Stomatal conductance models for ozone risk assessment at ... - Mottles

Agricultural and Forest Meteorology 234 (2017) 212–221

Contents lists available at ScienceDirect

Agricultural and Forest Meteorology journal homepage: www.elsevier.com/locate/agrformet

Original article

Stomatal conductance models for ozone risk assessment at canopy level in two Mediterranean evergreen forests Yasutomo Hoshika a , Silvano Fares b,∗ , Flavia Savi b,c , Carsten Gruening d , Ignacio Goded d , Alessandra De Marco e , Pierre Sicard f , Elena Paoletti a a

Institute of Sustainable Plant Protection, National Research Council of Italy, Via Madonna del Piano 10, I-50019 Sesto Fiorentino, Italy Council for Agricultural Research and Economics, Research Center for the Soil-Plant System, Via della Navicella 2-4, 00184 Rome, Italy c Department for Innovation in Biological, Agro-Food and Forest Systems (DIBAF), University of Tuscia, Viterbo, Italy d European Commission, Joint Research Centre, Ispra, Italy e Italian National Agency for New Technologies, Energy and Sustainable Economic Development, C.R. Casaccia, Via Anguillarese 301, 00123 S. Maria di Galeria, Rome, Italy f ACRI-HE, 260 Route Du Pin Montard BP234, 06904 Sophia-Antipolis Cedex, France b

a r t i c l e

i n f o

Article history: Received 15 April 2016 Received in revised form 26 December 2016 Accepted 4 January 2017

Keywords: Ground-level ozone Stomatal conductance models Pinus pinea Quercus ilex Mediterranean forest

a b s t r a c t Forests in Mediterranean Europe are located in a hot spot for tropospheric ozone formation. We applied two canopy-level stomatal conductance models (Jarvis-type and Ball-Woodrow-Berry models) to two Mediterranean evergreen forests (Umbrella pine at San Rossore, and Holm oak at Castelporziano, in central Italy), which is essential for assessing ozone impact on forests via the stomatal flux-based approach. Parameterizations of the models was carried out by the Eddy Covariance technique. Both Jarvis-type and Ball-Woodrow-Berry models well explained the observed stomatal conductance and stomatal ozone flux in both forests. Maximum stomatal conductance was 72% higher in Umbrella pine than in Holm oak, leading to higher stomatal ozone flux. Inclusion of a soil water function improved the performance of the Jarvis-type model for the estimation of stomatal ozone flux. We found contrasting results concerning the coefficient m (the slope of the photosynthesis-stomatal conductance relationship) in the Ball-WoodrowBerry model between the two forests: while m was constant for varying soil water status in the Holm oak forest, it declined with soil drying in the Umbrella pine forest. This may result from higher stomatal sensitivity to drought in Umbrella pine as a response to avoid drought stress. Overall, these results advance our understanding of ozone risk assessment in Mediterranean forests. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Concentrations of tropospheric ozone (O3 ) have now doubled in the northern hemisphere since the pre-industrial period (Hartmann et al., 2013). In the Mediterranean countries, current surface O3 levels are high enough to damage forest trees (Sicard et al., 2013). Ozone induces damages to plants, e.g., visible leaf injury, premature leaf senescence, decreased photosynthesis, reduced growth rates and altered allocation of carbon (Paoletti, 2007; Mills et al., 2011; Matyssek et al., 2013; Fares et al., 2013a; Sicard et al., 2016). Recent O3 risk assessment for forest trees has focused on a stomatal O3 flux basis (Hoshika et al., 2012a; CLRTAP, 2015), because

∗ Corresponding author. E-mail addresses: [email protected], [email protected] (S. Fares). http://dx.doi.org/10.1016/j.agrformet.2017.01.005 0168-1923/© 2017 Elsevier B.V. All rights reserved.

O3 enters the plant tissues through the stomata and directly damages cell proteins and membranes by oxidation (Omasa et al., 2002; Contran and Paoletti, 2007; Leisner and Ainsworth, 2012; Tiwari et al., 2016). To better understand O3 flux partitioning in stomatal and non-stomatal sinks over forest canopies, direct measurements of O3 fluxes were carried out using the Eddy Covariance technique (Cieslik, 2004; Mikkelsen et al., 2004; Fares et al., 2013a). Thanks to techniques such as the Evaporative/Resistance method, bulkcanopy stomatal conductance can be estimated from the energy balance of a forest canopy (Monteith and Unsworth, 1990; Cieslik, 2004; Fares et al., 2013b). In the last decade, it was intensively discussed in which way O3 damage to plants can best be estimated (CLRTAP, 2015). Eddy covariance provides an opportunity to parameterize canopy-level stomatal conductance for forests, where leaves may not respond to their environment equally throughout the canopy (Baldocchi, 1989). An accurate parameterization is

Y. Hoshika et al. / Agricultural and Forest Meteorology 234 (2017) 212–221

essential for a stomatal flux-based approach to assess O3 impact on forest trees (Zhang et al., 2003; Emberson et al., 2007). Traditionally, two main empirical approaches are used for estimating stomatal conductance (Damour et al., 2010): the empirical Jarvis-type model (Jarvis, 1976; Emberson et al., 2007) and the semi-empirical approach, called Ball-Woodrow-Berry (BWB) model (Ball et al., 1987). The former approach is based on a multiplicative algorithm that adjusts a reference value of stomatal conductance (i.e., maximum stomatal conductance, Gmax ) according to changes in phenology and environmental variables such as light intensity, air temperature, vapor pressure deficit and soil moisture. This model is recommended by UNECE-CLRTAP (United Nations Economic Commission for Europe Convention on Longrange Transboundary Air Pollution) to calculate stomatal O3 fluxes and estimate a metric for O3 risk assessment of forest trees in Europe (CLRTAP, 2015). The BWB method assumes that stomatal conductance is tightly coupled to photosynthetic rate because stomata open and close to keep a nearly constant ratio between intercellular and ambient CO2 concentration. This ratio may vary with atmospheric humidity. Therefore, Ball et al. (1987) elaborated a model that links stomatal conductance to leaf photosynthesis, humidity deficit and CO2 concentration at the leaf surface. The Mediterranean climate is characterized by a hot summer with low precipitation, which results in a limitation of stomatal O3 flux by stomatal closure due to drought stress (Paoletti, 2006). Southern Europe is representative of water-limited environments (dry and semi-dry habitats) that cover about 41% of Earth’s land surface (Reynolds et al., 2007). Soil moisture deficit represents a major limiting factor for stomatal conductance in the Mediterranean region (Chaves et al., 2002; Alonso et al., 2008). Several studies have tried to include a soil moisture function into the Jarvistype model because it is critical for water-limited environments (Stewart, 1988; Alonso et al., 2008; Büker et al., 2012; GonzálezFernández et al., 2013; De Marco et al., 2016). Also for the BWB model, recent studies have included the drought impact by using the modification of m (the slope of the photosynthesis-stomatal conductance relationship of the BWB model) with soil drying (van Wijk et al., 2000, 2002; He et al., 2014; Knauer et al., 2015). However, it is still under discussion whether drought stress may change the coefficient m (Sala and Tenhunen, 1996; Xu and Baldocchi, 2003; Baldocchi and Xu, 2005; Keenan et al., 2009; Fares et al., 2013b). In Mediterranean Europe, previous modeling studies for O3 risk assessment have mainly targeted the evergreen broadleaf Holm oak (Quercus ilex) representing the most typical climax forest (Emberson et al., 2007; Fares et al., 2013b). However, Mediterranean forests are also dominated by pioneer tree species such as pines (Pinus pinea, P. pinaster and P. halepensis). The crown condition of Mediterranean lowland pines is characterized by an increase in mean defoliation since 1991 (Fischer and Lorenz, 2011; Sicard and Dalstein-Richier, 2015). Stomatal conductance models may need different parameters for Mediterranean pine forests. For example, under water limitation, pine trees may avoid drought stress by earlier stomatal closure relative to oak species (Picon et al., 1996). The aims of this study were i) to characterize the difference in the stomatal conductance parameters of both Jarvis-type and BWB models in two representative Mediterranean forests (Q. ilex and P. pinea), ii) to examine whether the slope of the photosynthesisstomatal conductance relationship (i.e., the coefficient m) in the BWB parameterization changes under drought stress, and iii) to test the performance of both Jarvis-type and BWB models for the estimation of canopy-level stomatal O3 flux in both forests.

