Ozone uptake by adult urban trees based on sap flow measurement

Environmental Pollution 162 (2012) 275e286

Contents lists available at SciVerse ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Ozone uptake by adult urban trees based on sap flow measurement Hua Wang a, b, Weiqi Zhou a, c, Xiaoke Wang a, Fuyuan Gao d, Hua Zheng a, Lei Tong a, Zhiyun Ouyang a, * a

State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China Institute of Forestry and Pomology, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100093, China c Department of Plant Sciences, University of California, Davis, CA 95616, USA d Beijing Teaching Botanical Garden, Beijing 100061, China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 March 2011 Received in revised form 11 November 2011 Accepted 17 November 2011

The O3 uptake in 17 adult trees of six urban species was evaluated by the sap flow-based approach under free atmospheric conditions. The results showed very large species differences in ground area scaled whole-tree ozone uptake (FO3 ), with estimates ranging from 0.61
0.07 nmol m2 s1 in Robinia pseudoacacia to 4.80
1.04 nmol m2 s1 in Magnolia liliiflora. However, average FO3 by deciduous foliages was not significantly higher than that by evergreen ones (3.13 vs 2.21 nmol m2 s1, p ¼ 0.160). Species of high canopy conductance for O3 (GO3 ) took up more O3 than those of low GO3 , but that their sensitivity to vapour pressure deficit (D) were also higher, and their FO3 decreased faster with increasing D, regardless of species. The responses of FO3 to D and total radiation led to the relative high flux of O3 uptake, indicating high ozone risk for urban tree species. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Urban trees Thermal dissipation method Canopy conductance Ozone Ozone flux Critical levels

1. Introduction The accelerating global urbanization caused severe air pollution (Wu, 2008). In Beijing, the air pollution problem has been exacerbated due to the wide-spread construction activities and increased number of vehicles (Hao et al., 2005; Chan and Yao, 2008). Air pollution affects human health, damages vegetation and various anthropogenic materials, and reduces visibility (McPherson et al., 1994). Trees in cities can improve the urban air quality by removing gaseous air pollution (Nowak and Dwyer, 2007). Therefore, great effort is made to increase and maintain urban trees. Tropospheric ozone (O3; a list of all abbreviations is given in Table 1) is a potential risk factor in trees (Matyssek and Sandermann, 2003). Its negative effects include morphological and histological injuries, reduced photosynthesis, and altered water balance, among others (Landolt and Keller, 1985). Both the exposure-based approach and the flux-based approach are currently used to evaluate the risk of ozone damage to vegetation (Simpson et al., 2007). Since ozone injuries are more closely related to the ozone absorbed through the stomata than to ozone concentration in the atmosphere (Landolt and Keller, 1985; Musselman and Massman, 1998), the flux-based

* Corresponding author. E-mail address: [email protected] (Z. Ouyang). 0269-7491/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2011.11.026

approach is better suited for the establishment of reliable critical levels as risk assessment tools for vegetation (Ferretti et al., 2007). The flux-based indices, AFsty (accumulated stomatal flux over threshold of y nmol m2 s1) (UNECE, 2004), can be used to serve this purpose. O3 uptake by plants depends on both the ambient O3 concentration and the stomatal conductance (Wieser and Havranek, 1995; Nowak and Dwyer, 2007). At a given ambient O3 concentration, the rate of O3 uptake by leaves was effectively controlled by stomatal conductance, which is influenced by factors such as leaf characteristics, crown position, tree age, tree height, climate, and altitude (Beckett et al., 2000; Schafer et al., 2000; Wieser et al., 2000; Nunn et al., 2007). In urban settings, higher temperature (Landsberg, 1981), lower plant density (Hagishima et al., 2007), irrigation (Martin and Stabler, 2002), energy balance properties of urban surfaces (Montague and Kjelgren, 2004), and night illumination (Longcore and Rich, 2004) may allow for higher stomata aperture, hence, more O3 uptake. Considerable researches have been conducted to investigate ozone uptake by natural forests (Wieser et al., 2003, 2006; Nunn et al., 2007; Köstner et al., 2008; Braun et al., 2010). Fewer studies, however, have examined ozone uptake by urban trees. O3 uptake by urban trees has been largely examined based on modeling approaches that integrate vegetation information with meteorological and ambient O3 concentration data (McPherson

276

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

sap flow measurements, since transpiration and O3 uptake are coupled through the activity of the stomata (Wieser et al., 2003). This method takes into account the presence of real boundary layer, and was previously used in heterogeneous and mountainous landscapes (Wieser et al., 2003, 2006; Matyssek et al., 2007, 2008). Therefore, we chose this method to evaluate O3 uptake by urban trees. The objectives of this study are: (1) to quantify and compare whole-tree ozone uptake by different urban species; (2) to examine how O3 flux of urban trees is regulated by the stomata and environmental conditions; and (3) to provide accurate quantification of O3 risk assessment with species-specific flux data.

Table 1 Abbreviations used in the text. Abbreviation

Definition (Units)

AS Ai DBH Ac LAI JO Ji JS ET EC

sapwood area (cm2) sapwood area at depth i (cm2) diameter at the breast height (cm) canopy projected area (m2) leaf area index sap flux density in the outer 2 cm of sapwood (g cm2 d1) sap flux density at depth i (g cm2 d1) sap flux density across the active sapwood (g cm2 d1) tree transpiration (kg d1) canopy transpiration (mm d1) error estimate at depth i error estimate in the outer 2 cm of sapwood error estimate of the Ji/JO ratio canopy conductance for O3 (mmol m2 s1) canopy O3 uptake rate (nmol m2 s1) accumulated exposure (O3 accumulated over a threshold of 40 ppb) accumulated stomatal ozone flux over thresholds of 0,1.6 nmol m2 s1 (mmol m2, projected area) estimated during the growing season ozone concentration of the ambient air (ppb) vapor pressure deficit (kPa) total radiation (W m2) air temperature ( C) soil water content (%) density of water (998 kg m3) universal gas constant adjusted for water vapor (0.462 m3 kPa K1kg1)

si so sR

GO3 FO3 AOT40 AFst0,1.6

[O3] D RS Ta SWC

r GV

2. Method and materials 2.1. Study site and trees The study was carried out at the 116,500 m2 Beijing Teaching Botanical Garden (116 250 3700 to 116 250 5000 E, 39 520 2000 to 39 520 2800 N), located close to the urban core of the Beijing city, China. Annual average precipitation from 1978 to 2009 in Beijing is about 585 mm, with more than 70% occurring from June to August (Beijing Water Authority, 2010). Annual average temperature from 1978 to 2009 in Beijing is approximately 11e12  C (State Statistical Bureau, 2011). The experimental site is a typical green space embedded in densely populated commercial and residential areas, with heavy pedestrian and motor vehicle traffic. Four deciduous tree species (Ginkgo biloba, Aesculus chinensis, Magnolia liliiflora, Robinia pseudoacacia) and two evergreen tree species (Pinus tabulaeformis, Cedrus deodara) were selected in this research. In terms of number of trees in Beijing, G. biloba and R. pseudoacacia are among the top five deciduous tree species, while P. tabulaeformis and C. deodara are among the top five evergreen tree species (Beijing Gardening and Greening Bureau, 2005). A. chinensis is becoming more and more popular in urban settings with prospects of aesthetic enhancement and economic values (Wei et al., 2008). The trees selected for certain type of species had similar diameter at the breast height (DBH), height, projected canopy area (Ac), and sapwood cross-sectional area (As) (Wang et al., 2011). The size of the selected trees in terms of DBH was determined based on the field surveys conducted in Beijing in 2007. Comprehensive species field surveys were conducted in Beijing in July and August, 2003, and in August, 2007, covering an area of 670 km2 (Zhao, 2010). The distribution of DBH for urban trees in Beijing in 2003 is shown in Fig.4. We chose the DBH size of the trees by considering the increase in DBH during the past five years. We selected two or three individual trees of each species for sap flux measurement. Sap flux measurement in urban environment was challenging due to relatively long distance among trees, insufficient sample size, species

et al., 1994; Nowak et al., 2000). These studies primarily focused on O3 removal by urban trees (e.g., Nowak et al., 2000; Yang et al., 2005). While the modeling approach can provide important information on O3 removal by urban trees at the city scale, variations in canopy resistance due to trees with different height and crown configurations, and heterogeneous ambient O3 concentrations may introduce uncertainties into the estimates (Scott et al., 1998). Alternatively, O3 uptake by urban trees can be measured based on the whole-tree ozone flux into the leaves that can be estimated by

A

B

C

D

E

F

G

H

DBH Fig. 1. The relationship between sapwood depth, sapwood area and diameter at the breast height (DBH) for G.biloba, P. tabulaeformis, R. pseudoacacia, and C. deodara.

