Stomatal and hydraulic conductance and water use in a eucalypt plantation in Guangxi, southern China

Agricultural and Forest Meteorology 202 (2015) 61–68

Contents lists available at ScienceDirect

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

Stomatal and hydraulic conductance and water use in a eucalypt plantation in Guangxi, southern China L.W. Zhu a , P. Zhao a,∗ , Q. Wang b , G.Y. Ni a , J.F. Niu a , X.H. Zhao a , Z.Z. Zhang a , P.Q. Zhao a , J.G. Gao a , Y.Q. Huang c , D.X. Gu c , Z.F. Zhang c a Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China b Graduate School of Agriculture, Shizuoka University, Shizuoka 422-8529, Japan c Guangxi Institute of Botany, Chinese Academy of Sciences, Guilin 541006, China

a r t i c l e

i n f o

Article history: Received 11 April 2014 Received in revised form 2 December 2014 Accepted 2 December 2014 Keywords: Eucalyptus Canopy stomatal conductance Hydraulic conductance Whole-tree water use Vapor pressure deficit

a b s t r a c t Previous work has demonstrated that constant water use per unit leaf area results from the combined control of stomatal and hydraulic conductance holds for plant water use; however, few studies have ever explored the water use of a popularly planted fast-grown tree species, eucalyptus, in southern China. In this study, seasonal variations in hydraulic traits and ecophysiological parameters were monitored via the sap flux, leaf water potential ( leaf ) and associated environmental variables to investigate the water use of a five-year-old Eucalyptus grandis × Eucalyptus urophylla plantation in Guangxi province, China, in July and October 2012 and January and April 2013. The results show that predawn leaf was similar among all months, suggesting an abundance of soil water in the study site. There was a significant seasonal variation in midday leaf . Moreover, canopy stomatal conductance (Gs ) was higher in July than in October and linearly decreased with the natural logarithm of the vapor pressure deficit (ln(VPD)). Furthermore, the hydraulic conductance from soil to leaves (K) per unit sapwood area in the dry season (October) was relatively high compared to that in the wet season (July). Whole-tree water use per day was estimated to be 7.7 kg d−1 and 6.7 kg d−1 in July and October, respectively, and linearly increased with leaf area. However, the slope of the regression line between whole-tree water use per day and leaf area was similar in July and October, clearly indicating that this eucalyptus stand had a constant water use per unit leaf area, which confirms the generally accepted concept. The findings of this study should help address the increasing ecohydrological and water resource concerns related to the rapid expansion of Eucalyptus spp. plantations in southern China, which recently underwent a severe drought. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Eucalyptus trees, which represent a fast-growing species, have been planted extensively in many parts of the world to meet the increasing demand for timber and fiber. It is reported that the Eucalyptus plantation area reached 3.68 million ha in China by the end of 2010 (data from the China Eucalyptus Research Centre), most of which are located in southern China. In Guangxi province, the Eucalyptus planting area has expanded to 1.8 million ha in the recent 10 years, making this province the largest eucalypt plantation region in China (data from the State Forestry

∗ Corresponding author at: South China Botanical Garden, Chinese Academy of Sciences, Xingke Road 723, Tianhe District, Guangzhou 510650, China. Tel.: +86 20 37252881. E-mail address: [email protected] (P. Zhao). http://dx.doi.org/10.1016/j.agrformet.2014.12.003 0168-1923/© 2014 Elsevier B.V. All rights reserved.

Administration, China). Because plant productivity is typically related to water resource accessibility (Sands and Nambiar, 1984; Hunt et al., 2006), increasing concerns about the water balance in catchments has resulted from the high transpiration rates of Eucalyptus species and from uncertain ecohydrological effects on non-native environments (Bari and Schofield, 1992; Bleby et al., 2012). Particularly, the severe drought that began in late 2009 caused serious water shortages and adverse social consequences in Guangxi province (Cao et al., 2012; Guo et al., 2013). Therefore, quantifying water use and understanding ecophysiological factors of local eucalypt plantations appear to be critical issues, especially in an area that has experienced more serious drought conditions in recent years. A clear understanding of plant water use has been stressed for modeling water fluxes across large spatial scales (Zeppel, 2013). Several previous studies have reported that there is a constant relationship between water use and leaf area (Hatton et al., 1998;

62

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

Meinzer, 2003; O’Grady, 2009). Manzoni et al. (2013b) further demonstrated that transpiration is a conservative process across plant types. As for eucalypt species, Hatton et al. (1998) observed a constant water use per unit leaf area under limited soil water conditions. However, whether such relationship holds for the expanding Eucalyptus plantation in southern China remains unclear. Furthermore, to clearly explain this behavior, plant water use and the underlying regulating mechanisms, which range from leaf physiological parameters to the plant hydraulic structure, should be comprehensively studied. An improved understanding of the roles of stomatal conductance and hydraulic conductance on plant water use will help to understand ecohydrological relationships across large spatial scales because stomatal conductance responds to various environmental stimuli, such as light, vapor pressure deficit and water status at the root (Buckley, 2005). Furthermore, previous work has shown that stomatal closure avoids high transpiration rates that would otherwise occur due to increasing vapor pressure deficits (Saliendra et al., 1995). Whitehead and Beadle (2004) indicated that stomata close in response to a decreasing water potential and an increasing vapor pressure deficit. However, studies have shown that the response sensitivity of stomatal conductance to the vapor pressure deficit varies among tree species (Oren et al., 1999). Furthermore, adaptation of hydraulic architecture may contribute to survival in dry habitats (Stout and Sala, 2003). Several studies have indicated that plants evolve coordination mechanisms to regulate stomatal aperture and water transport in the xylem (Manzoni et al., 2013a; Domec et al., 2004, 2006, 2012). Bleby et al. (2012) has investigated the water use of eucalyptus in a seasonally dry forest and reported that hydraulic and physiological traits vary according to size and growth conditions. As environmental factors change, especially soil water availability, transpiration rates are synergistically affected by stomatal conductance and whole-tree hydraulic conductance (Cruiziat et al., 2002; Manzoni et al., 2013a; Hacke et al., 2000; Katul et al., 2003). In this study, we examined seasonal variations in leaf water potential, canopy stomatal conductance, soil-to-leaf hydraulic conductance and water use in five-year-old Eucalyptus grandis × Eucalyptus urophylla trees to explore whether a constant water use per unit leaf area is valid for this widely planted species in southern China. To accomplish his goal, an analysis based on the relationship between whole-tree water use and leaf area was performed in both the wet and dry seasons. The focal point of plant water use, i.e., the sensitivity of canopy stomatal conductance to the vapor pressure deficit, was explicitly investigated to evaluate the environmental control on water loss. Finally, the seasonal changes in canopy stomatal conductance and hydraulic conductance were used to explain characteristics of water use in the studied Eucalyptus plantation.

