Ecosystem water use efficiency for a sparse vineyard in arid northwest China

Agricultural Water Management 148 (2015) 24–33

Contents lists available at ScienceDirect

Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat

Ecosystem water use efficiency for a sparse vineyard in arid northwest China Sien Li a , Shaozhong Kang a,∗ , Lu Zhang b , Taisheng Du a , Ling Tong a , Risheng Ding a , Weihua Guo a , Peng Zhao a , Xia Chen a , Huan Xiao a a b

Center for Agricultural Water Research in China, China Agricultural University, Beijing, 100083, China CSIRO Land and Water, GPO Box 1666, Canberra, ACT 2601, Australia

a r t i c l e

i n f o

Article history: Received 29 August 2013 Received in revised form 9 August 2014 Accepted 12 August 2014 Keywords: Carbon dioxide flux Evapotranspiration Sparse vegetation Vineyard Water use efficiency

a b s t r a c t Ecosystem water use efficiency (WUE) can be defined as the ratio of net CO2 exchange to evapotranspiration, which implicates the interactions between carbon sequestration and water consumption. Previous studies mainly focused on ecosystem WUE for forests, grasslands and farmlands, but paid little attention to the sparse vineyard. How the vineyard WUE varied on daily and seasonal time scales remains uncertain. The vineyard CO2 and water fluxes were measured by the eddy covariance method during 2008 in arid northwest China to address the issues. Results indicate that the seasonal variation of vineyard WUE presented a downward-parabolic trend, with a mean value of 4 mg g−1 and a maximum value of 10 mg g−1 . Compared with other ecosystems, WUE for vineyard was lower than that for forests, maize, wheat and wetlands, but higher than grasslands and Savannas. The severely dry climate and the sparse vegetation led the results. Such factors as radiation, air temperature and humidity, soil moisture, canopy conductance and leaf area index all exerted significant influences on vineyard WUE. However, the vineyard WUE was highly sensitive to solar radiation and air temperature changes, and it decreased significantly with the rising radiation and temperature, which is remarkably different from previous studies. Such results were mainly due to the great impact on CO2 exchange exerted by soil layer in the sparse vineyard, and the high sensitivity of soil respiration to temperature changes induced by radiation and air temperature. The CO2 assimilation reduced with the increasing radiation and air temperature, however the vineyard evapotranspiration increased rapidly, thus the vineyard WUE declined significantly with the rising radiation and air temperature. These results provided a new insight for understanding the carbon and water cycles over the sparse vegetation. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Ecosystem water use efficiency (WUE, mg g−1 ) can be defined as the ratio of net ecosystem productivity (NEP, mg CO2 m−2 s−1 ) to evapotranspiration (ET, g H2 O m−2 s−1 ) (Baldocchi, 1994; Scanlon and Albertson, 2004; Kuglitsch et al., 2008). It connects the ecological processes and hydrological processes and implicates the interaction between carbon sequestration and water consumption. Currently, global climatic changes such as the rise in CO2 concentration and the global warming have significantly altered ecosystem WUE through influencing photosynthesis and transpiration (Tao et al., 2008; Guo et al., 2010; Allen et al., 2011; Keenan et al., 2013). Thus exploring the ecosystem WUE is critical in revealing

∗ Corresponding author. Fax: +86 10 62737611. E-mail addresses: [email protected], [email protected] (S. Kang). http://dx.doi.org/10.1016/j.agwat.2014.08.011 0378-3774/© 2014 Elsevier B.V. All rights reserved.

the response of ecological and hydrological processes to global climatic changes and optimizing the water and carbon managements in practice (Brouder and Volenec, 2008; Green et al., 2010; Grewal et al., 2011; Keenan et al., 2013; Liu and Tao, 2013). Ecosystem WUE is mainly controlled by such environmental factors as soil moisture, atmospheric CO2 concentration, air temperature and humidity and solar radiation and also by physiological factors such as canopy conductance and leaf area index. Its regulation and controlling mechanism is similar to that at leaf scale, but different in the aspect that ecosystem WUE involves in and is influenced by both vegetation and soil. It is the combined effects of processes like photosynthesis, respiration, evaporation and transpiration. Thus ecosystem WUE has a far more complicated controlling mechanism than that at leaf scale. So far, scientists have conducted many researches on ecosystem WUE. Baldocchi (1994) indicated that the maximum WUE for maize and wheat in the growing season could reach 15 mg CO2 g−1 H2 O.

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

25

Law et al. (2002) showed that the ratio of carbon gain to water loss was 3.4 mg g−1 for grasslands, 3.2 mg g−1 for deciduous broadleaf forests, 3.1 mg g−1 for crops, 2.4 mg g−1 for evergreen conifers and 1.5 mg g−1 for tundra vegetation. Ponton et al. (2006) indicated that the average WUE for grassland, aspen and Douglas-fir was 2.6, 5.4 and 8.1 mg g−1 , respectively. Zhao et al. (2007) indicated that the WUE of wheat in north China Plain reached a peak value of 14 mg g−1 . Clement et al. (2012) indicated that the forest in Scotland sequestered about 6 tonnes of C per hectare per annum using the 5 years of eddy covariance measurements. In a latest paper published in Nature, Keenan et al. (2013) found a substantial increase in water-use efficiency in temperate and boreal forests of the Northern Hemisphere over the past two decades, and indicated that the increase is most consistent with the strong CO2 fertilization effect. Vineyards are usually planted in the form of the single vertical trellis and wide row so as to ensure sufficient illumination and favorable ventilation. In this sense, the vineyard can be considered a sparse ecosystem. Many studies have indicated that soil evaporation could account for 50% of total ET over the entire growth stage (Zhang et al., 2008). By virtue of the notable difference in water and carbon transports between the soil layer and the vegetable layer in the vineyard, the sparse ecosystem like the vineyard needs substantial study on relationships between WUE and environmental and physiological factors and how such factors regulate WUE. To address these questions, the eddy covariance method is adopted in the vineyard in arid northwest China to measure water and carbon fluxes during the whole growth period, with the aim to: (1) analyze the daily and seasonal variations of ecosystem WUE for the sparse vineyard and its difference from the WUE for other ecosystems and (2) reveal the response patterns of vineyard WUE to environmental and physiological factors such as radiation, air temperature and humidity, CO2 concentration, soil moisture, canopy conductance and leaf area index.

