march/april 2013, Vol. 89, No. 2 — The ForesTry chroNicle 169

IntroductionWater shortages are a critical factor that constrains the distri-bution and abundance of plants in semi-arid areas of theworld. The capacity of plants access different water resourcesis associated with their life forms and distribution (Dodd et al.1998). Many perennial plants in semi-arid and seasonally dryareas have a dimorphic root system, which utilizes soil mois-ture in the upper soil layers during wet seasons and penetratesto deeper layers to take up water when the dry season arrives( Zapater et al. 2011, Kray et al. 2012). This seasonal shift of

Seasonal water use patterns of semi-arid plants in Chinaby Guodong Jia1, Xinxiao Yu2, Wenping Deng1

ABSTRACTWater sources of woody plants in semi-arid or seasonally dry areas of China are little known. This study investigated thedifferences in water sources for plants due to seasonal changes (wet/transitional and dry seasons) in semi-arid areas. Sta-ble isotope techniques were applied to determine plant water sources in different seasons. The results show that there isgenerally a switch of water sources from shallow depths in the rainy season to lower depths in the dry season. This studyhighlights how seasonal changes in climate in semi-arid China affect plant water uptake and suggests that further studywith replicated systematic experiments are needed to better understand the responses in water use patterns to changes inenvironmental conditions in drought-prone areas.

Keywords: water use patterns, Platycladus orientalis, multi-source mass balance, semi-arid China

RÉSUMÉLes sources d’eau disponibles pour les plantes ligneuses des zones semi-arides ou qui ont une saison sèche en Chine sontpeu connues. Cette étude se penche sur les différences entre les sources d’eau pour les plantes lors des changements de sai-sons (saison pluvieuse ou de transition, saison sèche) dans les zones semi-arides. Les techniques faisant appel à des iso-topes stables ont été utilisées pour identifier les sources d’eau des plantes à différentes saisons. Les résultats indiquent qu’ily a habituellement un passage des sources d’eau à faibles profondeurs au cours de la saison pluvieuse, vers des sources plusprofondes au cours de la saison sèche. Cette étude montre comment les changements saisonniers du climat dans la zonesemi-aride de la Chine affectent l’absorption de l’eau par les plantes et laissent croire qu’il faudra d’autres études avec desdispositifs répétés systématiquement mieux comprendre comment le modèle d’utilisation de l’eau change avec les saisonsdans les zones exposées à la sécheresse.

Mots clés : modèles d’utilisation de l’eau, Platycladus orientalis, bilan de masse multi-sources, zone semi-aride de la Chine

1PhD candidate, College of Soil and Water Conservation, Beijing Forestry University, Beijing, China.2Professor, College of Soil and Water Conservation, Beijing Forestry University, Beijing, China. Corresponding author. E-mail: yuxinxiao111@126.com

Guodong Jia

water resources is significantfor species growing in semi-arid areas (Moreno andBertiller 2012). Traditionalstudies to determine root dis-tribution have used excavationmethods, which are destruc-tive, time-consuming and notsuitable for long-term research(Meinzer et al. 2001). In addi-tion, root distribution may notreflect the water uptake depthsof plants (Moreira et al. 2000).Analysis of isotopic composi-

tions of water from plant stems and soil materials has beenapplied in ecological studies as an alternative approach tostudy water uptake by roots (Zencich et al. 2002, Nie et al.2011). There is limited isotopic fractionation by plant rootsduring water uptake (Dawson and Ehleringer 1991) and theisotopic compositions of xylem water are in accordance withthe isotopic values of soil water where the plants absorb thewater. By comparing hydrogen and oxygen isotopic composi-tions from xylem water with potential water sources, propor-tional contributions of water sources to the plants may bedetermined (Ehleringer and Dawson 1992). Snyder and

