Carbon, nitrogen, and hydrogen isotope ratios in creekside trees in western Kansas

Carbon, nitrogen, and hydrogen isotope ratios in creekside trees in western Kansas

Environmental and Experimental Botany 71 (2011) 1–9 Contents lists available at ScienceDirect Environmental and Experimental Botany journal homepage...

510KB Sizes 0 Downloads 0 Views

Environmental and Experimental Botany 71 (2011) 1–9

Contents lists available at ScienceDirect

Environmental and Experimental Botany journal homepage: www.elsevier.com/locate/envexpbot

Carbon, nitrogen, and hydrogen isotope ratios in creekside trees in western Kansas Brian R. Maricle a,∗ , Sam R. Zwenger a,1 , Raymond W. Lee b a b

Department of Biological Sciences, Fort Hays State University, 600 Park Street, Hays, KS 67601-4099, USA School of Biological Science, Washington State University, Pullman, WA 99164-4236, USA

a r t i c l e

i n f o

Article history: Received 17 December 2009 Received in revised form 23 September 2010 Accepted 28 September 2010 Keywords: Celtis occidentalis Lonicera tatarica Morus alba Riparian zone Stable isotopes

a b s t r a c t Three species of creekside trees were monitored weekly during the 2007 and 2008 growing seasons. The 2007 growing season was wet early, but became drier as the season progressed. In contrast, the 2008 growing season was dry early, but became wetter as the season progressed. Creekside trees were measured to determine effects of changing water regimes on leaf-level processes. Lonicera tatarica plants were compared to Morus alba and Celtis occidentalis trees. Leaves were monitored for changes in stomatal conductance, transpiration, ␦13 C, ␦15 N, ␦D, leaf temperature, and heat losses via latent, sensible, and radiative pathways. ␦D of creek water was more similar to ground water than to rain water, but the creek was partially influenced by summer rains. ␦D of bulk leaf material was significantly higher in individuals of C. occidentalis compared to the other species, consistent with source water coming from shallower soil layers. Despite decreasing water levels, none of these tree species showed signs of water stress. There were no significant differences between species in stomatal conductance or transpiration. Leaf ␦13 C was significantly lower in individuals of L. tatarica compared to the other species. Differences in ␦13 C were attributed to a lower carboxylation capacity, consistent with lower leaf nitrogen content in L. tatarica plants. Leaf ␦15 N was significantly lower in individuals of L. tatarica compared to the other species, consistent with uptake of a different N source. Two of the three sites appeared to be affected by inorganic N from fertilizer run-off. Evidence is presented that these species acquired water and nitrogen from different sources, resulting from differences in root uptake patterns. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Water status has long been an important issue in plant ecology. Many ecosystems have been examined with response to drought conditions, including floodplains (Cleverly et al., 1997), tallgrass prairies (Nippert and Knapp, 2007a), shortgrass prairies (Tomanek and Hulett, 1970), estuaries (Visser et al., 2002), wetland hummocks (Davis et al., 2005), flooded forests (Herrera et al., 2008), and crop systems (Neumann, 2008). In some cases, physiological processes such as photosynthesis and response to water status have been used to explain invasiveness (Kercher and Zedler, 2004). Water status is also useful in describing vegetation shifts. Major changes in vegetation communities in Kansas (Kuchler, 1974) and the Great Plains of North America (Lauenroth et al., 1999) are attributable to precipitation patterns and seasonal droughts. In addition, drought is the most common cause of crop failure. Histor-

∗ Corresponding author. Tel.: +1 785 628 5367; fax: +1 785 628 4153. E-mail address: [email protected] (B.R. Maricle). 1 Present address: School of Biological Sciences, University of Northern Colorado, Greeley, CO 80639, USA. 0098-8472/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.envexpbot.2010.09.015

ically, over 25% of soils in the USA have been affected by drought, and 40.8% of all insurance claims for crop loss in the USA from 1939 to 1978 resulted from drought (Boyer, 1982). Understanding responses of plants to conditions of decreasing water supply is clearly of great ecological and economic importance. Vegetation shifts in relation to water supply can also happen temporally, across years or within a season. During drought, plants of Pinus sylvestris L. were observed to acquire water from deeper soil layers compared to wetter periods (Brandes et al., 2007). Similarly, deeper-rooted prairie species are able to use water from deeper soil layers, while shallow-rooted grasses are restricted to surface soil moisture (Nippert and Knapp, 2007a). In some cases, soil water is partitioned by different rooting depths between species (Jackson et al., 1999). Plant root architecture is often optimized to acquire limiting nutrients (Drew, 1975) or water (Bucci et al., 2009). In the present study, three species of trees were studied near a creek during two growing seasons. Trees were compared to examine responses to drying soils, and how this might be different between trees with different rooting depths. Moisture in shallow soils is replenished quickly, even with small rainfall events. However, shallow soils often become dry when water is limiting (Nippert and Knapp,

2

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

2007a). In contrast, deeper soils remain moist, but require more water and time to replenish moisture (Bucci et al., 2009). Thus, species with deeper roots are expected to have a more reliable source of water, whereas shallow-rooted species have quick access to water from small rainfall events, but shallow soils are more subject to drying (Nippert and Knapp, 2007a). Stable isotope ratios are useful for determining hydrologic sources of water (Gat, 1996) and plant uptake of water (Ehleringer and Dawson, 1992). Precipitation and surface water ␦18 O and ␦D are strongly correlated with temperature (Yapp and Epstein, 1982), as this relates to increased evaporation. ␦18 O and ␦D also become higher with closer proximity to marine sources of heavy O and H (Gat, 1996). Thus, isotopic values in rain water are heaviest over the oceans, where ␦18 O and ␦D are high from the source sea water (Ehleringer and Dawson, 1992). Heavier rain water falls first (i.e., near the coasts), and lighter water remains to fall later as the clouds move farther inland (Gat, 1996) or northward to cooler temperatures (Yapp and Epstein, 1982; Bowen et al., 2005). Increases of ␦D and ␦18 O in plants occur in response to evaporation and from gradients in leaf to atmosphere isotope values of water vapor (Farquhar et al., 2007; Barbour, 2007). ␦18 O and ␦D of xylem water in plants represents ␦18 O and ␦D of soil water (Ehleringer and Dawson, 1992). Xylem water then enters leaves, and ␦18 O and ␦D increase from evaporative loss of lighter isotopes (Farquhar et al., 2007; Brandes et al., 2007). These values of ␦18 O and ␦D are then recorded in leaf organic material (Farquhar et al., 2007; Grams et al., 2007). Evaporative enrichment of water isotopes is especially notable in arid environments (e.g., English et al., 2007). Therefore, measures of ␦18 O and ␦D in leaf material of creekside trees can provide valuable information on water status and evaporative conditions during the formation of that leaf. This study was conducted to investigate patterns of water uptake in creekside trees during changing water conditions. Effects of soil drying on creekside trees were examined during the 2007 and 2008 growing seasons in Hays, Kansas, USA. The spring of 2007 was abnormally wet in western Kansas. Creeks were at or above their banks for much of late spring and early summer. As 2007 progressed, creek water receded to lower levels. The year 2008 was one of above-average rainfall, but much of this rain fell during late summer. In 2008, creek levels were low early in the summer, but then became higher from late-summer rains. In the present study, creekside trees were monitored to compare effects of decreasing water availability between years. Different rooting depths were expected to influence ability to acquire water during dry periods. Rooting depth has been shown to influence water source in streamside trees in Utah (Dawson and Ehleringer, 1991), in mangrove communities in Florida (Ewe et al., 2007), and in forest communities in the fog belt of northern California (Dawson, 1998). In the present study, shallow-rooted Tatarian honeysuckle (Lonicera tatarica L.) trees were expected to be affected by soil drying to a greater extent than deeper-rooted white mulberry (Morus alba L.) and hackberry (Celtis occidentalis L.) trees. Leaves were monitored for changes in stomatal conductance, transpiration, ␦13 C, ␦15 N, ␦D, leaf temperature, and heat losses via latent, sensible, and radiative pathways. These changes were monitored weekly and were related to changes in creek water ␦D.