of vineyard according to the prevailing wind direction. The least fetch exceeded 600 m, which can fully meet the requirements of EC measurement. Measurements were made continuously from May 1st to October 11th in 2008. Net radiation (Rn ) was measured by a net radiometer (model NR-LITE, Kipp & Zonen, Delft, Netherlands) at a height of 4.5 m above the ground. Four soil heat flux plates (model HFP01, Hukseflux, Netherlands) were used to measure soil heat flux. The procedures conducted for correcting the eddy covariance measurements included: (1) 10-min interval for eddy flux computation (Twine et al., 2000); (2) the signal asynchrony correction (Wolf et al., 2008); (3) the oxygen-correction proposed by Tanner and Greene (1989); (4) planar fit method for coordinate rotation (Finnigan et al., 2003; Paw et al., 2000); (5) density correction according to the method of Webb et al. (1980) and (6) filling data gaps using the mean diurnal variation (MDV) method (Falge et al., 2001). In this study, sum of vineyard (ET + H, w m−2 ) accounted for 95% of available energy (Rn − G, w m−2 ) over whole experimental period. For the daytime EC data, the measured energy budget components were forced to close using “Bowen-ratio closure” method proposed by Twine et al. (2000), which assumes that Bowen-ratio is correctly measured by the EC system. But for the nighttime EC data, especially when the available energy was below zero, another method – the “residual ET closure” method also proposed by Twine et al. (2000) was adopted to close the energy balance in our study. This method assumed that the EC-based H was accurately measured, and solved for ET as the residual to the energy-balance equation. After forcing the energy balance to be closed, the ET data by the EC system (ETEC ) were adopted in the following analysis (Li et al., 2013a,b).

2. Materials and methods

Soil moisture content was measured using portable device (Diviner 2000, Sentek Pty Ltd., Australia) (Li et al., 2013a,b). Fifteen PVC access tubes with the depth of 1.2 m were evenly inserted in the soil in the ditch, shaded and non shaded parts of the ridge, respectively. The measurements were calibrated by the oven drying method. The normalized soil water content of 0–1 m layer is calculated as: F() = ( −  w )/( f −  w ), where  is the measured soil water content,  f is the field capacity,  w is the wilting coefficient. Leaf area index was measured every 10 days using AM300 portable leaf area meter (ADC BioScientific Ltd., UK), respectively.

2.1. Experimental site and description The experiment was conducted at Shiyanghe Experimental Station for Water-saving in Agriculture and Ecology of China Agricultural University, located in Wuwei City, Gansu Province of northwest China (N37◦ 52 , E102◦ 50 , altitude 1581 m) during May 1st to September 25th, 2008. The experimental site is located in a typical arid zone where mean annual temperature is 8 ◦ C, annual accumulated temperature (>0 ◦ C) 3550 ◦ C, annual precipitation 164 mm, mean annual pan evaporation approximate 2000 mm, the average annual duration of sunshine 3000 h and the average number of frost free days 150 days. The groundwater table is 40–50 m below the ground surface (Li et al., 2008, 2012, 2013a,b) Measurements in the vineyard were made in a field with a length of 1650 m and a width of 1400 m in 2008. The area was planted with grapevines (Vitis vinifera L. cv Merlot Noir) in 1999 with row spacing of 270 cm and plant spacing of 100 cm. The trellis for grapevine was 1.5 m in height. The soil texture is sandy loam, with a mean dry bulk density of 1.47 g cm−3 , porosity of 52%, field capacity of 0.35 cm3 cm−3 and a permanent wilting point of 0.12 cm3 cm−3 for the 0–100 cm layers. Furrow irrigation was conducted five times on 18th May, 24th June, 18th July, 16th August and 9th September with total 420 mm in the vineyard. The precipitation was 79 mm during May 1st–September 25th, 2008. 2.2. Eddy covariance measurement and correction An opened eddy covariance system (Campbell Scientific Inc., USA) was installed at 4.2 m above the ground at the northwest

2.3. Other measurements

2.4. Calculation of canopy conductance using the re-arranged Penman-Monteith equation The Penman–Monteith (PM) model can be written as (Monteith, 1965):(1)ET =

(Rn −G)+Cp a VPD/ra ++ ( rs /ra )

where  is the latent heat of vaporization (J kg−1 ), ET the crop evapotranspiration,  the slope of the saturation water vapor pressure versus temperature curve (kPa K−1 ), Rn the net radiation (W m−2 ), G the soil heat flux (W m−2 ), Cp the specific heat of dry air at constant pressure (J kg−1 K−1 ), a the air density (kg m−3 ), VPD the water vapor pressure deficit (kPa), ra the aerodynamic resistance (s m−1 ),  the psychrometric constant (kPa K−1 ) and rs the canopy resistance (s m−1 ). The aerodynamic resistance ra can be ln((z−d)/(hc −d)) ln((z−d)/z0 ) calculated as (Thom, 1972):(2)ra = k2 u where z is the reference height (m), d the zero plane displacement (m), hc the mean crop height (m), z0 the roughness length of the crop relative to momentum transfer (m), k the von Karman constant (0.40) and u (m s−1 ) the wind speed at the reference height measured by eddy covariance. According to Monteith (1965), d can be calculated as 0.67 hc , z0 as 0.13 hc .

26

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

Thus the canopy conductance can be derived using the re-arranged PM equation (Stewart, 1988):(3)gc =   ra

ETEC (Rn −G)+Cp a VPD/ra −ETEC (+)

where ETEC is the latent heat flux measured by eddy covariance (W m−2 ) and gc is the canopy conductance (mm s−1 ). 2.5. Calculation of the ecosystem water use efficiency The ecosystem water use efficiency can be defined as (Kuglitsch et al., 2008; Yu et al., 2008):(4)WUE = NEP ET where NEP is net ecosystem productivity measured by eddy covariance system (mg CO2 m−2 s−1 ), ET the evapotranspiration using eddy covariance (g H2 O m−2 s−1 ). The positive value of NEP presents that the ecosystem absorbs CO2 from atmosphere, thus the positive WUE quantifies the rate of carbon uptake per unit of water loss.

