REGULAR ARTICLE

Soil water uptake by trees using water stable isotopes(δ2H and δ18O)−a method test regarding soil moisture, textureand carbonate

Meik Meißner & Michael Köhler & Luitgard Schwendenmann &

Dirk Hölscher & Jens Dyckmans

Received: 24 May 2013 /Accepted: 5 November 2013 /Published online: 30 November 2013# Springer Science+Business Media Dordrecht 2013

AbstractAims Stable isotopes of oxygen and hydrogen are oftenused to determine plant water uptake depths. We inves-tigated whether and to what extend soil moisture, claycontent, and soil calcium carbonate influences the waterisotopic composition.Methods In the laboratory, dried soil samples varying inclay content were rewetted with different amounts ofwater of known isotopic composition. Further, we re-moved soil carbonate from a subset of samples prior torewetting. Water was extracted from samples via cryo-genic vacuum extraction and analysed by massspectrometry.Results The isotopic composition of extracted soil waterwas similarly depleted in both 18O and 2H with

decreasing soil moisture and increasing clay and car-bonate content. Soil carbonate changed the δ18O com-position while δ2H was not affected.Conclusions Our results indicate that soil carbonate cancause artifacts for 18O isotopic composition of soil wa-ter. At low soil moisture and high carbonate content thiscould lead to conflicting results for δ18O and δ2H inplant water uptake studies.

Keywords Isotopes .Water . Soil . Clay. Carbonate

Introduction

The natural abundance of the stable isotopes of δ2H(deuterium) and δ18O are widely used to investigateecological and hydrological processes, such as plant-water relations. Using the “spatial isotope fingerprint ofwater” (West et al. 2006) in plant physiological studieshas significantly improved our understanding of ecolog-ical processes since its first emergence in the 1970s.

Several studies investigated the water sources ofplant species in a wide range of environments differingin climate, soil type and vegetation type (Dawson andEhleringer 1991; Ehleringer et al. 1991; Flanagan et al.1992; Dawson et al. 1993; Dawson 1993; Jackson et al.1995; Jackson et al. 1999; Meinzer et al. 1999). Waterstable isotopes can provide useful information on re-source complementarity or competition among individ-uals of a population, especially in studies on plantspecies diversity effects or neighborhood relationships(Schwendenmann et al. 2010).

Plant Soil (2014) 376:327–335DOI 10.1007/s11104-013-1970-z

Responsible Editor: Rafael S. Oliveira.

Electronic supplementary material The online version of thisarticle (doi:10.1007/s11104-013-1970-z) contains supplementarymaterial, which is available to authorized users.

M. Meißner (*) :M. Köhler :D. HölscherTropical Silviculture and Forest Ecology,Georg-August-Universität Göttingen,Büsgenweg 1, 37077 Göttingen, Germanye-mail: mmeissn3@gwdg.de

L. SchwendenmannSchool of Environment, The University of Auckland,Private Bag 92019, Auckland 1142, New Zealand

J. DyckmansCentre for Stable Isotope Research and Analysis,Georg-August-Universität Göttingen,Büsgenweg 2, 37077 Göttingen, Germany

It is assumed that fractionation of O and H isotopesfollow the same patterns during the evaporation process-es (Wershaw et al. 1966; Thorburn and Walker 1993).Thus, naturally occurring vertical gradients of δ2H andδ18O in soil water provide similar information aboutplant water uptake depth from soils (Plamboeck et al.1999). It is further assumed that the isotopic composi-tion of water is unchanged by root uptake and duringtransport through the stem xylem in most plants(Wershaw et al. 1966; White et al. 1984), with theexception of some coastal wetland species (Lin et al.1993) and certain woody xerophytes (Ellsworth andWilliams 2007), which fractionate 2H but not 18O duringroot water uptake. Still, δ2H and δ18O can be usedsimilarly in water uptake studies for a wide range ofwoody plant species where samples of different possiblewater sources are obtained together with plant (xylem)water samples.