was established at each oxbow lake. Three species of trees were growing at all three sites. One individual of each M. alba L. (white mulberry), C. occidentalis L. (hackberry), and L. tatarica L. (Tatarian honeysuckle) was identified at each site, and all were monitored weekly during the growing seasons of 2007 and 2008 in a repeated measures design. Individual plants were 2–3 m in height, and were similar in crown diameter. All plant individuals were within 3 m of the creek bed. 2.2. Plant measures Point measures of plants were taken during mid-day hours once per week. Stomatal conductance (gs ) was measured on abaxial surfaces of mature, non-senescent leaves with an SC-1 Leaf Porometer (Decagon Devices, Inc.; Pullman, WA, USA). Leaf temperature (TL ) was also measured on each plant with the leaf porometer. Maximum leaf width was recorded for each leaf, and the leaf was sampled for stable isotope analysis. Leaf transpiration rates were calculated after Campbell and Norman (1998) from measures of stomatal conductance, leaf size, leaf temperature, and wind speed. First, vapor pressure was calculated for air inside and outside the leaf. Dry bulb (Ta ) and wet bulb temperatures (Tw ) outside the leaf were measured with a digital psychrometer (Extech Instruments RH300; Waltham, MA, USA), and leaf temperature (TL ) was measured with the leaf porometer. Vapor pressure at ambient temperature [ea (Ta )] was calculated as: ea (Ta ) (kPa) = es (Tw ) − ((pa )(Ta − Tw ))

(1)

where ea (Ta ) is in kPa. In this equation,  is the psychrometer constant (6.66 × 10−4 ◦ C−1 ), pa is atmospheric pressure (94.96 kPa at the study location), and es (Tw ) is the saturation vapor pressure at the wet bulb temperature, calculated after Campbell and Norman (1998) as: es (Tw ) (kPa) = 0.611 exp

 17.502T  w Tw × 240.97

(2)

Vapor pressure inside the leaf was assumed to be saturated, calculated as: es (TL ) (kPa) = 0.611 exp

 17.502T  L TL × 240.97

(3)

Vapor pressure deficit (VPD) of the atmosphere was calculated after Campbell and Norman (1998) as: VPD (kPa) = es (Ta ) − ea (Ta )

(4)

Transpiration of water from leaves (E, in mol m−2 s−1 ) was calculated as: E (mol m−2 s−1 ) = gv ×

(es (TL ) − ea (Ta )) pa

(5)

where gv is the leaf conductance to water vapor (mol m−2 s−1 ), a combination of surface (stomatal) conductance (gs ) and boundary layer conductance to water vapor (gva ), calculated after Campbell and Norman (1998) as: gv (mol m−2 s−1 ) =

1 [(1/gs ) + (1/gva )]

(6)

Measures of gs were read from the porometer, and gva was calculated after Campbell and Norman (1998) as: 2. Materials and methods



gva (mol m−2 s−1 ) = 0.147 2.1. Study sites Three study sites were established on oxbow lakes of Big Creek on the campus of Fort Hays State University in Hays, Kansas, USA. Each oxbow lake is a separate body of water that fills and dries independently from the others. Thus, an independent field site

u d

(7)

where u is wind speed (m s−1 ), measured with an AM-4204 hot wire anemometer (Lutron Electronic Enterprise Co., Ltd.; Taipei, Taiwan), and d is the characteristic dimension of the leaf, calculated by multiplying the maximum leaf width (in meters) by 0.72 (Campbell and Norman, 1998). Air flow was laminar (Re < 5 × 105 ),

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

so a turbulence correction was not needed (Campbell and Norman, 1998). The Reynolds number (Re) was calculated as: Re (unitless) =

ud 

(8)

where  is the kinematic viscosity of air (1.55 × 10−5 m2 s−1 at 25 ◦ C). Exchanges of energy between the leaf and its surroundings were also calculated. Longwave (radiative) energy loss (Loe , in W m−2 ) from the leaf was calculated by: Loe (W m−2 ) = εL TL 4

(9)

where εL is the longwave emissivity of the leaf (taken to be 0.97, Campbell and Norman, 1998),  is the Stefan–Boltzmann constant (5.67 × 10−8 W m−2 K−4 ), and TL is leaf temperature (K). Latent heat loss (E, in W m−2 ) was calculated as E, where  is the latent heat of vaporization of water, equal to 44,000 J mol−1 . Sensible heat loss (H, in W m−2 ) was calculated after Campbell and Norman (1998) as: H (W m−2 ) = cp gHa (TL − Ta )

(10)

where cp is the specific heat of air (29.3 J mol−1 ◦ C−1 ), and gHa

is the conductance to heat loss through the boundary layer of the leaf, calculated as:



gHa (mol m−2 s−1 ) = 0.135

u d

Leaf ␦13 C values were determined relative to VPDB (Vienna Pee Dee Belemnite) and ␦15 N values were determined relative to atmospheric N2 (Ehleringer and Osmond, 1989). Leaves were sampled from each individual every week of the study. Leaves were dried at 60 ◦ C for 24 h, and milled to a fine powder in a Wiley Mill (model 3383-L10; Thomas Scientific; Swedesboro, NJ, USA). Ground leaf powder was able to pass through a 40-mesh screen. Two-mg samples were placed in tin capsules and combusted in a Eurovector elemental analyzer. N2 and CO2 gases were produced, separated by gas chromatography, and entered the inlet of a Micromass Isoprime isotope ratio mass spectrometer for determination of 13 C/12 C or 15 N/14 N ratios (R). Samples (0.35 mg) of bulk leaf tissue were placed in silver capsules and pyrolyzed for ␦D analysis. The resulting H2 gas entered the mass spectrometer for analysis. Values of ␦13 C, ␦15 N, or ␦D were determined as: Rsample Rstandard



−1

Species

Leaf ␦D (‰)

Water ␦D − leaf ␦D (‰)

