Tracing Changes in Ecosystem Function under …basin.yolasite.com/resources/Pataki et al. 2003...

September 2003 Vol 53 No 9 BioScience 805

Articles

The responses of terrestrial ecosystems to risingconcentrations of atmospheric carbon dioxide (CO2)

and the resulting global changes are still not fully understoodHumans emitted 6 gigatons of carbon per year into the at-mosphere from fossil fuel burning and cement productionduring the 1990s yet only about half of this carbon accu-mulated in the atmosphere Of the remainder about halfwas absorbed by the oceans and half by terrestrial ecosystems(IPCC 2001) The mechanisms that govern the magnitude ofnet carbon uptake by the terrestrial biosphere are unknownbut enhanced growth under elevated atmospheric CO2 iscommonly proposed (Schimel et al 2000) Although indi-vidual plants often respond to elevated CO2 by fixing morecarbon in photosynthesis and thereby increasing produc-tion of biomass this does not necessarily result in long-termremoval of CO2 from the atmosphere Long-term removal re-quires the transfer of fixed carbon into long-lived ecosystempools such as woody biomass and recalcitrant soil carbon(Koumlrner 2000)

Researchers have exposed a variety of plant communitiesto elevated CO2 levels to improve our understanding ofwhole-ecosystem responses Such studies have used open-topchambers closed chambers natural CO2 springs and free-airCO2 enrichment (FACE) These experiments are critical to ourability to determine the fate of carbon in the terrestrial bio-sphere In addition to increasing the rate of fixation of car-bon in photosynthesis elevated atmospheric CO2 levels mayindirectly cause other changes in ecosystem function Plantsmay for example reduce their uptake of water as a result ofclosing their stomata in response to elevated levels of at-mospheric CO2 (Morison 1998) The exchange of carbonand water with the atmosphere is also closely related to thecycling of nitrogen which generally limits growth in temperate systems (Zak et al 2000) Changes in the ratio of carbon tonitrogen in plant material may influence the quality of foodavailable for herbivores Thus elevated atmospheric CO2levels can affect populations at different trophic levels (Lindroth 1996a Owensby et al 1996)

Diane E Pataki (e-mail patakibiologyutahedu) is a science officer and James R Ehleringer is a Focus 1 leader for the International GeospherendashBiosphere ProgrammersquosGCTE (Global Change and Terrestrial Ecosystems) Focus 1 project Pataki and Ehleringer are also professors in the Department of Biology University of Utah SaltLake City UT 84112 David S Ellsworth is a professor in the School of Natural Resources and Environment University of Michigan Ann Arbor MI 48109 R DaveEvans is a professor in the School of Biological Sciences Washington State University Pullman WA 99164 Miquel Gonzalez-Meler is a professor in the Departmentof Biological Sciences University of Illinois Chicago IL 60607 John King is a professor in the School of Forest Resources and Environmental Science Michigan Tech-nological University Houghton MI 49931 Steven W Leavitt is a professor in the Laboratory of Tree-Ring Research University of Arizona Tuscon AZ 85721Guanghui Lin is an associate research scientist in the Biosphere 2 Center Columbia University Oracle AZ 85623 Roser Matamala is a researcher in the EnvironmentalResearch Division Argonne National Laboratory Argonne IL 60439 Elise Pendall is a professor in the Department of Botany University of Wyoming Laramie WY82071 Rolf Siegwolf is head of Stable Isotopes and Ecosystem Fluxes in the Laboratory of Atmospheric Chemistry Paul Scherrer Institut CH-5232 Villigen PSI Switzer-land Chris van Kessel is a professor and agronomist in the Department of Agronomy and Range Science University of California Davis CA 95616 copy 2003 Ameri-can Institute of Biological Sciences

Tracing Changes in EcosystemFunction under Elevated CarbonDioxide ConditionsDIANE E PATAKI DAVID S ELLSWORTH R DAVE EVANS MIQUEL GONZALEZ-MELER JOHN KINGSTEVEN W LEAVITT GUANGHUI LIN ROSER MATAMALA ELISE PENDALL ROLF SIEGWOLF CHRIS VAN KESSELAND JAMES R EHLERINGER