According to Tanner and Sinclair (1983), the ecosystem WUE   kd Es 1 − ET also can be expressed as follows:(5)WUE = VPD where kd is a constant and determined by crop variety, VPD the water vapor pressure deficit (kPa), Es the soil evaporation and ET the total evapotranspiration. The formula indicates the negative relationships between ecosystem WUE and VPD, Es /ET. This is important for us to understand why the ecosystem WUE for the sparse vineyard in arid northwest China is significantly lower than other vegetations in later discussion. 3. Results 3.1. Seasonal variation of meteorological, physiological and soil factors over the sparse vineyard In order to reveal the relationship between ecosystem WUE and environmental factors, the variation patterns of net radiation (Rn ),

500

Rn

(a) Rn

-2

R n (W m )

400 300 200 100 40.0 00 (b) Ta

Ta(

)

30.0 20.0 10.0 Ta

5.00 0.0 0.0

VPD

(c) VPD

VPD (kPa)

4.00 3.00 2.00 1.00

(d) CO2 concentration

-3

CO2 concentration (mg m )

650 0.00 600 550 500 CO2

450 2008-5-1

2008-5-31

2008-6-30

2008-7-30

2008-8-29

2008-9-28

Date (Year-Month-Day) Fig. 1. Seasonal variation of daytime net radiation (Rn , a), air temperature (Ta , b), water vapor pressure deficit (VPD, c) and atmospheric CO2 concentration (Ca , d) over the vineyard in 2008. These data were measured by eddy covariance system and only the data that ranged from 8:00 to 17:00 were adopted. The local standard time was adopted.

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

27

(a) Gg cc

-1

Canopy conductance(mm s)

20.0 15.0 10.0 5.0 Gc

2.50 0.0 (b) LAI

1.50

2

-2

LAI (m m )

2.00

1.00 0.50 LAI

1.00 0.00 0.00 (c) Normalized soil water content

F(θ)

0.75 0.50 0.25 F(θ)

0.00 2008-5-1

2008-5-31

2008-6-30

2008-7-30

2008-8-29

2008-9-28

Date (Year-Month-Day) Fig. 2. Seasonal variation of daytime canopy conductance (gc , a), leaf area index (LAI, b) and the normalized soil water content (F(), c) in 2008. Canopy conductance was obtained by the re-arranged Penman–Monteith model (Eq. (3)). The normalized soil water content is calculated as: F() = ( −  w )/( f −  w ), where ␪ is the measured soil water content,  f is the field capacity,  w is the wilting coefficient. The data that ranged from 8:00 to 17:00 were adopted.

air temperature (Ta ), water vapor pressure deficit (VPD), air CO2 concentration (Ca ), canopy conductance (gc ), leaf area index (LAI) and the normalized soil moisture (F()) were investigated primarily. Fig. 1 depicts the seasonal variation of daytime Rn , Ta , VPD and Ca during May–September in 2008. Daytime Rn varied from 30 W m−2 to 450 W m−2 with an average of 335 W m−2 . Daytime Ta changed from 10 ◦ C to 32 ◦ C with a mean of 23 ◦ C. Compared with Ta , VPD showed a similar trend. It had a peak value of 3.92 kPa, a minimum value of 0.10 kPa and an average of 2 kPa. Daytime CO2 concentration ranged from 472 mg g−1 to 625 mg g−1 with an average of 525 mg g−1 . Fig. 2 shows the seasonal variation of gc , LAI and F(). gc fluctuated intensively over the whole crop growth stage with a mean value of 5.6 mm s−1 . LAI presented a regular logistical curve and it ranged from 0 m2 m−2 to 2.04 m2 m−2 . The normalized soil water content varied from 0 to 1, controlled by water balance between irrigation, precipitation and evapotranspiration. 3.2. Daily and seasonal variation of the ecosystem WUE for the sparse vineyard The mean daily variation of net ecosystem productivity (NEP), evapotranspiration (ET) and ecosystem water use efficiency (WUE) for the vineyard during different months is shown in Fig. 3. Local

standard time was adopted in the analysis. The vineyard NEP got the peak value at midday and was lower than 0 during nighttime for the nighttime respiration (Fig. 3a). Compared with NEP, the vineyard ET presented a regular bell-shaped curve. It reached the peak in midday and was close to zero during nighttime (Fig. 3b). Different with NEP and ET, WUENEP was nearly constant during the daytime and fluctuated sharply in nighttime (Fig. 3c). Due to uncertainty with ET, the nighttime values of WUE have week interest. The seasonal variation of daytime NEP, ET and WUENEP is shown in Fig. 4. NEP presented a typical parabola pattern, which was smaller at the early growth stage, peak in the middle stage and smaller in the last stage (Fig. 4a). However, vineyard ET fluctuated irregularly during the whole experimental period, which was mainly effected by Rn , VPD, wind and temperature (Fig. 4b). Different with NEP and ET, WUENEP increased sharply in the early growth stage, and varied gently during the middle and last stage (Fig. 4c). In the early season, the respiratory rate exceeded photosynthetic rate, so that the ecosystem was a carbon source, and WUENEP was negative. With the growth of crop, the photosynthetic rate improved continuously, and the highest WUE reached was 10 mg g−1 . At the end of the period, a higher WUE was still maintained. During the whole season, the mean WUE was 4 mg g−1 . A significantly linear regression equation was established between NEP and ET. The slope of the equation was 3.25 mg g−1 , which is close to the mean value of WUENEP (Fig. 5). Table 1 also indicates

28

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

0.75

The canopy WUE depends on the photosynthesis and transpiration, while the photosynthetic rate is closely related to the radiant intensity and the radiant intensity also exerts important effect on crop transpiration. In the study we found that the vineyard NEP varied gently but ET increased significantly when radiation increased, which led to that WUE fell linearly along with the rising radiation. The linear regression equations were obtained between WUE and Rn with high F value under different VPD conditions, which indicates that the significant linear relationship existed between WUE and Rn either on low or high VPD conditions(Fig. 6a). Similar to Fig. 6a, Fig. 6b indicates the decrease of NEP and WUE and the increase in ET under different VPD conditions, as the temperature increased. Temperature’s influence on lead photosynthesis can be depicted by a bell curve (Yu, 2007). The photosynthetic rate is accelerated with temperature rise, but diminishes when an optimum temperature has given way to high temperature that can deactivate enzyme. The transpiration rate and the respiratory rate will linearly or exponentially increase with temperature rise. Accordingly, the relation between WUE and temperature varies under different temperatures. Our results are different from previous ones and will be discussed later. Similar to Fig. 6a and b, Fig. 6c shows that the vineyard NEP decreased and ET increased with the rising VPD, thus WUE fell rapidly when VPD increased under different Rn conditions. The vineyard WUE increased significantly with the rising CO2 concentration (Fig. 6d).