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

277

P Fig. 2. Daily sap flux density ( JO) for each tree in the selected species from November 1, 2008 to October 31, 2009. Days of missing data was filled by multiple regression or curve estimation considering the change of leaf area and soil water content (grey colour).

diversity, suitable sites, convenient power, and granted permission (Pataki et al., 2011). Hereby, it was not feasible to study many trees of each species over a long time period. Moreover, a careful literature survey on previous studies shows that 1 to 8 individuals are adequate to indicate transpiration characteristics and their response to environmental variables of homegarden tree species, irrigated grapevines, natural rain forest, wet tropical forest, deciduous forest etc. in diverse environments (Granier et al., 1996; Pataki and Oren, 2003; Lu et al., 2003; O’Brien et al., 2004; Burgess, 2006; Costa et al., 2006; Fernandez et al., 2009). 2.2. Environmental measurements An automated weather station was very close to the trees selected for the study. The station was located at an open area, with minimized influences from nearby

trees, buildings, and other obstacles. Environmental parameters, including air temperature, air relative humidity, wind speed, total radiation, and rainfall, were monitored by the meteorological station (HMP45C temperature/humidity sensors, Vaisala, Inc., Helsinki, Finland; 034B wind sensor, Met One Instruments, Grants Pass, Oregon, USA; CMP-11 pyranometer, Kipp and Zonen, Delft, Netherlands; TE525MM tipping bucket rain gauge, Campbell Scientific, Inc., Logan, UT). Vapour pressure deficit (D) was calculated from air temperature and relative humidity data (Campbell and Norman, 1998). Three soil water probes (ECH2O, Decagon Devices, Inc., Pullman, WA, USA) were placed at the depth of 30 cm among the sampled trees to monitor soil water content. These environmental parameters were continuously monitored and recorded every 10 min. Ambient ozone (O3) concentrations were monitored around the selected species and recorded every hour using Model 49i gas analyzer (Thermo Fisher Scientific Inc., Massachusetts, USA). The gas analyzer was calibrated every week.

278

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

Fig. 3. (AeC) Diurnal variation in sap flux density (JO) at four aspects in P.tabulaeformis trees on 24th, June, 2009, C. deodara trees on 14th, August, 2009, and R. pseudoacacia trees on P 15th, August, 2009; (DeF) Linear relationships between JO on north aspect (x-axis) and JO on south, east, west aspects (y-axis); (GeI) Daily means of sap flux densities ( JO) in azimuth categories from November 1, 2008 to October 31, 2009. Vertical lines represent
SE.

2.3. Sap flow measurements and scaled transpiration Instantaneous sap flux density, flow per unit sapwood cross-sectional area (JO, g H2O cm2 s1), is estimated by thermal dissipation probes (Dynamax, Houston, TX, USA) since April, 2008, which continuously measure the temperature difference between heated and unheated probes (DT) (Granier, 1987). The DT data were recorded as 10 min means (CR1000 data logger, Campbell Scientific Ltd., UK). Errors may be caused by uncertainties in the baseline position (Oishi et al., 2008). During conditions of zero sap flux density, such as during prolonged rain events or after several night time hours, the temperature difference between upper and lower thermocouples represents the steady state temperature difference due solely to the dissipation of heat into non-transporting sapwood (Phillips et al., 1997). We selected the highest temperature difference between heated and unheated probes during times of zero flux (DTM) as a baseline day by day, from which any sap flow causes a decrease in DTM (Granier, 1987; Phillips et al., 1997). Deviation from the baseline is used to estimate JO (Granier, 1987). Both axial and radial variations in sap flux density across the active sapwood (Js, g H2O cm-2 d-1) should be considered to reduce the errors that may occur when scaling-up sap flux density measurements to the whole-tree or stand scale (Oren et al., 1998; Oren et al., 1999a; Lu et al., 2000; Delzon et al., 2004; Ford et al., 2004; Tateishi et al., 2008). It is recommended to measure the axial variations in Js with 1 or 2 probes for small trees, and 4 probes for large trees (Herbst et al., 2007). Therefore, in this study, one 30 mm long probe (TDP 30) was deployed at the north aspect of each G. biloba, A. chinensis and M. liliiflora tree. Two TDP 30 probes were deployed on each P. tabulaeformis tree, with 1 at the north aspect and 1 at the south aspect in the trunk. Four TDP 30 probes were deployed on each C. deodara and R. pseudoacacia tree, with 1 at each of the

four aspects (i.e., east, south, west, and north). Based on the continuous measurements of axial variation, we calculated the mean values of sap flux density at the east, south, west, and north aspect in the trunk to represent the averaged sap flux in the outmost sapwood, which were used in our analyses. Direct measurements at multiple sapwood depths are required to calculate radial variations in Js (Lu et al., 2000; Delzon et al., 2004; Ford et al., 2004; Tateishi et al., 2008). In our study, for P. tabulaeformis No.3 and C. deodara No.3, radial variations in Js were measured by 80 mm long probes with thermocouples mounted at the 15 mm, 70 mm. For C. deodara No. 1 and No. 2, the radial variations in Js were measured by 100 mm long probes with thermocouples mounted at the 15 mm, 50 mm, and 90 mm (Dynamax, Houston, TX, USA). However, these were not feasible in each studied tree due to the following reasons: 1) it was too expensive. We had 17 selected trees, and three sets of sap flow measurement systems due to the low plant density of selected species; 2) we used three 100 Ahr batteries, rather than alternating current for safety reasons, which limited the probes increasement; and 3) the local government has strict regulations on trees in urban areas, which confined our experiments on trees. Therefore, we used a simplified function for scaling radially detailed in Pataki et al. (2011). Angiosperm

    x  0:09963 2 Ji =JO ¼ 1:033  exp  0:5 0:4263

(1)

gymnosperm

    x þ 0:3724 2 Ji =JO ¼ 1:257  exp  0:5 0:6620

(2)

where Ji/JO is the ratio of sap flux at the actual to the outermost sapwood depth, x is the relative sapwood depth based on previously published values.

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

A

B

DBH

279

C

DBH

D

DBH

F

E

DBH

DBH

DBH

Fig. 4. DBH distribution of some urban tree species in Beijing in 2003.

The whole-tree transpiration was estimated according to ET ¼

n X i¼1

Ji Ai 1000

2.5. Estimation of canopy conductance (3)

where ET is the total daily tree transpiration (kg d1), n is the number of 2 cm increments in sapwood depth, and Ji and Ai are the sap flux density (g cm2 d1) and sapwood area (cm2), respectively, at depth i. To estimate the error at each 2-cm-sapwood increment (si), we used the following equation (Pataki et al., 2011):

si ¼ Ji

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi   2  sO sR 2 þ JO Ji : JO

(4)

where JO is sap flux in the outer 2 cm of xylem, so is the standard deviation in JO, and sR is the error of the radial trends regression estimate, 0.2583 for angiosperms and 0.1714 for gymnosperms (Pataki et al., 2011). The error at each sapwood increment was then propagated to obtain the total error in estimating whole-tree transpiration (Pataki et al., 2011). 2.4. Sapwood depth and sapwood area A destructive harvest is required to determine the exact proportion of a particular sensor’s length that extends into non-conductive sapwood (Oishi et al., 2008). With other ongoing studies (long-term water use by urban tree species) at the site, we did not core trees at the end of the threeeyear round experiment. Alternatively, we determined the relationship between tree diameter and sapwood area for each study species at the same location and applied that relationship to our study trees (Fig.1). Trees of various diameter at the breast height (DBH) rank in G. biloba (n ¼ 33), R. pseudoacacia (n ¼ 23), P. tabulaeformis (n ¼ 22), and C. deodara (n ¼ 15) were cored at 1.30 m from the ground at the Beijing Teaching Botanical Garden right before we started the sap flux measurement. Sapwood depth was visually estimated. As there are only a few trees for A. chinensis and M. liliiflora in our study site, its sapwood depth was determined by removing cores from the nearby trees with DBH difference less than 2 cm.