Other studies however suggest that physicochemicalsoil properties may lead to fractionation of hydrogenand/or oxygen in soil water. For example, isotopic ex-change reactions with carbonates and water can occur atambient temperatures in the soil (Zeebe 2009). Thus,exchange reactions with soil carbonate originating fromlimestone parent rock could be a source for oxygenisotope variation in meteoric water additional to frac-tionation by evaporation (Gat 1996). In particular, ascarbonate has formed by deposition of marine organ-isms or precipitated inorganically from aqueous solu-tions at equilibrium conditions and incorporated theambient water isotope composition (Clayton 1961),which was most likely different from the prevalentcomposition at the time of exchange reaction. Isotopicdeviations of extracted soil water from the input waterwere previously also reported in particular for clayeysoils (Walker et al. 1994; Hsieh et al. 1998; Koenigeret al. 2011). Clay particles may interact with bulk water

and create pools of energetically differing water withvarying isotope compositions (Oerter et al. 2012).Further, the formation of hydration spheres around cat-ions in aqueous solutions fractionates oxygen isotopesof water (Sofer and Gat 1972).

In a prior study, we employed the natural abundance ofwater isotopes to assess the vertical water uptake patternsof Fagus sylvatica, Tilia sp. and Fraxinus excelsior dur-ing a summer desiccation period in a temperate forest inCentral Germany. Here we encountered a notable differ-ence between the isotopic composition of δ2H and δ18Oin soil compared to xylem water samples. Therefore,depending on whether 2H or 18O was used for the esti-mation of the uptake depth, the calculated depths differed.The difference was irrespective of tree species.

The aim of this study was to test which physical and/or chemical soil properties may influence the isotopiccomposition of extracted soil water to find possibleexplanations of our contradictory findings in the previ-ous study. Since it is likely that clay and carbonate canalter soil water isotopic composition and since weworked on clay and carbonate rich soils, we testedwhether these factors could explain the deviation wefound. Further, soil water content might play an impor-tant role in isotopic discrimination: Especially at lowsoil water content the effect might be more eminent.Therefore we included this factor in our study, too.

Methods

Study site description

Soil and plant material was collected from several re-search sites within the Hainich National Park close tothe village of Weberstedt (51°05′28″N, 10°31′24″E) inCentral Germany. Sites were located on a low mountainrange at an elevation of approximately 350 m a.s.l.,situated within a subatlantic climate zone. Mean annualprecipitation is 544–662 mm (average of 30 years pre-cipitation records from four climate stations around thenational park; DWD, 2008) with frequent soil desicca-tion periods occurring during summer months. The geo-logical substrate of the area is Triassic (shell-) limestoneoverlain by loess, which forms nutrient-rich Luvisols.Clay content was estimated by fractionation of particlesaccording to size limits by sieving and gravity sedimen-tation (Gee and Bauder 1986) and ranged from 17 % inthe surface soil to 45 % at 0.7 m depth (Table A1). Soil

Table 1 Comparison of mean isotopic deviation between inputwater and extracted water among the treatments. Significant dif-ferences between mean values (paired t-test) are marked with anasterisk (p<0.01)

Treatments Mean isotopic deviation (%)

δ18O δ2H

Input water - control –0.70* –6.40*

Input water - HCl 0.40 –4.24*

Control - HCl –1.31* –2.20

328 Plant Soil (2014) 376:327–335

clay minerals are base-rich vermiculites in 0–0.3 m andsmectites from 0.3–0.7 m soil depth (Butz-Braun 2001).Soil pH increases with soil depth, apparently influencedby the limestone parent rock underneath (Butz-Braun2001; Guckland 2009).

Sample preparation

During summer 2009, increment cores were taken at1.3 m from 48 trees (n=48 samples). Soil samples werecollected from five depth intervals (0–0.1, 0.1–0.2, 0.2–0.3, 0.3–0.5 and 0.5–0.7m) below the respective trees (n=240) with an auger. Mean gravimetric water content of thesamples was around 24 % (13–45 %). Water was extract-ed from all samples using the cryogenic vacuum extrac-tion method (Ehleringer and Osmond 1989; West et al.2006) at ~115 °C and 0.01 mbar. Prior to sample extrac-tion, we tested the effect of different extraction times (60,90, 180 and 240min) on the isotopic composition ofwaterextracts to determine the time required to obtainunfractionated water samples (West et al. 2006) from soiland stem. According to our tests, an extraction time of90 min is required to obtain unfractionated water samplesfrom soil and plant material, which is substantially longerthan the times recorded in a similar experiment for clayeysoil and stem samples by West et al. (2006).