Wenping DengXinxiao Yu

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

170 mars/aVril 2013, Vol. 89, No 2 — The ForesTry chroNicle

Williams (2000) found that Goodding’s willow (Salix good-dingii C.R. Ball) only utilized groundwater during the rainyseason. Duan et al. (2008) showed that Chinese pine (Pinustabulaeformis Carr.) derived 92.5% of its xylem water fromrain water during the growing season. McCole and Stern(2007) found that Ashe’s juniper (Juniperus ashei J. Buchholz)used groundwater in the summer and soil water during thewinter. However, these studies took advantage of a two- orthree-compartment linear mixing approach, which is limitedto calculating contributions of no more than three watersources (Burgess et al. 2000, Phillips and Gregg 2001). In fact,plants may have multiple sources of water, including differentsoil depths and rainwater. Phillips and Gregg (2003) proposeda multi-source mass balance approach, which resolved thisrestriction and could assess contributions of different watersources (Asbjornsen et al. 2007). Using the softwareIsoSource (Phillips and Gregg 2003), the fractional contribu-tion of each potential source is determined and the results arepresented as a distribution range of feasible solutions ratherthan unique solutions. This may provide a more realisticapproach in plant water uptake studies and avoid misinter-pretation of results (Benstead et al. 2006).

Chinese arborvitae (Platycladus orientalis [Linn.] Franco)is widely planted in semi-arid China (Chen et al. 2009). How-ever, how it survives and even thrives in semi-arid ordrought-prone areas is unknown and hence it is important tounderstand seasonal water use patterns and dynamics. In thisstudy, stable isotope techniques were applied with the multi-source mass balance analysis approach to examine the depthof water uptake in the soil horizon and determine the propor-tional contributions of different water sources to total plantwater uptake in different seasons. The goals of this studywere: 1) to determine the proportional contribution of waterfrom different soil depths to total plant water uptake; 2) toevaluate the utility of standard stable isotope analysis basedon single-source direct inference and multi-source mass bal-ance approaches; and 3) to enhance understanding of howseasonal change alters plant water use dynamics and theimplications for hydrological functioning in semi-arid areas.

Materials and MethodsSite descriptionThe study site was a 50-year-old plantation of Chinesearborvitae. Research was carried out July to November, 2011at the Jiufeng National Forest, Beijing (Fig. 1). This is anexperimental forest established in the 1960s to increase forestcover in the western part of Beijing and planted with a varietyof coniferous species, but predominantly P. orientalis. Standdensity is 1176 trees/ha with a mean tree height of 10.7 m anda DBH of 20.9 cm. Soil type is sandy loam, with a texture con-sisting mainly of sand, silt and clay. The maximum depth is120 cm above weathered bedrock. Soil pH ranges between 6.4and 8.4; mean total nitrogen and mean total phosphorus are4.87 g kg-1 and 101.50 mg kg-1, respectively. The climate is hotand rainy in summer (June to September), cold and dry inwinter (November to February). Mean annual temperature is11.6°C with an average of 193 frost-free days. Mean annualprecipitation is approximately 630 mm, 70% to 80% fallsbetween July and September.

Meteorological dataMean daily and monthly precipitation was monitored by theChinese Forest Ecosystem Research Network (CFERN)weather station located some 300 m from the sampling site(Fig. 1).

Collection of samples for isotopic analysesSamples of each rain event were collected into a plastic tankto avoid evaporation and immediately put into 50 ml cappedvials, sealed with Parafilm ‘M’® and frozen to -20°C until iso-topic analysis. Branch samples of P. orientalis were collectedfrom three plots at each of three sampling sites on July 16 andJuly 23 (wet season), October 16 and October 24 (transitionalseason), November 7 and November 19 (dry season) repre-senting different periods of water use. The area of each plotwas 10 m � 10 m. At each plot, three soil sampling points andthree to five adjacent plants of P. orientalis were selected ran-domly for testing variations in isotopic compositions. Soilsamples (two replicated samples/depth) were collected at 10-cm intervals to 120 cm using a soil auger. All plant and soilsamples were placed immediately into 50 ml capped vialsafter collection, sealed with Parafilm ‘M’® and frozen to -20°Cuntil isotopic analysis.

Isotopic analysesWater was extracted from branches and soil samples using acryogenic vacuum distillation approach (West et al. 2006).The δ2H and δ18O isotope ratios of the branch water, soilwater and precipitation were measured with a liquid waterisotope laser spectroscopy instrument (Model DLT-100, LGRInc.) (Lis et al. 2008) at the Key Laboratory of Soil and WaterConservation, Beijing Forestry University. The δ2H and δ18Ovalues are expressed as the 2H/H and 18O/16O relative toVienna Standard Mean Ocean Water (VSMOW) in ‰,respectively. VSMOW is a water standard defining the iso-topic composition of freshwater. This standard includes boththe established values of stable isotopes found in water andcalibration materials provided for standardization and inter-laboratory comparisons of instruments used to measure thesevalues in experimental materials.Fig. 1. Location of the study site.