2. Materials and methods 2.1. Site description The study was conducted in a five-year-old E. grandis × E. urophylla plantation at Huangmian Forest Farm in Liuzhou, Guangxi province, China (latitude: 24◦ 45.8 N; longitude: 109◦ 53.6 E; altitude: 226 m) in July and October 2012 and January and April 2013. The area is located in the transitional zone from China’s low subtropics to tropics, which is characterized by a monsoon climate. Annual rainfall varies from 1750 mm to 2000 mm and is seasonally unevenly distributed; most precipitation falls between April and August. The mean annual temperature is approximately 21 ◦ C; the minimum and maximum mean monthly temperatures are approximately 10 ◦ C in January and 29 ◦ C in July, respectively.

A 400-m2 experimental plot was established in the plantation on a steep northwest-facing slope of 28◦ for long-term ecological research. The plot density was 1425 trees per hectare, which were planted in a soil of heavy loam and a pH of 3.5; the organic content and total nitrogen content of the soil 2.8% and 0.1%, respectively. The understory vegetation, which primarily included Miscanthus floridulus and Rhus chinensis, was less developed. A 23.5-m-high tower with a horizontal area of 2 m × 2 m was constructed inside the experimental plot to provide an anchor station to mount various environmental sensors (described below) and as a platform for leaf water potential measurements. During the measurement period, the sampled trees were averaged to be approximately 11.5 m high. The mean leaf area index was estimated to be 1.8 m2 m−2 . Moreover, the maximum photosynthetically active radiation (PAR) and the average soil volumetric water content were 1552.6 (±167.8) ␮mol m−2 s−1 and 0.32 (±0.03) m3 m−3 , respectively.

2.2. Monitoring of environmental factors and leaf area index The soil volumetric water content at a depth of 30 cm was monitored using a SM300 sensor (Delta-T Devices, Ltd., Cambridge, UK). A SKP215 quantum sensor (Sky Instruments Ltd., Powys, U.K.), a wind speed sensor (AN4-05, Casella, Ltd., UK) and an AT2&RHT2 sensor (Delta-T Devices, Ltd., Cambridge, UK) were mounted over the tower for monitoring PAR, wind speed and both air temperature and relative humidity, respectively. Environmental factors were sampled every 30 s; their 10-min means were recorded in a data logger (DL2e, Delta-T Devices, Ltd., Cambridge, UK). The vapor pressure deficit (VPD) was calculated from the measured air temperature and air relative humidity as follows (Campbell and Norman, 1998): VPD = a exp

 bT  (T + c)

(1 − RH) ,

(1)

where T is the air temperature (◦ C), RH is the air relative humidity and a, b and c are constants set to 0.611, 17.502 and 240.97, respectively. In addition to these climatic variables, the leaf area index (LAI) was measured every month using an LAI 2000 plant canopy analyzer (Li-Cor, Inc., Lincoln, NE). We randomly captured image data using the analyzer at 10–15 sample points within the experimental plot. We subsequently calculated and recorded the average LAI values for the plantation.

2.3. Sap flux and canopy stomatal conductance The sap flux was monitored using self-made thermal dissipation probes following Granier’s prototype (1987). This type of sensor comprises two probes with a length of 20 mm and a diameter of 2.0 mm. The upper probe of the sensor contains a copperconstantan thermocouple and a heating element of constantan, which is continuously heated with a constant power of 0.2 W; the unheated lower probe serves as a temperature reference. During the field campaign, the two probes were typically inserted radially into the stem at 10–15 cm apart and 1.3 m above the ground on the northern side of a sampled tree. The probes were covered with plastic locket and insulated with aluminum film to avoid mechanical disturbance and direct solar heating. 15 trees were selected for sap flux measurements based on the distribution of tree diameters shown in Fig. 1. In addition, the characteristics of each sampled tree are also presented in Table 1. While obtaining the measurements, the voltage outputs of each pair of probes were synchronously recorded with environmental factors.

300

Precipitation Evapotranspiration

250

5.1-7.0

7.1-9.0

9.1-11.0 Diameter (cm)

>11.1

Table 1 Tree height, diameter at the breast height (DBH), sapwood area and leaf area of the sampled trees. No.

Tree height (m)

DBH (cm)

Sapwood area (m2 × 10−3 )

Leaf area (m2 )

1 2 3 4 5 6 7 8 9 10 11* 12* 13* 14* 15*

10.1 10.8 11.9 12.2 11.5 11.9 11.2 10.9 13.7 11.6 8.6 9.4 10.4 14.4 13.3

8.7 10.6 10.4 8.1 10.6 10.1 8.9 9.4 9.1 8.4 8.3 9.9 8.5 12.3 11.6

5.4 7.8 7.5 4.7 7.8 7.2 5.7 6.3 6.0 5.0 4.9 6.9 5.2 9.9 9.0

15.4 22.2 21.4 13.1 22.3 20.5 16.0 18.0 16.9 14.2 13.7 19.8 14.6 28.3 25.8

1.231

,

(2)

where TM is the maximum temperature difference under zeroflux conditions and T is the temperature difference between the heated and the unheated probes. The sapwood area and leaf area of the sampled trees were estimated from the allometric equation. The estimation was based on several felled trees with different diameters at the breast height (DBH, 1.3 m above the ground) outside the experimental plot. The sapwood depth was directly measured on the sawed stem disk from the felled trees. There was a large color difference in the heartwood. The leaf area was immediately measured using a leaf area meter (LI3000, Li-Cor, USA) for each leaf from which the total leaf area was determined. Finally, the allometric equations for the sapwood area and the leaf area using the DBH were respectively determined to be: As = 0.1102DBH 2 + 10.139DBH − 42.132 AL = 3.6027DBH − 15.985





R2 = 0.97,

R2 = 0.94,





n=7 ,

20 150 15 100

10 5

0

0 Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.