Responses of ecosystems to elevated levels of atmospheric carbon dioxide (CO2) remain a critical uncertainty in global change research Two keyunknown factors are the fate of carbon newly incorporated by photosynthesis into various pools within the ecosystem and the extent to which elevated CO2 is transferred to and sequestered in pools with long turnover times The CO2 used for enrichment in many experiments incorporatesa dual isotopic tracer in the sense that ratios of both the stable carbon-13 (13C) and the radioactive carbon-14 (14C) isotopes with respect to carbon-12 are different from the corresponding ratios in atmospheric CO2 Here we review techniques for using 13C and 14C abundances to followthe fate of newly fixed carbon and to further our understanding of the turnover times of ecosystem carbon pools We also discuss the application ofnitrogen oxygen and hydrogen isotope analyses for tracing changes in the linkages between carbon nitrogen and water cycles under conditions ofelevated CO2

Keywords elevated carbon dioxide stable isotopes radiocarbon global change carbon cycle

The need to develop methods to quantify ecosystem re-sponses to elevated CO2 in relatively small-scale and short-term experiments presents both challenges and opportunitiesfor ecologists Measurements of the abundance of various car-bon hydrogen oxygen and nitrogen isotopes in sampleshave long been used in various subdisciplines of ecology asunique tools that integrate physical and physiological processesover space and time (table 1 box 1) New developments inmass spectrometry allow stable isotope analyses to be con-ducted using very small samples of biomass soil organicmatter or water These advances have improved our abilityto make inferences about plant physiology biogeochemistryand ecosystem function This is particularly valuable in stud-ies of elevated CO2 in which destructive measurements mustbe limited to maintain the integrity of experimental plots forlong-term study Many studies of elevated CO2 effects are intrinsically isotopic tracer experiments because the ratios ofcarbon-13 (13C) to carbon-12 (12C) and of carbon-14 (14C)to 12C in the added CO2 are distinct from the ratios in at-mospheric CO2 Plants grown in conditions of elevated CO2that contain such isotopic tracers will incorporate the tracerin carbon-containing compounds The two labels can then beused to trace the fate of carbon newly incorporated into theecosystem under high CO2 conditions and to calculate the residence time of carbon in various pools

Here we discuss current and potential research that com-bines isotope ecology and experiments with elevated CO2 toimprove our understanding of interactions between eco-system function and global change

The carbon-13 labelThe mean d13C (see box 1) of atmospheric CO2 is currentlyabout ndash8 parts per mil (permil) (Keeling and Whorf 2002) In con-trast commercial CO2 originating from geologically ancientcarbon (eg carbon from petroleum byproducts or wells) isoften relatively depleted in 13C with d13C values typicallyranging from ndash27permil to ndash45permil In most experiments with elevated CO2 the increase in CO2 concentration is accom-plished by mixing commercial CO2 with ambient air As a result short-term fluctuations in canopy CO2 concentration

above and below the targeted concentration may occur andthe concentration may vary from one place to anotherAssessing actual exposure of plants to CO2 therefore be-comes a significant challengeA further complication is thatd13C of the commercial CO2 may vary if the source changesover time Periodic sampling of the added CO2 can track itsisotopic composition and subsequent modeling of CO2circulation in CO2 plots can provide estimates of the averageexposure of experimental plants to CO2 and 13CO2 how-ever the best way of measuring plantsrsquo CO2 exposure over timemay be to measure of isotopic ratios of carbon in C4 plants(box 2 Pepin and Koumlrner 2002) Such plants utilize a photo-synthetic pathway in which the initial fixation of carbon occurs through the action of the enzyme phosphoenolpyru-vate (PEP) carboxylase This enzyme has a much lower iso-topic fractionation factor than rubisco (ribulose bisphosphatecarboxylaseoxygenase) the initial carbon-fixing enzyme forplants utilizing the more common C3 photosynthetic pathway (OrsquoLeary 1981 Farquhar et al 1989) that is PEP carboxylase discriminates less against 13C when it fixes carbon The isotopic composition of C4 plants thereforechanges less in response to factors that influence stomatalopening (such as light temperature and humidity) than thatof C3 plants and its variation tracks the isotopic content ofatmospheric CO2

