Carbonyls compounds Twelve compounds were identified and quantified in this study: formaldehyde, acetaldehyde, acetoneeacrolein, propanal, crotonaldehyde, butanal, benzaldehyde, isopentanal, pentanal, o-tolualdehyde, m/p-tolualdehyde and hexanal (the sum of m/ptolualdehyde and acetoneeacrolein was reported because they could not be well separated by the analytical method). Minimum, maximum and average concentrations of carbonyl compounds, ozone and temperature during the two campaigns are listed in

Table 1 MDLs found in the study for each compound. MDLs (mgm3) e 7days Carbonyls Formaldehyde Acetaldehyde Acetone þ acrolein Propanal Crotonaldehyde Butanal Benzaldehyde Isopentanal Pentanal o-tolualdehyde m/p-tolualdehyde Hexanal 2,5-dimethylbenzaldehyde

0.09 0.26 0.06 0.12 0.05 0.05 0.20 0.01 0.10 0.02 0.01 0.12 0.01

Ozone

0.39

MDLs (mgm3) e 7days VOCs Pentane Hexane Heptane Benzene Methylciclohexane 1-butanol Octane Toluene Nonane Ethylbenzene m/p-xylene a-pinene o-xylene Styrene Decane 1,2,4-TMB Limonene Undecane

0.23

0.17

0.07 0.28

MDLs (mgm3) e 14days 0.02 0.13 0.11 0.05 0.06 0.18 0.06 0.28 0.17 0.14 0.23 0.57 0.28 0.59 0.23 0.19 0.31

260

Table 2 Minimum, maximum and average (in parenthesis) carbonyls and ozone concentrations (mg m3), and temperature (K) at Cabañeros National Park border.

(0.06) (0.17) (0.05) (0.04) (0.79) (7.57) (300)

(2.57) (4.69) (0.23) (0.08) (0.05) (0.13) (0.13) (0.88) (8.56) (69.6) (299) (2.12) (0.31) (1.71) (0.20) (0.20) (0.15)

(0.16) (0.09) (0.19) (0.57) (0.52) (0.13) (0.34)

(0.96) (7.30) (62.8) (300) (1.05) (5.22) (75.0) (292) (1.26) (6.22) (68.3) (289) (1.16) (4.53) (84.4) (284) (1.08) (4.07) (56.1) (282) (0.88) (3.62) (33.8) (280) (0.82) (3.73) (44.6) (288)

(0.27) (0.17) (0.55) (0.44) (0.29) (1.28) (8.19) (59.8) (295) (0.30)

(0.69) (0.42) (0.40) (1.33) (10.6) (72.0) (301)

0.90e1.64 0.99e1.87 2.20e4.34 0.48e0.71 0.49e0.63 e 0.27e0.33 N.de0.24 0.64e0.80 0.39e0.43 0.35e0.46 1.21e1.46 8.13e12.8 62.8e79.5 301e301

Fig. 3. Average concentrations of carbonyls compounds during the two campaigns together with the values of temperature.

Table 2. 2,5-dimethylbenzaldehyde was not detected in any sample. Crotonaldehyde, benzaldehyde, isopentanal and tolualdehydes were either not detected or below their detection limits in many samples. Total carbonyl concentrations ranged from 2.78 to 19.7 mg m3. Butanal appeared overlapped with an interference (no carbonyl compound) in the samples analysed from August 2010 to February 2011 and it was not quantified. Its high levels obtained with respect to the rest of carbonyls from March to August 2011 together with the UV spectrum indicate that butanal appears overlapped with other carbonyl compound may be 2-butanone. Both compounds appear together in the chromatogram when a standard is introduced under the analytical conditions. Therefore, the concentration of butanal in Table 2 is the sum of butanal and 2butanone, however, for the discussion butanal has not been considered. The most abundant carbonyls during the first period of measurement (AugusteNovember 2010) were generally either acetoneeacrolein or hexanal, the concentrations were in the range 0.43e4.34 mg m3 and 0.67e1.46 mg m3, respectively. Fig. 3 shows the plot of the monthly quantities measured for carbonyl compounds during the two sampling periods together with the averaged values of temperature. As Fig. 3 shows, August 2010 was the most carbonyl-polluted month followed by July 2011 and September 2010. October and November 2010 were the months with lower concentration of carbonyl compounds. Acetoneeacrolein mixing ratios ranged from 0.35 in February to 4.52 mg m3 in June 2011, with an average of 1.78 mg m3 during the two sampling periods, while ambient levels of hexanal varied from 0.67 mg m3 in October to 1.72 mg m3 in April; the average concentration during the period was 1.06 mg m3. Formaldehyde varied between values below detection limit and 2.56 mg m3 in October and June, respectively with an average concentration of 0.96 mg m3 and the range of acetaldehyde during the whole period was from below detection limits in February, March and April to 1.89 mg m3 in June. The concentrations of other carbonyls ranged from non-detected to 1.18 mg m3. This pattern was consistent with the monthly variations of atmospheric oxidants and ambient temperature, except for May and August 2011 where the behaviour is not the expected. The seasonal variation of the carbonyls is reported in Fig. 4. The concentrations of the main carbonyls analysed in this study (acetoneeacrolein, formaldehyde, hexanal and also acetaldehyde) reach the maximum levels in summer except hexanal whose maximum is observed in winter. It is important to note that the values of winter are the averaged values found only in February.