Fig. 2. Mean monthly changes in precipitation, air temperature and evapotranspiration during the period 1971–2000 in Liuzhou of Guangxi province, China (data were obtained from the China Meteorological Data Sharing Service System).

Gs =

The temperature difference between the two probes was used to calculate the sap flux density Js (g H2 O m−2 s−1 ) according to the empirical relationship of Granier (1985, 1987): (TM − T ) T

30

The canopy stomatal conductance (Gs ) was determined using a simplified model based on the work of Monteith and Unsworth (1990):

* Indicates the trees selected for leaf water potential measurements. The measurements were performed in June 2012.



35

25

200

50

Fig. 1. Frequency distribution of tree diameters at the experimental site. All diameters were measured at 1.3 m above the ground.

Js = 119

Air temperature

63

Air temperature ( )

40 35 30 25 20 15 10 5 0

Precipitation (mm)

Individual number of trees

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

(3)

n=7 ,

where As is the sapwood area (10−4 m2 ), AL is the individual tree leaf area (m2 ). The whole-tree sap flux (g s−1 ) was estimated by multiplying the sap flux density by the sapwood area, while the daily water use per tree was calculated as the integral of the daily courses of the whole-tree sap flux.

aEc , VPD

(4)

where Gs is the average canopy stomatal conductance (mmol m−2 s−1 ), and a is the constant obtained from the psychrometric constant, the latent heat of vaporization of water, the specific heat of air and the density of liquid water. Moreover, Ec is the canopy transpiration, which was estimated from the sap flux density and expressed on a ground area basis. Because the eucalypt stand canopy is relatively open, air can circulate with little obstruction in windy conditions. Therefore, it is logical to assume that there is no vertical gradient in VPD throughout the canopy, which suggests that Eq. (4) is applicable in this study. This simplification is strengthened by Kumagai et al. (2008), who integrated the response of the canopy stomatal conductance to environmental variables and proved that this simplification was applicable for evaluating the environmental control on canopy transpiration. Furthermore, empirical relationships between Gs and VPD provide a convenient approach to explore the response of stomatal conductance to varying atmospheric conditions (Jarvis, 1976; Whitehead, 1998). Accordingly, the response of Gs to VPD can be described by: Gs = −m ln VPD + b,

(5)

where b is a reference canopy stomatal conductance for VPD = 1 kPa and −m is the sensitivity of Gs to VPD, which is relatively constant over a range of VPD (Oren et al., 1999).

2.4. Leaf water potentials Leaf water potentials ( leaf ) were measured at both predawn and midday on leaf-bearing branches of five sampled trees that were near the tower using PMS-1000 pressure chambers (PMS Instrument, Corvallis, OR, USA) for 4–5 consecutive days in each of the four studied months (i.e., July and October in 2012 and January and April in 2013); seasonal variations in rainfall and air temperature were accounted for (Fig. 2). Branches with leaves were cut and subsequently transferred into a pressure chamber to determine  leaf ; all measurements were typically completed within 10–15 min. Three replicates were made for each of the sampled trees.

64

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

2.5. Hydraulic conductance Water movement from the soil to leaves was described according to Ohm’s law (Cochard et al., 1996), from which the hydraulic conductance from the soil to leaves can be determined as: K=

Es , (soil − leaf )

(6)

where K is the apparent hydraulic conductance from the soil to a plant’s leaves, Es is the transpiration rate during the period in which leaf water potential measurements were obtained and is expressed on a per unit sapwood area or leaf area basis (mmol m−2 s−1 ),  soil is the average soil water potential, which was represented by the predawn leaf water potential because the measurements were conducted in sunny days; therefore the soil water potential remained relatively constant throughout the entire day and  leaf is the average leaf water potential. Wullschleger et al. (1998) reviewed whole-tree water use and found that the range of the hydraulic conductance per unit sapwood area was smaller than the specific-leaf area hydraulic conductance. In addition, the hydraulic conductance per unit sapwood area was found to be useful measure of the hydraulic efficiency of stems (Manzoni et al., 2013b). Therefore, we adopted the sapwood area as the unit of the soil to leaf hydraulic conductance in this study. 2.6. Statistical analysis The correlation between whole-tree water use and either the leaf area or the vapor pressure deficit was analyzed by performing a linear regression; the differences between the slopes from two months were examined via univariate analysis of variance. Seasonal variations in the studied variables (i.e., environmental factors, leaf water potentials, canopy stomatal conductance and hydraulic conductance) were analyzed by one way ANOVA or a t-test for independent samples. 3. Results 3.1. Seasonal and diurnal variations in environmental factors As illustrated in Fig. 2, the observed air temperature, precipitation and evapotranspiration exhibited pronounced seasonal variations. Rainfall primarily occurred between April and August, accounting for approximately 72% of the annual precipitation. Precipitation appeared to outweigh evapotranspiration from April to July, which produced a relatively humid environment. On the contrary, evapotranspiration exceeded rainfall by 133% during the period September–January, resulting in relatively dry conditions. Therefore, the sap flux data for the four studied months (i.e., January, April, July and October) were clearly obtained under different air temperature and water conditions and were properly selected to analyze water relationships for E. grandis × E. urophylla trees over seasonal time scales. The soil volumetric water content () in October was significantly lower than in the other months (0.25 m3 m−3 vs.