This property of C4 photosynthesis is being used in a Swissexperiment in which a mature temperate forest is fumigatedwith CO2 from tubing distributed throughout the canopy C4plants placed in pots throughout the canopy are periodicallysampled to calculate the effective atmospheric CO2 concen-tration from d13C of C4 plant material (box 2 Pepin andKoumlrner 2002) This method agrees well with estimates ofaverage CO2 level derived from measuring the CO2 concen-tration in sampled air (figure 1) and greatly reduces the needfor air turbulence modeling or rapid gas sampling deviceswhile still providing spatial information about the CO2 gra-dients that may occur in these experiments Once the 13CO2exposure of experimental plants has been characterized the13C label can then be used to follow the fate of experimentallyadded CO2 in the ecosystem

The carbon- 14 labelIf commercial CO2 is derived from fossil sources it containsno measurable 14C because it has effectively all decayed to nitrogen-14 (14N) during its long isolation from the naturalsource of 14C the action of cosmic rays in the atmosphere andon Earthrsquos surface (The half-life of 14C is 5730 years) The ab-sence of 14C in fossil CO2 provides a useful isotopic tracer thatmay show less temporal variation than the 13C isotope ratioWhen the 14C-free commercial CO2 is added to ambient air14C concentrations in the air are diluted This signal is presentin the plots with elevated CO2 in many completed and ongoingFACE experiments but to date only one has attempted to takeadvantage of itAt the FACE experiment in Maricopa Arizonathe 14C-depleted CO2 added to experimental plots resulted incotton plants that contained only 70 of the 14C activity of

806 BioScience September 2003 Vol 53 No 9

Articles

Table 1 Isotopes of common elements and their naturalabundanceElement Isotope Abundance ()

Hydrogen 1H 999852H 0015

Carbon 12C 988913C 11114C lt 0001

Nitrogen 14N 996315N 037

Oxygen 16O 997617O 00418O 020

Note All the isotopes except 14C are stable

plants in ambient plotsmdasha change that is easily detectable withcurrent methods Organic carbon pools in soil from plots ex-posed to elevated CO2 were also depleted in 14C compared withsoil from plots exposed to ambient CO2 (Leavitt et al 1994)As with d13C any heterogeneity in plants and soils in the en-riched plots induced by inhomogeneous broadcast of thecommercial CO2 will lead to unevenness in the distributionof the tracer Furthermorebecause 14C is present in such smallconcentrations relative to 13C and because very small sam-ples are used in 14C analysis effects of heterogeneity andcontamination could be magnified in 14C relative to those for 13C

The depleted 14C signal present in studies of elevated CO2that used ldquodeadrdquo CO2 from petrochemical sources shouldnot be confused with the 14C signal that was artificially gen-erated during thermonuclear bomb testing and contributesto the current ldquonaturalrdquoatmospheric background This ldquobombspikerdquo created a background 14C time marker centered in1964 after which the atmospheric 14C activity declined as thepulse was incorporated into the ocean groundwater plantsand soils This time marker has been very helpful in under-standing carbon cycling in soils among other things (eg Dorrand Munnich 1986 Jenkinson et al 1992 Trumbore 1993)quite apart from the value of 14C in experiments that employfumigation with tracer CO2 If 14C depletion in plots with elevated CO2 is used as a tracer it is with reference to currentlevels of atmospheric 14C in ambient plots which because ofatmospheric tests of thermonuclear bombs are still above 1950levels

Resolving the fate of elevated carbon dioxideUnderstanding the future distribution of fixed atmosphericcarbon in the terrestrial biosphere is a major goal in many


Articles

Isotopes are atoms of an element that differ in thenumber of neutrons Heavier isotopes having moreneutrons form stronger chemical bonds have lowerreaction rates in enzymatic reactions diffuse moreslowly and are more often found in lower-energyphase states in equilibrium reactions such as thephase between vapor and liquid water These proper-ties can be exploited to investigate physical and bio-chemical processes and are the foundation for stableisotope approaches in ecological studies In additionradioactive isotopes such carbon-14 (14C) provideanother tracer particularly useful for dating biologicalmaterial Commonly used isotopes in ecology are given in table 1

Isotopic composition is often expressed as a ratiobetween the heavier and the lighter isotope Expressionof the absolute ratio is impractical because of the largedifferences in natural abundance between isotopes ofthe same element (table 1) so it is more convenient toexpress an isotope ratio relative to a standard