Celtis occidentalis Lonicera tatarica Morus alba

−94.1 ± 2.9‰ (a) −104.2 ± 1.7‰ (b) −102.9 ± 3.6‰ (b)

56.7 ± 3.0‰ (b) 66.6 ± 1.8‰ (a) 65.1 ± 3.6‰ (a)

was measured in a Micromass Isoprime isotope ratio mass spectrometer for determination of D/H ratios. Routine precision for the instrument was ±2.0‰ for water ␦D. 2.4. Data analysis All data were analyzed using repeated measures analysis of variance (ANOVA). Species, sites, and years were fixed effects, and individual plants were the repeated effect. Bivariate correlations were used to investigate relationships between measured parameters by using Spearman’s correlation coefficient (SPSS 12.0; 2003 SPSS Inc.; Chicago, IL, USA). All analyses were performed at ˛ = 0.05. 3. Results

2.3. Stable isotope measures



Table 1 ␦D of bulk leaf tissue (‰) and discrimination against D (source ␦D − product ␦D) of Celtis occidentalis, Lonicera tatarica, and Morus alba plants in 2007. Means of 17 repeated measures are presented ± SE. Letters in each column indicate significant differences at ˛ = 0.05.

(11)

The Bowen ratio (unitless), calculated as H/E, is a comparison of heat loss via sensible and latent pathways (Campbell and Norman, 1998).

ı (‰) = 1000 ×

3

(12)

Routine precision for the instrument was ±0.06‰ for ␦13 C, ±0.4‰ for ␦15 N, and ±0.7‰ for ␦D, calculated as the standard deviation of 10 duplicate standards. Water ␦D values were determined relative to V-SMOW (Vienna Standard Mean Ocean Water) (Ehleringer and Osmond, 1989). Creek water samples were collected from each of the three sites every week of the study. A 1.5 mL polypropylene, snap-top microcentrifuge tube was filled with surface water at each site (Clark and Fritz, 1997). Wind exposure caused moderate surface flow of water, which kept the upper layers of water well mixed. Rain water samples were also collected as available during the two years of the study. Rain water was collected flowing from a rain gutter during rainfall events in Hays, Kansas, USA. Additionally, alluvial ground water samples were obtained from a well north of Hays during the study. All water samples were kept frozen until analysis. ␦D

Average rainfall in Hays, Kansas is 586 mm year−1 (Kansas State University Research and Extension data). About 80% of annual precipitation in Hays occurs from April through September (Adler and HilleRisLambers, 2008). Both 2007 and 2008 were above average in precipitation. In 2007, 786 mm of rain were recorded. The early portion of the 2007 growing season was wet, but the later portion of the growing season became drier. Precipitation was above average six of the seven months from January through July 2007, but decreased to average or below average precipitation during August and September 2007. In 2008, 692 mm of rain were recorded. The 2008 growing season was dry early, but became wet later. Precipitation was below average for five of the seven months from January through July 2008. Abundant rainfall began in late July and continued through August and again in October 2008 (Kansas State University Research and Extension data). Additionally, afternoon point measures during sampling periods indicate air temperatures (Ta ) in 2008 were significantly higher (ANOVA, p = 0.002) and vapor pressure deficit (VPD) was significantly higher in 2008 compared to 2007 (ANOVA, p < 0.001). Relative humidity (hr ) was significantly higher in 2007 compared to 2008 (ANOVA, p < 0.001). Rain water ␦D ranged from −53 to 18‰ during the study (Fig. 1), and was not different between years (ANOVA, p = 0.884). Ground water remained constant near −42‰ throughout the study (Fig. 1). ␦D of creek water ranged from −47 to −21‰, with a seasonal oscillation of heavier water (i.e., higher ␦D) during summer and lighter water (i.e., lower ␦D) during spring (Fig. 1). ␦D in creek water was significantly lower in 2007 compared to 2008 (ANOVA, p < 0.001). There was a marginally significant difference in ␦D between sites, with site 3 trending toward being the lightest, and site 2 trending toward the heaviest (ANOVA, p = 0.057). Mean ␦D in bulk leaf tissue (only measured from 2007 samples) ranged from −94 to −104‰ (Table 1), and was not different between sites (ANOVA, p = 0.553). Leaf ␦D in plants of C. occidentalis was significantly higher than in L. tatarica and M. alba individuals (Fig. 2; ANOVA, p = 0.024). Discrimination against D (creek water ␦D minus leaf ␦D, i.e., source ␦ minus product ␦; O’Leary et al., 1992) ranged from 57 to 67‰ and was significantly lower in individuals of C. occidentalis compared to L. tatarica and M. alba plants

4 Table 2 Correlation coefficients (top) and p-values (bottom) for variables in the study. n = 324 individual measurements in each case, except n = 131 for water ␦D and water ␦D − leaf ␦D. Leaf ␦D was measured in 2007, but not 2008, so it could not be correlated with year. Significant correlations are in bold font. * indicates p < 0.05; ** indicates p < 0.01. Year