Urban trees generally have open, well-ventilated, aerodynamically rough canopies. Therefore, a strong coupling of canopy surface to the atmosphere can be assumed (Jarvis and Mcnaughton, 1986; Schafer et al., 2000). We validated this assumption for our selected deciduous tree species. Using a gas exchange system (LI-6400, Li-cor, USA), we measured the stomatal conductance (GS) of 9 healthy mature leaves from 3 branches in the outer middle canopy of each tree at noon (12:00e13:00) for three days per month from May to October. For comparison, averaged stomatal conductance at the leaf level was obtained by relating GC around 12:00 from May to October to the tree’s foliage area. Finally, the average calculated stomatal conductance values were compared to values obtained from gas exchange measurements for the same tree during the same time period. The difference between these two averages wasn’t significant for both G. biloba and A. chinensis (Table 4; p ¼ 0.513 for G. biloba; p ¼ 0.792 for A. chinensis). These results indicated that the boundary layer resistance was low under urban tree canopy, and the canopy surface is aerodynamically well coupled to the atmosphere. Therefore, the whole-tree canopy conductance (GC, mmol H2O m2 s1) of the selected trees can be calculated based on a simplified equation (Köstner et al., 1992): Gc ¼ ðEC rGV Ta Þ=D

(5)

EC is the canopy transpiration obtained by dividing ET by projected crown area (Granier, 1987); r is the density of water (998 kg m3); GV is the universal gas constant adjusted for water vapor (0.462 m3 kPa K1 kg1); Ta is the air temperature ( C); and D is vapour pressure deficit (kPa). We selected a subset of the data to calculate GC based on the following rules: 1) data collected during rainfall and 2 h after, were excluded to avoid the discrepancy between evaporation and tree transpiration (Granier et al., 2000); 2) data were excluded when global radiation, or vapour pressure deficit, or stand transpiration were too low (<5% of the maximum value), because of the large relative uncertainties in computing GC under these conditions (Granier et al., 2000); 3) data were excluded when D < 0.6 kPa for GC estimation, to keep errors in GC estimates to less than 10% (Ewers and Oren, 2000); and 4) only data collected during the leaf-on periods were included for the selected deciduous species. Days with leaf-on after full expansion for G. biloba, A. chinensis, M. liliiflora, R. pseudoacacia were DOY115-305, DOY 118-306, DOY 118-306, DOY

280

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

Table 2 Multiple regression or curve estimation between daily sap flux density and environmental factors among the studied trees. Species

Method

G. biloba 1 G. biloba 2 A. chinensis 1 A. chinensis 2 A. chinensis 3 M. liliiflora 1 M. liliiflora 2 M. liliiflora 3 R. pseudoacacia 1 R. pseudoacacia 2 R. pseudoacacia 3 P. tabulaeformis 1 P. tabulaeformis 2 P. tabulaeformis 3 C. deodara 1 C. deodara 2 C. deodara 3

Period

Curve estimation Curve estimation Curve estimation Curve estimation Curve estimation Curve estimation Curve estimation Curve estimation Multiple regression Multiple regression Multiple regression Multiple regression Multiple regression Multiple regression Curve estimation Curve estimation Curve estimation

MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. MayeOct. One year One year One year

118-306, respectively. The critical level for trees is defined for a six month growing season (UNECE, 2004). Hence, only data collected during DOY 122-305 (from May 1 to October 31) were used for the selected species for comparisons with critical level. Unlike the intermittent measurements, the continuous sap flow and environmental recordings over a rather small time scale are more reliable and less susceptible to measurement errors. Therefore, the variations of ozone uptake based on this method would be representative on how the environmental variables influence the ozone uptake by urban trees. Response functions of GC to vapour pressure deficit (D) and total radiation (RS) were derived from boundary line analysis (Webb, 1972; Jarvis, 1976; Dye and Olbrich, 1993; Rico et al., 1996; Martin et al., 1997), using the maximum of GC of 2e3 trees per species for different D classes (step width: 0.2 kPa) and RS classes (step width: 50 W m2). How GC responds to changes in D can be described by the following function (Oren et al., 1999b): GC ¼ mlnðDÞ þ b

(6)

where m and b were estimated using least squares regression with Sigma-Plot 10.0 (Systat Software Inc., San Jose, California). The relationship between GC and ln(D) provides convenient benchmarks for comparisons among species. The parameter em, the slope of the regression, quantifies stomatal sensitivity to D. The parameter b, the intercept, is the reference conductance (Gref) at D ¼ 1 kPa. 2.6. Calculation of O3 uptake The canopy O3 uptake was calculated according to the flux equation below, by assuming zero O3 concentration at the mesophyll surface (Laisk et al., 1989): FO3 ¼ ½O3 GC 0:613

(7)

where FO3 is the whole-tree canopy O3 uptake rate scaled to ground surface area, [O3] is the ozone concentration of the ambient air (Wieser et al., 2000), GC is the canopy conductance to water vapour, and 0.613 is the conversion factor representing the ratio of the diffusion coefficients of ozone to water vapour in air (c.f. Wieser and Havranek, 1995). The concentration-based critical levels (CLec) and flux-based critical levels (Clef) are typically used to indicate the risk of ozone damage to vegetation (Simpson et al., 2007). The CLe for crops, natural vegetation, and forest trees has been derived from a linear relationship between the accumulated exposure (O3 accumulated over a threshold of x ppb, AOTx) or the accumulated flux (accumulated O3 flux above a threshold y, AFsty) and relative yield or percentage reduction in biomass (UNECE, Table 3 Sap flow in the outer 2 cm of sapwood (JO) and whole-tree water use across total sapwood depth, expressed both as total sap flux per unit sapwood area (JS) and as water use per tree (ET). Tree species

JO (g cm2 d1)

G. biloba A. chinensis M. liliiflora R. pseudoacacia P. tabulaeformis C. deodara

124.84 139.11 97.27 86.85 141.30 111.32









9.03 21.12 12.48 13.83 20.61 22.92

JS (g cm2 d1) 85.35 104.58 78.88 86.85 95.09 69.98









33.36 79.23 33.67 23.96 58.21 58.19

ET (kg d1) 16.09 23.49 4.56 14.04 21.09 35.09









R2

Equation

4.15 7.06 1.42 4.76 6.77 14.44

Notes: Values
SE are given for JO, while table entries for JS and ET are values
the error term derived from Eq. (4) in the text, accounting for both uncertainty in JS and in the radial trend of sap flux.