0.34 m3 m−3 , P < 0.05). Based on archived data from 1971 to 2000, rainfall and evapotranspiration in July were 209.7 mm and 202.5 mm, respectively, compared with 65.3 mm and 161.1 mm in October. These conditions clearly indicate a large difference in the soil water status for these two months, which are usually selected to represent wet and dry seasons. Daily courses of environmental variables in July and October are illustrated in Fig. 3. The soil moisture exhibited only small diurnal changes; however, the season difference was much larger (i.e., between the wet and dry seasons). On the contrary, pronounced daily wind speed variations were observed. Furthermore, the evaporative demand in October was higher than in July (Fig. 3b and f, n = 4, P < 0.05), averaging 1.8 ± 0.1 kPa and 1.3 ± 0.1 kPa, respectively.

3.2. Canopy stomatal conductance, leaf water potentials and hydraulic conductance The mean canopy stomatal conductance (Gs ) for the condition VPD > 1 kPa in July was greater than that in October (100.5 ± 5.2 mmol m−2 s−1 vs. 57.2 ± 2.2 mmol m−2 s−1 , P < 0.01). Gs decreased with increasing VPD; a linear relationship between Gs and ln(VPD) was fitted (Fig. 4). According to Eq. (5), the sensitivity coefficient for Gs with respect to ln(VPD) was 63.5 mmol m−2 s−1 ln(kPa)−1 ; Gs was 102.8 mmol m−2 s−1 for VPD = 1 kPa. Although  at a depth of 30 cm varied seasonally, no significant difference in the predawn leaf was observed (Table 2, P > 0.05), which was estimated to have only a marginal difference of −0.2 MPa. There was no significant correlation between predawn leaf and  (P > 0.05, data not shown), indicating that the roots of the studied Eucalyptus trees may have extend beyond a depth of 30 cm and used additional water resources from deeper soils. The mean midday leaf was −0.89 MPa, −0.83 MPa, −1.20 MPa and −1.45 MPa in January, April, July and October, respectively, and the midday leaf exhibited a negative relationship with VPD (Fig. 5, P < 0.01). This relationship indicates that the midday leaf decreased with increasing VPD, decreasing to −1.6 MPa as VPD increased to approximately 2 kPa in October. By excluding the data for cloudy days, we found significant differences in the midday leaf among the four studied months, i.e., lower values (more negative) were observed in July and October, while higher values were detected in January and April (Table 2, P < 0.05). As a result, the water potential gradient between leaves and the soil ( ) also varied significantly between the seasons (P < 0.05). The mean hydraulic conductance per unit sapwood area (K) was estimated to be 2.18 ± 0.49 mmol m−2 s−1 MPa−1 , 1.48 ± 0.42 mmol m−2 s−1 MPa−1 , 1.74 ± 0.24 mmol m−2 s−1 MPa−1 and 2.00 ± 0.60 mmol m−2 s−1 MPa−1 in January, April, July and October, respectively. K significantly differed among the four studied months, exhibiting higher values in January than in April and July (Table 2, n = 4, P < 0.01). K in October was similar to that observed in January, remaining relatively high compared with the values measured in April and July; however, these differences were not statistically significant (P > 0.05).

Table 2 The predawn leaf water potential ( leaf ), midday  leaf , water potential gradient between leaves and the soil ( ) and hydraulic conductance per unit sapwood area (K) for Eucalyptus grandis × E. urophylla in January, April, July and October.

Jan. Apr Jul. Oct.

Predawn  leaf (MPa)

Midday  leaf (MPa)



−0.28(0.03)a −0.21(0.03)a −0.18(0.01)a −0.24(0.03)a

−0.89b −0.83(0.14)b −1.20(0.11)a −1.45(0.16)a

0.65(0.04)b 0.54(0.11)b 1.01(0.10)a 1.22(0.09)a

(MPa)

Values in parenthesis are standard deviations, while letters denote significant differences at the 0.05 level. The standard deviation for midday because there was only one sunny day in the five consecutive experimental days for midday leaf measurements.

K (mmol m−2 s−1 MPa−1 ) 2.18(0.49)a 1.48(0.42)b 1.74(0.24)b 2.00(0.60)ab leaf

is not shown in January

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

PAR (μmol m -2 s-1 )

2500

a

Jul.

e

65

Oct.

2000 1500 1000 500 0

4

b f

3.5 VPD (KPa)

3 2.5 2 1.5 1 0.5 0 Soil volumetric water content (%)

38

j

c

36 34 32 30 28 26 24

Wind speed (m s-1 )

22

4 3.5 3 2.5 2 1.5 1 0.5 0

h

d

Fig. 3. Daily changes in photosynthetically active radiation (PAR), vapor pressure deficit (VPD), soil volumetric water content () and wind speed (v) in July and October. The results are based on July 9–13 (a–d) and October 11–14 (e–h), respectively. Only 4 days were used in October because heavy rain fell on the fifth day.

3.3. Plant water use The leaf area index (LAI) did not significantly differ between July and October (n = 3, P > 0.05), having values of 1.75 ± 0.13 m2 m−2 and 1.92 ± 0.04 m2 m−2 , respectively. The mean whole-tree transpiration was estimated to be 7.4 ± 0.3 kg d−1 and 6.7 ± 0.3 kg d−1 in July and October, respectively; however, the difference was not significant (P > 0.05). The mean daily whole-tree transpiration of the sampled trees was found to linearly increase with increasing mean daily VPD (Fig. 6). VPD alone explained 78% and 93% of the variations in the whole-tree transpiration in July and October, respectively. However, there was no significant difference in the slopes of the mean daily whole-tree transpiration with respect to VPD in July and October. The intercepts were clearly different, which might be due to the large difference in soil moisture between the two months (Fig. 3c and j). Moreover, the whole-tree water use per day, although exhibiting a linear relationship with the leaf area, had slopes (based on linear regressions) of 0.45 kg d−1 m−2 and

0.42 kg d−1 m−2 in July and October, respectively; these slopes were not statistically significant (Fig. 7, P > 0.05). 4. Discussion 4.1. Water use in planted eucalypt The whole-tree transpiration in this study was estimated based on sap flux measurements. To obtain accurate sap flux-based whole-tree transpiration data, the radial variation in the sap flux density with the sapwood depth should be considered (Ewers et al., 2002; Zeppel et al., 2011). On the one hand, the mean sapwood depth of the studied Eucalyptus was deeper than the 2 cm probes (3.2 ± 0.07 cm), which may have avoided the 2 cm-long probes were placed into the heartwood. Besides, our previous studies demonstrated that the sap flux density at 0–2 cm depth of the sapwood was similar with that at 2–4 cm depth (data not shown). On the other hand, Clearwater et al. (1999) found that thermal dissipation probes