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

march/april 2013, Vol. 89, No. 2 — The ForesTry chroNicle 171

[1] δ2H or δ18O = ( Rsample

�1) � 1000Rstandard

where and are the molar ratios of 2H/H or 18O/16O of thesample and VSMOW. Measurement precision was better than±0.32 ‰ and ±0.17 ‰ for δ2H and δ18O, respectively(Berman et al. 2009).

Analysis and statisticsBased on similar δ18O values of different soil depths on eachsampling day, soil profiles (0–120 cm) were subdivided intofour or five depth intervals to facilitate root water uptake(Table 1). The isotopic composition for each depth wasaccording to the average value of samples within each inter-val. The isotope values of each interval and the stem waterwere analyzed by the IsoSource software (the multi-sourcemass balance approach) to evaluate the contribution of eachsoil depth to stem water. The multi-source mass balanceapproach analyzed the data more systematically and provideda more quantitative assessment of the estimated range of fea-sible contributions of potential water sources from differentsoil depths to total water uptake. The fractional incrementwas set at 1%, and the uncertainty level was set at 0.2 ‰.

ResultsIsotopic composition of rainwater Fig. 2 shows that δ2H and δ18O values ranged from -119.4 ‰to -19.3 ‰ and -15.5 ‰ to -1.1 ‰, respectively, with standarddeviations of 23.5 ‰ and 3.4 ‰, respectively. The larger stan-dard deviation of δ2H and δ18O values indicated the signifi-cant seasonal variations in rain water isotopic compositions.The Global Meteoric Water Line (GMWL) indicates the aver-age relationship between hydrogen and oxygen isotope ratiosin natural terrestrial waters, expressed as a worldwide average(Craig 1961). A meteoric water line may be calculated for agiven area, and used as a baseline within that area. Kineticfractionation will cause the isotope ratios to vary betweenlocalities within that area. This relationship is used within thefield of isotope hydrology. Craig’s original assertion is thatisotopic enrichments, relative to ocean water, display a linearcorrelation over the entire range for water that has not under-gone excessive evaporation. The local meteoric water line forrainwater was δ2H = 6.82δ18O - 2.94 (R2 = 0.90, n = 35), forwhich the slope and the intercept were smaller than theGMWL δ2H = 8δ18O + 10), indicating isotopic evaporativeenrichment in precipitation in this area.

Isotopic compositions of soil and branch waterStandard stable isotope analysis based on single-source directinference assumed that plants were extracting water from a

single dominant soil depth deter-mined by the intersection of δ18O val-ues of the stem water and the soil pro-file. Fig. 3 shows that in the wetseason, soil isotopic values of δ18Onear the surface did not show evapo-rative isotopic enrichment, and varia-tions of δ18O values for differentdepths were not significant. However,before and during the dry seasons,isotopic values of δ18O in the upperhorizon were less negative due toenrichment associated with greaterevaporation and little precipitation.Isotopic values of the deeper soil water(below 40 cm) stayed relatively con-stant during all seasons, ranging from-8.9 ‰ to -6.9 ‰ with a standarddeviation of 0.67 ‰. Isotopic plantsignatures corresponded to isotopiccompositions in the soil; conse-quently, the variation in isotopic com-positions of soil profiles in differentseasons provided a direct way to studythe depth plants were absorbing water.

Fig. 3 also shows that P. orientalismainly obtained water from the upper40 cm on July 16, October 16, andNovember 19. On other samplingdates, uptake was below 40 cm. How-ever, specific depths of water uptakewere less clear and not determineddue to the more complicated isotopicgradient for soil water.

Fig. 2. Isotopic compositions of rainwater during different sampling periods.

Table 1. Subdivided soil depth intervals on different sampling dates

Soil depth intervals Sampling dates

0-20 cm20-40 cm40-80 cm>80 cm July 16 July 23 October 16 October 24 November 19

0-20 cm20-30 cm30-40 cm November 740-80 cm>80 cm

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

172 mars/aVril 2013, Vol. 89, No 2 — The ForesTry chroNicle

Fig. 3. δ18O values (±SD) of soil water and stem water on different sampling dates. ( ) corresponds to the δ18O value of the stem.