After cryogenic extraction, samples were oven-driedat 110 °C for 72 h and stored for further analysis. In2012, some of these samples were then used to study theeffects of soil physical and chemical properties on theisotopic composition of soil water. A total of 25 sampleswere randomly selected with each sample beingremoistened with water of known isotopic composition,extracted and oven-dried again in order to treat allsamples equally before the onset of the experiment.

Fifteen of these 25 samples (5 g soil per sample) wereagain moistened with the same water to three differentextents: 0.6 ml (low water input), 1.5 ml (medium waterinput) and approximately 3 ml (saturating to water hold-ing capacity; high water input) (n=5 per group). Theisotopic composition of the water used for the rewettingexperiment was −56.0±1.8 ‰ (δ2H) and −9.5±0.2 ‰(δ18O). Rewetting of soil samples with differentamounts of water produced soil water potentials ofapproximately −2.5 MPa, −0.6 MPa and −0.08 MPa,respectively. Water potentials were calculated from wa-ter contents, using the Rosetta DLL (Dynamik LinkedLibrary) program by Schaap et al. (2001), implementedin the HYDRUS-1D model (Simunek et al. 2008).

From each of the remaining 10 samples, we separatedtwo subsamples (n=5 per group with 5 g soil per sub-sample). One subsample was remoistened with 1.0 mlwater of known isotopic composition (control sample; -56.2±0.4‰ δ2H and −9.4±0.03‰ δ18O) and the otherwas treated with hydrochloric acid (10 % HCl) to re-move carbonates and washed until neutral pH wasreached. Afterwards, the soil was dried to remove allwater from the sample and remoistened with the samewater as that used in the control sample. Water potentialof remoistened samples was about −0.8 MPa. Samplevials from both laboratory experiments were sealed withparaffin tape to avoid water evaporation and kept in therefrigerator for 72 h. Water was later extracted from thesoil samples as described above.

Extracted water from both experiments was weightedto determine the water recovery rate. In addition, re-maining sample material was also weighted, dried andweighted again for the same purpose. The water recov-ery rate during extraction of remoistened samples was>98 %, tested by drying and re-weighing (Fig. A1).According to findings of Araguás-Araguás et al.(1995) and West et al. (2006) this recovery rate shouldbe sufficient to obtain an unfractionated water sample.

Measurements of δ2H and δ18O isotopic compositionwere conducted on a high temperature conversion ele-mental analyzer (TC/EA, Thermo Electron Corporation,Bremen, Germany) coupled to a Delta V Plus isotoperatio mass spectrometer (Thermo Electron Corporation)with reversed sample flow (Gehre et al. 2004) giving ameasurement precision of ±2‰ for δ2H and ± 0.2‰ forδ18O, respectively.

Statistical analysis

An ANOVA was applied followed by a paired t-test totest for differences among treatments (low, medium andhigh water input; HCl and control). Statistical tests wereconsidered significant at p≤0.01. All statistical analyseswere done in R version 2.15.1 (R Core Team 2012).

Results

Soil water isotopic composition from the field study

The isotopic composition of soil water extracted fromsoil samples acquired at different depths during the soildesiccation period in 2009 ranged from −35.0 to −80.0

Plant Soil (2014) 376:327–335 329

‰ (δ2H) and −5.2 to −11.5 ‰ (δ18O). Due to a strongevaporation, an isotopic gradient developed in concertwith increasing soil depth (Fig. 1). As soil and stemsamples were taken simultaneously, the isotopic com-position of xylem water was expected to reflect thesignatures of soil water at the depth of soil water uptake.However, comparing the signatures of either δ2H orδ18O soil and xylem water with a simple graphicalinterference approach indicates a distinct difference be-tween both isotopes (Fig. 1). Though a graphical “bestmatch” approach is not suitable to estimate water uptakedepth as plants often withdraw water from more thanone distinct soil depth; it is used here to illustrate the biasfound between δ2H or δ18O.