JO JO JO JO JO JO JO JO JO JO JO JO JO JO JO JO JO

¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼

0.8423

147.4 D 150.92 D0.6785 205.66 D0.9248 112.91 D1.2214 149.77 D0.9225 103.39 D0.9038 119.51 D1.0657 56.649 D1.2512 22.089 þ 0.257PAR þ 5.677SWC  22.952D 27.386 þ 0.423PAR þ 8.091SWC 108.074 þ 0.309PAR  33.036D 81.684 þ 0.410PAR þ 7.577SWC 40.383 þ 0.265PAR þ 3.249SWC þ 15.926D 27.044 þ 0.494PAR þ 5.686SWC  18.533D 90.266 D1.2494 67.703 D1.3324 120.04 D1.3237

0.5781 0.4104 0.6125 0.7382 0.6691 0.617 0.661 0.6125 0.641 0.629 0.364 0.640 0.741 0.665 0.6207 0.6133 0.534

2004). New critical levels (AOT40 5 ppm h and AFst1.6 4 mmol m2) for ozone effects on trees have been proposed recently (UNECE, 2004). In this study, we used these two new critical levels to evaluate the potential risk of ozone damage on urban trees. 2.7. Data processing 2.7.1. Gap filling Missing data is a common problem when conducting sap flow studies over long periods of time. In order to estimate annual water loss and compare species differences for the studied species, it was necessary to devise gap filling procedures for sap flux during the study period. Previous studies have used a multiple regression approach by regressing daily sap flux against daily PAR and D (Wullschleger and Norby, 2001), or a trigonometric regression approach that accounts for independence in climate data (Hubbard et al., 2010). In our study, due to power-off or equipment maintenance, there were 20, 23, 23, 144, 60, 73 missing days during the one year period (365 days), and 13, 13, 14, 74, 0, 12 missing days during the growing period (183 days) for G. biloba, A. chinensis, M. liliiflora, R. pseudoacacia, P. tabulaeformis, and C. deodara. As for daily JO, we estimated daily JO for missing days by multiple regression or curve estimation to measured sap flux densities and climate data collected on site. Photosynthetically active radiation (PAR), vapor pressure deficit (D), and soil water content (SWC) were the best predictors of daily JO, accounting for approximately 60% of the observed variability (Table 2). Canopy leaf area index was close to maximum during this time. As for hourly canopy conductance (GC), a filter was employed to studied species to make nearly all individuals have the same gaps, except R. pseudoacacia with a twomonth gap. For R. pseudoacacia, Jarvis-style model was built (R2 ¼ 0.5476 for R. pseudoacacia 1, R2 ¼ 0.5235 for R. pseudoacacia 2, R2 ¼ 0.7248 for R. pseudoacacia 3), and used to fill the GC gap. These data, along with known estimates for the other hours, were used to calculate accumulative ozone uptake and compare species difference for the studied species during the growing season. 2.7.2. Statistical analyses Statistical analyses were performed by SPSS 11.5 (SPSS Inc., Chicago, USA). Paired sample t tests were conducted to test whether there were statistically significant differences at p < 0.05 in average stomatal conductance obtained from sap flow measurements and gas exchange measurements of the deciduous species. Independent t tests were used to test whether there were significant differences at p < 0.05 in the average GO3 , FO3 , annual average AFst0 between deciduous species and evergreen species, as well as JO at the North and South aspect of trunk in P. tabulaeformis. OneTable 4 Average canopy conductance (GC) (expressed on leaf area) obtained from sap flow measurements, compared to gas exchange data in the same tree at the same study site at noon during the investigation period MayeOctober, 2009. Species

G. biloba 1 G. biloba 2 A. chinensis 1 A. chinensis 2 A. chinensis 3

Sap flow measurements

Gas exchange measurements

GC (mmol m2 s1)

GS (mmol m2 s1)

86.41 60.03 109.68 80.08 137.68








2.88 2.47 3.23 2.06 4.09

65.45 67.21 116.92 123.02 63.33








18.71 22.33 28.94 29.29 13.73

Notes: Values
SE are given for GC and GS across replicates of the same tree from May to October.

H. Wang et al. / Environmental Pollution 162 (2012) 275e286 way ANOVA was used to test whether there were significant differences at p < 0.05 in JO at the different aspect of trunk in C. deodara and R. pseudoacacia, as well as the average GO3 , FO3 , and annual average AFst0 among species.

3. Results 3.1. Sap flux scaling The daily sap flux density in the outer 2 cm of sapwood (JO) for studied trees showed a seasonal change pattern, with flux densities increasing in April as canopy leaf area started growing, reaching a broad plateau between May and August, and then declining in September (Fig. 2). The high JO of summer coincided with the rapid growth season, strong solar radiation, and ample water supply, typical environment of trees in temperate ecosystems. During the winter months, even the JO of evergreen trees was markedly dampened (Fig. 2). In order to conduct species comparisons for the studied species from 1st, May to 31th, October in 2009, gap filling procedures for sap flux densities were devised. A. chinensis and P. tabulaeformis exhibited high JO (139.11
21.12; 141.30
20.61 g cm2 d1); G. biloba and C. deodara showed intermediate values (124.84
9.03; 111.32
22.92 g cm2 d1); and M. liliiflora and R. pseudoacacia had much lower densities (97.27
12.48; 86.85
13.83 g cm2 d1). Axial variation was substantial in the selected trees in P. tabulaeformis, C. deodara, and R. pseudoacacia (Fig. 3AeC). Due to more exposure to sun in the south aspect, the anatomy of the xylem structure and under branch height, there was apparent relationship between sap flux density and aspect in P. tabulaeformis and C. deodara, but no apparent relationship in R. pseudoacacia (Wang et al., 2010). Hourly JO over 24 h at different aspects were highly correlated (Fig. 3DeF). Therefore, mean JO could be accurately estimated based on the measurement obtained on one aspect. Daily means of JO at the different aspect of trunk were similar (Fig. 3GeI; for P. tabulaeformis, Independent t-test, p ¼ 0.972; for C. deodara, one-way ANOVA, p ¼ 0.913; for R. pseudoacacia, one-way ANOVA, p ¼ 0.422). This similarity in urban trees was different from findings

281

from natural forests (Oren et al., 1999a), which maybe because every aspect of trunk in urban trees exposes to similar radiation and humidity conditions, due to the low density of trees. JS (sap-flux density across the active sapwood) and ET (tree transpiration) estimated with the method proposed by Pataki et al. were shown in Table 3 (Pataki et al., 2011). The differences in tree transpiration among species were relatively large, with estimates ranging from 4.56
1.42 kg tree1 d1 in M. liliiflora to 35.09
14.44 kg tree1 d1 in C. deodara. The highest JS was found in A. chinensis, while the highest ET was found in C. deodara. The coefficient of variation (¼standard deviation/mean), as well as the difference between the coefficient of variation for JO and JS were also similar to the pattern in the Pataki paper, which was about 25% for JO and 50% for JS (Pataki et al., 2011). 3.2. Ozone load Daily average concentration of ambient O3 exhibited a seasonal cycle with maxima of 75 ppb during the summer and minima of 1 ppb during the winter (Fig. 5). The maximum O3 concentration occurred in the summer when the air temperature was high and the photochemical reaction was active. During our study period (from November 1, 2008 to October 31, 2009), the hourly average O3 concentration was 22 ppb, with the maximum of 138 ppb. Most of the hourly average concentrations of O3, appriximately 84%, were within the range of 0e40 ppb. Approximately 7% of the hourly average concentrations of O3 were greater than 60 ppb. The accumulated ozone exposure over a threshold of 40 ppb or nl l1(AOT40) during the growing season (from May 1, 2009 to October 31, 2009) was 20.12 ppm h at the study site. 3.3. Ozone uptake The seasonal variations of daily average canopy conductance to O3 (GO3 ), whole-tree O3 uptake rate (FO3 ), and accumulative

Fig. 5. Seasonal course of daily mean ambient ozone concentration (O3), ground area scaled canopy conductance for ozone (GO3 ), canopy O3 uptake rate (FO3 ), and accumulated stomatal ozone flux (AFst) of six tree species in urban environment from November 1, 2008 to October 31, 2009. Jarvis-style model was built and used to fill the canopy conductance gap in R. pseudoacacia (grey colour).