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

Sum of butanal and 2-butanone; b.d.l: below detection limit; N.d: not detected. a

(0.97) (0.91) (1.53) (0.38) 0.75e1.19 0.48e1.33 0.60e2.45 0.31e0.45 N.d 2.04e3.11 b.d.leb.d.l 0.04e0.08 0.05e0.28 b.d.le0.10 b.d.le0.08 0.69e0.89 3.52e6.47 e 300e300 (1.84) (1.31) (3.32) (0.60) 1.45e2.19 1.09e1.57 3.02e4.13 0.46e0.81 N.d 1.66e9.40 b.d.le0.35 0.06e0.10 b.d.le0.31 0.09e0.20 b.d.le0.27 0.69e1.08 8.04e10.31 67.1e72.0 298e301 (1.37) (1.09) (2.63) (0.63) 0.69e2.56 0.53e1.89 1.29e4.52 0.46e1.04 N.d 1.10e3.43 b.d.le0.44 N.de0.14 b.d.le0.34 b.d.le0.38 N.d 0.69e1.37 3.93e12.77 53.8e70.3 298e302 (1.15) (0.70) (1.36) (0.56) 0.87e1.27 0.49e0.87 1.02e1.66 0.48e0.64 N.d 1.13e2.02 b.d.le0.29 N.de0.06 b.d.le0.31 N.d N.d 0.84e1.40 4.30e6.13 69.6e80.9 290e295 (1.19) (0.49) (1.38) (0.81) (0.29) (2.28) (0.22) (0.06) (0.45) (0.06) 0.83e1.54 b.d.le0.91 0.85e1.93 0.63e0.95 b.d.le0.36 1.71e2.80 b.d.le0.29 b.d.le0.08 0.29e0.81 b.d.le0.13 N.d 0.89e1.72 5.53e6.83 45.1e84.5 287e291 (0.59) (0.29) (1.02) (0.48) (0.21) (2.36) (0.21) 0.51e0.74 b.d.le0.38 0.76e1.32 0.28e0.82 0.16e0.27 1.81e2.64 b.d.le0.25 N.d 0.46e0.73 N.d N.d 0.96e1.49 3.62e5.49 76.3e90.5 279e289 (0.51) (0.33) (0.69) (0.44) (0.35) 0.35e0.60 b.d.le0.46 0.35e1.25 0.31e0.60 0.29e0.39 e b.d.leb.d.l N.d 0.40e0.68 N.d N.d 0.82e1.21 2.78e5.32 42.7e76.3 280e285 (0.52) (0.43) (0.49) (0.40) (0.38) 0.33e0.78 0.26e0.65 0.43e0.58 0.34e0.48 0.30e0.52 e b.d.le0.27 N.d N.de0.64 N.de0.33 N.d 0.78e1.02 3.20e4.73 27.2e38.2 283e287 (0.32) (0.40) (0.80) (0.30) (0.48) b.d.le0.74 0.33e0.50 0.45e1.13 b.d.le0.47 0.30e0.76 e b.d.le0.23 N.d b.d.le0.54 0.32e0.36 N.d 0.67e0.98 2.98e4.77 40.7e46.8 285e290 (0.64) (1.01) (2.11) (0.49) (0.93)

0.27e1.15 0.79e1.64 1.09e2.98 0.4e0.58 0.52e1.18 e 0.24e0.30 b.d.le0.32 b.d.le0.80 0.38e0.58 b.d.le0.55 1.09e1.46 5.66e10.9 56.5e63.7 292e297 Formaldehyde Acetaldehyde Acetone þ acrolein Propanal Crotonaldehyde Butanala Benzaldehyde Isopentanal Pentanal o-tolualdehyde m/p-tolualdehyde Hexanal Total carbonyls Ozone Temperature

(1.25) (1.37) (3.38) (0.59) (0.55)

Aug-11 Jul-11 Jun-11 May-11 Apr-11 Mar-11 Feb-11 Nov-10 Oct-10 Aug-10

Sep-10 Period

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

Fig. 4. Seasonal variation of carbonyl compounds.