Effect of soil moisture on soil water isotopiccomposition

Extracted water from soil samples with low andmedium input water (0.6 and 1.5 ml, respectively)showed a significant deviation (p<0.01) from theinput water. The δ2H and δ18O of low water inputsamples was −68.9±7.2 ‰ and −11.2±1.5 ‰, respec-tively. The isotopic composition of medium water inputsamples was −62.7±4.1 ‰ (δ2H) and −10.3±0.9 ‰(δ18O). In contrast, water from samples with high waterinput (3.0 ml) had an isotopic composition similar(−56.6±3.0 ‰ (δ2H) and −9.2±0.2 ‰ (δ18O)) to thatof the input water (Fig. 2).

Effect of clay content on soil water isotopic composition

The isotopic composition of both isotopes showeda significant negative relationship with clay con-tent at low water content (Fig. 3, A1 and A2). Incontrast, no significant correlation was found forsamples rewetted with medium (Fig. 3, B1 andB2) and high (data not shown) amount of inputwater. Hence, the effect of clay on the isotopiccomposition of extracted water after rewettingstrongly depends on the amount of water usedfor rewetting. It should be noted that clay andcarbonate content are co-correlated. Therefore itis not possible to clearly separate effects of clayand carbonate

Effect of soil calcium carbonate on soil water isotopiccomposition

Control samples which were rewetted with 1 ml waterwere depleted in both δ2H and δ18O compared to inputwater, as was expected from the experiment with differ-ing quantities of input water. However, samples thatwere previously treated with hydrochloric acid (HCl)were significantly enriched in δ18O (p<0.01) com-pared to control samples (Fig. 4), but they did notdiffer significantly with respect to input water. Theeffect of HCl treatment on the average δ2H valuewas not significant (Table 1).

Fig. 1 Average stable isotope composition of δ18O and δ2H (Aand B, respectively) in soil and xylem samples. Values aremeans ±standard deviation (n=48). The grey bars indicate the depth ofwater uptake for a given element as indicated by graphical

interference. Dotted arrows indicate the isotope shift necessaryfor either O isotopes (towards higher values) or H isotopes (towardlower values) to bring in line results of O and H isotope analysiswith respect to plant water uptake depth (see discussion)

330 Plant Soil (2014) 376:327–335

Discussion

We found that the isotopic composition of δ2H or δ18Obetween soil and xylem water resulted in different soilwater uptake depths of the trees depending on which ofthe isotopes is considered (Fig. 1). Using graphicalinterference to derive plant water uptake depth is asimplistic approach assuming that water is withdrawnfrom only a single discreet soil layer. However, it ismore likely that plant xylem water is a mixture of waterfrom several soil depths. Multiple sourcemixingmodels(Phillips and Gregg 2001, 2003; Moore and Semmens2008; Parnell et al. 2010) account for water uptake frommore than one discrete soil layer and weigh the impor-tance of certain layers for water uptake by incorporatinge.g. soil water potentials into the calculation. Suchmodels can also account for a certain bias in the isotopiccomposition between δ2H and δ18O. However, due tothe magnitude of the bias presented here and its system-atic nature across all samples, even mixing modelswould not yield reliable results. The differences between

Fig. 2 Isotopic composition of extracted soil water (filled circles)in relation to input water (open circle). Samples were treated withdifferent quantities of input water, (low=0.6 ml; medium=1.5 ml;high=3.0 ml). Values are means ± standard deviation (n=15)

Fig. 3 Effect of clay content(%) in sample soil on theisotopic composition of ex-tracted soil water. Datashown for samples with low(A1 and A2) and medium(B1 and B2) input water,respectively