Loe

Max leaf width

gv

E

H

Bowen ratio

Tleaf − Ta

Leaf ␦13 C

Leaf ␦15 N

Leaf ␦D

Water ␦D

Water ␦D − leaf ␦D

Leaf C:N ratio B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

Loe

0.035 0.528

Max leaf width

0.057 0.307

−0.067 0.232

gv

−0.218** 0.000

0.095 0.086

−0.051 0.360

E

−0.009 0.877

0.450** 0.000

−0.041 0.460

0.778** 0.000

H

−0.178** 0.001

−0.001 0.992

0.143** 0.010

−0.134* 0.015

−0.065 0.244

Bowen ratio

−0.179** 0.001

0.033 0.551

0.156** 0.005

0.049 0.376

0.144** 0.010

0.928** 0.000

Tleaf − Ta

−0.201** 0.000

−0.020 0.722

0.097 0.082

−0.109 0.050

−0.041 0.467

0.964** 0.000

0.920** 0.000

Leaf ␦13 C

0.155** 0.005

−0.060 0.283

0.312** 0.000

−0.180** 0.001

−0.151** 0.006

−0.114* 0.041

−0.138* 0.013

−0.134* 0.016

Leaf ␦15 N

0.025 0.649

0.332** 0.000

0.175** 0.002

0.123* 0.027

0.176** 0.002

0.114* 0.041

0.116* 0.037

0.090 0.108

0.147** 0.008

Leaf ␦D



0.117 0.184

0.030 0.732

0.148 0.091

0.216* 0.013

0.159 0.070

0.195* 0.026

0.174* 0.047

0.022 0.805

0.172* 0.049

Water ␦D

0.578** 0.000

0.049 0.382

0.059 0.294

−0.221** 0.000

−0.118* 0.036

−0.089 0.113

−0.150** 0.008

−0.118* 0.036

0.226** 0.000

−0.054 0.341

−0.091 0.302

Water ␦D − leaf ␦D



−0.107 0.226

−0.039 0.662

−0.145 0.099

−0.199* 0.023

−0.203* 0.020

−0.241** 0.006

−0.221* 0.011

0.053 0.549

−0.188* 0.032

−0.962** 0.000

0.308** 0.000

Leaf C:N ratio

−0.078 0.160

−0.074 0.187

−0.521** 0.000

−0.118* 0.033

−0.138* 0.013

−0.100 0.073

−0.156** 0.005

−0.083 0.135

−0.301** 0.000

−0.446** 0.000

−0.138 0.116

−0.001 0.989

0.133 0.131

VPD

0.310** 0.000

0.581** 0.000

−0.095 0.088

0.138* 0.013

0.554** 0.000

−0.500** 0.000

−0.283** 0.000

−0.558** 0.000

0.018 0.748

0.095 0.089

0.203* 0.020

0.134* 0.018

−0.161 0.067

−0.045 0.445

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

5

Fig. 1. ␦D values (‰) of rainwater (open circles), groundwater (open triangles), and creek water (filled symbols) for 2007 and 2008. Creek water sampling occurred from 31 May to 27 September in 2007 and from 21 May to 24 September in 2008. Rain water was sampled as available from 1 June to 17 October in 2007 and from 17 March to 23 October in 2008. Mean groundwater ␦D is represented by a gray line.

(Table 1; ANOVA, p = 0.034). Leaf ␦D was positively correlated with E, Bowen ratio, TL –Ta , leaf ␦15 N, and VPD (Table 2; correlation, p ≤ 0.049). Leaf ␦D was negatively correlated to water ␦D − leaf ␦D (Table 2; correlation, p < 0.001). Leaf ␦13 C ranged from −31.9 to −28.1‰ across species throughout the study (Fig. 2). Leaf ␦13 C was significantly higher in 2008 compared to 2007 (ANOVA, p < 0.001). Moreover, there was a seasonal oscillation in leaf ␦13 C that paralleled creek water ␦D (cf. Fig. 1). Although there were consistent variations between sites (accounting for the large error bars in Fig. 2), there was no significant difference in ␦13 C between sites (ANOVA, p = 0.246). Leaf ␦13 C was significantly lower in plants of L. tatarica compared to C. occidentalis and M. alba plants (ANOVA, p = 0.001). Leaf ␦13 C was positively correlated with year, maximum leaf width, leaf ␦15 N, and water ␦D (Table 2; correlation, p ≤ 0.008). Leaf ␦13 C was negatively correlated to gv , E, H, Bowen ratio, TL –Ta , and leaf C/N ratio (Table 2; correlation, p ≤ 0.041). Leaf ␦15 N ranged from −0.1 to 6.5‰ across species and years (Fig. 3). Leaf ␦15 N was not different between years (ANOVA, p = 0.799). Leaf ␦15 N was significantly lower in individuals of L. tatarica compared to C. occidentalis and M. alba individuals (ANOVA, p < 0.001). Additionally, there was a significant difference between sites, with leaves from sites 2 and 3 having significantly higher ␦15 N compared to leaves from site 1 (ANOVA, p = 0.030). Leaf ␦15 N was positively correlated with Loe , maximum leaf width, gv , E, H, Bowen ratio, leaf ␦13 C, and leaf ␦D (Table 2; correlation, p ≤ 0.041). Leaf ␦15 N was negatively correlated to water ␦D–leaf ␦D and leaf C/N ratio (Table 2; correlation, p ≤ 0.032). Leaf C/N ratios ranged from 9.1 in some M. alba leaves to 26.1 in L. tatarica leaves (Fig. 4). Leaf C/N ratios were highest in L. tatarica plants, which were significantly greater than C. occidentalis individuals, in turn significantly higher than M. alba plants (ANOVA, p < 0.001). There was no difference in C/N ratio between years or

Fig. 3. Leaf ␦15 N (‰) of Celtis occidentalis, Lonicera tatarica, and Morus alba plants combined for 2007 and 2008, separated into three sites. Bars are means of 36 repeated measures ± SE.

sites (ANOVA, p ≥ 0.345). Leaf C/N ratio was negatively correlated to maximum leaf width, gv , E, Bowen ratio, leaf ␦13 C, and leaf ␦15 N (Table 2; correlation, p ≤ 0.033). Point measures of stomatal conductance (gs ) ranged from 0.037 to 0.123 mol m−2 s−1 across species and treatments (data not shown). These figures were used in calculating total leaf conductance to water vapor, gv . Mean gv ranged from 0.043 to 0.072 mol m−2 s−1 (data not shown). Measures of gv closely paralleled measures of E (Fig. 5). Measures of gv were not different between species (ANOVA, p = 0.810). gv was significantly higher in 2007 compared to 2008, and gv was significantly greater at site 2 compared to site 1 (ANOVA, p ≤ 0.048). There were significant positive relationships between gv and E, leaf ␦15 N, and VPD (Table 2; correlation, p ≤ 0.027). There were significant negative relation-

Fig. 2. Leaf ␦13 C (‰) of Celtis occidentalis, Lonicera tatarica, and Morus alba plants in 2007 and 2008. Points are means of 3 replicates ± SE.

6

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

Fig. 4. Leaf C/N ratio of Celtis occidentalis, Lonicera tatarica, and Morus alba plants in 2007 and 2008. Points are means of 3 replicates ± SE.

ships between gv and H, leaf ␦13 C, water ␦D, and leaf C/N ratio (Table 2; correlation, p ≤ 0.033). Point measures of leaf temperature (TL ) ranged from 24 to 33 ◦ C across species during the study (data not shown). TL was not different between species, sites, or years (ANOVA, p ≤ 0.470). The difference between leaf temperature and air temperature (TL –Ta ) ranged from +2.3 to −6.7 ◦ C (data not shown), with leaves at site 1 being significantly cooler in relation to air compared to leaves at sites 2 and 3 (ANOVA, p = 0.009). Accordingly, leaf Bowen ratios at site 1 were significantly lower than at sites 2 and 3 (ANOVA, p = 0.032) (data not shown). TL –Ta was positively correlated to H, Bowen ratio, and leaf ␦D (Table 2; correlation, p ≤ 0.047), TL –Ta was negatively correlated to year, leaf ␦13 C, water ␦D, water ␦D–leaf ␦D, and VPD (Table 2; correlation, p ≤ 0.036). Latent heat flux (E) ranged from 7 to 124 W m−2 across species and years (Fig. 5). No significant differences in E were detected between years or sites (ANOVA, p ≥ 0.307). E was positively correlated with Loe , gv , Bowen ratio, leaf ␦15 N, leaf ␦D, and VPD (Table 2; correlation, p ≤ 0.013). E was negatively correlated with leaf ␦13 C, water ␦D, water ␦D–leaf ␦D, and leaf C/N ratio (Table 2; correlation, p ≤ 0.036). Some weak patterns in E emerged between species. Whereas E was not significantly different between species (ANOVA, p = 0.199), leaves of L. tatarica trended toward the lowest