Ambient levels of acetoneeacrolein account for 34% of the total carbonyls concentration in summer with an average value of 2.74 mg m3, follow by formaldehyde (17%; 1.35 mg m3), acetaldehyde (15%; 1.19 mg m3), hexanal (13%; 1.00 mg m3) and propanal (7%; 0.52 mg m3). Acetone is often the most abundant carbonyl in semirural or rural environments due to being largely from biogenic origin (Goldstein and Schade, 2000). For the rest of compounds the concentrations are lower than 0.30 mg m3. Benzaldehyde, isopentanal and hexanal do not experience large variations between seasons what suggests the emission source to be constant during all the year. Crotonaldehyde and o-tolualdehyde reach the maximum levels in autumn, 0.60 and 0.32 mg m3, respectively. The different seasonal cycle of formaldehyde, acetaldehyde and acetoneeacrolein with respect to the other carbonyls suggests different formation mechanisms and sinks compared to the others. These seasonal cycles for the main carbonyls could be caused by photochemical processes (oxidation of biogenic and even anthropogenic hydrocarbons under determined meteorological conditions as demonstrated by the air mass trajectories analysis that will be explained in Section 3.2) and also the direct emission from vegetation as it is discussed below. The results of formaldehyde and acetaldehyde for other forest and rural areas are shown in Table 3 for comparison. The carbonyl concentrations measured at Cabañeros National Park border are in the same range or lower than the levels determined in other forest and rural areas. Our data are comparable to those found in the small village of Covelo considered a rural/forest site from Portugal (Evtyugina et al., 2006).

261

The formaldehyde/acetaldehyde (C1/C2) and acetaldehyde/ propanal (C2/C3) concentration ratios were also measured. C1/C2 concentration ratio has been proposed as an indicator of the biogenic source of formaldehyde (Shepson et al., 1991). C1/C2 ratio can vary between w1 (urban) and 10 (deciduous forested, in which isoprene chemistry often plays a dominant role). Likewise, C2/C3 concentration ratio may serve as an effective indicator of the presence of anthropogenic pollution because propanal is associated only with anthropogenic hydrocarbon precursors, whereas acetaldehyde has both anthropogenic and natural hydrocarbon precursors. C2/C3 would be high in forest and rural atmospheres and low in polluted urban air. The ratios for this and other studies are shown in Table 3. The C1/C2 ratios in this study vary from 0.86 in autumn to 2.65 in spring while the C2/C3 ratios ranged from 0.77 in winter to 2.26 in summer. This variation agrees with what one might expect, higher ratios in summer, due to oxidation of hydrocarbons. However, these ratios would reflect typical values for urban air, so care must be taken if using such ratios to identify the source of carbonyl compounds because our results are similar to those obtained for other forest or rural areas as demonstrated in Table 3 (see the studies referenced in this table). It must be taken into account that some plants could emit many kinds of carbonyls including formaldehyde, acetaldehyde and propanal (VillanuevaFierro et al., 2004; Martin et al., 1999), in addition, the main oaks present in the park are not a source of isoprene, thus for these reasons, these ratios might not be what one expects to find. Also, it is important to note that our data are the average concentration of one week. Finally, the correlation analysis between carbonyl compounds, ozone and temperature during the two campaigns (the total number of matching pairs is 42), is presented in Table 4. Carbonyls are moderately correlated between themselves except the correlations between formaldehyde and acetaldehyde, acetoneeacrolein and formaldehyde, and acetaldehyde and acetoneeacrolein that were relatively strong, R ¼ 0.73, R ¼ 0.75 and R ¼ 0.92 what suggests that these aldehydes have common sources and sinks. Most carbonyls correlated weakly with temperature, except formaldehyde, acetaldehyde and acetoneeacrolein what seems to indicate the existence of ambient conditions favourable to oxidation of hydrocarbons or direct emission from vegetation. Ozone did not show significant correlations with carbonyls suggesting that the ozone could come from remote areas as explained below. Therefore, biogenic contributions to the atmospheric concentrations of aldehydes can be derived from the knowledge of direct or indirect release from vegetation. The main contribution to the atmospheric budget is seen in the secondary reaction of biogenic (or anthropogenic) hydrocarbons with OH and NO3 radicals and ozone as well as in photolysis (Kesselmeier and Staudt, 1999).

Table 3 Range and average (in parenthesis) formaldehyde and acetaldehyde mixing ratios and C1/C2 and C2/C3 concentration ratios in the present study and other rural and forest sites. Data are given in mg m3. Location Forest Cabañeros National Park, Spain Forest Park, South of China Bavaria, coniferous forest, Germany Tijuca Forest, Brazil Langmuir, New Mexico Rural Sagalhos, Portugal Lota, Portugal Covelo, Portugal Socorro NM, New Mexico b.d.l: below detection limit.

Date

Formaldehyde

Acetaldehyde

C1/C2

C2/C3

Reference

AugusteNovember 2010 FebruaryeAugust 2011 Summer 2004 June, 2001,2002 JanuaryeAugust 2008 JuneeAugust, 1997

b.d.le2.56 (0.96)

0.13e1.89 (0.79)

1.51

1.65

This work

0.24e17.93 (3.70) 4.89 0.046e6.24 (2.86) 2.82  2.08

0.02e12.91 (3.33) 2.81 b.d.le7.34 1.80  1.26

1.11 1.74 1.57

3.70 e 1.88 0.63

Yu et al., 2008 Müller et al., 2006 Custódio et al., 2010 Villanueva-Fierro et al., 2004

June, June, June, June,

0.13e1.54 (0.58) 0.04e1.16 (0.43) 0.22e2.51 (1.13) 4.17  1.72

0.08e1.01 (0.44) 0.06e3.05 (0.52) 0.35e1.31 (0.70) 2.52  1.62

1.32 0.82 1.61 1.65

2.93 5.77 4.12 0.38

Evtyugina et al., 2006 Evtyugina et al., 2006 Evtyugina et al., 2006 Villanueva-Fierro et al., 2004

July 2001,2002 July 2001,2002 July 2001,2002 August, 1997

262

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

Table 4 Correlation coefficients for carbonyl compounds, ozone and temperature.