66

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

120

50

Gs (mmol m-2 s-1)

100

y = -63.521x + 102.84 R² = 0.941 1

0 0

80

0.2

0.4 0.6 ln (VPD)

0.8

1

60

Water use (kg d-1)

140

14

Jul. E=0.4471AL-0.0421, R2=0.74

12

Oct. E=0.4223AL-3.1475, R2=0.71

10 8 6 4 2

40

0

20

0

0 1

0.5

1.5 VPD (KPa)

2

0.5

1

VPD (KPa) 1.5

2

2.5

3

3.5

Midday leaf water potential (MPa)

0 -0.2 y = -0.4763x - 0.327 R² = 0.8989

-0.4 -0.6 -0.8 -1 -1.2 -1.4 -1.6 -1.8

Fig. 5. Correlation between the midday leaf water potential ( sure deficit (VPD) during the study periods.

leaf )

and vapor pres-

Mean water use (Kg d-1)

can correctly estimate the sap flux density for diffuse-porous tree species. Therefore, although the variations in the sap flux with the sapwood depth were not monitored, we assumed that the radial gradients in the sap flux density within the sapwood for this E. grandis × E. urophylla plantation could be ignored. As a result, the whole-tree transpiration could be estimated using the sap flux density multiplied by the sapwood area. The Huber value, which is often used to study the hydraulic efficiency, is expressed as the ratio of the sapwood area to the leaf area. Previous studies have reported Huber values (As :AL )

10 9 8 7 6 5 4 3 2 1 0

Oct. y = 2.9911x + 1.4212 R2=0.9273 Jul. y = 2.3057x + 4.3511 R2=0.7835

0

0.5

1 1.5 VPD (KPa)

2

20

30

40

AL (m2)

2.5

Fig. 4. Relationship between the canopy stomatal conductance (Gs ) and the vapor pressure deficit (VPD). Gs data are the daytime maximum values estimated from the sap flux measurements on fine days (July 9–13 and October 11–14) The inset figure denotes the relationship between Gs and ln (VPD) (data for VPD < 1 kPa were excluded).

0

10

Fig. 7. The relationship between the leaf area (AL ) and the whole-tree water use per day. Each data point is the daily sum of the sap flux for each sampled tree on sunny days when the daily VPD exceeded 1 kPa.

of 2 × 10−4 –16 × 10−4 m2 m−2 based on four habitats in Sydney (heathland, two woodlands and a mangrove) (Macinnis-Ng et al., 2004). Previous work has also found that for desert tree species and mountainous tree species, the leaf area to sapwood area ratio is 0.12 m2 cm−2 and 0.20 m2 cm−2 (As :AL values of 5 × 10−4 –8.3 × 10−4 m2 m−2 ), respectively (Maherali and DeLucia, 2001). In our study, Huber values were obtained only in July and October; the values in these two months were both approximately 3.3 × 10−4 m 2 m−2 (P > 0.05, n = 15, data not shown in Section 3). The values observed in the current study are relatively small compared with those determined by Maherali and DeLucia (2001) and fall within the range of values observed by Macinnis-Ng et al. (2004). Bleby et al. (2012) reported that water use of Eucalyptus in a seasonally dry forest with a mixed-aged tree species fell between 0.5 kg d−1 and 6 kg d−1 . In this E. grandis × E. urophylla plantation, the whole-tree water use was found to be 7.7 kg d−1 and 6.7 kg d−1 in July and October, respectively. One of the reasons for the relatively small water use may be the less efficient transport system resulting from smaller Huber values. The water demands of the plant are determined by LAI and environmental conditions (Zeppel et al., 2008). The constant LAI indicated the ecohydrological equilibrium between plant water use and soil water availability (Bleby et al., 2012; Specht, 1972; Nemani and Running, 1989). In our study, the similar LAI between the two months might demonstrate an equilibrium state with the soil environment. Furthermore, Ewers et al. (2002) found that the daily sums of transpiration could be exponentially explained by VPD. Similar studies were also conducted by Oren et al. (1999) and Pataki et al. (2000), indicating that the variations in transpiration with VPD were influenced by soil moisture. We also observed a similar increase in the daily wholetree transpiration with increasing VPD between the two months, which might have resulted from the abundant soil water availability. In Maherali and DeLucia (2001), VPD reached 6 kPa, which probably resulted in relatively high Huber values. In our study, the maximum VPD was approximately 3 kPa, which is relatively small. This difference in VPD is another possible reason for the relatively small whole-tree water use per day found in the current work. 4.2. Conservative water use per unit leaf area of planted eucalypt

2.5

Fig. 6. The relationship between the daily average vapor pressure deficit during daytime hours (VPD) and the mean whole-tree water use per day. Each point shows the mean of the sampled trees (n = 13, the data from two sampled trees can’t be used due to sensor failure); the bars denote standard deviations.