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

march/april 2013, Vol. 89, No. 2 — The ForesTry chroNicle 173

Fig. 4. Frequency histograms at different sampling dates showing the range of water sources at different soil depths to total wateruptake. P.F. is proportional frequency and S.P. source proportion.

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

174 mars/aVril 2013, Vol. 89, No 2 — The ForesTry chroNicle

Feasible contributions of potential soil water sourcesEstimates of the contributions of water sources at differentsoil depths to total water uptake are provided by frequencyhistograms (Fig. 4). On each sampling day, the frequency his-togram was clear for a certain subdivided soil depth, whileothers were broad and diffuse. P. orientalis received a signifi-cant portion (72% to 79%) of water from the 20–40 cm depth.As the wet season progressed, histogram patterns also becamebroad and diffuse. In the transitional and dry seasons, treesobtained moisture mainly from lower soil levels (40–80 cm)and the proportions of soil water to total plant water uptakewere 41% to 84%. As the dry season increased, plants utilizeddeeper soil water. However, in the transitional and dry sea-sons with earlier rainfall, P. orientalis shifted to the surface soil(0–20 cm) to withdraw a relatively large proportion of water:28% to 37% (mean 31.8 %) and 53% to 61% (mean 57%) onOctober 16 and November 19, respectively (Fig. 4). Althoughthere was rainfall on October 23 (Fig. 5), surface soil watercontributed little to total plant water uptake (Fig. 4).

DiscussionSeasonal isotopic variations in soil waterVariations in isotopic composition of upper soil water weredue to combined effects of evaporation and earlier rainfall,while compositions in deeper soil layers were essentiallydetermined by previous rainfalls (Song et al. 2009). Isotopicsoil profiles varied significantly in the upper soil layers ( 0–40cm) between seasons, while soil layers below 40 cm remainedrelatively constant (Fig. 3). The dissimilarities between upperand lower soil depths could be due to differences in soil mois-ture evaporation rates and infiltration rates at different depthsin different seasons. Soil moisture evaporation ratesdecreased as soil depth increased (Barnes and Allison 1983,Gazis and Feng 2004), an important factor that led to isotopicfractionation of the soil profile. Infiltration rates at differentsoil depths could be partially due to the soil moisture deficit(Robinson 1999) before precipitation in different seasons. Inthe wet season, the isotopic values of δ18O at all soil depthsdid not vary much (Fig. 3b). This could be due to 1) relatively

Fig. 5. Mean daily relative humidity (RH), precipitation and temperature in 2011. (•) corresponds to the δ18O value of the rainwater.

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

march/april 2013, Vol. 89, No. 2 — The ForesTry chroNicle 175

higher relative humidity (Fig. 5), which led to a decrease invapor pressure deficit (Wilson et al. 2001, Farquhar et al.2007) so that evaporation at the surface soil was not signifi-cant; 2) frequent precipitation recharged the soil water, whichled to no significant variations (Zencich et al. 2002); 3) thesoil moisture content was above the capillary breaking pointdue to frequent precipitation and the movement of soil waterto the evaporation surface was mainly as liquid, which slowedthe evaporation rate (Benoit and Kirkham 1963); and/or 4)the possibility that water moved through the profile throughlarge pores or “preferential flow” paths, which would havepreserved the isotopic composition of the rain (Gazis andFeng 2004, Brooks et al. 2009). In contrast, the correlationbetween soil moisture evaporation rate and temperature wasnot significant (p > 0.1). Isotopic compositions of all soildepths were determined to be a mixture of previous rainfalls(Fig. 5 and Fig. 3, July 16 and July 23), so there were noremarkable variations in the δ18O values of the vertical gradi-ents of the soil profile.

In the transitional (October 16 and 24) and dry (Novem-ber 7 and 19) seasons, with long periods of drought and lim-ited rain, the soil moisture deficit of the upper soil layers(0–40 cm) increased and the mean relative humiditydecreased (Fig. 5). Thus, rainwater could stay near the soilsurface for longer periods with longer exposure times forevaporation, resulting in a significant evaporative enrich-ment in the upper soil horizon (Fig. 3). However, the rela-tively less negative δ18O values of all depths on October 24are due to the precipitation on October 23 after soil waterrecharged on October 13 (Fig. 5). Another finding was thatdeeper soil (below 40 cm) was barely affected by evaporationand the δ18O values were a mixture of previous rainfall, sothe values of δ18O below 40 cm were more negative in thetransitional and dry seasons than in the wet season (Fig. 3).As there was no rain between October 24 and November 16,δ18O values below 40 cm should be negative; however, theδ18O value at 40–60 cm November 7 was more negative thanon October 24 (Fig. 3), which suggests the possibility of“hydraulic redistribution” (Brooks et al. 2002). Soil wateruptake depth by roots on November 7 also supports thisfinding (Fig. 3e and Fig. 4e).