1. Temperature 2. Ozone 3. Formaldehyde 4. Acetaldehyde 5. Acetone þ acrolein 6. Propanal 7. Crotonaldehyde 8. Butanalc 9. Benzaldehyde 10. Isopentanal 11. Pentanal 12. Tolualdehydes 13. Hexanal

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

1 0.44b 0.61b 0.76b 0.80b 0.31a 0.50 0.36 0.41b 0.30 0.05 0.48 0.10

1 0.45b 0.24 0.39a 0.42b 0.07 0.05 0.25 0.21 0.40a 0.02 0.47b

1 0.73b 0.75b 0.68b 0.04 0.40 0.49b 0.23 0.03 0.10 0.24

1 0.92b 0.37a 0.57 0.50 0.49b 0.48 0.05 0.55 0.12

1 0.44b 0.50 0.60 0.60b 0.43 0.20 0.56 0.27

1 0.05 0.16 0.41b 0.28 0.27 0.04 0.55b

1 0.11 0.43b 0.67 0.23 0.72 0.17

1 0.10 0.22 0.24 0.56 0.13

1 0.27 0.41b 0.45 0.52b

1 0.44b 0.54 0.30a

1 0.66 0.66b

1 0.34a

1

Note: aSignificant with a probability over 95%; bSignificant with a probability over 99%; csum of butanal and 2-butanone.

Ozonolysis occurring at vegetation surfaces, for instance, was recognized as a source of acetone between other carbonyls (Fruekilde et al., 1998). In addition, several studies have demonstrated that plants directly emit a large spectrum of aldehydes, for example formaldehyde and acetaldehyde can be emitted from two Mediterranean tree species, Q. ilex (holm oak), one of the most abundant oak in the study area, and Pinus pinea (Italian stone pine), (Schäfer et al., 1995; Kesselmeier et al., 1997). Kreuzwieser et al. (2002) report that the main carbonyl species emitted in considerable amounts by the leaves of Q. ilex were acetaldehyde, formaldehyde and acetone with emission rates ranging between c. 1 nmol m2 min1 (acetone) and 45 nmol m2 min1 (acetaldehyde), then it is reasonable to think that the good correlation between these species might be mainly due to direct emission from vegetation. On the other hand, according to Kesselmeier and Staudt (1999) not only exists biogenic sources but also biogenic sinks. Several experiments have demonstrated that vegetation can be both a source of and a sink for short-chain aldehydes, depending on several factors, such as the environmental conditions, metabolic activities rates, and leaf surface and age. Jork (1996) demonstrated a bidirectional exchange of formaldehyde and acetaldehyde for several crops species. In addition, butanal and hexanal can be emitted directly from agricultural and natural plant species. Butanal is emitted from grass land and hexanal was detected in many plants such as grape, blossoming rye, rape, beech, hornbeam, birch, oak and grass land (König et al., 1995). Ciccioli et al. (1993) speculated that the ubiquitous occurrence of C4eC10 alkanals is due to vegetative emissions and showed that alkanals are found in

essential oils from various plants, fruits and blossoms. In our study hexanal correlates well with pentanal (R ¼ 0.66) which might indicate a common source from biogenic emissions. Also benzaldehyde can be released from biological sources such as grasses (Kirstine et al., 1998). Finally, based on the results it can be concluded that some carbonyls such as formaldehyde, acetaldehyde, acetoneeacrolein and propanal seem to have their origin in the photooxidation of hydrocarbons but also the direct emission from vegetation could be important. The rest of carbonyls could be formed from the emission from different plants. The data are certainly insufficient to assure the main source of the most abundant carbonyl compounds found in this study and further experimental studies are necessary in order to determine the role of vegetation in the formation of these species. 3.2. VOCs Table 5 summarizes the results obtained in the two sampling campaigns. During the first campaign only benzene, toluene and xylenes were monitored. The mentioned hydrocarbons were not detected or the values of concentration were below detection limit. Due to these results, it was decided to extend the exposition period of the cartridges to two weeks instead of one in the second campaign, monitoring more compounds including aliphatic (pentane, hexane, octane, nonane, decane, undecane and ciclohexane), other aromatic hydrocarbons (ethylbenzene, estyrene, 1,2,4-trimethylbenzene), terpenes (a-pinene and limonene) and