282

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

Table 5 Average canopy conductance for O3 (GO3 , mmol m2 s1), canopy O3 uptake rate (FO3 , nmol m2 s1), and accumulated stomatal ozone flux over thresholds of 0,1.6 nmol m2 s1 (AFst0,1.6, mmol m2, projected area) estimated during the growing season (from May, 1 to October, 31). Species

GO3 mmol m2 s1

FO3 nmol m2 s1

AFst0 mmol m2

G. biloba A. chinensis M. liliiflora R. pseudoacacia P. tabulaeformis C. deodara

80.27
104.16
129.10
17.35
58.08
62.76


3.09
4.01
4.80
0.61
2.18
2.24


15.77 19.15 20.97 3.43 11.12 11.29

0.84 16.61 29.22 2.14 9.16 5.21

0.01 0.61 1.04 0.07 0.31 0.17









0.02 3.16 5.63 0.41 1.66 1.02

AFst1.6 mmol m2 9.21 12.38 14.70 0.15 5.11 5.21









0.11 2.89 5.13 0.08 1.28 0.81

Notes: To make nearly all individuals have the same gaps, a filter was employed to studied species except R. pseudoacacia. For R. pseudoacacia, Jarvis-style model was built and used to fill the canopy conductance gap.

stomatal O3 flux (AFst0) of the 17 selected trees were shown in Fig. 5. The highest levels of GO3 and FO3 were observed in summer months, while the lowest levels were observed in winter months (Fig. 5). During the growing season, the average GO3 (p ¼ 0.004), FO3 (p ¼ 0.002), and annual average AFst0 (p ¼ 0.013) differed significantly among studied species. A. chinensis and M. liliiflora had the high FO3 (4.01
0.61; 4.80
1.04 nmol m2 s1), while R. pseudoacacia had the lowest FO3 (0.61
0.07 nmol m2 s1). The other studied species showed intermediate values. In particular, the

A

average GO3 , FO3 , and annual accumulative stomatal O3 flux AFst0 of deciduous species (G. biloba, A. chinensis, M. liliiflora, R. pseudoacacia) by contrast was not significantly higher than those of evergreen species (P. tabulaeformis and C. deodara) (82.94 vs 60.42 mmol m2 s1 p ¼ 0.322; 3.13 vs 2.21 nmol m2 s1 p ¼ 0.160; 14.74 vs 11.20 mmol m2; p ¼ 0.364) (Table 5). 3.4. Effects of environmental factors on ozone uptake O3 uptake was determined by interaction of canopy conductance and ambient O3 concentration. The resulting O3 uptake rate (FO3 ) followed a close pattern to ambient O3 concentration (Fig. 6). At a given ambient O3 concentration, FO3 was effectively controlled by canopy conductance to O3 (GO3 ) and, hence, by vapour pressure deficit (D) and total radiation (RS). D and RS were correlated. At the range of D between 0.8 kPa and 1 kPa, RS was the dominant factor of GO3 response, so while both D and RS increased, so did GO3 (Fig. 6). When D was approaching 1 kPa and beyond, D became the dominant factor, the increase of D leading to a decrease in GO3 , despite of the increase of RS, because of the need to protect the hydraulic system (Fig. 6). An analysis of hourly data confirmed that the high D led to varying degrees of decline in GO3 and thus contrasting reduction in FO3 , especially with more pronounced reductions in G. biloba, A. chinensis, and M. liliiflora (Fig. 6A). At low D, however, GO3 was

B

kP

W

Fig. 6. Means of ambient air ozone concentration (O3), canopy conductance for ozone (GO3 ), and canopy ozone uptake rate (FO3 ) in relation to (A) vapour pressure deficit (D) and (B) total radiation (RS). Symbols represent the results of boundary line analysis selecting the maximum GC of 2 or 3 trees per species for different D classes (step width: 0.2 kPa) and RS classes (step width: 50 W m2).

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

high and thus allowed a high O3 flux even at low O3 concentrations (Fig. 6A). Using equation (6), we analyzed data selected from the boundary line analysis for each species. All the least-square regressions between GC and D were significant (p < 0.0001), with the regression coefficient of determination (R2) ranging from 0.97 to 0.99. The sensitivity of GC to D (the slope m in equation (6)) increases linearly with Gref (the b in equation (6)) for study species (Table 6). The slope of this linear relationship was 0.50 (R2, 0.99). Furthermore, the thresholds at which GC began to decrease also occurred at higher D values (1.2 kPa) for G. biloba, A. chinensis, and M. liliiflora than that for other species (1.0 kPa) (Table 6). Hence, the higher was the Gref, the more sensitive to D, and their absolute absorption of O3 would decrease faster with increasing D, regardless of species (Fig. 6A). It’s important to note that sensitivities were estimated based on data of GC in the range in which GC decreases exponentially with D, but not on all the data. For all the data, the reduction in GC was less sensitive to the increase of D. A further analysis on the hourly data showed that only small changes occurred for both GO3 and O3 concentration, and thus for FO3 , with the increase of RS (Fig. 6B). For G. biloba, A. chinensis, and M. liliiflora, GO3 , and in turn for FO3 , tended to increase with the increase of RS, and then decrease rapidly after approximately 400 W m2 at which the stomata presumably became light saturated. However, the O3 uptake by the other studied species remained largely constant with the increase of RS during most of the time. 4. Discussion 4.1. Whole-tree O3 uptake of contrasting species The sap flow based approach for estimating ozone uptake at the whole-tree or the stand level covered the effect of boundary layers on ozone flux (Wieser et al., 2003). Both average canopy O3 uptake rate and accumulated O3 uptake estimated during the growing season (from May, 1 to October, 31) of the studied species in urban environment were comparable with values obtained in a Pinus canariensis forest in Canary Islands (Wieser et al., 2006), Norway spruce (Picea abies) and cembran pine (Pinus cembra) at six rural sites between 580 and 1950 m a.s.l (Wieser et al., 2000), and P. abies, Fagus sylvatica, Larix decidua in stands at high and low altitude and contrasting water availability (Nunn et al., 2007). The studied species in urban environment exhibited higher O3 uptake in summer months, when higher temperature, radiation, vapor pressure deficit, and precipitation occurred. The seasonal variation in FO3 in our study was similar to that observed in a pine forest at high elevation (Matyssek et al., 2007, 2008). However, our results were different from that observed in a Pine forest in Canary Island (Mediterranean type climate) (c.f. Luis et al., 2005; Wieser et al., 2006). Due to the combination of strong fog impact on air humidity and access to deep soil water reserves, O3 uptake did not Table 6 Stomatal sensitivity to vapor pressure deficit (D) estimated as the slope of the relationship between canopy conductance and ln(D) (Oren et al., 1999b), applied to those data in which gc decreased exponentially with D (see Fig. 6A). Species

Sensitivity(m)

gcref(b)

m/b

D thresholda

G. biloba A. chinensis M. liliiflora R. pseudoacacia P. tabulaeformis C. deodara

233.57. 279.67 416.6 44.35 116.34 160.19

451.75 562.69 873.20 82.14 241.43 302.53

0.52 0.50 0.48 0.54 0.48 0.53

1.00 1.20 1.40 0.80 1.00 1.00

a

D threshold is the value of leaves to air vapor pressure difference from which gc begins to decrease in an exponential form. Reference canopy conductance (gcref) is gc at D ¼ 1 kPa.