The predawn  leaf has been typically regarded as an approximate indicator of soil water availability (Baldocchi et al., 2004). In this study, the predawn  leaf was similar among the four months, revealing that the studied plants accessed water in soil layers where little change in the soil water content occurred (White et al., 2000). Mielke et al. (2000) indicated that the distribution of the active root system can be determined by the regression coefficient between the predawn  leaf and the soil water content at different soil depths

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

for a clonal eucalypt plantation. In our study, although the predawn  leaf was not correlated with soil moisture at a depth of 30 cm, profile measurements of the soil water content at different soil depths were not performed. Therefore, we could not arbitrarily illustrate the active root locations. White et al. (2000) found that the stomatal conductance was related to the predawn  leaf when the studied trees were exposed to water stress. Bucci et al. (2009) found that the predawn  leaf varied from −4 MPa to −1 MPa for shallow-rooted species and deep-rooted species. In our studied eucalypt plantation, the predawn  leaf was less negative (i.e., −0.2 kPa). Additionally, we did not find a relationship between the stomatal conductance and the predawn  leaf (data not shown). These results further indicate that this eucalypt plantation was not under water-stressed conditions. Many eucalypt species have been reported to have constant water use per unit leaf area (Eberbach, 2003; Hatton et al., 1998; Zeppel and Eamus, 2008). In our study, similar variations in the whole-tree transpiration with the leaf area for July and October also imply a constant water use per unit leaf area. Hatton and Wu (1995) proposed a conceptual model for the relationship between the whole-tree water use and the leaf area, in which a linear regression can be used when the soil water is freely available. Hatton et al. (1998) explored the water use behavior of eucalypts and found that the tree transpiration per unit leaf area was constant among their studied trees. In their study, two theories were employed to explain the uniformity of the water use per unit leaf area. The first theory is the nutrients theory, which suggests that the water function of plants varies little once the plants have adapted to local nutrient conditions. The second theory explains that plants maintain constant cost in all leaves according to optimum theory. Eucalyptus plantations have been present for more than 50 years in southern China. Moreover, long-term observations have shown that their water use is relatively stable (Qi, 2002; Lane et al., 2004). In our experimental site, soil nutrient status was at a normal level (organic content: 2.8%; total nitrogen content: 0.1%). Therefore, environmental variables exerted no restriction on plant transpiration where abundant soil water was available. Moreover, optimization could be achieved by similar leaf water efficiencies, which were maintained by adjusting the leaf area index between the two months (1.75 m2 m−2 in July and 1.92 m2 m−2 in October). In addition to the aforementioned theories, this effect might be observed because the hydraulic conductance per unit sapwood area at noon in October was relatively high compared with that in July even though Gs was larger in July than in October. This point is discussed in the following paragraph concerning isohydry and anisohydry. The constant water use per unit leaf area could be achieved by stomatal regulation in coordination with hydraulic adaption between the seasons. Plants respond differently to fluctuating environmental conditions; therefore, plants are generally divided into two categories: isohydry and anisohydry. Isohydric plants have strong stomatal control that results in a relatively constant midday  leaf , while anisohydric plants often have weak stomatal sensitivity to VPD and soil water content, resulting in large fluctuations in the midday  leaf (Turner et al., 1984; Franks et al., 2007). Bucci et al. (2009) indicated that tree species with deeper roots exhibited an isohydric response in dry-wet seasons. Franks et al. (2007) compared these two plant types and concluded that isohydric plants had a strong stomatal control on transpiration, while anisohydric plants had high leaf gas exchange rates. In our study, there were significant seasonal variations in the midday leaf and in the soil to leaf water potential gradients; however, the predawn leaf was constant. The reason for the steady predawn  leaf might be that deep root system of E. grandis × E. urophylla enabled the trees to gain access to abundant soil water in deep soil layers. The midday leaf decreased to −1.5 MPa in the dry season (October). Therefore, this plantation

67

is assumed to be an anisohydric form of water status control. Additionally, regarding Gs in this experiment, the daytime maximum Gs was used to examine sensitivity of E. grandis × E. urophylla to VPD, which could avoid the effects of limited solar irradiance. Körner and Cochrane (1985) found that the sensitivity coefficient of Gs to VPD for Eucalyptus pauciflora was 194 mmol m−2 s−1 ln(kPa)−1 at the leaf level. Oren et al. (1999) reviewed the sensitivity coefficient of Gs to VPD for tree species with different vascular structures (e.g., diffuse-porous trees and ring-porous trees) and showed that it varied from 23 mmol m−2 s−1 ln(kPa)−1 to 92 mmol m−2 s−1 ln(kPa)−1 for the diffuse-porous tree species. In our study, the sensitivity coefficient of E. grandis × E. urophylla, which is a diffuse-porous tree species, was 63.5 mmol m−2 s−1 ln(kPa)−1 , which is within the range suggested by Oren et al. (1999). Under the condition of nonlimiting light, ln(VPD) accounted for the variations in Gs (R2 = 0.94). Oren et al. (1999) revealed a consistent relationship between Gs for VPD = 1 kPa (Gsres ) and the sensitivity of the stomatal response to increasing VPD (−dGs /ln VPD). They studied a wide variety of mesic adapted species and found that the slope of −dGs /ln VPD versus Gsres was 0.6. For this Eucalyptus plantation, the slope of −dGs /ln VPD versus Gsres was 0.62, which is very close to 0.6. Our result again generalized this finding that the sensitivity of the stomata to VPD could be calculated using Gs for VPD = 1 kPa. Therefore, the above discussion has demonstrated that there were not only large seasonal fluctuations in the midday leaf but also relatively strong stomatal regulation of water loss for this E. grandis × E. urophylla plantation. In other words, this plantation is not only an isohydric form, but also an anisohydric form of water status control. Franks et al. (2007) also found anisohydric but isohydrodynamic regulation of the plant water status when they studied the seasonal variations in plant water use and hydraulic properties in a Eucalyptus gomphocephala natural stand. In their study, midday  leaf values largely varied among the studied seasons; however, the water potential gradient from the soil to leaves was relatively constant. In Bleby et al. (2012), the slope of the linear relationship between the water use per day and the leaf area was 0.27 kg d−1 m−2 for eucalyptus at natural sites in southwestern Australia. Hatton et al. (1998) reviewed this topic and found slopes between 0.86 kg d−1 m−2 and 3.80 kg d−1 m−2 . They attributed the small slope to limited soil water conditions. Our results are inconsistent with those presented in Hatton et al. (1998). Although the plantation in our study did not suffer soil water stress, the slope was relatively small (i.e., 0.44 kg d−1 m−2 ). Therefore, it is necessary to explore the water use efficiency on a large spatial scale to generalize these results. We conclude that both the water use of E. grandis × E. urophylla and the leaf area exhibit similar variations in different seasons, which facilitates foresters to regulate tree planting density and hydrologists to assess hydrological behavior based on canopy development. Moreover, biometric scalars, such as the leaf area, are often applied to convert sap flux measurements to the stand level (Wullschleger et al., 1998). A constant relationship between plant water use and leaf area would allow modelers to predict water fluxes across large regions. However, a challenge for scaling plant water use to the stand level is the variability in water use as a function of tree age (Irvine et al., 2004; Ryan et al., 2000, 2006). Thus, further studies are required to test this hypothesis across a wide range of tree ages.