Table 5 Ambient air mean levels of VOCs (mg m3) at Cabañeros National Park border. Period

Aug-10

Sep-10

Oct-10

Nov-10

Feb-11

Mar-11

Apr-11

May-11

Jun-11

Pentane Hexane Heptane Benzene Methylciclohexane 1-butanol Octane Toluene Nonane Ethylbenzene m/p-xylene a-pinene Styrene Decane 1,2,4-TMB Limonene Undecane

N.m N.m N.m b.d.l N.m N.m N.m b.d.l N.m N.m N.d N.m N.m N.m N.m N.m N.m

N.m N.m N.m b.d.l N.m N.m N.m b.d.l N.m N.m N.d N.m N.m N.m N.m N.m N.m

N.m N.m N.m b.d.l N.m N.m N.m b.d.l N.m N.m N.d N.m N.m N.m N.m N.m N.m

N.m N.m N.m 0.24  0.08 N.m N.m N.m b.d.l N.m N.m N.d N.m N.m N.m N.m N.m N.m

0.04  0.05 b.d.l N.d 0.32  0.45 N.d N.d N.d 0.09  0.05 N.d N.d N.d N.d N.d b.d.l N.d b.d.l b.d.l

b.d.l b.d.l b.d.l 0.47  0.01 N.d N.d b.d.l b.d.l N.d N.d N.d b.d.l N.d b.d.l N.d N.d N.d

0.13  0.06 b.d.l N.d 0.21  0.09 N.d N.d N.d 0.09  0.04 N.d N.d N.d b.d.l N.d b.d.l N.d b.d.l b.d.l

N.d b.d.l N.d 0.11  0.02 N.d N.d b.d.l b.d.l N.d N.d N.d 0.35  0.13 N.d b.d.l N.d b.d.l b.d.l

0.54 0.34 N.d 0.07 N.d N.d N.d 0.51 N.d N.d N.d 0.61 N.d N.d N.d b.d.l b.d.l

N.m: not measured; b.d.l. below detection limit; N.d: not detected.

 0.76  0.49  0.01

 0.072

 0.37

Jul-11

Aug-11

0.04  0.06 b.d.l N.d 0.07  0.04 N.d N.d N.d 0.07  0.03 N.d N.d N.d b.d.l N.d N.d N.d N.d N.d

b.d.l b.d.l N.d 0.07  0.03 N.d N.d N.d 0.09  0.01 N.d N.d N.d b.d.l N.d N.d N.d b.d.l N.d

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

alcohols such as 1-butanol. Between all the monitored compounds, only some of them were quantified above detection limit occasionally such as toluene, pentane, hexane and a-pinene. Only benzene was detected and quantified during the whole period from February to August although from May to August the levels were near the detection limits. Heptane, 1-butanol, methylcyclohexane, octane, nonane, ethylbenzene, xylenes, styrene, decane and limonene were either not detected or below detection limits. The concentration range of toluene during the second campaign was from below detection limit to 0.51 mg m3. Benzene was only detected in November 2010 (0.24 mg m3) and the range in the second campaign was from 0.07 mg m3 (June, July and August) to 0.47 mg m3 (March). a-pinene was detected only in spring with values of 0.35 and 0.61 mg m3 in May and June, respectively. This compound is one of the main species emitted by holm oaks (Q. ilex) (Kesselmeier et al., 1997) and cork oaks (Q. suber) (Pio et al., 2012), however, its low concentration can be interpreted as a result of low emission rates and/or its fast decomposition after emission (Stewart et al., 2013). The same conclusion can be made for limonene also one of the main compounds emitted by cork oaks (troposphere lifetime is very short). Toluene might be of biogenic origin, Heiden et al. (1999) found that toluene is synthesized by the plants, but the presence of benzene cannot be explained in terms of biogenic emissions and the low levels could indicate transport from other areas. In general, the benzene concentrations in rural sites are higher than toluene or other alkylated benzenes, due to the accelerated atmospheric degradation of alkylated benzenes relative to benzene (Atkinson, 1990; Clarkson et al., 1996). Air from the urban area could undergo chemical oxidation before reaching this area. The presence of

263

anthropogenic VOCs such as benzene may suggest that this area is certainly affected by the transport of anthropogenic air masses under certain meteorological conditions, which is confirmed by the analysis of back trajectories data. The highest concentration of benzene was measured in the last week of March (Fig. 2b) where the main contribution of air masses came from Mediterranean basin. On the other hand, benzene is emitted from the same primary sources as formaldehyde and is considered a good vehicle exhaust marker. The formaldehyde/benzene ratio would not change appreciable in the absence of photochemical formation of carbonyls in the atmosphere. The ratio observed at the sampling point for the last week of March is 13.6 confirming that formaldehyde was mostly produced by photochemical reaction during this time. But also, the direct emission of formaldehyde from vegetation could contribute to increase this ratio. Benzene and toluene concentrations measured at Cabañeros National Park border were lower or much lower than other forest or rural sites around the world such as a rural area of Northern Spain in an area far from sources of pollution (1.8e2.4 mg m3 for benzene; Parra et al., 2006), Montelibretti located at about 30 km from Rome and characterized by very weak traffic (0.5 mg m3 for benzene and 0.9 mg m3 for toluene; Yassaa et al., 2006) or the small village of Covelo Baixo considered a rural/forest zone in Portugal (b.d.le1.73 mg m3 for benzene, 0.81e1.14 mg m3 for toluene; Evtyugina et al., 2006). On the other hand, a parallel sampling to the chemically desorbable RadielloÒ was carried out with the thermally desorbable RadielloÒ, from May 22 to June 19. This study was only qualitative, and much more VOCs were identified, demonstrating the higher sensibility of the thermal desorption. Fig. 5 shows the