-2 -1

NEP (mg CO2 m s )

(a) NEP 2008 0.50 0.25 0.00 May July September

-0.25 -0.50 -0.50 0.12

-2 -1

ET (g H2 O m s )

0.10

June August

(b) ET 20 08 May

0.08

June July

0.06

August September

0.04 0.02

-1

WUENEP (mgCO 2 g H2 O)

0.00 0.00 10

5

(c) WUE NEP 200 8

0 -5 -10

May July September

-15

June August

-20 0:00

4:00

8:00

12:00

16:00

20:00

0:00

Time (hh:mm) Fig. 3. Mean daily variation of net ecosystem productivity (NEP, a), evapotranspiration (ET, b) and ecosystem water use efficiency (WUE, the ratio of NEP to ET) for the sparse vineyard during different months in 2008.

that the sparse vineyard WUE was higher than that for Savanna and grassland, but lower than that for evergreen coniferous, wetland, wheat and maize. 3.3. Response of the ecosystem WUE for the sparse vineyard to climatic changes The variability in ecosystem WUE is controlled by many factors. The relationships between vineyard WUE and meteorological factors will be analyzed primarily.

3.4. Response of the ecosystem WUE for the sparse vineyard to physiological and soil moisture changes The effect of the calculated canopy conductance on vineyard WUE was investigated (Fig. 7a). Results show that vineyard WUE increased notably before canopy conductance reached 13 mm s−1 . However, when canopy conductance exceeded 13 mm s−1 , WUE varied slightly with the conductance increase. This is similar to the photosynthesis-stomata-evapotranspiration relation at the leaf level (Yu, 2007). We also found that LAI played an important role in driving NEP, ET and WUE (Fig. 7b). The vineyard NEP presented a significant increase with the rising LAI. The vineyard ET also had a close relationship with LAI. As a result, the vineyard WUE showed a strong linear relationship with LAI. The response of NEP, ET and WUE to soil moisture is similar to that of gc (Fig. 7c). A parabolic function between WUE with soil moisture was established. Soil moisture influences WUE in an extremely complex process and mechanism, which will be discussed later.

Table 1 Comparison of the ecosystem water use efficiency for the sparse vineyard to that for forest, savanna, grassland, wheat and maize and wetland.

*

Ecosystem type

Mean annual temperature (◦ C)

Mean annual precipitation (mm)

VPD (kPa)

Maximum ecosystem WUE

Literature resource

Temperate broad leaf forest Savanna (in Mongu) Savanna (in Maun) Savanna (in Okwa river) Savanna (in Tshane) Savanna Temperate grassland Wheat and maize Wetland Evergreen coniferous Evergreen coniferous Sparse vineyard

8.1 21.1 23.5 24.2 25.3 12.0 NA NA NA NA NA 8

697 879 460 407 365 646

About 1.8 About 1.0 About 1.8 About 1.8 About 2.0 About 1.5 NA NA About 1.5 About 1.5 About 1.5 2.5

24.4 mg CO2 g−1 H2 O (NEP/ET) 1.2 mg CO2 g−1 H2 O (NEP/ET) 1.4 mg CO2 g−1 H2 O (NEP/ET) 1.8 mg CO2 g−1 H2 O 2.2 mg CO2 g−1 H2 O 1.2 mg CO2 g−1 H2 O 4 mg CO2 g−1 H2 O 15 mg CO2 g−1 H2 O 11 mg CO2 g−1 H2 O 15 mg CO2 g−1 H2 O 18 mg CO2 g−1 H2 O 10 mg CO2 g−1 H2 O

Herbst et al. (2002) Scanlon and Albertson (2004) Scanlon and Albertson (2004) Scanlon and Albertson (2004) Scanlon and Albertson (2004) Verhoef et al. (1996) Hunt et al. (2002) Baldocchi (1994) Mahrt and Vickers (2002) Mahrt and Vickers (2002) Mahrt and Vickers (2002) Our study

The part date was obtained from Hu et al. (2009).

164

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

29

0.60 -2 -1

NEP (mg CO2 m s )

(a) NEP 2008 0.40 0.20 0.00 NEP

0.16 -0.2 0 -0.20

ET

-2 -1

ET (g H2 O m s )

(b) ET 200 8 0.12 0.08 0.04

10 0.00 0.00 WUENEP (mgCO 2 g H2 O)

(c) WUE NEP 2008

-1

6 2 -2 WUENEP

-6 2008-5-1

2008-5-31

2008-6-30

2008-7-30

2008-8-29

2008-9-28

Date (Year-Month-Day) Fig. 4. Seasonal variation of daytime net ecosystem productivity (NEP, a), evapotranspiration (ET, b) and ecosystem water use efficiency (WUE, the ratio of NEP to ET) for the sparse vineyard during 2008. The data from 8:00 to 17:00 were adopted.

4. Discussion 4.1. Comparison of the ecosystem WUE for the sparse vineyard and that for other ecosystems Our results indicate that ecosystem WUE of sparse vineyard was higher than that of savanna and grasslands, but lower than that of evergreen coniferous forests, wetlands, wheat and maize (Table 1). This is primarily caused by the dry climate and the sparse

-2 -1

NEP (mg CO2 m s )

0.60 0.40 0.20 0.00

NEP = 3.25 ET- 0.00 5 2

R = 0.40, F=87, p=0.01<0.05 -0.20 0.00

0.04

0.08

0.12

0.16

-2 -1

ET (g H 2 O m s ) Fig. 5. Relationship between daytime net ecosystem productivity (NEP) and evapotranspiration (ET) for the sparse vineyard during 2008.

vegetation cover over the vineyard. According to the formula of WUE = kd (1−Es /ET)/VPD given by Tanner and Sinclair (1983), it can be inferred that the ecosystem WUE was mainly dependent on crop variety, VPD and Es /ET. In our study, the vineyard is near to the Tengger desert. The climate is dry and average annual rainfall is only 164 mm, average VPD reached to 2.5 kPa The lower precipitation and higher VPD against other studies significantly increased ET and reduced vineyard WUE. Additionally, the average Es /ET was near to 50% in the vineyard based on our measurements (Zhang et al., 2008). These also significantly lowered the ecosystem WUE. Many studies have also confirmed that WUE varies drastically among different ecosystems (Lu et al., 2007; Yu et al., 2008; Hu et al., 2009). Ponton et al. (2006) found that WUE of Douglas-fir forest, aspen forest and grassland fell in sequence, so did the maximum evapotranspiration rate, which was the critical cause for WUE variation among the three ecosystems. Lu et al. (2007) analyzed annual average WUE of major ecosystems in west China, in sequence of forests in the mountainous regions > arbor and shrubs in deserts > irrigated farmland > grassland > deserts and Gobi. Yu et al. (2008) indicated that WUE of mixed broadleaf-conifer forest in the temperate zone in north China was higher than that in the coniferous forest and broad leaved evergreen forest in subtropics in south China and that WUE of the two ecosystems had different responses to climatic changes. Compared with the vineyard WUE at leaf scale, the value of WUE at ecosystem was significantly lower than that at leaf scale.