5. Conclusions In this study, we explored water use of a E. grandis × E. urophylla plantation by investigating seasonal variations in the canopy stomatal conductance, leaf water potential, sap flux and hydraulic traits. Without soil moisture limitations, seasonal variations in the midday leaf were consistent with typical anisohydric behavior.

68

L.W. Zhu et al. / Agricultural and Forest Meteorology 202 (2015) 61–68

Moreover, a linear correlation between Gs and ln VPD demonstrated a stomatal control of plant water status. There were seasonal variations in Gs and K. Gs was smaller due to higher VPD, while K was relatively high in the dry season (October), providing strong support for seasonal characteristics of whole-tree water use. A linear relationship between the whole-tree water use per day and the leaf area was observed in this study; the degree of variation was constant among the studied months. This result facilitates modelers for selecting appropriate scalars to predict water balance at larger scales. Conflict of interest statement The authors declare that there are no conflicts of interest. Acknowledgments This study was sponsored by the National Natural Scinece Foundation of China (41030638, 41275169, 31170673, 31300335), the Natural Science Foundation of Guangdong province, China (S2012020010933) and the Doctoral Starting up Foundation of South China Botanical Garden, Chinese academy of sciences (201231). References Baldocchi, D.D., Xu, L.K., Kiang, N., 2004. How plant functional-type, weather, seasonal drought, and soil physical properties alter water and energy fluxes of an oak-glass savanna and annual grassland. Agric. For. Meteorol. 123, 13–39. Bari, M.A., Schofield, N.J., 1992. Lowering of a shallow, saline water table by extensive eucalypt reforestation. J. Hydrol. 133, 273–291. Bleby, T.M., Colquhoun, I.J., Adams, M.A., 2012. Hydraulic traits and water use of Eucalyptus on restored versus natural sites in a seasonally dry forest in southwestern Australia. For. Ecol. Manage. 274, 58–66. Bucci, S.J., Scholz, F.G., Goldstein, G., et al., 2009. Soil water availability and rooting depth as determinants of hydraulic architecture of Patagonian woody species. Oecologia 160, 631–641. Buckley, T.N., 2005. The control of stomata by water balance. New Phytol. 168, 275–291. Campbell, G.S., Norman, J.M., 1998. An Introduction to Environmental Biophysics, second ed. Springer-Verlag, New York, NY, pp. 286. Cao, C.X., Zhao, J., Gong, P., et al., 2012. Wetland changes and droughts in southwestern China. Geomat. Nat. Hazard. Risk 3, 79–95. Clearwater, M.J., Meinzer, F.C., Andrade, J.L., et al., 1999. Potential errors in measurement of nonuniform sap flow using heat dissipation probes. Tree Physiol. 19, 681–687. Cochard, H., Breda, N., Granier, A., 1996. Whole tree hydraulic conductance and water loss regulation in Quercus during drought: evidence for stomatal control of embolism? Ann. For. Sci. 53, 197–206. Cruiziat, P., Cochard, H., Ameglio, T., 2002. Hydraulic architecture of trees: main concepts and results. Ann. For. Sci. 59, 723–752. Domec, B., Jansen, S., Brodribb, T.J., et al., 2012. Global convergence in the vulnerability of forest to drought. Nature 491, 752–755. Domec, J.C., Warren, J.M., Meinzer, F.C., et al., 2004. Native root xylem embolism and stomatal closure in stands of Douglas-fir and ponderosa pine: mitigation by hydraulic redistribution. Oecologia 141, 7–16. Domec, J.C., Scholz, F.G., Bucci, S.J., et al., 2006. Diurnal and seasonal variation in root xylem embolism in neotropical savanna woody species: impact on stomatal control of plant water status. Plant Cell Environ. 29, 26–35. Eberbach, P.L., 2003. The eco-hydrology of partly cleared, native ecosystems in southern Australia: a review. Plant Soil. 257, 357–369. Ewers, B.E., Mackay, D.S., Gower, S.T., et al., 2002. Tree species effects on stand transpiration in northern Wisconsin. Water Resour. Res. 38 (8–1), 8–11. Franks, P.J., Drake, P.L., Froend, R.H., 2007. Anisohydric but isohydrodynamic: seasonally constant plant water potential gradient explained by a stomatal control mechanism incorporating variable plant hydraulic conductance. Plant Cell Environ. 30, 19–30. Granier, A., 1987. Evaluation of transpiration in a Douglas fir stands by means of sap flow measurements. Tree Physiol. 3, 309–319. Guo, F., Jiang, G.H., Yuan, D.X., et al., 2013. Evolution of major environmental geological problems in karst areas of southwestern China. Environ. Earth Sci. 69, 2427–2435. Hacke, U.G., Sperry, J.S., Ewers, B.E., et al., 2000. Influence of soil porosity on water use in Pinus taeda. Oecologia 124, 495–505. Hatton, T.J., Wu, H.I., 1995. Scaling theory to extrapolate individual tree water use to stand water use. Hydrol. Processes 9, 527–540. Hatton, T., Reece, P., Taylor, P., et al., 1998. Does leaf water efficiency vary among eucalypts in water-limited environments? Tree Physiol. 18, 529–536.