213

Table 1 Daily mean temperature (T, in ◦ C), daily mean volumetric soil water content (SWC, in m3 m−3 ) and daily mean ozone concentration (in ppb) in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore) and associated standard deviations (±SE). Sites and seasons

T (◦ C)

SWC (m3 m−3 )

O3 (ppb)

Castelporziano Spring Summer Autumn Winter

18.3 (±0.2) 26.7 (±0.2) 23.2 (±0.2) 13.1 (±0.3)

0.187 (±0.004) 0.100 (±0.002) 0.126 (±0.005) 0.194 (±0.006)

38.7 (±0.7) 40.2 (±0.9) 29.9 (±0.8) 23.2 (±1.0)

San Rossore Spring Summer Autumn Winter

17.0 (±0.4) 24.4 (±0.2) 19.5 (±0.3) 7.8 (±0.6)

0.085 (±0.002) 0.051 (±0.001) 0.032 (±0.002) 0.174 (±0.011)

36.1 (±1.5) 50.8 (±1.6) 34.2 (±2.3) 23.8 (±1.2)

At Castelporziano (in 2013–2014), spring: March to May, summer: June to August, autumn: September to November, winter: January, February and December. At San Rossore (in 2013), spring: 22 April-12 May, summer: 8–27 July, autumn: 9–29 September, winter: 20 January-10 February.

2. Materials and methods 2.1. Site descriptions The first site is a Holm oak forest at Castelporziano (41◦ 70 42 N, 12◦ 35 72 E), 15 m a.s.l., 1.5 km from the seashore of the Thyrrenian sea, and 25 km SW from the center of Rome (Italy). The soil has a sandy texture (sand content above 60%) and low water-holding capacity according to a detailed study carried out by Pinzari et al. (1999). The dominant tree species is Q. ilex with an average height of 14.9 m and a Leaf Area Index of 3.7 m2 leaf m−2 ground . The forest stand is uneven-aged with the oldest trees planted more than 80 years ago. Further details are described in Fares et al. (2014). The second site is a Pinus pinea (Umbrella pine) forest at San Rossore (43◦ 43 55 N, 10◦ 17 27 E), 12 m a.s.l., approximatively 1200 m from the seashore of the Thyrrenian sea and 8 km from Pisa (Italy). The soil is a sandy calcaric regosol. Soil texture was as follows (% on weight): clay 3%, silt 2% and sand 95% (Gruening, person. comm.). It is an almost pure, even-aged about 93 years old forest. The average canopy height is 19 m, with a Leaf Area Index of 3.3 m2 leaf m−2 ground . Further information on the general area of the measurement site is given in Matteucci et al. (2015). At both sites, the climate is Thermo-Mediterranean, characterized by prolonged stress aridity during the summer, and a moderate cold stress during the winter. The meteorological data during the measurement campaign were shown in Table 1. 2.2. Measurements of environmental parameters and fluxes At Castelporziano, values of air temperature, precipitation (Davis vantage pro meteorological station, Davis Instruments Corp. CA, USA) and volumetric soil moisture content in the soil profile (10, 50, 100 cm depth, CS 650, Campbell scientifics, Shepshed, UK) were measured at 1-min resolution, averaged at half-hour intervals and recorded by a data logger (CR3000, Campbell scientifics, Shepshed, UK). Flux measurements above canopy started on 1st January, 2013 and ended on 31st December, 2014. Instantaneous wind speed and temperature fluctuation were measured by a three-dimensional sonic anemometer (Gill Windmaster, Gill Instruments, Lymington, UK). Closed-path analytical equipment was installed in an air conditioned cabin, below a 19-m tall tower. Fast response measurements of O3 were made by chemiluminescence using coumarin dye with an instrument custom developed by the National Oceanic and Atmospheric Administration (NOAA, Silver Spring, MD). Air was sampled continuously at one inlet at the top of the tower through Teflon tubes with 4 mm internal diameter and a Teflon filter (PFA

214


holder, PTFE membrane, pore size 2 mm) at 30 cm from the inlet. The filters were replaced every two weeks to avoid contamination or flow problems. O3 detector was calibrated against 30 min average O3 concentrations from a second UV absorption monitor (Thermo scientifics, USA), which measured air sampled from the same inlet height. H2 O and CO2 concentrations were measured with a close path infrared gas analyzer (LI-7200, LICOR, Lincoln, NE, USA). The raw analog data were recorded in a datalogger (LI7550, LICOR, Lincoln, NE, USA) at 10 Hz for all gases. Ozone, CO2 and H2 O fluxes were calculated as the covariance of gas concentration and vertical wind velocity according to the eddy covariance technique, as described in Fares et al. (2012). The wind data were rotated according to the planar fit method (Wilczak et al., 2001). Errors due to sensor separation and damping of high frequency eddies were corrected using spectral analysis techniques as outlined by Rissmann and Tetzlaff (1994). Data were discarded if results from the stationary test were above 60% (Foken and Wichura, 1996), or low turbulence (u* < 0.15 m s−1 ) occurred, which was found during >80% of night hours. At San Rossore, four campaigns of flux measurements were conducted on 20th January-10th February, 22nd April-12th May, 8th-27th July and 9th-29th September, 2013. Water and CO2 concentrations were measured using an enclosed infra-red gas analyzer (LI-7200, Li-Cor, Lincoln, NE, USA). Ozone concentration was measured with a UV absorption monitor (2 B Technologies). A 3-dimensional sonic anemometer (HS-50, Gill Instruments, Lymington, UK) was used to instantaneously measure wind speed and directions. These instruments were installed at a height of 24 m. For the flux calculations, the EdiRe software package was used (R. Clement, University of Edinburgh). For CO2 and water fluxes, gap filling and partitioning were performed with the online Eddy Covariance data gap-filling and flux-partitioning tool (Reichstein et al., 2005), including u*-filtering to replace unreliable data due to low-turbulence conditions with gap-filled data. Stomatal conductance from the Eddy Covariance measured evapotraspiration (Gsw ) was calculated using the Monteith equation also called Evaporative/Resistance method extensively discussed in previous research (Monteith and Unsworth, 1990; Cieslik, 2004; Fares et al., 2012). To minimize soil evaporation effect on total evapotranspiration (and therefore overestimate Gsw ), we did not use measurements for three days after a rainfall event. Photosynthesis was assumed to be equal to Gross Primary Production (GPP), calculated by adding the ecosystem respiration term (Reco ) to net ecosystem exchange (NEE) directly measured with Eddy Covariance. Reco was calculated using the respiration measurements made at night and extrapolated to the daytime based on air temperature according to the model formulation explained by Lasslop et al. (2010).