30

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

Chaves et al. (2010) indicated that the vineyard WUE at leaf scale varied from 45 mg g−1 to 93 mg g−1 for five vineyard cultivars. This is due to the great impact imposed by soil layer, and the reducing effect in carbon sequestration produced by soil respiration. Thus, the crop variety, climate type and vegetation cover determine the ecosystem WUE value. 4.2. Response of the ecosystem WUE for the sparse vineyard to climatic changes How the ecosystem WUE varies with the global climate changes is the hot topic at present (Keenan et al., 2013; Medlyn and De Kauwe, 2013). In the study we found that the vineyard NEP varied gently and WUE decreased remarkably when net radiation increased (Fig. 6a). These results disagreed with the previous studies (Law et al., 2002; Zhao et al., 2007; Yu, 2007; Hu et al., 2008). Law et al. (2002) indicated a positive relation between gross photosynthesis and photosynthetic photon flux density for a deciduous forest. Zhao et al. (2007) also found that the NEP for a wheat field showed a strong increase when radiation went up. Yu (2007) illustrated that the function between WUE at leaf scale and radiation is similar to the light-photosynthesis curve. WUE also has the light saturation point. Below the critical point, WUE will rise with increase of light intensity, whereas, when it has exceeded the critical point, it will remain unchanged. This study, however, is different from the foregoing pattern, mainly due to that the vineyard falls into the category of sparse vegetation, in which case, soil respiration has important influence on CO2 exchange process and is highly sensitive to radiation and the ground temperature change caused by radiation change. With the rising radiation, respiration increases rapidly, which will counteract the carbon gain by photosynthesis and reduce the net CO2 absorbing capacity (NEE). However the vineyard ET will increase under the rising radiation condition. As a result, WUE decreases with a significantly linear function with Rn . Our study also indicates that the vineyard WUE was negatively related to air temperature (Fig. 6b), which disagrees with previous studies (Zhao et al., 2007; Liu and Tao, 2013). Temperature rise may increase or decrease the photosynthetic rate of crops, thereby to influence accumulation of dry matter. At the same time, temperature change will influence the evapotranspiration process at the crop level through influencing leaf stomatal conductance and soil evaporation (Collino et al., 2005; Avola et al., 2008; Gratani et al., 2009). As predicted in some studies, air temperature rise would increase the maximum evapotranspiration of farmland by 50–70 mm where winter wheat grows during its growth period in north China, 8–12% higher than that under the current climate. The actual ET may increase by 1–2%, which thereby will reduce WUE of wheat (Guo et al., 2010, Liu and Tao, 2013). Therefore, the relationship between WUE and temperature varies under different environments. Whereas, this research confirms that WUE of sparse vineyard presented a notably linear trend along with the rising air temperature, which is obviously distinct from previous studies. This may be because that soil respiration increased significantly with the temperature rise. The percentage of coverage on the surface of the vineyard was lower than 50%, so soil respiration was the critical constituent for biosphere–atmosphere CO2 exchange. Besides, soil respiration was extremely sensitive to temperature. These factors interacted to yield results in this study. Our study also confirmed a negative relationship between vineyard WUE and VPD (Fig. 6c), which is in line with previous studies (Tanner and Sinclair, 1983; Law et al., 2002; Ponton et al., 2006; Kuglitsch et al., 2008; Yang et al., 2010). Previous studies indicate that VPD mainly affects the transpiration rate, but insignificantly photosynthesis (Hu et al., 2008). The higher the

VPD is, the greater the driving force of atmospheric air will be, which will remarkably increase the transpiration rate, resulting in rapid fall of WUE. Our results confirmed these studies (Tanner and Sinclair, 1983; Law et al., 2002; Ponton et al., 2006; Kuglitsch et al., 2008). We found that vineyard WUE showed a significantly positive relationship against air CO2 concentration (Fig. 6d), which agrees with previous studies (Keenan et al., 2013; Medlyn and De Kauwe, 2013). In a latest paper published by Nature magazine, Keenan et al. (2013) reported that the water-use efficiency for temperate and boreal forests of the Northern Hemisphere increased substantially over the past two decades, and the observed increase was most consistent with a strong CO2 fertilization effect. It is generally accepted that when CO2 concentration rises, stomata close partially, which will increase resistance and restrain spread of H2 O and CO2 . However, the reduced transpiration is greater than the reduced photosynthesis. The concentration gradient of CO2 is increased, which can promote photosynthesis to gain effect and at the same time inhibit water consumption. Thus, WUE remarkably rises along with the rising CO2 concentration. Our research agrees well with these studies (Hu et al., 2008; Keenan et al., 2013; Medlyn and De Kauwe, 2013).