Seasonal isotopic variations in plant waterThis study employed two different approaches in an effort tosee how P. orientalis depends on upper or lower water sources,and as a result, interpretations are limited. In the wet season,both the direct inference approach and the multi-source massbalance approach suggest that water sources for P. orientaliswas dominated by surface soil water (Fig. 3a and Fig. 4a). Asthe wet season progressed with greater precipitation, it wasdifficult to determine which depth of soil was the main watersource. There was a possibility that a mixture of soil waterfrom different depths was the principal water source. How-ever, plants tend to utilize soil water that is more accessible(higher soil moisture content). There is also the possibilitythat the surface soil still dominated the water source up to90% (Fig. 4b) due to frequent precipitation that fullyrecharged the surface layers (Fig. 5), suggesting that P. orien-talis depends more on summer rains. P. orientalis is a coniferwith a relatively shallow root system (Zhao et al. 2006). In therainy season, coniferous species with developed lateral roots

largely rely on rain water (Flanagan et al. 1992, Valentini et al.1992). Li et al. (2007) found that Larix sibirica Ledeb. onlyused recent rainfall in July (the growing season) in semi-aridareas of Mongolia, which is similar to the results here. Thehigh dependency on summer rains by P. orientalis suggests itshigh water use efficiency as well (Ehleringer et al. 1991, Wanget al. 2007).

In the dry season, the direct inference approach showeddifferential water uptake (Fig. 3c–f). IsoSource calculationsshowed the same patterns (Fig. 4c–f). This suggests that themain water source for P. orientalis was at lower soil depthsafter long periods without rain. This is similar to the findingsof Romero-Saltos et al. (2005). Roots utilized water in theupper soil layers that was recharged by the earlier rainfall (Fig.5). Furthermore, the contributions of upper soil layers to totalwater uptake were different on October 24 and November 19,although with similar previous precipitation (Fig. 4). Thesedata suggest that previous summer rains that penetrated intothe deeper soil layers contributed more to total plant wateruptake in the transitional season than in the dry season (Fig.4c). After a lengthy period of drought, deeper water fromsummer rains decreased and contributed less. White et al.(1985) found that Pinus strobus L. absorbed water from lowerdepths during a drought period but obtained water from theupper soil when rain occurred. This supports the findings inthis study. However, rainfall on October 23 contributed littleto plant water utilization (Fig. 4d), which suggests that precip-itation reaches a minimum threshold to recharge the soil forroots in the upper layers. Soil moisture in drought-proneareas varies significantly even in adjacent areas and plantsgenerally change their water acquisition spatial patterns inresponse (Schlesinger et al. 1990). The root zone can obtainmore water where soil moisture availability is normal thanwhere it is lacking to maintain the growth of plants(Simoneau and Habib 1994, Green and Clothier 1995). Thefindings in this study are consistent with other studies thatshow that a number of plants in arid and semi-arid areasobtain surface soil water during the wet season and shift todeeper soil layers during the dry season. They possess a“dimorphic root”, which is a root system with two distinctroot forms adapted to perform different functions (Chimnerand Cooper 2004, Hasselquist and Allen 2009). One of themost common manifestations is in plants with both lateralroots and a taproot that grows straight down to the watertable. Dimorphic root systems are advantageous for largewoody species where the acquisition of seasonally limitedwater is a premium.

The two approaches applied in this study have differentcharacteristics. The direct inference approach provides visualassessment and indicates possible water sources but ignoresthe possibility that plants obtain water at different soil depths(Wang et al. 2010). The multi-source mass balance approachovercomes this. There are three advantages of this approach:1) it provides a method that analyzes the data more systemat-ically, which reduces misinterpretations from observers’ bias;2) it permits sensitivity analysis by applying different stan-dards; and 3) it provides a more quantitative assessment of therange of feasible contributions of water sources from differentsoil depths to total water uptake (Asbjornsen et al. 2007).