283

show a significant reduction when the air and soil were dry, hence no clear seasonal trend was observed in a pine forest in Canary Island (c.f. Luis et al., 2005; Wieser et al., 2006). There were very large species differences in ground area scaled whole-tree ozone uptake (FO3 ), with estimates ranging from 0.61
0.07 nmol m2 s1 in R. pseudoacacia to 4.80
1.04 nmol m2 s1in M. liliiflora. This was mainly due to the influence of management, disturbance, species composition, and many other aspects of decision-making in urban forests as compared to their natural counterparts (Pataki et al., 2011). However, average FO3 by deciduous foliages was not significantly higher than that by evergreen ones (3.13 vs 2.21 nmol m2 s1, p ¼ 0.160). In contrast, Wieser et al. (2003) found that when scaled to the ground surface area, the O3 uptake by evergreen foliages (Norway spruce, Cembran pine) was significantly higher than that by deciduous ones (European larch), which was due to differences in stomatal behavior, leaf area index, and canopy structure. Notably, we found the annual O3 uptake by trees estimated by our sap flow measurements in Beijing was 0.78 g m2 (i.e., an average of AFst0 during growing season 16.20 mmol m2  48 g mol1O1000 ¼ 0.78 g m2), which is much lower than values estimated from the Urban Forest Effects Model (e.g., McPherson et al., 1994; Yang et al., 2005; Nowak and Dwyer, 2007; Table 7). Nowak and Dwyer (2007) found that the average annual O3 removal rate by canopy cover in several USA cities was about 2.96 g m2 (i.e., total pollution removal 9.3 g m2 x the ratio of O3 to total pollution removal 31.8% ¼ 2.96 g m2). Yang et al. (2005) found that the annual O3 removal rate per canopy area in Beijing was approximately 5.50 g m2 (i.e., 27.5 g m2  20% ¼ 5.50 g m2). McPherson et al. (1994) showed that annual O3 removal rate by tree canopy cover in Chicago was 2.86 g m2. Since the length of the season over which estimates were made (in-leaf, daytime, yearly) were same, this large discrepancy between estimates from our field data and those from modeling may be due to the following reasons. First, the canopy resistance (Rc) of approximately 1.74 cm s1 used in UFORE model was significantly smaller than 2.81 cm s1, the average Rc of the selected species measured in our study (Table 7). This difference in canopy resistance may be the primary cause of the large discrepancy. In modeling approaches, canopy resistance has been typically calculated either based on a multi-layer canopy deposition model, or derived from literature on individual trees and forests (Baldocchi et al., 1987; McPherson et al., 1994; Nowak and Crane, 2000; Yang et al., 2005). Second, leaf area index that was used to calculate canopy stomatal resistance and canopy cuticle resistance in UFORE model was much higher than that in our study (Table 7). Third, ambient O3 concentrations measured in our study were lower than those used in modelling (Table 7). Fourth, while sap flow measurements only estimate the O3 flux into the leaves, modeling methods calculate the total dry deposition that includes both stomatal O3 uptake and adsorption onto non-transpiring external tree surfaces,

Table 7 Ozone removal/uptake values per square meter of tree canopy cover per year (g m2 year1) from the literatures. Site

O3 concentration Tree Canopy LAI (ppb) cover resistance (%) (cm s1)

Chicago

Reference Removal rate g m2 year1

Hourly 27 (13e38) Beijing Monthly 70(40e120) USA cities e

11

1.74

4.30 2.86

17

e

e

5.50

e

e

e

2.96

Beijing

e

2.81

2.34 0.85

Hourly 22 (0e138)

McPherson et al., 1994 Yang et al., 2005 Nowak and Dwyer, 2007 Our results

284

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

SWC SWC SWC

Fig. 7. Seasonal course of daily mean soil water content (SWC) amongst the studied trees from November 1, 2008 to October 31, 2009. Symbols GB, AC, ML, CD denote Ginkgo biloba, Aesculus chinensis, Magnolia liliiflora, Cedrus deodara, respectively.

which may lead to higher ozone removal. Fifth, deposition velocity in UFORE model was calculated as the inverse of the sum of the aerodynamic, quasi-laminar boundary layer and canopy resistances (Nowak and Crane, 2000), but in our study we only calculated canopy resistances, which would also contributed to the difference. In conclusion, this difference may suggest the necessity to consider the species-specific flux in O3 risk assessment. 4.2. Environmental control of ozone uptake in urban trees Variations in O3 uptake are partly due to the variations in both ambient O3 concentrations and stomatal conductance. At our study site, there was a pronounced seasonal pattern in O3 concentrations. O3 concentrations were high in summer months, but much lower during the winter months. The resulting FO3 showed a close pattern to ambient air O3 concentrations (Fig. 5). Stomatal control is a critical factor in limiting whole-tree ozone uptake (Emberson et al., 2000; Wieser et al., 2006; Nunn et al., 2007). The degree of stomatal opening that regulated the whole-tree ozone uptake in our selected trees was mainly influenced by the vapour pressure deficit and radiation. The relative high O3 uptake by urban trees may be due to the reasons below. On the one hand, at low D, stomata were widely open allowing a high flux of O3 into the leaves even at low O3 concentrations. Similar response was also found in a planted exotic species (Pseudotsuga menziesii) in South America, which replaced native vegetation resulting in increased vegetation water use (Fernandez et al., 2009). Secondly, the slope in the strong linear relationship between the sensitivity of GC to D (the slope m in equation (6)) and Gref (the b in equation (6)) for studied species was of 0.50 (Table 6). This value was much smaller than that for the mesic adapted species (0.60), much larger than that for desert species (0.42 and 0.38), but closer with that for plantations species (0.46) (Oren et al., 1999b; Fernandez et al., 2009). Thirdly, high water availability at our study site might increase O3 uptake. Chamber experiments with seedlings or twigs gave evidences that drought stress protected plants from O3 injury mainly by regulating stomatal aperture (Wieser and Havranek, 1993). In our study site, the soil water content were not in deficit condition for nearly all the studied species, except for R. pseudoacacia (Fig.7), which may not limit the stomata opening, and hence contributed to the relative high ozone uptake by urban trees. 4.3. Do ozone concentrations exceeding critical levels cause probable corresponding effects on urban trees? Both the exposure-based and flux-based critical levels (AOT40 and AFst1.6) were used to evaluate the potential impacts of ozone on

trees in our study site. Our results indicated that there was great potential O3 risk for trees in our study site, when using the exposurebased critical level. The AOT40 (accumulated exposure over a threshold of 40 ppb or nl l1) during the growing season at our study site was 20.12 ppm h, which far exceeded the AOT40 5 ppm h, the current critical level for ozone impacts on trees (UNECE, 2004). When using the flux-based critical level, the AFst1.6 values for most of the studied species were above 4 mmol m2, the current fluxbased critical level for ozone impacts on trees (UNECE, 2004). These results indicated that urban trees species, including both deciduous species and evergreen species, are susceptible to O3 damage. However, AFst1.6 values observed for the four deciduous species ranged from 14.70
5.13 mmol m-2 in M. liliiflora to 0.15
0.08 mmol m-2 in R. pseudoacacia, suggesting that the tolerating capacity to O3 was diverse. By contrast, AFst1.6 values observed for the two evergreen species approached the current flux-based critical level, indicating that they may be less susceptible to O3 damage. These results are similar to findings from previous studies that Mediterranean evergreen shrubs have a higher capacity to tolerate ozone stress than their deciduous relatives (Calatayud et al., 2010). In addition, our results suggest that the flux-based measurement is better suited for evaluating the risk of ozone impacts on urban trees than the exposure-based method. These results are also consistent with those from previous studies that ozone fluxes are more accurate than AOT40 exposure-based measurement in predicting ozone visible foliar injury (Gerosa et al., 2008). Acknowledgments This study was supported by the National Natural Science Foundation of China (Grant Nos. 41030744, 31170424), and Special Foundation of State Key Laboratory of Urban and Regional Ecology. We thank the editor, two anonymous reviewers, Dr. Heather R. McCarthy (University of Oklahoma), and Dr. Ping Zhao (South China Institute of Botany, Chinese Academy of Sciences) for their constructive comments and suggestions. We also thank all the members of Beijing Urban Ecosystem Research Station and Beijing Teaching Botanical Garden for their assistance in the field. References Baldocchi, D.D., Hicks, B.B., Camara, P., 1987. A canopy stomatal resistance model for gaseous deposition to vegetated surfaces. Atmospheric Environment 21, 91e101. Beckett, K.P., Freer-Smith, P., Gail, T., 2000. Effective tree species for local air quality management. Journal of Arboriculture 26, 12e18. Beijing Gardening and Greening Bureau, 2005. Collection of Beijing Urban Gardening and Afforestation Survey, Monograph. Beijing Publishing House, Beijing. Beijing Water Authority, 2010. Beijing Water Resources Bulletin 2009. China Water and Hydropower Publisher, Beijing.