Fig. 5. Latent heat flux (E, W m−2 ) and transpiration rates (E, mmol m−2 s−1 ) from leaves of Celtis occidentalis, Lonicera tatarica, and Morus alba plants in 2007 and 2008. Points are means of 3 replicates ± SE.

transpiration rates (Fig. 5). But in a related parameter, leaves of L. tatarica were significantly narrower than leaves in C. occidentalis or M. alba (ANOVA, p < 0.001). Maximum leaf width averaged 35.9 mm in L. tatarica plants, 52.3 mm in C. occidentalis plants, and 56.9 mm in M. alba plants. Maximum leaf width correlated with H, Bowen ratio, leaf ␦13 C, and leaf ␦15 N (Table 2; correlation, p ≤ 0.010). Maximum leaf width was negatively correlated with leaf C/N ratio (Table 2; correlation, p < 0.001). Although E was not different between sites, there was a significant difference in leaf temperature between sites (ANOVA, p = 0.009). 4. Discussion In this study, stable isotopes of carbon, nitrogen, and hydrogen were used to investigate patterns of water use in three species of creekside trees in western Kansas, USA. Trees were measured over two growing seasons that had different timings of rainfall. Lonicera tatarica trees generally have shallow root systems, so these individuals were expected to be affected to a greater degree by soil drying compared to deeper-rooted Celtis occidentalis and M. alba trees, which had access to water in deeper soil layers. ␦D of rain in the present study ranged from −53 to 17‰, with most measurements between −20 and −40‰. These values are similar to ␦D in precipitation maps presented by Bowen et al. (2005) and Bowen (2009). Creek water samples ranged from −47 to −21‰ across sites, slightly higher than previously published measures of river water in Arizona (Busch et al., 1992). ␦D values in Big Creek water were much closer to values for ground water (measured from well water) compared to rain water for much of the study. Creek water ␦D was normally slightly higher than ground water ␦D. This could result from evaporative enrichment of ground water, or by mixing of ground water with rain water with higher ␦D. Creek water ␦D showed a seasonal oscillation, changing from lighter in spring to heavier in summer. Moreover, ␦D in creek water was higher in 2008, which relates to significantly higher Ta and vapor pressure deficit (VPD) in 2008 compared to 2007. Rain water samples also showed a seasonal trend in ␦D, but these data were more erratic than creek water samples. Normally the seasonal pattern of rainfall would be heaviest in the summer, with a peak near July or August, and becoming lighter thereafter, with a minimal value in mid-winter (Dawson and Ehleringer, 1991; Gat, 1996). Repeated collections of rain water at early, middle, and late times during rain events might provide better data for this graph, as ␦18 O and ␦D values can vary over the course of a rain shower (Gat, 1996). In contrast, creek water ␦D remained more consistent from week to week, most likely because these samples represent well-mixed sources. Values for creek water showed a slight seasonal oscillation, with the highest ␦D values during August. This seasonal oscillation of creek water either suggests a mixing of rain and ground water, or increased evaporation of water during summer months. A signifi-

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

cant positive correlation between VPD and creek water ␦D suggests an evaporative enrichment of water isotopes in creek water. Thus, Big Creek is most likely ground water fed, but is also influenced by rain water and evaporative enrichment during summer. The present study, conducted on riparian trees in a semi-arid environment, shows numerous leaf-level parameters significantly correlated with VPD and evaporation. During uptake of water from soil, there have been no reported cases of 18 O/16 O fractionation (Barbour, 2007), although D/H fractionation has been observed in mangroves with salt water (Lin and Sternberg, 1992). Nonetheless, xylem water in plants is normally considered to represent soil water isotopically for both ␦18 O and ␦D (Ehleringer and Dawson, 1992). Xylem water then enters leaves, where it is subject to evaporative enrichment of 18 O and D (Farquhar et al., 2007; Brandes et al., 2007). Leaf water is the source of all organically bound hydrogen in plant tissue (Yakir and DeNiro, 1990), including cellulose. ␦18 O and ␦D of leaf cellulose represents ␦18 O and ␦D of water in the leaf as cellulose was forming (Farquhar et al., 2007; Grams et al., 2007), with an enrichment in both 18 O and 13 C (Grams et al., 2007). ␦18 O of leaf cellulose tends to be 25 to 29‰ higher than the leaf water from which it formed (Siegwolf et al., 2001). Cellulose is preferred as the most valuable plant material for measures of ␦18 O, since oxygens in cellulose do not exchange with oxygens in water. Indeed, Grams et al. (2007) found very poor correlations between ␦18 O of leaf cellulose and ␦18 O of bulk leaf material. Moreover, ␦18 O of bulk leaf material is not correlated with ␦18 O of source water (Epstein et al., 1976). But in other cases, however, ␦18 O of bulk leaf material has strong relationships with stomatal conductance (Sullivan and Welker, 2007). Carbon-bound hydrogens in cellulose are not exchangeable (Epstein et al., 1976), which represent the majority of hydrogen in cellulose. Therefore, ␦D might be preferable to ␦18 O for measures comparing isotope ratios in bulk tissues with source water. Even though hydroxyl hydrogens are exchangeable in plant cellulose, ␦D of bulk leaf tissue has a moderately strong relationship to ␦D of cellulose in raw plant material (cf. Fig. 4 in Epstein et al., 1976). However, no such relationship exists when dealing with ␦D of bulk wood (Epstein et al., 1976). Thus, this relationship must be approached with caution, especially when dealing with woody samples. In the present study, leaf ␦D averaged −94‰ for individuals of C. occidentalis, and −103 and −104‰ for L. tatarica and M. alba plants, respectively. These values are lower than ␦D of salt marsh plants from southern California (mostly ca. 55–75‰, Smith and Epstein, 1970), consistent with inland versus coastal trends of precipitation ␦D (Yapp and Epstein, 1982). ␦D of bulk leaf tissue was 57–67‰ lower than ␦D of source water in the present study. This is slightly lower than the 55‰ depletion observed by Smith and Epstein (1970) for salt marsh plants. Greater enrichment of D and 18 O is often correlated with low humidity (Yapp and Epstein, 1982), as low humidity leads to greater evaporative enrichments. Decreasing humidity led to increased ␦18 O of leaf cellulose in numerous tree species measured by Epstein et al. (1977). Similarly, low relative humidity in the semi-arid setting of the present study might account for a slightly higher ␦D compared to plants in a coastal setting. In the present study, the deep-rooted C. occidentalis individuals had a higher ␦D compared to the shallow-rooted L. tatarica plants and individuals of the other deep-rooted species, M. alba. Higher ␦D values sometimes indicate a shallower water source in the soil, i.e., one subjected to evaporative enrichment (Jackson et al., 1999; Nippert and Knapp, 2007b). Alternatively, increasing evaporative demand leads to increased ␦18 O in leaves (CabreraBosquet et al., 2009); a similar relationship is seen with ␦D (Dawson and Ehleringer, 1991). However, values of ␦D and ␦18 O are more complex than simple evaporation rates. Yakir et al. (1990) mea-