In addition, the data suggest that P. orientalis dependslargely on summer rainfall and competes for water with

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

176 mars/aVril 2013, Vol. 89, No 2 — The ForesTry chroNicle

annual shallow-rooted herbs and shrubs. Undergrowth vege-tation in this forest ecosystem may grow deep roots to surviveduring the growing season. Water is a key factor that con-strains the distribution and abundance of plants in semi-aridareas. From fieldwork carried out earlier in 2010 the ecologi-cal biodiversity index of the P. orientalis forest ecosystem waslow, suggesting the lack of available water for undergrowthvegetation and an inability to compete for water with P. orien-talis. Furthermore, the high dependence on summer rainfallimplies the low leaf water potential of P. orientalis (Ehleringeret al. 1991), suggesting the ability to survive from seriouswater stress.

ConclusionsThis study examined seasonal water use patterns of a 50-year-old P. orientalis plantation. The results have important impli-cations for plant species in semi-arid China. Both the directinference approach and the multi-source mass balanceapproach demonstrated similar water use patterns despite thedifferent characteristics between these two methods. There isgenerally a switch of water sources from shallow soil depths inthe rainy season to lower depths in the dry season. Thisswitch may be due to their dimorphic root systems and var-ied rainfall distribution between seasons. In addition, there isa minimum threshold of precipitation for P. orientalis toobtain water by roots in the upper soil in the dry season afterlong periods of drought. Furthermore, the ability to competeand high dependence on summer rainfall of P. orientalisshould be one of the more widely planted species in water-limited areas of China.

We acknowledge that no replications were made. Furtherstudies with more sampling periods and more plant speciesare needed to determine seasonal water use patterns andplant–soil–water cycling processes to better understand dif-ferential responses in water use patterns among species indrought-prone areas.

AcknowledgementsThis research was financially supported by grant No.41171028 from the National Natural Science Foundation ofChina, and was also supported by CFERN & GENE AwardFunds on Ecological Paper. Funding was also provided byBeijing Forestry University for preparing this manuscript.Thanks to Jianbo Jia,Yajun Li, Yujie Liu and Yanjing Bai fortheir assistance with the field work and lab measurements.

ReferencesAsbjornsen, H., G. Mora and M.J. Helmers. 2007. Variation inwater uptake dynamics among contrasting agricultural and nativeplant communities in the Midwestern U.S. Agric. Ecosyst. Environ.121(4): 343–356.Barnes, C.J. and G.B. Allison. 1983. The distribution of deuteriumand 18O in dry soils. 1. Theory. J. Hydrol. 60(1–4): 141–156.Benoit, G.R. and D. Kirkham. 1963. The effect of soil surface con-ditions on evaporation of soil water. Soil Sci. Soc. Amer. 27(5):495–498.Benstead, J.P., J.G. March, B. Fry, K.C. Ewel and C.M. Pringle.2006. Testing IsoSource: stable isotope analysis of a tropical fisherywith diverse organic matter sources. Ecology. 87(2): 328–338.Berman, E., M. Gupta, C. Gabrielli, T. Garland and J.J. McDon-nell. 2009. High-frequency field-deployable isotope analyzer forhydrological applications. Water Resour. Res. 45: W10201, DOI:10.1029/2009WR008265