H. Wang et al. / Environmental Pollution 162 (2012) 275e286 Braun, Sabine, Schindler, Christian, Leuzinger, S., 2010. Use of sap flow measurements to validate stomatal functions for mature beech (Fagus sylvatica) in view of ozone uptake calculations. Environmental Pollution, 1e10. Burgess, S.S.O., 2006. Measuring transpiration responses to summer precipitation in a Mediterranean climate: a simple screening tool for identifying plant wateruse strategies. Physiologia Plantarum 127, 404e412. Calatayud, V., Marco, F., Cerveróa, J., Sánchez-Peñab, G., Sanz, M.J., 2010. Contrasting ozone sensitivity in related evergreen and deciduous shrubs. Environmental Pollution 158, 3580e3587. Campbell, G.S., Norman, J.M., 1998. An Introduction to Environmental Biophysics, second ed. Springer-Verlag, New York. Chan, C.K., Yao, X.H., 2008. Air pollution in mega cities in China. Atmospheric Environment 42, 1e42. Costa, W., Amaratunga, K., Udumullage, R., 2006. Transpiration characteristics of some homegarden tree species in Central Sri Lanka. In: Kumar, B.M., Nair, P.K.R. (Eds.), Tropical Homegardens: a Time-Tested Example of Sustainable Agroforestry. Springer, Netherlands, pp. 251e267. Delzon, S., Sartore, M., Granier, A., Loustau, D., 2004. Radial profiles of sap flow with increasing tree size in maritime pine. Tree Physiology 24, 1285e1293. Dye, P.J., Olbrich, B.W., 1993. Estimating transpiration from 6-year-old Eucalyptus grandis trees: development of a canopy conductance model and comparison with independent sap flux measurements. Plant, Cell & Environment 16, 45e53. Emberson, L.D., Wieser, G., Ashmore, M.R., 2000. Modelling of stomatal conductance and ozone flux of Norway spruce: comparison with field data. Environmental Pollution 109, 393e402. Ewers, B.E., Oren, R., 2000. Analyses of assumptions and errors in the calculation of stomatal conductance from sap flux measurements. Tree Physiology 20, 579e589. Fernandez, M.E., Gyenge, J., Schlichter, T., 2009. Water flux and canopy conductance of natural versus planted forests in Patagonia, South America. Trees-Structure and Function 23, 415e427. Ferretti, M., Fagnano, M., Amoriello, T., Badiani, M., Ballarin-Denti, A., Buffoni, A., Bussotti, F., Castagna, A., Cieslik, S., Costantini, A., De Marco, A., Gerosa, G., Lorenzini, G., Manes, F., Merola, G., Nali, C., Paoletti, E., Petriccione, B., Racalbuto, S., Rana, G., Ranieri, A., Tagliaferri, A., Vialetto, G., Vitale, M., 2007. Measuring, modelling and testing ozone exposure, flux and effects on vegetation in southern European conditionseWhat does not work? A review from Italy. Environmental Pollution 146, 648e658. Ford, C.R., McGuire, M.A., Mitchell, R.J., Teskey, R.O., 2004. Assessing variation in the radial profile of sap flux density in Pinus species and its effect on daily water use. Tree Physiology 24, 241e249. Gerosa, G., Marzuoli, R., Desotgiu, R., Bussotti, F., Ballarin-Denti, A., 2008. Visible leaf injury in young trees of Fagus sylvatica L. and Quercus robur L. in relation to ozone uptake and ozone exposure. An open-top chambers experiment in South Alpine environmental conditions. Environmental Pollution 152, 274e284. Granier, A., 1987. Evaluation of transpiration in a Douglas-fir stand by means of sap flow measurements. Tree Physiology 3, 309e320. Granier, A., Huc, R., Barigah, S.T., 1996. Transpiration of natural rain forest and its dependence on climatic factors. Agricultural and Forest Meteorology 78, 16e29. Granier, A., Loustau, D., Breda, N., 2000. A generic model of forest canopy conductance dependent on climate, soil water availability and leaf area index. Annals Forest Science 57, 755e765. Hagishima, A., Narita, K., Tanimoto, J., 2007. Field experiment on transpiration from isolated urban plants. Hydrological Processes 21, 1217e1222. Hao, J.M., Wang, L.T., Li, L., Hu, J.N., Yu, X.C., 2005. Air pollutants contribution and control strategies of energy-use related sources in Beijing. Science in China, Series D 48 (Suppl. II), 138e146. Herbst, M., Roberts, J.M., Rosier, P.T.W., Taylor, M.E., Gowing, D.J., 2007. Edge effects and forest water use: a field study in a mixed deciduous woodland. Forest Ecology and Management 250, 176e186. Hubbard, R.M., Stape, J., Ryan, M.G., Almeida, A.C., Rojas, J., 2010. Effects of irrigation on water use and water use efficiency in two fast growing Eucalyptus plantations. Forest Ecology and Management 259, 1714e1721. Jarvis, P.G., 1976. The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field. Philosophical Transactions of the Royal Society of London 273, 593e610. Jarvis, P.G., Mcnaughton, K.G., 1986. Stornatal Control of Transpiration: Scaling up from Leaf’ to Region. Advances in Ecological Research, vol. 15. Academic Press, London. Köstner, B.M.M., Schulze, E.D., Kelliher, F.M., Hollinger, D.Y., Byers, J.N., Hunt, J.E., McSeveny, T.M., Meserth, R., Weir, P.L., 1992. Transpiration and canopy conductance in a pristine broad-leaved forest of Nothofagus: an analysis of xylem sap flow and eddy correlation measurements. Oecologia 91, 350e359. Köstner, B., Matyssek, R., Heilmeier, H., Clausnitzer, F., Nunn, A.J., Wieser, G., 2008. Sap flow measurements as a basis for assessing trace-gas exchange of trees. Flora 203, 14e33. Laisk, A., Kull, O., Moldau, H., 1989. Ozone concentration in leaf intercellular air spaces is close to zero. Plant Physiology 90, 1163e1167. Landolt, W., Keller, T., 1985. Uptake and effects of air pollutants on woody plants. Experientia 41, 301e309. Landsberg, H.E., 1981. The Urban Climate. Academic Press, New York. Longcore, T., Rich, C., 2004. Ecological light pollution. Frontiers in Ecology and the Environment 2, 191e198.