7

sured lowered transpiration rates in dry-treatment cotton plants. Lowered transpiration resulted in drier air surrounding the leaves, which led to increased enrichment of D and 18 O in leaf water. Thus, increased transpiration rates do not always mean increased ␦D and ␦18 O. Instead, the isotopic gradient between leaf water and atmospheric water, within the context of leaf temperature and transpiration, all must be considered (Yakir et al., 1990). In the present study, point measures of gs and transpiration were not different between species. Further, water use efficiencies, derived from ␦13 C, do not support the idea of greater transpiration from C. occidentalis leaves. ␦13 C values in C. occidentalis trees were similar to M. alba trees, both of which were significantly higher than L. tatarica plants. Thus, it seems likely that the deep-rooted C. occidentalis individuals are drawing water from a different source than the other species. Rooting depth has been shown to influence water source in several plant systems, including redwood forests in northern California (Dawson, 1998), streamside trees in Utah (Dawson and Ehleringer, 1991), and tallgrass prairie in Kansas (Nippert and Knapp, 2007b). In the present study, different rooting depths on trees also appeared to influence source water. However, all trees were still able to take up water from Big Creek, as there was no evidence of water stress. Carbon isotope discrimination in C3 plants (including the three species in the present study) depends largely on Ci /Ca , the ratio of internal to ambient partial pressures of CO2 . Specifically, conditions that lower Ci /Ca result in increased ␦13 C (Farquhar et al., 1982). Previous studies have documented increases in ␦13 C in C3 plants in response to numerous conditions that lower gs (and thus lower Ci /Ca ), including drought (Farquhar and Richards, 1984), salinity (Neales et al., 1983), and even low temperature (Wang et al., 2008). Pairing measures of leaf ␦13 C with leaf ␦18 O or ␦D allows separation of stomatal versus carboxylation limitations on photosynthesis (Ramírez et al., 2009). Both CO2 uptake and transpiration rely on stomatal conductance. Low stomatal conductance in a C3 plant would lead to higher ␦13 C but lower ␦D. By contrast, limits by carboxylation capacity would decrease leaf ␦13 C but would have no effect on ␦18 O or ␦D (Grams et al., 2007). In the present study, both leaf ␦13 C and leaf ␦D were negatively correlated with gv , indicating a dependence on stomata. Leaf ␦13 C and leaf ␦D were not correlated with each other, however, suggesting influences of carboxylation limitations on ␦13 C. In the present study, leaf ␦13 C ranged from −31.9 to −28.1‰ across species, consistent with well-watered C3 plants (O’Leary, 1988). Leaf ␦13 C was significantly lower in L. tatarica plants compared to the other two species, which indicates higher Ci /Ca and suggests lower water use efficiency. This is interesting, since the shallow-rooted L. tatarica plants would be affected first during times of receding creek water. But gs and E were not different between species. This apparent discrepancy can be explained by lower N content of L. tatarica leaves. Lower N results in (among other things) a decreased carboxylation capacity (Turner et al., 2008). This results in higher Ci /Ca and hence lower ␦13 C. Therefore, differences in ␦13 C between these species occur without difference in gs or E. In vivo gas exchange measurements will be needed to confirm differences in Ci /Ca between species. Leaf ␦13 C became lower as the growing season progressed. Normal growth of leaves will involve a large mass of cellulose laid down early in the growing season, which will largely reflect heavier carbon stored over winter in a perennial plant (Helle and Schleser, 2004). After the leaf becomes self-supporting in terms of carbon, leaf ␦13 C often decreases during a growing season. Cellulose carbon is inert (Gaudinski et al., 2005), but the ␦13 C signal in leaves changes as carbon from winter storage is diluted by newly fixed carbon with lower ␦13 C (Helle and Schleser, 2004). Leaf nitrogen at site 1 was significantly lighter than at sites 2 and 3. Higher ␦15 N values most likely indicate differing amounts

8

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9

of fertilizer run-off. Anaerobic conditions in wetlands allow denitrification of inorganic nitrogen from fertilizer. Denitrification discriminates against 15 N, resulting in lighter N2 volatilized and heavier N remaining in sediments (Handley and Raven, 1992). As a result, wetland sites subjected to fertilizer run-off tend to have higher ␦15 N (Diebel and Vander Zanden, 2009). In the present study, sites 2 and 3 were in close proximity to manicured lawns, whereas site 1 was surrounded by less-maintained areas. Different ␦15 N values between sites most likely reflect inorganic nitrogen entering sites 2 and 3 from fertilizer run-off, but leaf ␦15 N correlated with many other variables in the study. ␦15 N increased with increasing transpiration (and related variables). ␦15 N also increased with increasing leaf temperature (and related parameters). Leaf ␦15 N values in L. tatarica individuals were significantly lighter compared to the other two species. This is consistent with a different source of N. Shallow-rooted L. tatarica plants apparently accessed a different source of inorganic N than the other two species. Potentially N in fertilizer runoff was assimilated by L. tatarica individuals earliest, which was then subject to enrichment by denitrification, and finally reached the deeper roots of C. occidentalis and M. alba plants. Leaf C/N ratios were not significantly different between sites, but there was a trend toward increasing leaf C/N during the 2008 growing season. This could result from normal senescence patterns of leaves (i.e., mobilization of N from leaves), or from a limitation of N in the environment. Since the same pattern was not observed in 2007, this might indicate a deficiency of N in the environment. During the transition to winter dormancy, much leaf nitrogen will be reabsorbed by the plant (Larcher, 2001), which would lead to increases in leaf C/N ratio. Such an increase was seen in leaf C/N ratios during 2008. Measures of gs in this study were very similar to measures presented by Gries et al. (2003) for Populus euphratica plants under field conditions. Measures in the present study occurred in a riparian setting, indicating a reliable source of creek water available for plants that can reach water with roots. As expected, gv was strongly related to E, since transpiration largely depends on stomatal conductance. Similarly, gv was negatively correlated with H, since decreasing stomatal conductance causes a decrease in latent cooling. Under conditions of decreasing water availability during soil drying, latent heat loss will decrease, and leaf temperature will increase. This shifts a greater portion of the heat load into sensible and radiative pathways, and the Bowen Ratio would be expected to increase (Maricle et al., 2007). Indeed, Yakir et al. (1990) found cotton leaves in dry-treatment plants were 2 ◦ C warmer in relation to air when compared to wet-treatment plants. In the present study, latent heat loss, leaf temperature, and Bowen ratios were not different between species. This indicates species responded similarly to seasonal changes in water status. However, gv , leaf C/N ratio, and E were significantly higher in 2007 compared to 2008, since more of the 2007 growing season was wet compared to the 2008 growing season. As a result, leaf ␦13 C was significantly lower in 2007 compared to 2008. Leaf transpiration rates relate to leaf size according to Eq. (7). Smaller leaves have a smaller boundary layer, which would normally facilitate transpiration. Transpiration rates depend on many factors, including behavior of the leaf (e.g., stomatal conductance or leaf orientation, Larcher, 2001), leaf size (Campbell and Norman, 1998), atmospheric conditions (e.g., wind speed or vapor pressure deficit, Campbell and Norman, 1998), soil water potential (Maricle et al., 2007), and root-soil interactions (Wahbi and Sinclair, 2007). Leaves of L. tatarica were significantly smaller than leaves of C. occidentalis and M. alba, creating a correspondingly smaller boundary layer and a lower resistance to heat and water loss. However, point measures of transpiration rates were not different between