10, 50, 100 cm depths at Castelporziano and 10 and 50 cm depths at San Rossore were tested in this study), respectively. The parameters can be found in Table 2. The response of stomatal conductance to phenology (fphen ) is described as follows:for Astart ≤ DOY < (Astart + fphen a ),

fphen = 1 − fphen

for (Aend − fphen b ) < DOY ≤ Aend ,

fphen = 1 − fphen

,

(1)

where Gs is stomatal conductance for O3 at canopy-level (mol m−2 s−1 ) and is expressed per unit of ground area (Wieser et al., 2008), Gmax is the maximum value of canopy stomatal conductance (mol m−2 s−1 ). The other functions are limiting factors of Gmax and are expressed in relative terms (i.e. values between 0 and 1). Here, fmin is a fraction to express the minimum stomatal conductance, fphen, flight , ftemp , fVPD , and fSWC are the variation in Gmax with leaf age, photosynthetic photon flux density at the leaf surface (PPFD, ␮mol photons m−2 s−1 ), temperature (T, ◦ C), vapor pressure deficit (VPD, kPa), and volumetric soil water content (SWC, m3 m−3 ;

d

· (Aend − DOY) /fphen

b

+ fphen

d

flight = 1 − exp (a · PPFD)

(2)

(3)

where a is a species-specific parameter defining the shape of the exponential relationship. The function of air temperature (T, ◦ C) is expressed

as:

ftemp =

T − Tmin Topt − Tmin

Tmax − T Tmax − Topt

Tmax −Topt Topt −Tmin

(4)

where Topt , Tmin , and Tmax denote the optimal, minimum, and maximum temperature (◦ C) for stomatal conductance, respectively. The response of stomatal conductance to vapor pressure deficit (VPD, kPa) is given by: fVPD =

(1 − fmin ) · (VPDmin − VPD) + fmin VPDmin − VPDmax

(5)

where VPDmin and VPDmax denote the threshold of VPD (kPa) for attaining minimum and full stomatal aperture, respectively. If VPD exceeds VPDmin then fVPD is set to fmin . If VPD is lower than VPDmax then fVPD is 1. Following Stewart (1988) and van Wijk et al. (2000), the standard function of stomatal conductance response to volumetric soil water content (SWC, m3 m−3 ) is given as: fSWC = 1 − k1 exp (k2 · S)

0≤S≤1

fSWC = 1 − k1

S < 0 (6)

S=

Gs = Gmax · fphen · flight · max fmin , ftemp · fVPD · fSWC

+ fphen c ;

where DOY is the day of the year. Here Astart and Aend are the year days for start and end of the growing season, respectively. The parameters fphen a and fphen b represent the number of days of fphen to reach its maximum and the number of days during the decline of fphen to the minimum value. The parameters fphen c and fphen d represent maximum fraction of fphen at Astart and Aend , respectively. The response of stomatal conductance to PPFD, i.e., flight , is specified as:

Canopy-level Jarvis-type stomatal conductance model is based on the multiplicative algorithm described by Jarvis (1976) and Emberson et al. (2007):

a

fphen = 1 ;

where

· (DOY-Astart ) /fphen

for (Astart + fphen a ) ≤ DOY ≤ (Aend -fphen b ),

2.3. Modelling of stomatal conductance

c

f − SWC f − w

(7)

and f and w are the volumetric soil water contents (m3 m−3 ) at field capacity and wilting point, respectively. k1 and k2 are the empirical parameters determining the saturating curve for Gs response to SWC. In the present study, we set those parameters according to Mintz and Walker (1993) (i.e., f = 0.22 m3 m−3 , w = 0.07 m3 m−3 at Castelporziano as fine sandy loam soil type; f = 0.12 m3 m−3 , w = 0.03 m3 m−3 at San Rossore as fine sand soil type). A parameter estimation was carried out using a boundary line analysis (Alonso et al., 2008; Hoshika et al., 2012a,b). First, in order to limit errors in calculating stomatal conductance for water vapor (Gsw ), data were only analyzed for VPD > 0.6 kPa (Ewers and Oren, 2000). Then the stomatal conductance data were divided


215

Table 2 Summary of Jarvis-type model parameters in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore). Parameter

(mol O3 m−2 s−1 )

Gmax fmin

Castelporziano (Holm oak forest)

San Rossore (Umbrella pine forest)

0.22

0.38

(fraction)

0.05

0.03

fphen

Astart Aend fphen a fphen b fphen c fphen d

(day of year) (day of year) (days) (days) (fraction) (fraction)

0 365 120 120 0.5 0.5

0 365 130 130 0.5 0.5

flight

a

(constant)

−0.0022

−0.0032

ftemp

Topt Tmin Tmax

(◦ C) (◦ C) (◦ C)

20 6 50

20 6 39

fVPD

VPDmax VPDmin

(kPa) (kPa)

0.6 3.9

0.6 4.2

fSWC

k1 k2 k1 k2 k1 k2

At 10 cm depth

0.58 1.90 0.67 1.88 0.70 1.22

0.46 1.27 0.47 2.84 – –

At 50 cm depth At 100 cm depth

Gmax , maximum stomatal conductance; fmin , minimum stomatal conductance; fphen , flight , ftemp , fVPD and fSWC are the variation in Gmax with leaf age, photosynthetic photon flux density (PPFD, ␮mol m−2 s−1 ), temperature (T, ◦ C), vapor pressure deficit (VPD, kPa), and volumetric soil water content (m3 m−3 ), respectively; Astart and Aend : the year days for start and end of the growing season, fphen a and fphen b : the number of days for fphen to reach its maximum and the number of days during the decline of fphen for the minimum to again be reached, fphen c and fphen d : fphen at Astart and Aend , a: is a parameter determining the shape of the hyperbolic relationship, Topt , Tmin , and Tmax : optimal, minimum and maximum air temperature for stomatal opening, VPDmin and VPDmax : vapor pressure deficit for attaining minimum and full stomatal aperture, k1 and k2 are the empirical parameters determining the Gs response to soil water content at each depth of soils (10, 50, 100 cm).

into classes with the following step-wise increases for each variable (Braun et al., 2010; Hoshika et al., 2012a,b): 200 ␮mol photons m−2 s−1 for PPFD (when PPFD values were less than 200 ␮mol photons m−2 s−1 , PPFD classes at 50 ␮mol photons m−2 s−1 steps were adopted), 2 ◦ C for T, 0.4 kPa for VPD and 0.01 m3 m−3 for SWC. A function was fitted against each model variable based on the 95th percentile value per each class of environmental factors. Values of Gmax and fmin were calculated from the maximum value of hourly canopy stomatal conductance data when VPD > 0.6 kPa and as 5th percentile of the data, respectively (Pleijel et al., 2002; Wever et al., 2002; Hoshika et al., 2012b). The equation of canopy-level BWB model (Ball et al., 1987) is given by: Gsc

m·A = · RH + G0 [CO2 ]