285

Lu, P., Muller, W.J., Chacko, E.K., 2000. Spatial variations in xylem sap flux density in the trunk of orchard-grown, mature mango trees under changing soil water conditions. Tree Physiology 20, 683e692. Lu, P., Yunusa, I.A.M., Walker, R.R., Müller, W.J., 2003. Regulation of canopy conductance and transpiration and their modelling in irrigated grapevines. Functional Plant Biology 30, 689e698. Luis, V.C., Jiménez, M.S., Morales, D., Kucera, J., Wieser, G., 2005. Canopy transpiration of a Canary Islands pine forest. Agricultural and Forest Meteorology 135, 117e123. Martin, C.A., Stabler, L.B., 2002. Plant gas exchange and water status in urban desert landscapes. Journal of Arid Environments 51, 235e254. Martin, T.A., Brown, K.J., Cermak, J., Ceulemans, R., Kucera, J., Meinzer, F.C., Rombold, J.S., Sprugel, D.G., Hinckley, T.M., 1997. Crown conductance and tree and stand transpiration in a second-growth Abies amabilis forest. Canadian Journal of Forest Research 27, 797e808. Matyssek, R., Sandermann, H.J., 2003. Impact of ozone on trees: an ecophysiological perspective. Progress in Botany 64, 349e404. Matyssek, R., Bytnerowicz, A., Karlsson, P.E., Paoletti, E., Sanz, M., Schaub, M., Wieser, G., 2007. Promoting the O3 flux concept for European forest trees. Environmental Pollution 146, 587e607. Matyssek, R., Sandenmann, H., Wieser, G., Booker, F., Cieslik, S., Musselman, R., Ernst, D., 2008. The challenge of making ozone risk assessment for forest trees more mechanistic. Environmental Pollution 156, 567e582. McPherson, E.G., Nowak, D.J., Rowntree, R.A. (Eds.), 1994. Chicago’s Urban Forest Ecosystem: Results of the Chicago Urban Forest Climate Project. U. S. Department of Agriculture, Forest Service, Northeastern Forest Experiment Station, Radnor, PA. Montague, T., Kjelgren, R., 2004. Energy balance of six common landscape surfaces and the influence of surface properties on gas exchange of four containerized tree species. Scientia Horticulturae 100, 229e249. Musselman, R.C., Massman, W.J., 1998. Ozone flux to vegetation and its relationship to plant response and ambient air quality standards. Atmospheric Environment 33, 65e73. Nowak, D.J., Crane, D.E., 2000. The Urban Forest Effects (UFORE) model: quantifying urban forest structure and functions. In: Hansen, Mark, Burk, Tom (Eds.), Integrated Tools for Natural Resources Inventories in the 21st Century. Gen. Tech. Rep. NC-212. U.S. Dept. of Agriculture, Forest Service, North Central Forest Experiment Station, St. Paul, MN, pp. 714e720. Nowak, D.J., Dwyer, J.F., 2007. Understanding the benefits and costs of urban forest ecosystems. In: Kuser, J.E. (Ed.), Urban and Community Forestry in the Northeast. Springer Science and Business Media, New York, pp. 25e46. Nowak, D.J., Civerolo, K.L., Rao, S.T., Sistla, G., Luley, C.J., Crane D, E., 2000. A modeling study of the impact of urban trees on ozone. Atmospheric Environment 34, 1601e1613. Nunn, A.J., Wieser, G., Metzger, U., Löw, M., Wipfler, P., Häberle, K.-H., Matyssek, R., 2007. Exemplifying whole-plant ozone uptake in adult forest trees of contrasting species and site conditions. Environmental Pollution 146, 629e639. O’Brien, J.J., Oberbauer, S.F., Clark, D.B., 2004. Whole tree xylem sap flow responses to multiple environmental variables in a wet tropical forest. Plant, Cell & Environment 27, 551e567. Oishi, A.C., Oren, R., Stoy, P.C., 2008. Estimating components of forest evapotranspiration: a footprint approach for scaling sap flux measurements. Agricultural and Forest Meteorology 148, 1719e1732. Oren, R., Phillips, N., Katul, G., Ewers, B.E., Pataki, D.E, 1998. Scaling xylem sap flux and soil water balance and calculating variance: a method for partitioning water flux in forests. Annals of Forest Science 55, 191e216. Oren, R., Phillips, N., Ewers, B.E., Pataki, D.E., Megonigal, J.P., 1999a. Sap-flux-scaled transpiration responses to light, vapor pressure deficit, and leaf area reduction in a flooded Taxodium distichum forest. Tree Physiology 19, 337e347. Oren, R., Sperry, J.S., Katul, G.G., Pataki, D.E., Ewers, B.E., Phillips, N., Schäfer, K.V.R., 1999b. Survey and synthesis of intra- and interspecific variation in stomatal sensitivity to vapour pressure deficit. Plant, Cell & Environment 22, 1515e1526. Pataki, D.E., Oren, R., 2003. Species differences in stomatal control of water loss at the canopy scale in a mature bottomland deciduous forest. Advances in Water Resources 26, 1267e1278. Pataki, D.E., McCarthy, H.R., Litvak, E., Pincetl, S., 2011. Transpiration of urban forests in the Los Angeles metropoli tan area. Ecological Applications 21, 661e677. Phillips, N., Nagchaudhuri, A., Oren, R., Katul, G., 1997. Time constant for water transport in loblolly pine trees estimated from time series of evaporative demand and stem sapflow. Trees-Structure and Function 11, 412e419. Rico, M., Gallego, H.A., Moreno, G., Santa Regina, I., 1996. Stomatal response of Quercus pyrenaica Willd to environmental factors in two sites differing in their annual rainfall (Sierra de Gata, Spain). Annals of Forest Science 53, 221e234. Schafer, K.V.R., Oren, R., Tenhunen, J.D., 2000. The effect of tree height on crown level stomatal conductance. Plant, Cell & Environment 23, 365e375. Scott, K.I., McPherson, E.G., Simpson, J.R., 1998. Air pollutant uptake by Sacramento’s urban forest. Journal of Arboriculture 24, 224e234. Simpson, D., Ashmore, M.R., Emberson, L., Tuovinen, J.P., 2007. A comparison of two different approaches for mapping potential ozone damage to vegetation. A model study. Environmental Pollution 146, 715e725. State Statistical Bureau, 2011. Statistical Yearbook of Beijing 2010. China Statistics Press, Beijing. Tateishi, M., Kumagai, T., Utsumi, Y., Umebayasi, T., Shiiba, Y., Inoue, K., Kaji, K., Cho, K., Otsuki, K., 2008. Spatial variations in xylem sap flux density in

286

H. Wang et al. / Environmental Pollution 162 (2012) 275e286

evergreen oak trees with radial-porous wood: comparisons with anatomical observations. Trees-Structure and Function 22, 23e30. UNECE, 2004. Manual on Methodologies and Criteria for Modelling and Mapping Critical Loads & Levels and Air Pollution Effects, Risks, and Trends, United Nations Economic Commission for Europe (UNECE) Convention on Long-range Transboundary Air Pollution, Geneva, Switzerland, p. 54. Wang, H., Ouyang, Z.Y., Zheng, H., Wang, X.K., Ni, Y.M., Ren, Y.F., 2010. Characteristics of spatial variations in xylem sap flow in urban greening tree species Pinus tabulaeformis, Cedrus deodara and Robinia pseudoacacia in Beijing, China. Chinese Journal of Plant Ecology 34, 924e937 (in Chinese with English abstract). Wang, H., Ouyang, Z.Y., Chen, W.P., Wang, X.K., Zheng, H., Ren, Y.F., 2011. Water, heat, and airborne pollutants effects on transpiration of urban trees. Environmental Pollution 159 (8e9), 2127e2137. Webb, R.A., 1972. Use of the boundary line in the analysis of biological data. Journal of Horticultural Science 47, 309e319. Wei, J.X., Jiang, W.B., Wong, M.L., 2008. Cultural connotations of Aesculus chinensis and their application in landscape architecture. Chinese Agricultural Science Bulletin 24, 356e359. Wieser, G., Havranek, W.M., 1993. Ozone uptake in the sun and shade crown of spruce: quantifying the physiological effects of ozone exposure. Trees-Structure and Function 7, 227e232.

Wieser, G., Havranek, W., 1995. Environmental control of ozone uptake in Larix decidua Mill.: a comparison between different altitudes. Tree Physiology 15, 253e258. Wieser, G., Häsler, R., Götz, B., Koch, W., Havranek, W.M., 2000. Role of climate, crown position, tree age and altitude in calculated ozone flux into needles of Picea abies and Pinus cembra: a synthesis. Environmental Pollution 109, 415e422. Wieser, G., Matyssek, R., Köstner, B., Oberhuber, W., 2003. Quantifying ozone uptake at the canopy level of spruce, pine and larch trees at the alpine timberline: an approach based on sap flow measurement. Environmental Pollution 126, 5e8. Wieser, G., Luis, V.C., Cuevas, E., 2006. Quantification of ozone uptake at the stand level in a Pinus canariensis forest in Tenerife, Canary Islands: an approach based on sap flow measurements. Environmental Pollution 140, 383e386. Wu, J.G., 2008. Making the case for landscape ecology an effective approach to urban sustainability. Landscape Journal 27, 41e50. Wullschleger, S.D., Norby, R.J., 2001. Sap velocity and canopy transpiration in a sweetgum stand exposed to free-air CO2 enrichment (FACE). New Phytologist 150, 489e498. Yang, J., McBride, J., Zhou, J., Sun, Z., 2005. The urban forest in Beijing and its role in air pollution reduction. Urban Forestry & Urban Greening 3, 65e78. Zhao, J.J., 2010. Species Composition and Spatial Distribution of Urban Plants within the Built-up Areas of Beijing, China. Research Center for Eco-Environmental Sciences. Chinese Academy of Sciences, Beijing.