Plant Soil (2014) 376:327–335 331

δ2H and δ18O were independent of the observed treespecies (Fagus, Tilia, Fraxinus). Thus it seems unlikelythat our findings are the result of a discriminationagainst 2H in plant roots as described by Lin et al.1993) for halophytes and Ellsworth and Williams(2007) for xerophytic plants. Results similar to ourswere also reported in a study by Brunel et al. (1991),where the isotopic composition of xylem water samplesdid not match with those of soil samples in a δ2H-δ18Oplot (xylem δ‰ values arranged notably above the bestfit regression line of soil samples). However, this dis-crepancy was relatively small and did not appear in allobservations during the study, and further investigationinto the difference was not pursued by the authors(Brunel et al. 1991). A considerable difference betweenδ2H and δ18O isotopic composition in soil and xylemwater was also observed by Li et al. (2007), who inves-tigated the water uptake of Larix sibirica in a semiaridforest region. The authors concluded that an indepen-dent application of both isotopes did not yield a consis-tent estimate of soil water uptake depth for several oftheir measurements (Li et al. 2007). The discrepancywas however explained as an artifact of either the ana-lytical procedure (cryogenic vacuum extraction andanalysis with a Finnigan MAT252 isotope ratio massspectrometer) or related to natural processes, which

could not be explained with the available data (Li et al.2007). A visual survey of graphs from other dual waterisotope studies (Midwood et al. 1998; Eggemeyer et al.2009; Rosatto et al. 2011) suggests that comparabledifferences due to isotopic shifts of δ2H and δ18O mayalso be found in other datasets. However, such incon-sistencies in studies of plant water uptake often remainundiscussed, since they rarely have significant impactson the results if soil depths are pooled to larger soildepth intervals (e.g. upper and lower soil profile).

It has previously been shown that high clay content insoil samples may hinder complete water extraction(West et al. 2006) but also affect the isotopic composi-tion of extracted water, especially if soil water content islow (Walker et al. 1994; Hsieh et al. 1998; Koenigeret al. 2011). Our findings corroborate both effects, sincewe observed that the deviation of the signature of inputwater from extracted water increases with both decreas-ing soil water content (Fig. 2) and increasing clay con-tent (at low soil water contents) (Fig. 3). This is also inline with the results of Savin and Epstein (1970), whoshowed that adsorbed and interlayer water of clays canexchange with atmospheric water at ambient tempera-tures. However, these findings cannot explain the discrep-ancy between δ2H and δ18O isotopic signatures in ourstudy (Fig. 1), as all observed processes affect both δ2Hand δ18O. The possibility that hygroscopic water stored inthe interstices between clay particles exchanged withambient soil water, causing a so called “memory effect”(Koeniger et al. 2011) is unlikely as in our laboratoryexperiment samples were already processed in vacuumextraction before rewetting, and we repeatedly addedwater of the same isotopic composition during our exper-iments. Moreover, an exchange between hygroscopicwater and soil water would probably affect both δ2Hand δ18O (Araguás-Araguás et al. 1995).

Our study revealed that the presence of carbonatessignificantly alters the isotopic composition of 18O ofadded water, whereas the shift of δ2H between addedand extracted water was independent from carbonatecontent (Fig. 4). Water from soil samples with carbonatewere more depleted in δ18O compared to samples wherecarbonate has been removed before rewetting (~ 1.3‰).This implies that, at least for our samples, differentmechanisms affect the isotopic composition of waterextracted from soil. The δ18O is presumably altered byexchange of oxygen with carbonates in soil, whichcould serve as an explanation for the difference between18O and 2H shown in Fig. 1. The observed increase of

Fig. 4 Isotopic composition (δ2H and δ18O) of extracted soilwater from control samples and samples pre-treated with hydro-chloric acid (HCl), in relation to original input water. Values aremeans ± standard deviation

332 Plant Soil (2014) 376:327–335

isotopic deviation with increasing clay content also sup-ports this finding, as clay and carbonate content in thisarea increase with soil depth (Butz-Braun 2001;Guckland 2009). Although soil carbonate contentcan serve as an explanation for the isotopic shiftin our samples, it is still possible that other pro-cesses are responsible for the alteration of oxygenisotopic composition of input water in the otherstudies mentioned.

The isotopic composition of hydrogen may also bealtered by water exchange with soil particles; this couldeither be exchangeable hydrogen on the clay surface orexchangeable O-bonded hydrogen in the organic matter.As isotopic depletion in our samples was correlated withclay (and therefore co-correlated with soil depth), weassume that the latter process only plays a minor role inour soils, as organic matter content decreased with in-creasing soil depth and hydrogen exchange with clayminerals is probably responsible for the observed shifts.