(8)

where Gsc is stomatal conductance for CO2 (mol m−2 s−1 ), A is photosynthetic assimilation rate (␮mol m−2 s−1 ), RH is relative humidity (fraction), [CO2 ] is CO2 concentration at leaf surface (␮mol mol−1 ), G0 is the residual stomatal conductance when A tends to zero (mol m−2 s−1 ), and m is an empirical coefficient which represents the composite sensitivity of stomatal conductance to photosynthesis. In this study, the following assumptions were employed for the application at canopy level: i) GPP was used for A (Wang et al., 2009; Fares et al., 2013b), ii) G0 was assumed to be zero, and iii) changes in RH and CO2 concentration between the measuring height and the leaf surface were assumed to be negligible. The performance of the BWB model applied to forest canopies was discussed by Wang et al. (2009) and Fares et al. (2013b). They proposed that the term (m·RH) in Eq. (8) can be considered a humidity response function (f(H)) according to the following equation: f (H) =

[CO2 ] · FLE

· ı · L · a · qsat(Ta) − qa · GPP

−

[CO2 ] · G0 GPP

where f(H) is a dimensionless function representing the response of stomatal conductance to the humidity of air at the leaf surface, is the volume of gas per mole (m3 ), ı is a parameter accounting for the difference of stomatal conductance to water vapor and to CO2 (=1.6), L is the latent heat of vaporization of water (J kg−1 ), a and qa are the density (kg m−3 ) and specific humidity (kg kg−1 ) of the ambient air, respectively, qsat(Ta) is the saturated specific humidity (kg kg−1 ) at air temperature, and FLE is the ecosystem level latent heat flux (W m−2 ) assuming that soil contribution to latent heat flux is negligible for the measuring period. To simplify the model parameters of stomatal conductance, an exponential relationship of f(H) was calculated at canopy level, in the simple form of: f (H) = exp (b · RH)

(10)

where b is the coefficient of the exponential equation. Stomatal conductance was therefore calculated as: Gsc =

exp (b · RH) · A + G0 [CO2 ]

(11)

van Wijk et al. (2000, 2002) tried to consider the impact of drought stress for the BWB model using the saturation curve (Eqs. (6), (7)). Based on this standard response function, we discussed whether the coefficient m in the BWB parameterization changes under drought stress. The dependency of f(H), which relates to m (i.e., m*RH), on SWC was tested. Two approaches were examined here: i) no change of f(H) throughout the season, and ii) a change of f(H) with soil water condition. Estimation of stomatal ozone flux Here a simplified equation was used to estimate canopy stomatal ozone flux according to CLRTAP (2015) as follows: Fst = Gs · [O3 ]

(12) (nmol m−2 s−1 ),

(9)

where Fst is the canopy stomatal flux of ozone Gs is stomatal conductance for O3 at canopy level (mol m−2 s−1 ), and [O3 ] is ozone concentration at canopy height. Gs was estimated by

216


both Jarvis and BWB models in the Holm oak and Umbrella pine forests. 2.4. Data analysis Simple correlation analysis was used to test the relationship between measured and estimated stomatal conductance. The exponential regression analysis was used to derive f(H) for the BWB parameterization. The analysis was performed using SPSS software (SPSS, Chicago, USA). 3. Results 3.1. Ozone concentration and ozone flux The daily O3 mean concentration was highest in summer at both sites (Table 1). Hourly maxima O3 concentrations at Castelporziano and San Rossore were 120 and 92 ppb, respectively, and were recorded during summer (not shown). Maximum temperature was recorded in summer at both sites. Mean temperatures were few degrees lower at San Rossore than at Castelporziano. SWC at 50 cm declined from winter to summer at both sites, while in autumn it increased at Castelporziano and decreased further at San Rossore. Fig. 1 shows total and stomatal O3 fluxes in Holm oak and Umbrella pine forests. Most of O3 removed by the forests was in the canopy region, because a portion up to 64% of the total O3 deposition was by stomatal uptake. Both total and stomatal O3 fluxes were higher in the Umbrella pine forest than in the Holm oak forest. Although O3 concentration in summer was as high as in spring in the Holm oak forest, stomatal O3 flux was lower in summer than in spring (110 and 198 ␮mol m−2 , respectively, as daily average accumulated stomatal O3 flux). And then stomatal O3 flux decreased in autumn and winter (76 and 79 ␮mol m−2 , respectively). On the other hand, in the Umbrella pine forest, although O3 concentration was higher in summer than in spring (+41%, Table 1), stomatal O3 flux was similar in both seasons (560 and 559 ␮mol m−2 , in spring and summer, respectively, as daily average accumulated stomatal O3 flux). The lowest stomatal O3 flux was recorded in winter (60 ␮mol m−2 ) at San Rossore. 3.2. Application of the jarvis-type model Parameters of limiting functions in the Jarvis-type stomatal conductance model (i.e., Gmax , fmin , flight , ftemp , fVPD , fSWC and fphen ) are listed in Table 2. The Gmax value was higher in Umbrella pine than in Holm oak, while fmin was four times higher in Holm oak. The parameters of fphen were similar in the two forests, although Gs needed 10 days more for increasing or declining in Umbrella pine. The response of stomatal conductance to PPFD (flight ) followed a typical light-response curve, with a light saturation point above 500 ␮mol m−2 s−1 in both forests (1050 ␮mol m−2 s−1 in Holm oak and 720 ␮mol m−2 s−1 in Umbrella pine i.e. the PPFD when Gs reached 90% of Gmax ). The response of stomatal conductance to air temperature (ftemp ) indicated that the optimal temperature for stomatal aperture was similar in both species. Similar stomatal responses to VPD were observed in both species, indicating that Gs was 30–40% of Gmax at 3.0 kPa of VPD. The saturation curve of Gs response to SWC was similar at both sites. Gs declined by 50–70% at w . Table 3 shows the results of a comparison between modelled and observed Gs in both forests. Inclusion of fSWC in the Jarvis-type model increased the model predictive capacity. We did not find a different performance of the model if SWC was calculated over different soil depths (data not shown). Therefore just the result using the SWC data down to 50 cm is shown in the following section.