Fig. 5. Chromatograms obtained from a blank (A) and an expose cartridge (B) using thermal desorption.

264

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265

chromatogram obtained from a blank and a sample. More than forty peaks have been identified comparing the measured spectrum to the databases. Generally, peaks with a similarity less than 90% have not been included. It is important to note that although many peaks appear using the thermal desorption, the concentrations must be very low taking into account that concentrations using chemically desorbable cartridges, were low or near detection limit. In addition, very few peaks appear in the chromatograms obtained from chemically desorbable cartridges and with a very low signal to noise ratio. 3.3. Ozone The mean ozone concentrations during the whole period are shown in Table 2. Comparison of the monthly average values indicates that ozone concentrations decreased from August up to November, and increased during the early months of the year. Ozone shows a clear seasonal variation with the lowest monthly value registered in November, 33.8 mg m3, and the maximum monthly value registered in March, 84.4 mg m3. The maximum value measured for ozone was 90.5 mg m3 registered during the last week of March. The O3 seasonal behaviour is in agreement with that found by Notario et al. (2012, 2013) but in our case the highest levels are registered in March, followed by May 2011, August 2010 and July 2011. The O3 average behaviour is also similar to that observed at other rural sites in Spain (García et al., 2005). In the study of García et al. (2005) the maximum ozone concentration was also reached in spring, exactly in April 2000 (80.3 mg m3) and May 2001 (87.2 mg m3), instead of summer. Due to O3 levels are high taking into account the extremely low levels of NOx that must have in the area, the meteorological conditions must play a crucial role in the formation of ozone. The analysis of air mass back trajectories on the sampling site indicates from 12 to 20 November, when the concentration of O3 is the lowest, that air mass comes from W and NW sector (Fig. 2a). However, the air masses coming from the eastern Mediterranean basin (Fig. 2b) during 20e27 March should have contributed significantly, together with the higher solar radiation, to the high levels of ozone observed during that week in the sampling point. This region may be affected by air masses loaded with air pollutants coming from large urban areas because of the mesoscale processes developed in the coastal area as commented in site description section. 4. Conclusions Volatile organic compounds, including carbonyl compounds have been measured at Cabañeros National Park border together with O3 levels from AugusteNovember 2010 and FebruaryeAugust 2011 using passive samplers. Total carbonyl concentrations ranged from 2.8 to 19.7 mg m3 being acetoneeacrolein, hexanal, formaldehyde and acetaldehyde the most abundant carbonyls. The results obtained seem to indicate that these carbonyls are produced by photochemical reaction although the contribution of direct biogenic emissions could be important. Most VOCs measured using chemically desorbable cartridges were either not detected or the concentration were below their detection limits although the qualitative study carried out with thermal desorbable cartridges, revealed the presence of much more VOCs. However, the presence of the anthropogenic VOC benzene suggests that this area is certainly affected by the transport of anthropogenic air masses, which is confirmed by the analysis of back trajectories data. These polluted air masses from remote areas, mainly from the eastern Mediterranean basin should be the responsible of the high levels of