Our data from the field experiment indicate thatresults from 18O and 2H measurements with regard toplant water uptake depth can only be brought in line ifsoil water is either lower in δ2H values or higher in δ18Othan the cryogenically extracted water (see Fig. 1). Onthe other hand, our laboratory experiments presentedhere show that both O and H are shifted to lower valuesin cryogenically extracted water as compared to inputwater. We therefore conclude, that in the field experi-ment H isotopes probably reflect the plant water uptake(i.e. plants take up water with the same H isotopiccomposition as is cryogenically extracted) whereas Oisotopes in extracted water are shifted to lower values ascompared to plant water uptake (see Fig. 1). As wecould also show that carbonates may be responsiblefor the O isotope shift we propose the following hypoth-esis to explain our findings: Soil water exchanges Owith soil carbonate in an equilibrium reaction which issubject to isotopic fractionation. This equilibrium frac-tionation will shift extracted water to higher δ18O valuesand is temperature dependent, with higher fractionationat lower temperatures (Beaudoin and Therrien 2009,http://www2.ggl.ulaval.ca/cgi-bin/isotope/generisotope.cgi). Given that the soil temperature in the field isconsiderably lower than during cryogenic vacuumextraction in the laboratory (since samples are heatedduring extraction to ensure complete extraction), waterin equilibrium with carbonates may exhibit higher δ18Ovalues in the field than in the laboratory extraction unit.As a result, O isotope composition measured by

cryogenic vacuum distillation does not reflect plantwater uptake for our data.

Therefore, plant water uptake (or acquisition) studiesapplying the natural abundance of water isotopes inmost cases yield valid results as both water taken upby plants and obtained by cryogenic vacuum extractionare subject to the same exchange/alteration processesthat may occur in the soil. Only for soils with highcarbonate and low water content, extraction artefactsmay compromise O isotope results.

Conclusion

We conclude that soil physical properties (e.g. clay con-tent) and/or chemical properties (e.g. carbonate content)may lead to isotopic fractionation of water added to soil,especially in soils with low soil water content. In mostcases cryogenic vacuum extraction and plant water up-take rely on the same – altered – water reservoir and thusthe fractionation observed in laboratory experiments willnot affect results of water uptake depth studies in general.However we could show that under special circumstancesartifacts during water extraction in the laboratory mayoccur. It is noteworthy that under such circumstances theanalysis of both O and H isotopes may help to reveal andcorrect for such artifacts and lead to more reliable data.

Acknowledgments This study was conducted within the frame-work of the Research Training Group “Graduiertenkolleg 1086:The role of biodiversity for biogeochemical cycles and bioticinteractions in temperate deciduous forests”, funded by the DFG(German Research Foundation). We thank Christina Langenbruchand Anja Guckland for providing data on soil properties andReinhard Langel for excellent technical assistance.

References

Araguás-Araguás L, Rozanski K, Gonfiantini R, Louvat D (1995)Isotope effects accompanying vacuum extraction of soil wa-ter for stable isotope analyses. J Hydrol 168(1):159–171

Beaudoin G, Therrien P (2009) The web stable isotope fractionationcalculator. In: De Groot PA (ed) Handbook of stable isotopesanalytical techniques, vol. 2, Elsevier, pp 1054–1047, http://store.elsevier.com/product.jsp?isbn=9780080930114&_requestid=410576

Brunel JP, Walker GR, Walker CD, Dighton JC, Kennett-Smith A(1991) Using stable isotopes of water to trace plant wateruptake. Stable isotopes in plant nutrition, soil fertility andenvironmental studies, Pro. Int. Conf., IAEA, Vienna, 1–5October 1990, IAEA, Vienna, 543–551

Plant Soil (2014) 376:327–335 333

Butz-Braun R (2001) Forstliches umweltmonitoring in Thüringen.In: Mitteilungen der Landesanstalt für Wald undForstwirtschaft, Gotha, pp 44–45

Clayton RN (1961) Oxygen isotope fractionation between calciumcarbonate and water. J Chem Phys 34:724–726

Dawson TE (1993) Hydraulic lift and water use by plants:Implications for water balance, performance and plant-plantinteractions. Oecologia 95(4):565–574