Brooks, J.R., F.C. Meinzer, R. Coulombe and J. Greg. 2002.Hydraulic redistribution of soil water during summer drought intwo contrasting Pacific Northwest coniferous forests. Tree Physiol.22: 1107–1117.Brooks, J.R., H.R. Barnard, R. Coulombe and J.J. McDonnell.2009. Ecohydrologic separation of water between trees and streamsin a Mediterranean climate. Nature. 3: 100–104.Burgess, S., M.A. Adams, N.C. Turner and B. Ward. 2000.Charac-terisation of hydrogen isotope profiles in an agroforestry system:implications for tracing water sources of trees. Agr. Water Manage.45: 229–241.Chen, L.D., J.P. Wang, W. Wei, B.J. Fu and D.P. Wu. 2009. Effectsof landscape restoration on soil water storage and water use in theLoess Plateau Region, China. Forest Ecol. Manag. 259(7):1291–1298.Chimner, R.A. and D.J. Cooper. 2004.Using stable oxygen isotopesto quantify the water source used for transpiration by native shrubsin the San Luis Valley, Colorado USA. Plant. Soil. 260(1): 225–236.Craig, H. 1961. Isotopic variations in meteoric waters. Science. 133:1702–1703.Dawson, T.E. 1993. Hydraulic lift and water use by plants: implica-tions for water balance, performance and plant–plant interactions.Oecologia 95: 565–574.Dawson, T.E. and J.R. Ehleringer. 1991. Streamside trees that donot use stream water. Nature. 350: 335–337.Dodd, M.B., W.K. Lauenroth and J.M. Welker. 1998. Differentialwater resource use by herbaceous and woody plant life-forms in ashortgrass steppe community. Oecologia 117(4): 504–512.Duan, D.Y., H. Ouyang, M.H. Song and Q.W. Hu. 2008. Watersources of dominant species in three alpine ecosystems on theTibetan Plateau, China. J. Integr. Plant Biol. 50(3): 257–264.Ehleringer, J.R., S.L. Philips, W. Schuster and D.R. Sandquist.1991. Differential utilization of summer rains by desert plants.Oecologia 88: 430–434.Ehleringer, J.R., T.E. Dawson. 1992. Water uptake by plants: Per-spectives from stable isotope composition. Plant Cell Environ. 15:1073–1082.Farquhar, G.D., L.A. Cernusak and B. Barnes. 2007. Heavy waterfractionation during transpiration. Plant Physiol. 106: 153–168.Flanagan, L.B., J.R. Ehleringer and J.D. Marshall. 1992. Differen-tial uptake of summer precipitation among co-occurring trees andshrubs in a pinyon–juniper woodland. Plant Cell Environ. 15(7):831–836.Gazis, C. and X.H. Feng. 2004. A stable isotope study of soil water:evidence for mixing and preferential flow paths. Geoderma. 119:97–111.Green, S.R. and B.E. Clothier. 1995.Root water uptake by kiwifruitvines following partial wetting of root zone. Plant. Soil. 173(2):317–328.Hasselquist, N.J. and M.F. Allen. 2009. Increasing demands on lim-ited water resources: consequences for two endangered plants inAmargosa Valley, USA. Am. J. Bot. 96(3): 620–626.Kray, J.A., D.J. Cooper and J.S. Sanderson. 2012.Groundwater useby native plants in response to changes in precipitation in an inter-mountain basin. J. Arid Environ. 83: 25–34.Li, S.G., H. Romero-Saltos, M. Tsujimura, A. Sugimoto, L. Sasaki,G. Davaa and D. Oyunbaatar. 2007. Plant water sources in the coldsemi-arid ecosystem of the upper Kherlen River catchment in Mon-golia: A stable isotope approach. J. Hydrol. 333(1): 109–117.Lis, G., L.I. Wassenaar and M.J. Hendry. 2008. High-PrecisionLaser Spectroscopy D/H and 18O/16O Measurements of MicroliterNatural Water Samples. Anal. Chem. 80(1): 287–293.McCole, A.A. and L.A. Stern. 2007. Seasonal water use patterns ofJuniperus ashei on the Edwards Plateau, Texas, based on stable iso-topes in water. J. Hydrol. 342: 238–248.Meinzer, F.C., M.J. Clearwater and G. Goldstein. 2001. Watertransport in trees: current perspectives, new insights and some con-troversies. Environ. Exp. Bot. 45: 239–262.

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.