ozone observed. Ozone showed a clear seasonal variation with the lowest monthly value registered in November with 33.8 mg m3 and the maximum monthly value registered in March with 84.4 mg m3 instead of summer as it would be expected. Acknowledgements We would like to thank the Junta de Comunidades de Castilla-La Mancha (project PCI08-0123-0381) for their financial support of this study. Authors thank Miguel Villanueva Rodriguez for his collaboration during the sampling period. References Atkinson, R., 1990. Gas-phase tropospheric chemistry of organic compounds: a review. Atmos. Environ. 24A, 1e41. Bergweiler, C., Carreras, H., Wannaz, E., Rodríguez, J., Toselli, B., Olcese, L., Pignata, M.L., 2008. Field surveys for potential ozone bioindicator plant species in Argentina. Environ. Monit. Assess. 138, 305e312. Ciccioli, P., Brancaleoni, E., Frattoni, M., Cecinato, A., Brachetti, A., 1993. Ubiquitous occurrence of semi-volatile carbonyl compounds in tropospheric samples and their possible sources. Atmos. Environ. 27A, 1891e1901. Clarkson, T.S., Martin, R.J., Rudolph, J., Graham, B.W.L., 1996. Benzene and toluene in New Zealand air. Atmos. Environ. 30, 569e577. Coll, I., Rousseau, C., Barletta, B., Meinardi, S., Blake, D.R., 2010. Evaluation of an urban NMHC emission inventory by measurements and impact on CTM results. Atmos. Environ. 44, 3843e3855. Custódio, D., Guimaraes, C.S., Varandas, L., Arbilla, G., 2010. Pattern of volatile aldehydes and aromatic hydrocarbons in the largest urban rainforest in the Americas. Chemosphere 79, 1064e1069. Draxler, R.R., Stunder, B., Rolph, G., Taylor, A., 2009. Hysplit 4 User’s Guide. NOAA Air Resources Laboratory, Silver Spring, MD. Evtyugina, M.G., Nunes, T., Pio, C., Costa, C.S., 2006. Photochemical pollution under sea breeze conditions, during summer at the Portuguese West Coast. Atmos. Environ. 40, 6277e6293. Figueruelo, J., Dávila, M., 2004. Química Física del Ambiente y de los procesos ambientales. Reverté, Barcelona. Finlayson-Pitts, B.J., Pitts Jr., J.N., 2000. Chemistry of the Upper and Lower Atmosphere: Theory, Experiments, and Applications. Wiley, New York. Fishman, J., Creilson, J.K., Parker, P.A., Ainsworth, E.A., Vining, G.G., Szarka, J., Booker, F.L., Xu, X., 2010. An investigation of widespread ozone damage to the soybean crop in the upper Midwest determined from ground-based and satellite measurements. Atmos. Environ. 44, 2248e2256. Fondazione Salvatore Maugeri. Instruction manual for Radiello sampler. Available online: Website http://www.radiello.com. Fruekilde, P., Hjorth, J., Jensen, N.R., Kotzias, D., Larsen, B., 1998. Ozonolysis at vegetation surfaces: a source of acetone, 4-oxopentanal, 6-methyl-5-hepten-2one, and geranyl acetone in the troposphere. Atmos. Environ. 32, 1893e1902. Gallego, E., Roca, F.X., Guardino, X., Rosell, M.G., 2008. Indoor and outdoor BTX levels in Barcelona City metropolitan area and Catalan rural areas. J. Environ. Sci. 20, 1063e1069. García, M.A., Sánchez, M.L., Pérez, I.A., de Torre, B., 2005. Ground level ozone concentrations at a rural location in northern Spain. Sci. Total Environ. 348, 135e 150. Goldstein, A.H., Schade, G.W., 2000. Quantifying biogenic and anthropogenic contributions to acetone mixing ratios in a rural environment. Atmos. Environ. 34, 4997e5006. Heiden, A.C., Kobel, K., Komenda, M., Koppmann, R., Shao, M., Wildt, J., 1999. Toluene emissions from plants. Geophy. Res. Lett. 26 (9), 128e1286. Jork, E.M., 1996. Laboruntersuchungen zum Austausch organischer Säuren und Aldehyde zwischen landwirtschaftlichen Pflanzen und der Atmosphäre. Diplomarbeit. Fachbereich Biologie, Universität Mainz. Kesselmeier, J., Staudt, M., 1999. Biogenic volatile organic compounds (VOC): an overview on emission, physiology and ecology. J. Atmos. Chem. 33, 23e88. Kesselmeier, J., Bode, K., Hofmann, U., Müller, H., Schäfer, L., Wolf, A., Ciccioli, P., Brancaleoni, E., Cecinato, A., Frattoni, M., Foster, P., Ferrari, C., Jacob, V., Fugit, J.L., Dutaur, L., Simon, V., Torres, L., 1997. The BEMA-Project: emission of short chained organic acids, aldehydes and monoterpenes from Quercus ilex L. and Pinus pinea L. in relation to physiological activities, carbon budget and emission algorithms. Atmos. Environ. 31, 119e134. Kirstine, W., Galbally, I., Ye, Y., Hooper, M., 1998. Emissions of volatile organic compounds (primarily oxygenated species) from pasture. J. Geophys. Res. 103 (D9), 10605e10619. König, G., Brunda, M., Puxbaum, H., Hewitt, C.N., Duckham, S.C., Rudolph, J., 1995. Relative contribution of oxygenated hydrocarbons to the total biogenic VOC emission of selected mid-European agricultural and natural plant species. Atmos. Environ. 29, 861e874. Kreuzwieser, J., Cojocariu, C., Jüssen, V., Rennenberg, H., 2002. Elevated atmospheric CO2 causes seasonal changes in carbonyl emissions from Quercus ilex. New Phytol. 154, 327e333.