4.3. Response of vineyard WUE to physiological and soil moisture changes The study indicates that a curvilinear function existed between vineyard WUE and gc (Fig. 7a). This may be attributed to the fact that when canopy conductance was lower, CO2 flux is more sensitive to conductance variation in contrast with water vapor flux, while when the canopy conductance exceeded the critical value, CO2 flux varied gently owing to the control from non-stomatal factors in company with increase in canopy conductance (Fig. 7a1), but the water vapor flux went up linearly when the canopy conductance increased. As a result, the ratio declined. Besides, our results also confirmed the important role of LAI in regulating vineyard WUE (Fig. 7b). However, many studies indicated that the variation in WUE was mainly controlled by VPD rather than LAI (Baldocchi, 1994; Scanlon and Albertson, 2004; Ponton et al., 2006). However, in another study, Hu et al. (2008) showed that the development of LAI exerted great effect in improving the WUE for grasslands. Our study agrees well with Hu et al. (2008). The reason is that the vineyard belongs to sparse ecosystem, and the increase in LAI would significantly lower Es /ET, enhance photosynthetic ability and reduce soil respiration simultaneously (Tanner and Sinclair, 1983). Our study also indicates that an approximately parabola relationship exist between the vineyard WUE and soil moisture (Fig. 7c). The leaf WUE may increase properly, mainly owing that to the ratio of intercellular CO2 concentration to atmospheric CO2 concentration (Ci /Ca ) which will decrease accordingly to maintain the photosynthetic rate when stomatal conductance reduces, thereby to raise WUE (Chaves et al., 2002). In addition, under the decreased stomatal conductance, if the internal impedance resistance remains unchanged, WUE will also rise. So far, a lot of experiments have confirmed that WUE of leaves will rise moderately under proper drought conditions (Chaves et al., 2002, 2009, 2010;; Medrano et al., 2010). Many studies at the ecosystem level have also confirmed that under moderate drought, the canopy conductance will fall and ecosystem WUE rise (Hu et al., 2009). Yet if drought is further intensified, the stomatal conductance will be reduced and also the photosynthetic capacity of leaves, thereby to increase the internal impedance, in which case, WUE may remain unchanged or fall, which was confirmed by this study(Fig. 7c3).

0.60

NEP (mg CO2 m s )

a1 NEP VS Rn

b1 NEP VS Ta

NEP =-E-04Rn +0.25

-2 -1

2

0.40

c1 NEP VS VPD

NEP =2.6E-03Ta +0.26

NEP = 3E-05Ca +0.18

d1 NEP VS Ca 2

R = 0.05, F=11, p=0.001<0.05

R = 0.01, F=0.80, p=0.37>0.05

R = 0.01, F=0.04, p=0.84>0.05

0.20 0.00

-2 -1

ET (g H2 O m s )

a2 ET VS Rn

ET= 2E-04 Rn +0.01

0.12

d2 ET VS Ca

2

2

R = 0.15, F=22, p=0.01<0.05

R = 0.26, F=48 , p=0.01 <0.05

ET= -E-04Ca +0.14 2

R = 0.19, F=30.0, p=0.01<0.05

0.08 0.04 0.00 0.00 10.0

WUENEP (mgCO 2 g H2 O)

ET= 0.01VPD +0.03

c2 ET VS VPD

ET= 3E-03 Ta -0.01

b2 ET VS Ta

2

R = 0.33, F=66, p=0.01<0.05

a3 WUE VS Rn 7.5

b3 WUE VS Ta

VPD<2: WUE= -0.02Rn +8.5 R = 0.46, F=32, p=0.00<0.05 VPD>2: WUE = -0.01 Rn + 5.0 R = 0.12, F=8.75, p=0.04<0.05

c3 WUE VSVPD Rn<300: WUE= -2.25 VPD +8.47

VPD<2: WUE= -0.38Ta +12 R = 0.36 , F=22, p=0.00<0.05

d3 WUE VS Ca

R = 0.56 , F=130 , p=0.00 <0.05

VPD>2: WUE= -0.21Ta +8 R = 0.20, F=15, p=0.00<0.05

WUE= 0.02Ca -5.86 R = 0.27, F=38 , p=0.01 <0.05

Rn>300: WUE= -1.32VPD +6.14 R = 0.48 , F=130, p=0.00 <0.05

5.0 2.5 0
0

150

0
0
0.0 300

450 -2

Rn (W m )

10 60 10

16

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

-0.20 0.16 -0.20

-1

NEP =-0.05VPD +0.30 2

2

R = 0.01, F=1.28, p=0.26>0.05

300
28

30

1

2 VPD (kPa)

3

4 500

530

560

590

620

-3

CO2 concentration (mg m )

Fig. 6. Response of vineyard net ecosystem productivity (NEP), evapotranspiration (ET), and ecosystem water use efficiency (WUE) to variation in net radiation (a1, a2 and a3), air temperature (b1, b2 and b3), water vapor pressure deficit (c1, c2 and c3) and atmospheric CO2 concentration (d1, d2 and d3). F and p represent the significance of the regression equation (P < 0.05, F > 0.05, significant). The statistical parameters were obtained from SPSS13.0.

31

32

S. Li et al. / Agricultural Water Management 148 (2015) 24–33

0.60 -2 -1

NEP (mg CO2 m s )

g cc a1 NEP VS G

c1 NEP VS Soil moisture

b1 NEP VS LAI

0.40 0.20 0.00

NEP =0.20LAI -0.045 2

R = 0.90, F=1176, p=0.0<0.05 2.00 -0.20

b2 ET/EI0 VS LAI

gc a2 ET/EI0 VS Gc

ET/ET0

1.50

c2 ET VS Soil moisture

ET/ET0 = 0.40LAI +0.30 2

R = 0.70, F=165, p=0.00<0.05 1.00

-1

WUENEP (mgCO 2 g H2O)

0.50 0.00 10.0 0.00

c3 WUE VS Soil moisture

b3 WUE VS LAI

g cc a3 WUE VS G

WUE = 2.2 LAI + 0.5

7.5

2

WUE = -6.3 F(θ) + 8.6 F(θ) + 1.5

2

R = 0.40, F=70, p=0.01 <0.05

2

R = 0.1

5.0 2.5

2

WUE= -0.00 4Gc +0.36 Gc+1.7 2

R = 0.37 0.0 0

5

10

15

20

20.0 0.0

0.5

gGc c (mm s -1 )

1.0

1.5 2

2.0

-2

LAI (m m )

0.0 2.

0.2

0.4

0.6

0.8

1.0

F(θ)

Fig. 7. Response of vineyard net ecosystem productivity (NEP), evapotranspiration (ET), and ecosystem water use efficiency (WUE) to variation in canopy conductance (a1, a2 and a3), leaf area index (b1, b2 and b3), soil moisture (c1, c2 and c3). F and p represent the significance of the regression equation (p < 0.05, F > 0.05, significant). The statistical parameters were obtained from SPSS13.0.