march/april 2013, Vol. 89, No. 2 — The ForesTry chroNicle 177

Moreira, M.Z., L.S.L. Sternberg and D.C. Nepstad. 2000. Verticalpatterns of soil water uptake by plants in a primary forest and anabandoned pasture in the eastern Amazon: an isotopic approach.Plant. Soil. 222(1–2): 95–107.Moreno, L. and M.B. Bertiller. 2012. Variation of morphologicaland chemical traits of perennial grasses in arid ecosystems. Are thesepatterns influenced by the relative abundance of shrubs? Acta.Oecol. 41: 39–45.Nie, Y.P., H.S. Chen, K.L. Wang, W. Tan, P.Y. Deng and J. Yang.2011. Seasonal water use patterns of woody species growing on thecontinuous dolostone outcrops and nearby thin soils in subtropicalChina. Plant. Soil. 341(1–2): 399–412.Phillips, D.L. and J.W. Gregg. 2001. Uncertainty in source parti-tioning using stable isotopes. Oecologia 127: 171–179.Phillips, D.L. and J.W. Gregg. 2003. Source partitioning using sta-ble isotopes: coping with too many sources. Oecologia 136: 261–269.Robinson, D. 1999. A comparison of soil-water distribution underridge and bed cultivated potatoes. Agr. Water Manage. 42: 189–204.Romero-Saltos, H., L.S.L. Sternberg, M.Z. Moreira and D.C. Nep-stad. 2005. Rainfall exclusion in an eastern Amazonian forest alterssoil water movement and depth of water uptake. Am. J. Bot. 92(3):443–455.Schlesinger, W.H., J.F. Reynolds, G.L. Cunningham, L.F. Huen-neke, W.M. Jarrell, R.A. Virginia and W.G. Whitford. 1990. Bio-logical feedbacks in global desertification. Science 247(4946):1043–1048.Simoneau, T. and R. Habib. 1994.Water uptake regulation in peachtrees with split-root systems. Plant Cell Environ. 17: 379–388.Snyder, K.A. and D.G. Williams. 2000. Water sources used byriparian trees varies among stream types on the San Pedro River,Arizona. Agr. Forest Meteorol. 105: 227–240.Song, X.F., S.Q. Wang, G.Q. Xiao, Z.M. Wang, X. Liu and P. Wang.2009. A study of soil water movement combining soil water poten-tial with stable isotopes at two sites of shallow groundwater areas inthe North China Plain. Hydrol. Process. 23: 1376–1388.

Valentini, R., D.E. Scarasia-Mugnozza and J.R. Ehleringer. 1992.Hydrogen and carbon isotope ratios of selected species of a Mediter-ranean macchia ecosystem. Funct. Ecol. 6: 627–631.Wang, M.C., J.X. Wang, Q.H. Shi and J.S. Zhang. 2007. Photosyn-thesis and water use efficiency of Platycladus Orientalis and RobiniaPseudoacacia saplings under steady soil water stress during differentstages of their Annual growth period. J. Integr. Plant Biol. 49(10):1470–1477.Wang, P., X.F. Song, D.M. Han, Y.H. Zhang and X. Liu. 2010. Astudy of root water uptake of crops indicated by hydrogen and oxy-gen stable isotopes: A case in Shanxi Province, China. Agr. WaterManage. 97: 475–482.West, A.G., S.J. Patrickson and J.R. Ehleringer. 2006.Water extrac-tion times for plant and soil materials used in stable isotope analysis.Rapid Commun. Mass Spectrom. 20: 1317–1321.White, J.W.C., E.R. Cook, J.R. Lawrence and W.S. Broecker. 1985.The D/H ratios of sap in trees: Implications for sources and tree ringD/H ratios. Geochim. Cosmochim. Ac. 173(2): 317–328.Wilson, K.B., P.J. Hanson, P.J. Mulholand, D.D. Baldocchi andS.D. Wullschleger. 2001.A comparison of methods for determiningforest evapotranspiration and its components: sap-flow, soil waterbudget, eddy covariance and catchment water balance. Agr. ForestMeteorol. 106: 153–168.Zapater, M., C. Hossann, N. Breda, C. Brechet, D. Bonal and A.Granier. 2011. Evidence of hydraulic lift in a young beech and oakmixed forest using 18O soil water labelling. Trees-Struct. Funct.25(5): 885–894.Zencich, S.J., R.H. Froend, J.V. Turner and V. Gailitis. 2002. Influ-ence of groundwater depth on the seasonal sources of water accessedby Banksia tree species on a shallow, sandy coastal aquifer. Oecolo-gia 131(1): 8–19.Zhao, Z., P. Li, W.P. Xue and S.W. Guo. 2006. Relation betweengrowth and vertical distribution of fine roots and soil density in theWeibei Loess Plateau. Front. Forest. China. 1(1): 76–81.

The

For

estr

y C

hron

icle

Dow

nloa

ded

from

pub

s.ci

f-if

c.or

g by

Bei

jing

Fore

stry

Uni

vers

ity o

n 06

/16/

13Fo

r pe

rson

al u

se o

nly.