(1) d = (RsampleRstandard ndash 1) acute 1000permil

where R is the molar ratio of the heavy to the light iso-tope and permil is parts per mil

Isotope ratios are expressed relative to common inter-national standards such as the PDB (Pee Dee Belem-nite) formation a standard for the ratio of carbon-13(13C) to carbon-12 with an R value of 00112372

Isotopic composition can also be expressed as a dis-crimination (D) against heavy isotopes in the reactionbetween source and product

(2) D = RsourceRproduct ndash 1

To calculate D from d

(3) D = (dsource ndash dproduct)(1 + dproduct1000)

In contrast 14C is expressed as an activity (concentra-tion) Using conventional radiocarbon (gas propor-tional counting or liquid scintillation) or acceleratormass spectrometer 14C dating methods the 14C activityis measured and expressed as percent modern carbonor as fraction modern carbon in which by conventionmodern carbon activity is defined as the backgroundactivity in 1950 (Stuiver and Polach 1977) All activitymeasurements are corrected for any fractionation tak-ing place in fixation by plants soil processes or labora-tory procedures by normalizing to a standard d13C value of ndash25permil

Box 1 Isotopes in ecology

Figure 1 Determination of experimental carbon dioxide(CO2) concentration at the Swiss Canopy Crane site usingtwo methods air sampling and the carbon isotope com-position of C4 plants placed in the canopy (plusmn 1 standarddeviation) The square on the right refers to the averagedaytime CO2 concentration across all enrichment daysSource Pepin and Koumlrner (2002)

CO2

conc

entra

tions

(mic

rom

oles

per

mol

)

experiments with elevated CO2 Carbon fixed in photosyn-thesis can be transferred into numerous ecosystem pools(figure 2) Will the increase in carbon fixed under elevated atmospheric CO2 conditions be transferred into long-livedpools or will the ecosystem carbon cycle merely be acceler-ated resulting in no long-term net increase in stored carbonTracing the fate of isotopically distinct carbon fixed under elevated atmospheric CO2 offers a unique means of resolv-ing some of these uncertainties at the ecosystem scale

Carbon inputs from isotopically labeled plants to the soilscan be revealed by shifts in the isotopic composition of soilorganic carbon The larger the difference between the carbonisotopic composition of the plants and soils the greater thepotential for success in quantifying inputs and losses Thistracer signal has now been used in many FACE studies (Allison et al 1983 Leavitt et al 1994 1996 Nitschelm et al1997 van Kessel et al 2000 Leavitt et al 2001 Schlesinger andLichter 2001) and chamber studies (Hungate et al 1997Torbert et al 1997) to determine the amount of new carbonthat has been incorporated into soils during the course of anexperiment Such isotopic tracers whether originating fromancient-source CO2 introduced into experimental plots inFACE experiments or from contrived shifts between C3 andC4 vegetation that cause plants and soils to have distinct iso-tope ratios (the traditional approach) can then be further usedto identify the specific organic carbon pools that new carbonhas entered (Jastrow et al 1996) One commonly cited

limitation of this method in studies of elevated CO2 is thatthere is typically no similarly strong isotope labeling in the con-trol plots so that comparing the difference in results between the elevated and control CO2 treatments is notstraightforward This has been overcome by using

middot small subplots of soils within the control plots whosecarbon isotope composition reflects previous growth ofC4 plants and is therefore distinct from that resultingfrom currently growing C3 plants (Allison et al 1983Ineson et al 1996 Cheng and Johnson 1998 Leavitt etal 2001)

middot small subplots exposed to pulsed pure 13CO2 withincontrol plots (Hungate et al 1997 Leavitt et al 2001)

middot carbon dioxide labeled with 13C or 14C and used tofumigate entire control plots in chamber experiments(Lin et al 1999 2001)

middot the small but quantifiable natural difference between13C in control plants and in the local soil organic car-bon (Nitschelm et al 1997)

All of these methods distinguish between the newly fixedcarbon in control plots and the existing soil carbon in orderto follow the transfer of new carbon into various soil poolsIn this way the allocation of new carbon in control plotscan be compared to the fate of new carbon in the elevated CO2treatment It should be noted that some of these methods