F. Villanueva et al. / Atmospheric Environment 85 (2014) 256e265 Martin, R.S., Villanueva, I., Zhang, J., Popp, C.J., 1999. Nonmethane hydrocarbon, monocarboxylic acids, and low molecular weight aldehyde and ketone emissions from vegetation in Central New Mexico. Environ. Sci. Technol. 33, 2186e 2192. Martin, P., Cabañas, B., Villanueva, F., Gallego, M.P., Colmenar, I., Salgado, S., 2010. Ozone and nitrogen dioxide levels monitored in an urban area (Ciudad Real) in central-southern Spain. Water Air Soil Pollut. 208, 305e316. Millán, M., Sanz, M.J., Salvador, R., Mantilla, E., 2002. Atmospheric dynamics and ozone cycles related to nitrogen deposition in the western Mediterranean. Environ. Pollut. 118, 167e186. Morknoy, D., Khummongkol, D., Prueaksasit, T., 2011. Seasonal and diurnal concentrations of ambient formaldehyde and acetaldehyde in Bangkok. Water Air Soil Pollut. 216, 693e702. Müller, K., Haferkorn, S., Grabmer, W., Wisthaler, A., Hansel, A., Kreuzwieser, J., 2006. Biogenic carbonyl compounds within and above a coniferous forest in Germany. Atmos. Environ. 40, S81eS91. Notario, A., Díaz-de-Mera, Y., Aranda, A., Adame, J.A., Parra, A., Romero, E., Parra, A., Muñoz, F., 2012. Surface ozone comparison conducted in two rural areas in central-southern Spain. Environ. Sci. Pollut. Res. 19, 186e200. Notario, A., Bravo, I., Adame, J.A., Díaz-de-Mera, Y., Aranda, A., Rodríguez, A., Rodríguez, D., 2013. Variability of oxidants (OX ¼ O3 þ NO2), and preliminary study on ambient levels of ultrafine particles and VOCs, in an important ecological area in Spain. Atmos. Res. 128, 35e45. Parra, M.A., González, L., Elustondo, D., Garrigó, J., Bermejo, R., Santamaría, J.M., 2006. Spatial and temporal trenes of volatile organic compounds (VOC) in a rural area of northern Spain. Sci. Total Environ. 370, 157e167. Pilidis, G.A., Karakitsios, P., Kassomenos, P.A., 2005. BTX measurements in a médium-sized European city. Atmos. Environ. 39, 6051e6065. Pio, C.A., Silva, P.A., Cerqueira, M.A., Nunes, T.V., 2012. Diurnal and seasonal emissions of volatile organic compounds from cork oak (Quercus suber) trees. Atmos. Environ. 39, 1817e1827.

265

Possanzini, M., Tagliacozzo, G., Cecinato, A., 2007. Ambient levels and sources of lower carbonyls at Montelibretti, Rome (Italy). Water Air Soil Pollut. 183, 447e454. Schäfer, L., Gabriel, R., Müller, H., Wolf, A., Kesselmeier, J., 1995. Emission of short chained organic acids and aldehydes in relation to physiological activities and apoplastic ion concentration in Mediterranean tree species during the B.E.M.A. field campaign in May 1994. In: Biogenic Emissions in the Mediterranean Area, BEMA-Project, Report on the 1st BEMA measuring campaign at Castelporziano, Rome (Italy), May 1994, EUR 16293 EN, Brussels, pp. 233e247. Seinfeld, J.H., Pandis, S.N., 1998. Atmospheric Chemistry and Physics e from Air Pollution to Climate Change. Wiley, New York. Shepson, P.B., Hastie, D.R., Schiff, H.I., Polizzi, M., Bottenheim, J.W., Anlauf, K., Mackay, G.I., Karecki, D.R., 1991. Atmospheric concentrations and temporal variations of C1eC3 carbonyl compounds at two rural sites in central Ontario. Atmos. Environ. 25A, 2001e2015. Stewart, D.J., Almabrok, S.H., Lockhart, J.P., Mohamed, O.M., Nutt, D.R., Pfrang, C., Marston, G., 2013. The kinetics of the gas-phase reactions of selected monoterpenes and cyclo-alkenes with ozone and the NO3 radical. Atmos. Environ. 70, 227e235. Villanueva, F., Notario, A., Adame, J.A., Millán, M.C., Mabilia, R., Albaladejo, J., 2013. A preliminary study on ambient levels of carbonyls, benzene, toluene and xylene in the south-west of the Iberian Peninsula (Huelva coast), Spain. Environ. Technol. 34, 289e299. Villanueva-Fierro, I., Popp, C.J., Martin, R.S., 2004. Biogenic emissions and ambient concentrations of hydrocarbons, carbonyl compounds and organic acids from ponderosa pine and cottonwood trees at rural and forested sites in Central New Mexico. Atmos. Environ. 38, 249e260. Yassaa, N., Brancaleoni, E., Frattoni, M., Ciccioli, P., 2006. Isomeric analysis of BTEX in the atmosphere using b-cyclodextrin capillary chromatography coupled with thermal desorption and mass spectrometry. Chemosphere 63, 502e508. Yu, Y., Wen, S., Lü, H., Feng, Y., Wang, X., Sheng, G., Fu, J., 2008. Characteristics of atmospheric carbonyls and VOCs in Forest Park in South China. Environ. Monit. Assess. 137, 275e285.