Dawson TE, Ehleringer JR (1991) Streamside trees that do not usestream water. Nature 350(6316):335–337

Dawson TE, Ehleringer JR, Hall AE, Farquhar GD (1993) Watersources of plants as determined from xylem-water isotopiccomposition: Perspectives on plant competition, distribution,and water relations. In: Ehleringer JR, Hall AE, Farquhar GD(ed) Stable isotopes and plant carbon-water relations,Academic Press Inc., pp 465–496

Eggemeyer KD, Awada T, Harvey FE, Wedin DA, Zhou X,Zanner CW (2009) Seasonal changes in depth of wateruptake for encroaching trees Juniperus virginiana and Pinusponderosa and two dominant C4 grasses in a semiarid grass-land. Tree Physiol 29(2):157–169

Ehleringer JR, Osmond CB (1989) Stable isotopes. In: Pearcy RW,Ehleringer JR, Mooney HA, Rundel PW (eds) Plant physio-logical ecology: Field methods and instrumentation.Chapman and Hall, New York, pp 281–300

Ehleringer JR, Phillips SL, Schuster WS, Sandquist DR (1991)Differential utilization of summer rains by desert plants.Oecologia 88(3):430–434

Ellsworth PZ, Williams DG (2007) Hydrogen isotope fraction-ation during water uptake by woody xerophytes. Plant Soil291(1–2):93–107

Flanagan LB, Ehleringer JR, Marshall JD (1992) Differentialuptake of summer precipitation among co-occurring treesand shrubs in a pinyon-juniper woodland. Plant CellEnviron 15(7):831–836

Gat JR (1996) Oxygen and hydrogen isotopes in the hydrologicalcycle. Annu Rev Earth Planet Sci 24:225–262

Gee GW, Bauder JW (1986) Particle-size analysis. In KluteA (ed) Methods of soil analysis: Part 1. Physical andmineralogical methods, 2nd ed. Agronomy Monograph9, pp 383–411

Gehre M, Geilmann H, Richter J, Werner RA, Brand WA (2004)Continuous flow 2H/1H and 18O/16O analysis of watersamples with dual inlet precision Rapid Commun. Mass Sp18(25):2650–2660

Guckland A (2009) Nutrient stocks, acidity, processes of N trans-formation and net uptake of methane in soils of a temperatedeciduous forest with different abundance of beech (Fagussylvatica L.). Dissertation, University of Göttingen

Hsieh JC, Savin SM, Kelly EF, Chadwick OA (1998)Measurement of soil-water δ18O values by direct equilibra-tion with CO2. Geoderma 82(1):255–268

Jackson PC, Cavelier J, Goldstein G, Meinzer FC, Holbrook NM(1995) Partitioning of water resources among plants of alowland tropical forest. Oecologia 101:197–203

Jackson PC, Meinzer FC, Bustamante M, Goldstein G, Franco A,Rundel PW, Caldas L, Igler E, Causin F (1999) Partitioningof soil water among tree species in a Brazilian Cerradoecosystem. Tree Physiol 19:717–724

Koeniger P, Marshall JD, Link T,Mulch A (2011) An inexpensive,fast, and reliable method for vacuum extraction of soil and

plant water for stable isotope analyses by mass spectrometry.Rap Commun Mass Spectr 25:3041–3048

Li S-G, Romero-Saltos H, Tsujimura M, Sugimoto A, Sasaki L,Davaa G, Oyunbaatar D (2007) Plant water sources in the coldsemiarid ecosystem of the upper Kherlen River catchment inMongolia: A stable isotope approach. J Hydrol 333:109–117

Lin GH, Sternberg L, Ehleringer JR, Hall AE, Farquhar GD (1993)Hydrogen isotopic fractionation by plant roots during wateruptake in coastal wetland plants. In: Ehleringer JR, Hall AE,Farquhar GD (ed) Stable isotopes and plant carbon-waterrelations, Academic Press Inc., pp 497–510

Meinzer FC, Andrade JL, Goldstein G, Holbrook NM, Cavelier J,Wright SJ (1999) Partitioning of soil water among canopy treesin a seasonally dry tropical forest. Oecologia 121:293–301