5. Conclusion

References

The ecosystem WUE for the sparse vineyard showed a downward-parabolic trend with a mean value of 4 mg g−1 under the dry climate and sparse vegetation cover conditions. Radiation, air temperature and humidity, soil moisture, canopy conductance and leaf area index all exerted direct and indirect influences on vineyard WUE. However, the vineyard WUE decreased significantly with the rising radiation and temperature, which is different from some studies. The main reason is that the increasing soil respiration would significantly reduce the carbon gained by vegetation under these conditions. These results are important for understanding the interactions between water and carbon cycles over this sparse ecosystem. The response of water and carbon cycles to environmental and physiological changes is complex and of great challenge for hydrologists and ecologists. How the WUE for sparse ecosystem varied with the global climate changes still remains to be investigated.

Allen, L.H., Kakani, V.G., Vu, J.C.V., Boote, K.J., 2011. Elevated CO2 increases water use efficiency by sustaining photosynthesis of water-limited maize and sorghum. J. Plant Physiol. 168, 1909–1918. Avola, G., Cavallaro, V., Patanè, C., Riggi, E., 2008. Gas exchange and photosynthetic water use efficiency in response to light, CO2 concentration and temperature in Vicia faba. J. Plant Physiol. 165, 796–804. Baldocchi, D., 1994. A comparative study of mass and energy exchange rates over a closed C3 (wheat) and an open C4 (corn) crop: II. CO2 exchange and water use efficiency. Agric. For. Meteorol. 67, 291–321. Brouder, S.M., Volenec, J.J., 2008. Impact of climate change on crop nutrient and water use efficiencies. Physiol. Plant. 133, 705–724. 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. Chaves, M.M., Flexas, J., Pinheiro, C., 2009. Photosynthesis under drought and salt stress: regulation mechanisms from whole plant to cell. Ann. Bot. 103, 551– 560. Chaves, M.M., Zarrouk, O., Francisco, R., Costa, J.M., Santos, T., Regalado, A.P., Rodrigues, M.L., Lopes, C.M., 2010. Grapevine under deficit irrigation: hints from physiological and molecular data. Ann. Bot. 105, 661–676. Clement, R.J., Jarvis, P.G., Moncrieff, J.B., 2012. Carbon dioxide exchange of a Sitka spruce plantation in Scotland over five years. Agric. For. Meteorol. 153, 106–123. Collino, D.J., Dardanelli, J.L., De Luca, M.J., Racca, R.W., 2005. Temperature and water availability effects on radiation and water use efficiencies in alfalfa (Medicago sativa L.). Anim. Prod. Sci. 45, 383–390. Falge, E., Baldocchi, D.D., Olson, R., Anthoni, P., Aubinet, M., Bernhofer, C., Burba, G., Ceulemans, R., Clement, R., Dolman, H., Granier, A., Gross, P., Grünwald, T., Hollinger, D., Jensen, N.-O., Katul, G., Keronen, P., Kowalski, A., Ta Lai, C., Law, B.E., Meyers, T., Moncrieff, J., Moors, E., Munger, J.W., Pilegaard, K., Rannik, Ü., Rebmann, C., Suyker, A., Tenhunen, J., Tu, K., Verma, S., Vesala, T., Wilson, K., Wofsy, S., 2001. Gap filling strategies for long term energy flux data sets. Agric. For. Meteorol. 107, 71–77. Finnigan, J.J., Clement, R., Malhi, Y., Leuning, R., Cleugh, H.A., 2003. A re-evaluation of long-term flux measurement techniques. Part I: Averaging and coordinate rotation. Bound. Layer Meteorol. 107, 1–48.

Acknowledgments We greatly appreciate the careful and precise comments by the reviewers and editors, and the literatures provided by the reviewers, especially the book written by Tanner and Sinclair (1983). This work was financially supported by Chinese National Natural Science Fund (51321001, 51379206 and 51009137), the National High Tech Research Plan (2011AA100502), and 111 Project funded by the Ministry of Education of the People’s Republic of China (B14002).

S. Li et al. / Agricultural Water Management 148 (2015) 24–33 Gratani, L., Varone, L., Crescente, M.F., 2009. Photosynthetic activity and water use efficiency of dune species: the influence of air temperature on functioning. Photosynthetica 47, 575–585. Green, T.R., Yu, Q., Ma, L., Wang, T.D., 2010. Crop water use efficiency at multiple scales. Agric. Water Manag. 97, 1099–1101. Grewal, H.S., Maheshwari, B., Parks, S.E., 2011. Water and nutrient use efficiency of a low-cost hydroponic greenhouse for a cucumber crop: an Australian case study. Agric. Water Manag. 98, 841–846. Guo, R., Lin, Z., Mo, X., Yang, C., 2010. Responses of crop yield and water use efficiency to climate change in the North China Plain. Agric. Water Manag. 97, 1185–1194. Herbst, M., Kutsch, W.L., Hummelshøj, P., Jensen, N.O., Kappen, L., 2002. Canopy physiology: interpreting the variations in eddy fluxes of water vapour and carbon dioxide observed over a beech forest. Basic Appl. Ecol. 3, 157–169. Hu, Z., Yu, G., Fu, Y., Sun, X., Li, Y., Shi, P., Wang, Y., Zheng, Z., 2008. Effects of vegetation control on ecosystem water use efficiency within and among four grassland ecosystems in China. Glob. Change Biol. 14, 1609–1619. Hu, Z., Yu, G., Wang, Q., Zhao, F., 2009. Ecosystem level water use efficiency: a review. Acta Ecol. Sin. 29, 1498–1507. Hunt, J.E., Kelliher, F.M., McSeveny, T.M., Byers, J.N., 2002. Evaporation and carbon dioxide exchange between the atmosphere and a tussock grassland during a summer drought. Agric. For. Meteorol. 111, 65–82. Keenan, T.F., Hollinger, D.Y., Bohrer, G., Dragoni, D., Munger, J.W., Schmid, H.P., Richardson, A.D., 2013. Increase in forest water-use efficiency as atmospheric carbon dioxide concentrations rise. Nature 499, 324–327. Kuglitsch, F.G., Reichstein, M., Beer, C., Carrara, A., Ceulemans, R., Granier, A., Janssens, I.A., Koestner, B., Lindroth, A., Loustau, D., Matteucci, G., Montagnani, L., Moors, E.J., Papale, D., Pilegaard, K., Rambal, S., Rebmann, C., Schulze, E.D., Seufert, G., Verbeeck, H., 2008. Characterisation of ecosystem water-use efficiency of European forests from eddy covariance measurements. Biogeosci. Discuss. 5, 4481–4519. Law, B.E., Falge, E., Gu, L., et al., 2002. Environmental controls over carbon dioxide and water vapor exchange of terrestrial vegetation. Agric. For. Meteorol. 113, 97–120. Li, S., Kang, S., Li, F., Zhang, L., 2008. Evapotranspiration and crop coefficient of spring maize with plastic mulch using eddy covariance in northwest China. Agric. Water Manag. 95, 1214–1222. Li, S., Kang, S., Zhang, L., Li, F., Du, T., Tong, L., Wang, S., Lu, H., 2012. Greater effect of canopy conductance in regulating the energy partition above the maize field in arid northwest China. Hydrol. Process., http://dx.doi.org/10.1002/hy p.9413. Li, S., Kang, S., Zhang, L., Li, F., Hao, X., Ortega-Farias, S., Guo, W., Ji, S., Wang, J., Jiang, X., 2013a. Quantifying the combined effects of climatic, crop and soil factors on surface resistance in a maize field. J. Hydrol. 489, 124–134. Li, S., Hao, X., Du, T., Tong, L., Zhang, J., Kang, S., 2013b. A coupled surface resistance model to estimate crop evapotranspiration in arid region of northwest China. Hydrol. Process., http://dx.doi.org/10.1002/hyp.9768. Liu, Y., Tao, F., 2013. Response of crop water use efficiency to elevated temperature and CO2 concentration—a review. Progress Geogr. 32, 416–424. Lu, L., Li, X., Huang, C., Frank, V., 2007. Analysis of the spatio-temporal characteristics of water use efficiency of vegetation in west China. J. Glaciol. Geocryol. 29, 777–784. Mahrt, L., Vickers, D., 2002. Relationship of area-averaged carbon dioxide and water vapour fluxes to atmospheric variables. Agric. For. Meteorol. 112, 195–202. Medlyn, B., De Kauwe, M., 2013. Biogeochemistry: carbon dioxide and water use in forests. Nature 499, 287–289. Medrano, H., Flexas, J., Ribas-Carbó, M., Gulías, J., 2010. Measuring water use efficiency in grapevines. In: Delrot, S., Medrano, H., Or, E., Bavaresco, L., Grando,