fSWC increased R2 values, and reduced RMSE of the model estimation at both sites: 10% at the Holm oak forest, and 33% at the Umbrella pine forest. Fig. 2 shows a comparison between modelled and observed Gs in both forests in each season. Measured Gs peaked in spring in both forests. The midday averaged Gs (10:00 to 14:00) reached 0.12 mol m−2 s−1 at Castelporziano, and 0.27 mol m− 2 s−1 at San Rossore. However, the midday Gs was 20% lower in summer and 30% in autumn than in spring at Castelporziano, and by 25% in summer and 50% in autumn than in spring at San Rossore. The model estimates without fSWC overestimated Gs for both Holm oak and Umbrella pine during summer and autumn, leading to an overestimation of canopy stomatal O3 flux (Table 4). By considering fSWC , the Jarvis-type model reproduced the seasonal trend of canopy stomatal O3 flux. 3.3. Application of the BWB model Fig. 3 shows the relationship of b to SWC in both forests (see Eq. 11), which is a determinant for f(H). b is closely related to m of the BWB model (i.e., b = ln[f(H)]/RH = m*{ln[f(H]/f(H)}). Thus the declines of b indicates a decrease of m. At the Holm oak forest, b was not dependent on SWC, indicating that m could be assumed to be constant even under soil drying. On the other hand, at the Umbrella pine forest, b decreased with decreasing SWC, implying that m may have declined with soil drying. The BWB model reproduced the diurnal trend of Gs in both forests similarly well as the Jarvis-type model (Fig. 2). At the Umbrella pine forest, the BWB model with change of f(H) depending on SWC tended to overestimate Gs in spring. However, during summer and autumn, the error of the model estimation was smaller when considering a change of f(H) depending on SWC than without. As a result, the model considering soil water availability better explained the variation of Gs (Table 3, +11%). 4. Discussion 4.1. Jarvis-type model performances We provide a new parameterization of the Jarvis-type model at canopy-level for two representative Mediterranean evergreen tree species (Table 2). Previously, Zhang et al. (2003) set a lower Gmax for evergreen needle forests compared to evergreen broadleaf forest based on a literature review when simulating dry deposition to forests. In contrast, we found a higher Gmax for the evergreen Umbrella pine forest than for the evergreen Holm oak forest (+52%). This potentially leads to the higher stomatal O3 flux in Umbrella pine than in Holm oak (Fig. 1). In the Holm oak forest at Castelporziano, a previous study showed that the single-leaf level maximum stomatal conductance for O3 (gmax ) was 0.043 mol(O3 ) m−2 s−1 in sunlit leaves averaged for a mixed Mediterranean forest, with individual contributions from Holm oak leaves of 0.058 mol(O3 ) m−2 s−1 (Fares et al., 2013b). By comparing Gmax with gmax , we obtained a Gmax /gmax ratio of ∼3.8 for Holm oak. Kelliher et al. (1995) suggested that a ratio of 3 exists according to a global database from a literature survey. This ratio, however, may vary with different weighting of the sunlit leaves in the canopy. In forest canopies, stomatal conductance of leaves is known to respond differently to a given stimulus due to different leaf age, species, and acclimation to the local environment (Baldocchi, 1989). Comparison of Gmax and gmax poses several questions on the best criteria to be adopted to estimate O3 damage to forests. In the current European standard (CLRTAP, 2015), single-leaf level stomatal O3 flux was used for the assessment of O3 -induced decline of carbon acquisition in forest trees. For global, landscape and single-tree level assessments, adopting stomatal O3 flux at canopy level, however,


217

Fig. 1. Daily course of total (black circle) and stomatal (gray circle) O3 flux values (mean ± SE) in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore). For the seasons, see Table 1.

should be recommended because: i) ozone is present at each layer of the canopy (Fares et al., 2014) and therefore its toxic effects occur throughout the canopy profile, ii) ozone damage depends on environmental conditions at the leaf scale, in particular light, and these conditions are non-linearly distributed within the canopy (Kitao et al., 2012; Watanabe et al., 2014), and iii) a growing number of experimental sites provide long-term measurements of Gmax . Such measurements often obey rigorous international standards (i.e. ICOS and FLUXNET networks), are non-destructive and are representative of mature ecosystems. In contrast, existing formulations

of gmax often rely on sunlit leaves from saplings usually growing in controlled conditions that differ from a natural environment. Although the highest O3 concentration was recorded in summer, Gs was reduced due to limited soil water availability, leading to limited stomatal O3 fluxes (Fig. 1). The greater reduction of Gs due to limitation of available soil water was observed in Umbrella pine relative to Holm oak (Fig. 2). The different stomatal sensitivity to drought between tree species has been recognized before (e.g., Castell and Terradas, 1995; Picon et al., 1996). This variability may be partly associated with the existence of drought adaptation strategies. Jensen et al. (1989) suggested two main groups of species

218


Table 3 Results of the correlation analysis between measured and modelled Gs using the Jarvis-type model (with or without soil water limitation) and the BWB model (with or without change of f(H) depending on soil drying) in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore). R2 denotes the determinant coefficient, and RMSE denotes root mean square error (n = 9334 at Castelporziano, and n = 3361 at San Rossore). Soil water content at 50 cm was used for the analyses because we did not find a different performance of the model between using the data of SWC at different depths. BWB model with changes of f(H) was not applied at Castelporziano because we did not find a clear relationship between f(H) and soil water content. Sites and seasons

Jarvis model

BWB model

with fSWC

without fSWC

no change of f(H)

changes of f(H) with soil drying

Castelporziano R2 RMSE (mol m−2 s−1 )

0.23 0.063

0.17 0.069

0.20 0.061

– –

San Rossore R2 RMSE (mol m−2 s−1 )

0.57 0.076

0.50 0.101

0.53 0.087

0.64 0.082

Table 4 Comparison of daily canopy stomatal ozone flux between measured and modelled values (mean ± SE) using the Jarvis-type model (with or without soil water limitation) and the BWB model (with or without change of f(H) depending on soil drying) in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore). Soil water content at 50 cm was used for the analyses. BWB model with changes of f(H) was not applied at Castelporziano because we did not find a clear relationship between f(H) and soil water content. Sites and seasons

Castelporziano Spring Summer Autumn Winter San Rossore Spring Summer Autumn Winter

Observation (␮mol m−2 )

Jarvis model (␮mol m−2 )

BWB model (␮mol m−2 )

with fSWC

without fSWC

no change of f(H)

changes of f(H) with soil drying

198 ± 7 110 ± 7 76 ± 4 79 ± 6

219 ± 12 114 ± 13 65 ± 4 45 ± 5

223 ± 12 183 ± 13 114 ± 6 49 ± 4

162 ± 7 109 ± 9 74 ± 4 60 ± 4

– – – –

560 ± 36 559 ± 24 476 ± 30 60 ± 9

528 ± 36 580 ± 33 443 ± 32 47 ± 6

536 ± 37 758 ± 36 836 ± 61 47 ± 6

391 ± 43 700 ± 47 638 ± 58 34 ± 10

534 ± 64 612 ± 46 414 ± 36 50 ± 15

i.e. i) drought avoiding species, which have high stomatal sensitivity to drought, and ii) drought tolerant species, which have lower stomatal sensitivity but structural and functional adaptive traits allow tolerating the reduced water availability. Such tolerance mechanisms may be associated with morphological/anatomical adjustments, leading to an increase in the apoplastic water fraction ˜ 2005; Serrano et al., 2005). In general, pines (Serrano and Penuelas, could be classified as drought avoiding species, which need to drastically reduce water loss by stomatal closure during drought (Castel and Terradas, 1995). On the other hand, Holm oak could be classified as a rather drought tolerant species (Fusaro et al., 2016), which maintains a relatively high stomatal conductance during waterstressed conditions (Tenhunen et al., 1987). As reported in previous studies in Mediterranean climate (Alonso et al., 2008; Büker et al., 2012; De Marco et al., 2016), the SWC function improved the performance of the Jarvis-type Gs model at both the Umbrella pine and Holm oak forests (Table 3). As Büker et al. (2012) pointed out, the SWC function is crucial for the O3 risk assessment in Mediterranean Europe. In the present study, 36–75% overestimation of canopylevel stomatal O3 flux was found during summer drought without considering the SWC function into the Jarvis-type model (Table 4). 4.2. BWB model performances The BWB model well explained the in-field measured Gs similarly to the Jarvis-type model (Table 3), although the BWB model reproduced the dynamics of Gs of Umbrella pine better than the Jarvis-type model, especially during morning hours (Fig. 2). Recent studies tried to include the impacts of drought stress in the BWB model using the modification of m (van Wijk et al., 2000; He et al., 2014). However, it is still under discussion whether the coefficient m in the BWB model decreases with soil drying ˜ or not (Sala and Tenhunen, 1996; Penuelas et al., 1998; Keenan