MidwoodAJ, Boutton TW, Archer SR,Watts SE (1998)Water useby woody plants on contrasting soils in a savanna parkland:Assessment with δD and δ18O. Plant Soil 205:13–24

Moore JW, Semmens BX (2008) Incorporating uncertainty andprior information into stable isotope mixing models. EcolLett 11(5):470–480

Oerter E, Finstad K, Schaefer J, Goldsmith GR, Dawson TE,Amundson R (2012) Oxygen isotope fractionation effectsin soil water via cations adsorbed to high-CEC clays. AGUFall Meeting, San Francisco, 3–7 December 2012, http://fallmeeting.agu.org/2012/eposters/eposter/h11b-1171/

Parnell A, Inger R, Bearhop S, Jackson AL (2010) Sourcepartitioning using stable isotopes: Coping with too muchvariation. Plos One 5(3):e9672

Phillips DL, Gregg JW (2001) Uncertainty in source partitioningusing stable isotopes. Oecologia 127(2):171–179

Phillips DL, Gregg JW (2003) Source partitioning using stableisotopes: Coping with too many sources. Oecologia 136(2):261–269

Plamboeck AH, Grip H, Nygren U (1999) A hydrological tracerstudy of water uptake depth in a Scots pine forest under twodifferent water regimes. Oecologia 119:452–460

R Core Team (2012) R: A Language and Environment forStatistical Computing. R Foundation for StatisticalComputing,Vienna, Austria, 2012. http://www.R-project.org. Accessed 01 May 2013

Rosatto DR, Silva LDCR, Villalobos-Vega R, Sternberg LDSL,Franco AC (2011) Depth of water uptake in woody plantsrelates to groundwater level and vegetation structure along atopographic gradient in a neotropical savanna. Environ ExpBot 77:259

Savin SM, Epstein S (1970) The oxygen and hydrogen isotopegeochemistry of clay minerals. Geochim Cosmochim Acta34(1):25–42

Schaap MG, Leij FJ, Van Genuchten MT (2001) ROSETTA: Acomputer program for estimating soil hydraulic parameters withhierarchical pedotransfer functions. J Hydrol 251:163–176

Schwendenmann L, Veldkamp E, Moser G, Hölscher D, KöhlerM, Clough Yet al (2010) Effects of an experimental droughton the functioning of a cacao agroforestry system, Sulawesi,Indonesia. Glob Chang Biol 16(5):1515

Simunek J, VanGenuchtenMT, SejnaM (2008) Development andapplications of the HYDRUS and STANMOD softwarepackages and related codes. Vadose Zone J 7(2):587–600

Sofer Z, Gat JR (1972) Activities and concentrations of oxygen-18in concentrated aqueous salt solutions: Analytical and geo-physical implications. Earth Planet Sci Lett 15(3):232–238

334 Plant Soil (2014) 376:327–335

Thorburn PJ, Walker GR (1993) The source of water transpired byeucalyptus camaldulensis: Soil, groundwater, or streams? In:Ehleringer JR, Hall AE, Farquhar GD (eds) Stable isotopes andplant carbon-water relations. Academic Press, pp 511–527

Walker GR, Woods PH, Allison GB (1994) Interlaboratory compar-ison of methods to determine the stable isotope composition ofsoil water. Chem Geol (Isot Geosci Sec) 111:297–306

Wershaw RL, Friedman I, Heller SJ (1966) Hydrogen isotopefractionation of water passing through trees. In: Hobson F,Speers M (eds) Advances of organic geochemistry.Pergamon, New York, pp 55–67

West AG, Patrickson SJ, Ehleringer JR (2006) Water extrac-tion times for plant and soil materials used in stableisotope analysis. Rapid Commun Mass Spectrom 20:1317–1321

White JWC, Cook ER, Lawrence JR, Broecker WS (1984)The D/H ratios of sap in trees: Implications for watersources and tree ring D/H ratios. Geochim CosmochimActa 49:237–246

Zeebe RE (2009) Hydration in solution is critical for stable oxygenisotope fractionation between carbonate ion and water.Geochim Cosmochim Acta 73:5283–5291

Plant Soil (2014) 376:327–335 335