species, nor do measures of ␦13 C or ␦D support the idea of greater transpiration rates in L. tatarica plants. Measures that related to leaf transpiration influenced leaf isotope composition to a large degree. Leaf ␦13 C was negatively related to gv and E, as decreased ␦13 C is normally associated with high stomatal conductance and low water use efficiency (Farquhar et al., 1982). However, leaf ␦13 C was not related to VPD. By contrast, measures of leaf ␦D were positively correlated with E and VPD, illustrating the utility of pairing measures of ␦13 C and ␦D on the same leaf. Increased transpiration resulted in enrichment of D in leaves, but ␦13 C was also influenced by carboxylation limitations. In conclusion, differing precipitation patterns in the 2007 and 2008 growing seasons allowed comparisons of changing water availability on creekside trees. None of the three species in the present study appeared to be stressed by drying of creekside soils. Interestingly, deep-rooted C. occidentalis trees appeared to take up water from a shallower source compared to the other species, as evidenced by different hydrogen isotope values in leaves. ␦D of creek water was more similar to ground water than to rain water, but the creek was also influenced by climate, since a seasonal oscillation was apparent in creek water. None of these tree species showed signs of water stress; differences in ␦13 C were consistent with lower leaf nitrogen content in L. tatarica plants. Leaves of shallow-rooted L. tatarica had lower ␦15 N compared to the deeprooted species, owing to different sources of inorganic N for uptake. Two of the three sites appeared to be affected by inorganic N from fertilizer run-off. Acknowledgments The authors thank Aaron M. Pfeifer and Jordan J. Brungardt for help with field sampling, Jessica A. Marshall for help with sample preparation, and Jordge LaFantasie for help with statistics. This study was partially funded by the Department of Biological Sciences and the Fort Hays State University Graduate School Small Research Grant. References Adler, P.B., HilleRisLambers, J., 2008. The influence of climate and species composition on the population dynamics of ten prairie forbs. Ecology 89, 3049–3060. Barbour, M.M., 2007. Stable oxygen isotope composition of plant tissue: a review. Funct. Plant Biol. 34, 83–94. Bowen, G.J., Wassenaar, L.I., Hobson, K.A., 2005. Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia 143, 337–348. Bowen, G.J., 2009. The Online Isotopes in Precipitation Calculator, version 2.2. http://www.waterisotopes.org. Boyer, J.S., 1982. Plant productivity and environment. Science 218, 443–448. Brandes, E., Wenninger, J., Koeniger, P., Schindler, D., Rennenberg, H., Leibundgut, C., Mayer, H., Gessler, A., 2007. Assessing environmental and physiological controls over water relations in a Scots pine (Pinus sylvestris L.) stand through analyses of stable isotope composition of water and organic matter. Plant Cell Environ. 30, 113–127. Bucci, S.J., Scholz, F.G., Goldstein, G., Meinzer, F.C., Arce, M.E., 2009. Soil water availability and rooting depth as determinants of hydraulic architecture of Patagonian woody species. Oecologia 160, 631–641. Busch, D.E., Ingraham, N.L., Smith, S.D., 1992. Water uptake in woody riparian phreatophytes of the southwestern United States: a stable isotope study. Ecol. Appl. 2, 450–459. Cabrera-Bosquet, L., Molero, G., Nogués, S., Araus, J.L., 2009. Water and nitrogen conditions affect the relationships of 13 C and 18 O to gas exchange and growth in durum wheat. J. Exp. Bot. 60, 1633–1644. Campbell, G.S., Norman, J.M., 1998. An Introduction to Environmental Biophysics. Springer-Verlag, New York, NY, USA. Clark, I.D., Fritz, P., 1997. Environmental Isotopes in Hydrogeology. Lewis Publishers, New York, NY, USA. Cleverly, J.R., Smith, S.D., Sala, A., Devitt, D.A., 1997. Invasive capacity of Tamarix ramosissima in a Mojave Desert floodplain: the role of drought. Oecologia 111, 12–18. Davis, S.M., Gaiser, E.E., Loftus, W.F., Huffman, A.E., 2005. Southern marl prairies conceptual ecological model. Wetlands 25, 821–831. Dawson, T.E., 1998. Fog in the California redwood forest: ecosystem inputs and use by plants. Oecologia 117, 476–485.