et al., 2009). Fares et al. (2013b) reported that BWB model with constant f(H) explained the observed Gs in a mixed Mediterranean forest during warm and cold days during late summer to autumn in 2011. In addition to their conclusion, our result indicates that f(H) parameters in Holm oak forest at Castelporziano did not change throughout the year (Fig. 3). No change of f(H) suggests that m is constant with variable soil water status in the BWB model. Xu and Baldocchi (2003) suggested that m is constant under limited available soil water as a direct effect of drought stress on photosynthetic capacity with a tight photosynthesis-stomatal conductance relationship. On the other hand, in the Umbrella pine forest, f(H) parameters varied with soil drying, suggesting that m declined with soil drying (Fig. 3). Tenhunen et al. (1990) similarly reported a decrease in m under soil drying in response to changes in stomatal patchiness or endogenous factors (root signals) and to the constant mesophyll carboxylation efficiency for Quercus coccifera. This change of m indicates that photosynthesis under limited available soil water is reduced by a decrease in CO2 concentration to the carboxylation sites through stomatal closure (Chaves et al., 2002). Umbrella pine appears to have a higher sensitivity of stomata to drought, i.e., greater reduction of Gs under drought stress as mentioned above (Fig. 2), relative to Holm oak. When pines suffer a drought stress, stomatal closure may therefore be a main factor to limit carbon assimilation (Schwanz et al., 1996). These different water use strategies between the two forests may explain the ˜ et al., 1998). contrasting results of f(H) with soil drying (Penuelas 5. Conclusions In conclusion, both Jarvis-type and BWB models were valid modeling tools to estimate stomatal O3 flux for O3 risk assessment at canopy-level in two Mediterranean forests, thanks to parameterization with field data. The Umbrella pine forest had a 72% higher


219

Fig. 2. Daily course of measured stomatal and estimated conductance (Gs ) (mean ± SE). Estimates usedthe Jarvis-type model (with or without considering soil water limitation fSWC ) and the BWB model (with or without change of f(H) depending on soil drying) in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore). BWB model with changes of f(H) was not applied at Castelporziano because we did not find a clear relationship between f(H) and soil water content. For the seasons, see Table 1.

Gmax than the Holm oak forest, leading to potentially higher stomatal O3 flux. The soil water function of the Jarvis-type model is crucial for O3 risk assessment also at the canopy level. Our results demonstrated that species-specific ecophysiological characteristics especially regarding drought avoidance and tolerance as suggested by Jensen et al. (1989) may closely relate to the sensitivity of m in the BWB model under water limited conditions. Finally, the presented parameters of stomatal conductance models at canopy level will contribute to the estimation of stomatal

O3 fluxes for O3 risk assessment in Mediterranean forests using an approach which diverges from single-leaf observation and moves towards a canopy-based estimation.

Acknowledgements We thank the financial support by the LIFE projects FO3REST (LIFE10 ENV/FR/208) and MOTTLES (LIFE15ENV/IT/000183). We also thank the staff of Regional Park of San Rossore and Castel-

220


Fig. 3. The dependency of b, which is a determinant for f(H), on the volumetric soil water content (m3 m−3 ) in a Holm oak forest (Castelporziano) and an Umbrella pine forest (San Rossore). b is closely related to m of the BWB model (i.e., b = ln[f(H)]/RH = m*{ln[f(H)]/f(H)}). Thus the declines of b indicates a decrease of m. The analysis was made by weekly value of f(H) and weekly averaged SWC data. In the case of no change of f(H) with soil drying, the averaged value of b for all data was applied. In the case of changing f(H) with soil drying, the standard response curve by Stewart (1988) was applied for the data in Umbrella pine forest. Parameter values of the regression line in Umbrella pine forest were: f = 0.12 m3 m−3 , w = 0.03 m3 m−3 . b = 2.85, when SWC > f ; b = 2.85 × [1 − 0.41*exp(1.96*)], when f > SWC > w ; b = 1.68, when SWC < w .

porziano for their constant and helpful support in carrying out experimental activities.

References Alonso, R., Elvira, S., Sanz, M.J., Gerosa, G., Emberson, L.D., Bermejo, B., Gimeno, B.S., 2008. Sensitivity analysis of a parameterization of the stomatal component of the DO3SE model for Quercus ilex to estimate ozone fluxes. Environ. Pollut. 155, 473–480. Büker, P., Morrissey, T., Briolat, A., Falk, R., Simpson, D., Tuovinen, J.-P., Alonso, R., Barth, S., Baumgarten, M., Grulke, N., Karlsson, P.E., King, J., Lagergren, F., ˜ J., Rhea, L., Schaub, M., Uddling, J., Matyssek, R., Nunn, A., Ogaya, R., Penuelas, Werner, W., Emberson, L.D., 2012. DO3 SE modeling of soil moisture to determine ozone flux to forest trees. Atmos. Chem. Phys. 12, 5537–5562. Baldocchi, D., Xu, L., 2005. Carbon exchange of deciduous broad-leaved forests in temperate and Mediterranean regions. In: Griffiths, H., Jarvis, P.G. (Eds.), The Carbon Balance of Forest Biomes. Taylor and Francis Group, Abingdon, New York, pp. 187–215. Baldocchi, D., 1989. Canopy-atmosphere water vapour exchange: can we scale from a leaf to a canopy? In: Estimation of Areal Evapotranspiration. IAHS Publication No.177, pp. 21–41. Ball, J.T., Woodrow, I.E., Berry, J.A., 1987. A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions. In: Iggens, J. (Ed.), Progress in Photosynthesis Research, vol. 4. Martinus-Nijhoff Publishers, Dordrecht, Netherlands, pp. 221–224. Braun, S., Schindler, C., Leuzinger, S., 2010. Use of sap flow measurements to validate stomatal functions for mature beech (Fagus sylvatica) in view of ozone uptake calculations. Environ. Pollut. 158, 2954–2963. CLRTAP, 2015. Mapping Critical Levels for Vegetation, Chapter III of Manual on methodologies and criteria for modelling and mapping critical loads and levels and air pollution effects, risks and trends. UNECE Convention on Long-range Transboundary Air Pollution (Accessed on 14 April 2016 on Web at) www.icpmapping.org. Castell, C., Terradas, J., 1995. Atmospheric vs soil water deficit as leaf conductance regulator in Mediterranean trees: a correlational approach. Orsis 10, 43–50.