Hunt, R.J., Strand, M., Walker, J.F., 2006. Measuring groundwater surface water interaction and its effect on wetland stream benthic productivity, Trout Lake watershed, northern Wisconsin, USA. J. Hydrol. 320, 370–384. Irvine, J., Law, B.E., Kurpius, M.R., et al., 2004. Age-related changes in ecosystem structure and function and effects on water and carbon exchange in ponderosa pine. Tree Physiol. 24, 753–763. Jarvis, P.G., 1976. Interpretation of variations in leaf water potential and stomatal conductance found in canopies in field. Philos. Trans. R. Soc. B 273, 593–610. Katul, G., Leuning, R., Oren, R., 2003. Relationship between plant hydraulic and biochemical properties derived from a steady-state coupled water and carbon transport model. Plant Cell Environ. 26, 339–350. Kumagai, T., Tateishi, M., Shimizu, T., et al., 2008. Transpiration and canopy conductance at two slope positions in a Japanese cedar forest watershed. Agric. For. Meteorol. 148, 1444–1455. Körner, C., Cochrane, P.M., 1985. Stomatal responses and water relations of Eucalyptus pauciflora in summer along an elevational gradient. Oecologia 66, 443–455. Lane, P.N.J., Morris, J., Zhang, N.N., 2004. Water balance of tropical eucalypt plantations in south-eastern China. Agric. For. Meteorol. 124, 253–267. Macinnis-Ng, C., McClenahan, K., Eamus, D., 2004. Convergence in hydraulic architecture, water relations and primary productivity amongst habitats and across seasons in Sydney. Funct. Plant Biol. 31, 429–439. Maherali, H., DeLucia, E.H., 2001. Influence of climate-driven shifts in biomass allocation on water transpiration and storage in ponderosa pine. Oecologia 129, 481–491. Manzoni, S., Vico, G., Katul, G., et al., 2013a. Biological constraints on water transport in the soil plant atmosphere system. Adv. Water Res. 51, 292–304. Manzoni, S., Vico, G., Katul, G., et al., 2013b. Hydraulic limits on maximum plant transpiration and the emergence of the safety-efficiency trade-off. New Phytol. 198, 169–178. Meinzer, F.C., 2003. Functional convergence in plant responses to the environment. Oecologia 134, 37–45. Mielke, M.S., Oliva, M.A., deBarros, N.F., et al., 2000. Leaf gas exchange in a clonal eucalypt plantation as related to soil moisture, leaf water potential and microclimate variables. Trees-struct. Funct. 14, 263–270. Monteith, J.L., Unsworth, M.H., 1990. Principles of Environmental Physics. Arnold, London. Nemani, R.R., Running, S.W., 1989. Testing a theoretical climate-soil-leaf area hydrologic equilibrium of forests using satellite data and ecosystem simulation. Agric. For. Meteorol. 44, 245–260. O’Grady, A.P., Cook, P.G., Eamus, D., et al., 2009. Convergence of tree water use within an arid-zone woodland. Oecologia 160, 643–655. Oren, R., Sperry, J.S., Katul, G.G., et al., 1999. Survey and synthesis of intra- and interspecific variation in stomatal sensitivity to vapor pressure deficit. Plant Cell Environ. 22, 1515–1526. Pataki, D.E., Oren, R., Smith, W.K., 2000. Sap flux of co-occurring species in a western subalpine forest during seasonal soil drought. Ecology 81, 2557–2566. Qi, S.X., 2002. Eucalyptus in China, second ed. Chinese Forestry Press, Beijing, pp. 517p. Ryan, M.G., Bond, B.J., Law, B.E., et al., 2000. Transpiration and whole-tree conductance in ponderosa pine trees of different heights. Oecologia 124, 553–560. Ryan, M.G., Phillips, N., Bond, B.J., 2006. The hydraulic limitation hypothesis revisited. Plant Cell Environ. 29, 367–381. Saliendra, N.Z., Sperry, J.S., Comstock, J.P., 1995. Influence of leaf water status on stomatal response to humidity, hydraulic conductance, and soil drought in Betula occidentalis. Planta 196, 357–366. Sands, R., Nambiar, E.K.S., 1984. Water relations of Pinusradiata in competition with weeds. Can. J. For. Res. 14, 233–237. Specht, R.L., 1972. Water use by perennial evergreen plant communications in Australia and Papus New Guinea. Aust. J. Bot. 20, 273–299. Stout, D.L., Sala, A., 2003. Xylem vulnerability to cavitation in Pseudotsuga menziesii and Pinus ponderosa from contrasting habitats. Tree Physiol. 23, 43–50. Turner, N.C., Schulze, E.D., Gollan, T., 1984. The response of stomata and leaf gas exchange to vapor pressure deficits and soil water content I. Species comparisons at high soil water contents. Oecologia 63, 338–342. White, D.A., Turner, N.C., Galbraith, J.H., 2000. Leaf water relations and stomatal behavior of four allopatric Eucalyptus species planted in Mediterranean southwestern Australia. Tree Physiol. 20, 1157–1165. Whitehead, D., 1998. Regulation of stomatal conductance and transpiration in forest canopies. Tree Physiol. 18, 633–644. Whitehead, D., Beadle, C.L., 2004. Physiological regulation of productivity and water use in Eucalyptus: a review. For. Ecol. Manage. 193, 113–140. Wullschleger, S.D., Meinzer, F.C., Vertessy, R.A., 1998. A review of whole-plant water use studied in trees. Tree Physiol. 18, 499–512. Zeppel, M., Eamus, D., 2008. Coordination of leaf area, sapwood area and canopy conductance leads to species convergence of tree water use in a remnant evergreen woodland. Aus. J. Bot. 56, 97–108. Zeppel, M., Macinnis-Ng, C., Palmer, A., Taylor, D., Whitey, R., Fuentes, S., Yunusa, I., Williams, M., Eamus, D., 2008. An analysis of the sensitivity of sap flux to soil and plant variables assessed for an Australian woodland using a soil-plantatmosphere model. Funct. Plant Biol. 35, 509–520. Zeppel, M.J.B., Lewis, J.D., Medlyn, B., Barton, C.V.M., Duursma, R.A., Eamus, D., Adams, M.A., Philips, N., Ellsworth, D.S., Forster, M.A., Tissue, D.T., 2011. Interactive effects of elevated CO2 and drought on nocturnal water fluxes in Eucalyptus saligna. Tree Physiol. 31, 932–944. Zeppel, M., 2013. Convergence of tree water use and hydraulic architecture in waterlimited regions: a review and synthesis. Ecohydrology 6, 889–900.