B.R. Maricle et al. / Environmental and Experimental Botany 71 (2011) 1–9 Dawson, T.E., Ehleringer, J.R., 1991. Streamside trees that do not use stream water. Nature 350, 335–337. Diebel, M.W., Vander Zanden, M.J., 2009. Nitrogen stable isotopes in streams: effects of agricultural sources and transformations. Ecol. Appl. 19, 1127–1134. Drew, M.C., 1975. Comparison of the effects of a localized supply of phosphate, nitrate, ammonium and potassium on the growth of the seminal root system, and the shoot, in barley. New Phytol. 75, 479–490. Ehleringer, J.R., Dawson, T.E., 1992. Water uptake by plants: perspectives from stable isotope composition. Plant Cell Environ. 15, 1073–1082. Ehleringer, J.R., Osmond, C.B., 1989. Stable isotopes. In: Pearcy, R.W., Ehleringer, J.R., Mooney, H.A., Rundel, P.W. (Eds.), Plant Physiological Ecology: Field Methods and Instrumentation. Chapman & Hall, London, UK, pp. 281–300. English, N.B., Dettman, D.L., Sandquist, D.R., Williams, D.G., 2007. Past climate changes and ecophysiological responses recorded in the isotope ratios of saguaro cactus spines. Oecologia 154, 247–258. Epstein, S., Yapp, C.J., Hall, J.H., 1976. The determination of the D/H ratio of nonexchangeable hydrogen in cellulose extracted from aquatic and land plants. Earth Planet. Sci. Lett. 30, 241–251. Epstein, S., Thompson, P., Yapp, C.J., 1977. Oxygen and hydrogen isotopic ratios in plant cellulose. Science 198, 1209–1215. Ewe, S.M.L., Sternberg, L.d.S.L., Childers, D.L., 2007. Seasonal plant water uptake patterns in the saline southeast Everglades ecotone. Oecologia 152, 607– 616. Farquhar, G.D., O’Leary, M.H., Berry, J.A., 1982. On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust. J. Plant Physiol. 9, 121–137. Farquhar, G.D., Richards, R.A., 1984. Isotopic composition of plant carbon correlates with water-use efficiency of wheat genotypes. Aust. J. Plant Physiol. 11, 539–552. Farquhar, G.D., Cernusak, L.A., Barnes, B., 2007. Heavy water fractionation during transpiration. Plant Physiol. 143, 11–18. Gat, J.R., 1996. Oxygen and hydrogen isotopes in the hydrologic cycle. Ann. Rev. Earth Planet. Sci. 24, 225–262. Gaudinski, J.B., Dawson, T.E., Quideau, S., Schuur, E.A.G., Roden, J.S., Trumbore, S.E., Sandquist, D.R., Oh, S.W., Wasylishen, R.E., 2005. Comparative analysis of cellulose preparation techniques for use with 13 C, 14 C, and 18 O isotopic measurements. Anal. Chem. 77, 7212–7224. Grams, T.E.E., Kozovits, A.R., Häberle, K.H., Matyssek, R., Dawson, T.E., 2007. Combining ␦13 C and ␦18 O analyses to unravel competition, CO2 and O3 effects on the physiological performance of different-aged trees. Plant Cell Environ. 30, 1023–1034. Gries, D., Zeng, F., Foetzki, A., Arndt, S.K., Bruelheide, H., Thomas, F.M., Zhang, X., Runge, M., 2003. Growth and water relations of Tamarix ramosissima and Populus euphratica on Taklamakan desert dunes in relation to depth to a permanent water table. Plant Cell Environ. 26, 725–736. Handley, L.L., Raven, J.A., 1992. The use of natural abundance of nitrogen isotopes in plant physiology and ecology. Plant Cell Environ. 15, 965–985. Helle, G., Schleser, G.H., 2004. Beyond CO2 -fixation by Rubisco – an interpretation of 13 C/12 C variations in tree rings from novel intra-seasonal studies on broad-leaf trees. Plant Cell Environ. 27, 367–380. Herrera, A., Tezara, W., Marín, O., Rengifo, E., 2008. Stomatal and non-stomatal limitations of photosynthesis in trees of a tropical seasonally flooded forest. Physiol. Plant. 134, 41–48. Jackson, P.C., Meinzer, F.C., Bustamante, M., Goldstein, G., Franco, A., Rundel, P.W., Caldas, L., Igler, E., Causin, F., 1999. Partitioning of soil water among tree species in a Brazilian Cerrado ecosystem. Tree Physiol. 19, 717–724. Kercher, S.M., Zedler, J.B., 2004. Flood tolerance in wetland angiosperms: a comparison of invasive and noninvasive species. Aquat. Bot. 80, 89–102. Kuchler, A.W., 1974. A new vegetation map of Kansas. Ecology 55, 586–604.

9

Larcher, W., 2001. Physiological Plant Ecology. Springer, Berlin, Germany. Lauenroth, W.K., Burke, I.C., Gutmann, M.P., 1999. The structure and function of ecosystems in the central North American grassland region. Great Plains Res. 9, 223–259. Lin, G., Sternberg, L.d.S.L., 1992. Effect of growth form, salinity, nutrient and sulfide on photosynthesis, carbon isotope discrimination and growth of red mangrove (Rhizophora mangle L.). Aust. J. Plant Physiol. 19, 509–517. Maricle, B.R., Cobos, D.R., Campbell, C.S., 2007. Biophysical and morphological leaf adaptations to drought and salinity in salt marsh grasses. Environ. Exp. Bot. 60, 458–467. Neales, T.F., Fraser, M.S., Roksandic, Z., 1983. Carbon isotope composition of the halophyte Disphyma clavellatum (Haw.) Chinnock (Azioaceae), as affected by salinity. Aust. J. Plant Physiol. 10, 437–444. Neumann, P.M., 2008. Coping mechanisms for crop plants in drought-prone environments. Ann. Bot. 101, 901–907. Nippert, J.B., Knapp, A.K., 2007a. Linking water uptake with rooting patterns in grassland species. Oecologia 153, 261–272. Nippert, J.B., Knapp, A.K., 2007b. Soil water partitioning contributes to species coexistence in tallgrass prairie. Oikos 116, 1017–1029. O’Leary, M.H., 1988. Carbon isotopes in photosynthesis. BioScience 38, 328–336. O’Leary, M.H., Madhavan, S., Paneth, P., 1992. Physical and chemical basis of carbon isotope fractionation in plants. Plant Cell Environ. 15, 1099–1104. Ramírez, D.A., Querejeta, J.I., Bellot, J., 2009. Bulk leaf ␦18 O and ␦13 C reflect the intensity of intraspecific competition for water in a semi-arid tussock grassland. Plant Cell Environ. 32, 1346–1356. Siegwolf, R.T.W., Matyssek, R., Saurer, M., Maurer, S., Günthardt-Goerg, M.S., Schmutz, P., Bucher, J.B., 2001. Stable isotope analysis reveals differential effects of soil nitrogen and nitrogen dioxide on the water use efficiency in hybrid poplar leaves. New Phytol. 149, 233–246. Smith, B.N., Epstein, S., 1970. Biogeochemistry of the stable isotopes of hydrogen and carbon in salt marsh biota. Plant Physiol. 46, 738–742. Sullivan, P.F., Welker, J.M., 2007. Variation in leaf physiology of Salix arctica within and across ecosystems in the High Arctic: test of a dual isotope (13 C and 18 O) conceptual model. Oecologia 151, 372–386. Tomanek, G.W., Hulett, G.K., 1970. Effects of historical droughts on grassland vegetation in the central Great Plains. In: Dort, W., Jones, J.K. (Eds.), Pleistocene and Recent Environments of the Central Great Plains. The University Press of Kansas, Lawrence, KS, USA, pp. 203–210. Turner, N.C., Schulze, E.D., Nicolle, D., Schumacher, J., Kuhlmann, I., 2008. Annual rainfall does not directly determine the carbon isotope ratio of leaves of Eucalyptus species. Physiol. Plant. 132, 440–445. Visser, J.M., Sasser, C.E., Chabreck, R.H., Linscombe, R.G., 2002. The impact of a severe drought on the vegetation of a subtropical estuary. Estuaries 25, 1184–1195. Wahbi, A., Sinclair, T.R., 2007. Transpiration response of Arabidopsis, maize, and soybean to drying of artificial and mineral soil. Environ. Exp. Bot. 59, 188–192. Wang, G.A., Han, J.M., Faiia, A., Tan, W.B., Shi, W.Q., Liu, X.Z., 2008. Experimental measurements of leaf carbon isotope discrimination and gas exchange in the progenies of Plantago depressa and Setaria viridis collected from a wide altitudinal range. Physiol. Plant. 134, 64–73. Yakir, D., DeNiro, M.J., 1990. Oxygen and hydrogen isotope fractionation during cellulose metabolism in Lemna gibba L. Plant Physiol. 93, 325–332. Yakir, D., DeNiro, M.J., Gat, J.R., 1990. Natural deuterium and oxygen-18 enrichment in leaf water of cotton plants grown under wet and dry conditions: evidence for water compartmentation and its dynamics. Plant Cell Environ. 13, 49–56. Yapp, C.J., Epstein, S., 1982. Climatic significance of the hydrogen isotope ratios in tree cellulose. Nature 297, 636–639.