Chaves, M.M., Pereira, J.S., Maroco, J., Rodrigues, M.L., Ricardo, C.P.P., Osorio, M.L., Carvalho, I., Faria, T., Pinheiro, C., 2002. How plants cope with water stress in the field. Photosynthesis and growth. Ann. Bot. 89, 907–916. Cieslik, S.A., 2004. Ozone uptake by various surface types: a comparison between dose and exposure. Atmos. Environ. 38, 2409–2420. Contran, N., Paoletti, E., 2007. Visible foliar injury and physiological responses to ozone in Italian provenances of Fraxinus excelsior and F. ornus. The Sci. World J. 7, 90–97. Damour, G., Simonneau, T., Cochard, H., Urban, L., 2010. An overview of models of stomatal conductance at the leaf level. Plant Cell Environ. 33, 1419–1438. De Marco, A., Sicard, P., Fares, S., Tuovinen, J.P., Anav, A., Paoletti, E., 2016. Assessing the role of soil water limitation in determining the Phytotoxic Ozone Dose (PODY ) thresholds. Atmos. Environ. 147, 88–97. Emberson, L.D., Büker, P., Ashmore, M.R., 2007. Assessing the risk caused by ground level ozone to European forest trees A case study in pine, beech and oak across different climate regions. Environ. Pollut. 147, 454–466. Ewers, B.E., Oren, R., 2000. Analyses of assumptions and errors in the calculation of stomatal conductance from sap flux measurements. Tree Physiol. 20, 579–589. Fares, S., Weber, R., Park, J.H., Gentner, D., Karlik, J., Goldstein, A.H., 2012. Ozone deposition to an orange orchard: partitioning between stomatal and nonstomatal sinks. Environ. Pollut. 169, 258–266. Fares, S., Vargas, R., Detto, M., Goldstein, A.H., Karlik, J., Paoletti, E., Vitale, M., 2013a. Tropospheric ozone reduces carbon assimilation in trees: estimates from analysis of continuous flux measurements. Glob. Chan. Biol. 19, 2427–2443, http://dx.doi.org/10.1111/gcb.12222. Fares, S., Matteucci, G., Scarascia Mugnozza, G., Morani, A., Calfapietra, C., Salvatori, E., Fusaro, L., Manes, F., Loreto, F., 2013b. Testing of models of stomatal ozone fluxes with field measurements in a mixed Mediterranean forest. Atmos. Environ. 67, 242–251. Fares, S., Savi, F., Muller, J., Matteucci, G., Paoletti, E., 2014. Simultaneous measurements of above and below canopy ozone fluxes help partitioning ozone deposition between its various sinks in a Mediterranean Oak Forest. Agric. For. Meteorol. 198–199, 181–191, http://dx.doi.org/10.1016/j.agrformet. 2014.08.014. Fischer, R., Lorenz, M., 2011. Forest condition in europe, 2011 technical report of ICP forests and FutMon. In: Work Report of the Institute for World Forestry 2011/1. ICP Forests, Hamburg (212 pp.). Foken, T., Wichura, B., 1996. Tools for quality assessment of surface-based flux measurements. Agric. For. Meteorol. 78, 83–105. Fusaro, L., Gerosa, G., Salvatori, E., Marzuoli, R., Monga, R., Kuzminsky, E., Fares, S., 2016. Early and late adjustments of the photosynthetic traits and stomatal density in Quercus ilex L. grown in an ozone-enriched environment. Plant Biol. 18, 13–21. González-Fernández, I., Bermejo, V., Elvira, S., de la Torre, D., González, A., Navarrete, L., Sanz, J., Calvete, H., García-Gómez, H., López, A., Serra, J., Lafarga, A., Armest o, A.P., Calvo, A., Alonso, R., 2013. Modelling ozone stomatal flux of wheat under Mediterranean conditions. Atmos. Environ. 67, 149–160. Hartmann, D.L., Klein Tank, A.M.G., Rusticucci, M., Alexander, L.V., Brönnimann, S., Charabi, Y., Dentener, F.J., Dlugokencky, E.J., Easterling, D.R., Kaplan, A., Soden, B.J., Thorne, P.W., Wild, M., Zhai, P.M., 2013. Observations: atmosphere and surface. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. He, L., Chen, J.M., Liu, J., Mo, G., Bélair, S., Zheng, T., Wang, R., Chen, B., Croft, H., Arain, M.A., Barr, A.G., 2014. Optimization of water uptake and photosynthetic parameters in an ecosystem model using tower flux data. Ecol. Model. 294, 94–104. Hoshika, Y., Paoletti, E., Omasa, K., 2012a. Parameterization of Zelkova serrata stomatal conductance model to estimate stomatal ozone uptake in Japan. Atmos. Environ. 55, 271–278. Hoshika, Y., Watanabe, M., Inada, N., Koike, T., 2012b. Modeling of stomatal ozone conductance for estimating ozone uptake of Fagus crenata under experimentally enhanced free-air ozone exposure. Water Air Soil Pollut. 223, 3893–3901. Jarvis, P.G., 1976. Interpretation of variations in leaf water potential and stomatal conductance found in canopies in field. Phil. Trans. Royal Soc. London B 273, 593–610. Jensen, C.R., Henson, L.E., Turner, N.C., 1989. Leaf gas exchange and water relations of lupins and wheat. II. Root and shoot water relations of lupin during drought-induced stomatal closure. Aust. J. Plant Physiol. 16, 415–428. Keenan, T., Garcia, R., Friend, A.D., Zaehle, S., Gracia, C., Sabate, S., 2009. Improved understanding of drought controls on seasonal variation in Mediterranean forest canopy CO2 and water fluxes through combined in situ measurements and ecosystem modelling. Biogeoscience 6, 1423–1444. Kelliher, F.M., Leuning, R., Raupach, M.R., Schulze, E.-D., 1995. Maximum conductances for evaporation from global vegetation types. Agric. For. Meteorol. 73, 1–16. Kitao, M., Winkler, J.B., Löw, M., Nunn, A.J., Kuptz, D., Häberle, K.-H., Reiter, I.M., Matyssek, R., 2012. How closely does stem growth of adult beech (Fagus sylvatica) relate to net carbon gain under experimentally enhanced ozone stress? Environ. Pollut. 166, 108–115. Knauer, J., Werner, C., Zaehle, S., 2015. Evaluating stomatal models and their atmospheric drought response in a land surface scheme: a multibiome analysis. J. Geophys. Res. Biogeosci. 120, 1894–1911.