33

S. (Eds.), Methodologies and Results in Grapevine Research. Springer, Netherlands, pp. 123–134. Monteith, J.L., 1965. Evaporation and environment. In: Fogg, G.E. (Ed.), The State and Movement of Water in Living Organisms: Symposium no. 19 of the Society for Experimental Biology. , pp. 205–234. Paw, U.K.T., Baldocchi, D.D., Meyers, T.P., Wilson, K.B., 2000. Correction of eddy covariance measurements incorporating both advective effects and density fluxes. Bound. Layer Meteorol. 97, 487–511. Ponton, S., Flanagan, L.B., Alstad, K.P., Johnson, B.G., Morgenstern, K.A.I., Kljun, N., Black, T., Barr, A.G., 2006. Comparison of ecosystem water use efficiency among Douglas fir forest, aspen forest and grassland using eddy covariance and carbon isotope techniques. Glob. Change Biol. 12, 294–310. Scanlon, T.M., Albertson, J.D., 2004. Canopy scale measurements of CO2 and water vapor exchange along a precipitation gradient in southern Africa. Glob. Change Biol. 10, 329–341. Stewart, J.B., 1988. Modelling surface conductance of pine forest. Agric. For. Meteorol. 43, 19–35. Tanner, B.D., Greene, J.P., 1989. Measurement of Sensible Heat and Water-Vapor Fluxes using Eddy-Correlation Methods. Final Report Prepared for U.S. Army Dugway Proving Grounds. U.S. Army, Dugway, Utah, pp. 17. Tanner, C.B., Sinclair, T.R., 1983. Efficient water use in crop production: research or re-search? In: Taylor et al, H. (Ed.), Limitations to Efficient Water Use in Crop Production. American Society Agronomy, Madison, WI, pp. 1–28. Tao, F., Hayashi, Y., Zhang, Z., Sakamoto, T., Yokozawa, M., 2008. Global warming, rice production and water use in China: developing a probabilistic assessment. Agric. For. Meteorol. 148, 94–110. Thom, A.S., 1972. Momentum, mass and heat exchange of vegetation. Q. J. R. Meteorol. Soc. 98, 124–134. Twine, T.E., Kustas, W.P., Norman, J.M., Cook, D.R., Houser, P.R., Meyers, T.P., Prueger, J.H., Starks, P.J., Wesely, M.L., 2000. Correcting eddy-covariance flux underestimates over a grassland. Agric. For. Meteorol. 103, 279–300. Verhoef, A., Allen, S.J., De Bruin, H.A., Jacobs, C.M., Heusinkveld, B.G., 1996. Fluxes of carbon dioxide and water vapour from a Sahelian savanna. Agric. For. Meteorol. 80, 231–248. Webb, E.K., Pearman, G.I., Leuning, R., 1980. Correction of flux measurements for density effects due to heat and water vapor. Q. J. R. Meteorol. Soc. 106, 85–100. Wolf, A., Saliendra, N., Akshalov, K., Johnson, D.A., Laca, E., 2008. Effects of different eddy covariance correction schemes on energy balance closure and comparisons with the modified Bowen ratio system. Agric. For. Meteorol. 148, 942–952. Yang, B., Pallardy, S.G., Meyers, T.P.G.U.L.H., Hanson, P.J., Wullschleger, S.D., Heuer, M., Hosman, K.P., Riggs, J.S., Sluss, D.W., 2010. Environmental controls on water use efficiency during severe drought in an Ozark Forest in Missouri, USA. Glob. Change Biol. 16, 2252–2271. Yu, G., Song, X., Wang, Q., Liu, Y., Guan, D., Yan, J., Sun, X., Zhang, L., Wen, X., 2008. Water use efficiency of forest ecosystems in eastern China and its relations to climatic variables. New Phytol. 177, 927–937. Yu, Q., 2007. Modeling the Agro-Ecological Process in the Field. Science Press, Beijing. Zhang, B., Kang, S., Li, F., Zhang, L., 2008. Comparison of three evapotranspiration models to Bowen ratio-energy balance method for a vineyard in an arid desert region of northwest China. Agric. For. Meteorol. 148, 1629–1640. Zhao, F.H., Yu, G.R., Li, S.G., Ren, C.Y., Sun, X.M., Mi, N., Ouyang, Z., 2007. Canopy water use efficiency of winter wheat in the North China Plain. Agric. Water Manag. 93, 99–108.