Y. Hoshika et al. / Agricultural and Forest Meteorology 234 (2017) 212–221 Lasslop, G., Reichstein, M., Papale, D., Richardson, A.D., Arneth, A., Barr, A., Stoy, P., Wohlfahrt, G., 2010. Separation of net ecosystem exchange into assimilation and respiration using a light response curve approach: critical issues and global evaluation. Glob. Chan. Biol. 16, 187–208. Leisner, C.P., Ainsworth, E.A., 2012. Quantifying the effects of ozone on plant reproductive growth and development. Glob. Chan. Biol. 18, 606–616. Matteucci, M., Gruening, C., Goded Ballarin, I., Seufert, G., Cescatti, A., 2015. Components, drivers and temporal dynamics of ecosystem respiration in a Mediterranean pine forest. Soil Biol. Biochem. 88, 224–235. Matyssek, R., Clarke, N., Cudlin, P., Mikkelsen, T.N., Tuovinen, J.-P., Wieser, G., Paoletti, E., 2013. Climate change, air pollution and global challenges: understanding and perspectives from forest research. Developments in Environmental Science, vol. 13. Elsevier, Amsterdam, pp. 622. Mikkelsen, T.N., Ro-Poulsen, H., Hovmand, M.F., Jensen, N.O., Pilegaard, K., Egeløv, A.H., 2004. Five-year measurements of ozone fluxes to a Danish Norway spruce canopy. Atmos. Environ. 38, 2361–2371. Mills, G., Pleijel, H., Braun, S., Büker, P., Bermejo, V., Calvo, E., Danielsson, H., Emberson, L., Grünhage, L., Harmens, H., Hayes, F., Karlsson, P.E., Simpson, D., 2011. New stomatal flux-based critical levels for ozone effects on vegetation. Atmos. Environ. 45, 5064–5068. Mintz, Y., Walker, G.K., 1993. Global fields of soil moisture and land surface evapotranspiration derived from observed precipitation and surface air temperature. J. Appl. Meteorol. 32, 1305–1334. Monteith, J.L., Unsworth, M.H., 1990. Principles of Environmental Physics. Edward Arnold, London, pp. 291. Omasa, K., Saji, H., Youssefian, S., Kondo, K., 2002. Air Pollution and Plant Biotechnology. Springer-Verlag, Tokyo (455pp.). Paoletti, E., 2007. Ozone impacts on forests. CAB Rev.: Perspect. Agric. Vet. Sci. Nutr, Nat. Resour. 2 (68) (13pp.). Paoletti, E., 2006. mpact of ozone on Mediterranean forests: A review. Environ. Pollut. 144, 463–474. ˜ ˜ Penuelas, J., Filella, I., Llusiá, J., Siscart, D., Pinol, J., 1998. Comparative field study of spring and summer leaf gas exchange and photobiology of the mediterranean trees Quercus ilex and Phillyrea latifolia. J. Exp. Bot. 49, 229–238. Picon, C., Guehl, J.M., Ferhi, A., 1996. Leaf gas exchange and carbon isotope composition responses to drought in a drought-avoiding (Pinus Pinaster) and a drought-tolerant (Quercus petraea) species under present and elevated atmospheric CO2 concentrations. Plant Cell Environ. 19, 182–190. Pinzari, F., Trinchera, A., Benedetti, A., Sequi, P., 1999. Use of biochemical indices in the mediterranean environment: comparison among soils under different forest vegetation. J. Microbiol. Methods 36, 21–28. Pleijel, H., Danielsson, H., Vandermeiren, K., Blum, C., Colls, J., Ojanperä, K., 2002. Stomatal conductance and ozone exposure in relation to potato tuber yield −results from the European CHIP programme. Eur. J. Agronom. 17, 303–317. Reichstein, M., Falge, E., Baldocchi, D., Papale, D., Aubinet, M., Berbigier, P., Bernhofer, C., Buchmann, N., Gilmanov, T., Granier, A., Grünwald, T., Havrankova, K., Ilvesniemi, H., Janous, D., Knohl, A., Laurila, T., Lohila, A., Loustau, D., Matteucci, G., Meyers, T., Miglietta, F., Ourcival, J.-M., Pumpanen, J., Rambal, S., Rotenberg, E., Sanz, M., Tenhunen, J., Seufert, G., Vaccari, F., Vesala, T., Yakir, D., Valentini, R., 2005. On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Glob. Chan. Biol. 11, 1424–1439. Reynolds, J.F., Stafford Smith, D.M., Lambin, E.F., Turner II, B.L., Mortimer, M., Batterbury, S.P., Downing, T.E., Dowlatabadi, H., Fernandéz, R.J., Herrick, J.E., Huber-Sannwald, E., Jiang, H., Leemans, R., Lynam, T., Maestre, F.T., Ayarza, M., Walker, B., 2007. Global desertification: building a science for dryland development. Science 316, 847–851. Rissmann, J., Tetzlaff, G., 1994. Application of a spectral correction method for measurements of covariances with fast-response sensors in the atmospheric boundary layer up to a height of 130 m and testing of the corrections. Boundary-Layer Meteorol. 70, 293–305.

221

Sala, A., Tenhunen, J.D., 1996. Simulations of canopy net photosynthesis and transpiration in Quercus ilex L. under the influence of seasonal drought. Agric. For. Meteorol. 78, 203–222. Schwanz, P., Picon, C., Vivin, P., Dreyer, E., Guehl, J.M., Polle, A., 1996. Responses of antioxidative systems to drought stress in pendunculate oak and maritime pine as modulated by elevated CO2 . Plant Physiol. 110, 393–402. ˜ Serrano, L., Penuelas, J., 2005. Contribution of physiological and morphological adjustments to drought resistance in two Mediterranean tree species. Biol. Plant. 49, 551–559. ˜ Serrano, L., Penuelas, J., Ogaya, R., Savé, R., 2005. Tissue-water relations of two co-occurring evergreen Mediterranean species in response to seasonal and experimental drought conditions. J. Plant Res. 118, 263–269. Sicard, P., Dalstein-Richier, L., 2015. Health and vitality assessment of two common pine species in the context of climate change in Southern Europe. Environ. Res. 137, 235–245. Sicard, P., De Marco, A., Troussier, F., Renou, C., Vas, N., Paoletti, E., 2013. Decrease in surface ozone concentrations at Mediterranean remote sites and increase in the cities. Atmos. Environ. 79, 705–715. Sicard, P., De Marco, A., Dalstein-Richier, L., Tagliaferro, F., Paoletti, E., 2016. An epidemiological assessment of stomatal ozone flux-based critical levels for visible ozone injury in Southern European forests. Sci. Tot. Environ. 541, 729–741. Stewart, J.B., 1988. Modelling surface conductance of pine forest. Agric. For. Meteorol. 43, 19–35. Tenhunen, J.D., Pearcy, R.W., Lange, O.L., 1987. Diurnal variations in leaf conductance and gas exchange in natural environments. In: Tenhunen, J.D., Catarino, F.M., Lange, O.L., Oechel, W.C. (Eds.), Plant Response to Stress: Functional Analysis in Mediterranean Ecosystems. Springer-Verlag, Berlin, pp. 305–327. Tenhunen, J.D., Sala Serra, A., Harley, P.C., Dougherty, R.L., Reynolds, J.F., 1990. Factors influencing carbon fixation and water use by Mediterranean sclerophyll shrubs during summer drought. Oecologia 82, 381–393. Tiwari, S., Grote, R., Churkina, G., Butler, T., 2016. Ozone damage, detoxification and the role of isoprenoids − new impetus for integrated models. Funct. Plant Biol. 43, 324–326. Wang, S., Yang, Y., Trishchenko, A.P., 2009. Modeling the response of canopy stomatal conductance to humidity. J. Hydrometeorol. 10, 521–532. Watanabe, M., Hoshika, Y., Inada, N., Koike, T., 2014. Canopy carbon budget of Siebold’s beech (Fagus crenata) sapling under free air ozone exposure. Environ. Pollut. 184, 682–689. Wever, L.A., Flanagan, L.B., Carlson, P.J., 2002. Seasonal and interannual variation in evapotranspiration: energy balance and surface conductance in a northern temperate grassland. Agric. For. Meteorol. 112, 31–49. Wieser, G., Matyssek, R., Then, C., Cieslik, S., Paoletti, E., Ceulemans, R., 2008. Upscaling ozone flux in forests from leaf to landscape. Ital. J. Agron. 1, 35–41. Wilczak, J.M., Oncley, S.P., Stage, S.A., 2001. Sonic anemometer tilt correction algorithms. Boundary-Layer Meteorol. 99, 127–150. Xu, L., Baldocchi, D.D., 2003. Seasonal trends in photosynthetic parameters and stomatal conductance of blue oak (Quercus douglasii) under prolonged summer drought and high temperature. Tree Physiol. 23, 865–877. Zhang, L., Brook, J.R., Vet, R., 2003. A revised parameterization for gaseous dry deposition in air quality models. Atmos. Chem. Phys. 3, 2067–2082. van Wijk, M.T., Dekker, S.C., Bouten, W., Bosveld, F.C., Kohsiek, W., Kramer, K., Mohren, G.M.J., 2000. Modeling daily gas exchange of a Douglas-fir forest: comparison of three stomatal conductance models with and without a soil water stress function. Tree Physiol. 20, 115–122. van Wijk, M.T., Bouten, W., Verstraten, J.M., 2002. Comparison of different modelling strategies for simulating gas exchange of a Douglas-fir forest. Ecol. Model. 158, 63–81.