Articles

Discrimination of carbon-13 (13C) during photosynthesis in C3 plants is caused by fractionation during diffusion throughstomatal pores and fractionation by carbon-fixing enzymes primarily rubisco (ribulose bisphosphate carboxylaseoxy-genase) These effects were quantified by Farquhar and colleagues (1982)

(1) D13CP = a + (b ndash a) acute cica

where cica is the ratio of intercellular to ambient carbon dioxide (CO2) partial pressure a is the fractionation associatedwith diffusion (44 parts per mil [permil]) and b is the net enzymatic fractionation associated with carboxylation (27permil)Therefore discrimination is a function of cica which is sensitive to a variety of factors that influence the balance of sto-matal conductance and assimilation rate Equation 1 (above) can also be combined with equation 3 from box 1 to predictthe isotopic composition of biomass if the isotopic composition of the air is known

For C4 plants we must separate fractionation factors for rubsico and phosphoenolpyruvate (PEP) carboxlyase The latterfixes carbon in the mesophyll for transport into bundle sheath cells In these cells rubsico is physically isolated from thestomatal cavity There is no fractionation associated with a product if all of the substrate is utilized Therefore we needonly consider the fractionation factor for rubsico to the extent that CO2 or HCO3

ndash (hydrocarbonate) leaks out of bundlesheath cells back into the stomatal cavity These effects were quantified by Farquhar (1983)

(2) D13CP = a + (b4 + b3f ndash a) acute cica

where b4 is the fractionation associated with PEP carboxylase (ndash57permil) b3 is the fractionation associated with rubisco(30permil) and f is the leakiness of the bundle sheath For many species the term (b4 + b3f ndash a) is close to zero such thatD13C may show little environmental variation in C4 plants (Farquhar et al 1989) In this case variability in d13CP which isrelated to D13CP (box 1 eq 3) can be interpreted as a function of dsource (see box 1) In using this approach researchersshould take care that f has been characterized in the species of interest

Box 2 The Farquhar model of photosynthetic discrimination

require the application of different experimental methods inthe elevated CO2 treatments and the control so results shouldbe interpreted with care

Many experiments with elevated CO2 have shown in-creased photosynthetic gain and larger biomass under elevatedCO2 conditions than under ambient conditions (Acklerlyand Bazzaz 1995 Drake et al 1997 DeLucia et al 1999)However another common response is an increase in respi-ratory losses of CO2 from the soil (Zak et al 2000) A criti-cal element in understanding the significance of increased soilrespiration is the partitioning between the autotrophic (plant-derived) and heterotrophic (soil organismndashderived) origin ofthe respired carbon When plants are larger a proportionalincrease in autotrophic respiration is expected as a result ofincreased root biomass Increased heterotrophic respirationmay however be an indicator of an accelerated carbon cyclein which plant carbon inputs are quickly decomposed afterdeposition as litter or senesced roots Stable isotopes are in-creasingly being used to distinguish the origins of soil fluxes(Rochette and Flanagan 1997 Cheng and Johnson 1998Andrews et al 1999 Lin et al 1999 2001 Pendall et al 2001)and provide an alternative to traditional harvest or disturbancemethods (Edwards and Norby 1999 Houmlgberg et al 2001) Itshould be noted that exact partitioning of respiration betweenheterotrophic and autotrophic components is not possiblewith a single isotopic label as the microbial community

immediately surrounding the roots (the rhizosphere) mayrapidly acquire the isotopic signature of labile root exudatesHowever the substrate-induced respiration method in whichsugar derived from C4 plants is applied to soil in an ecosys-tem containing C3 plants (Houmlgberg and Ekblad 1996 Ekbladand Houmlgberg 2001) provides good estimates of the relativecontributions of heterotrophic and autotrophic respirationas autotrophic respiration will have a C3 carbon isotope ratio distinct from microbial decomposition of the C4 sugarFurthermore the use of C4 plants and elevated CO2 isotopetracers allows partitioning of the soil CO2 flux into ldquooldrdquo orpre-label and ldquonewrdquo or current growing season compo-nents (eg Pendall et al 2001)

The elevated 13CO2 label has been used at the Duke FACEexperiment in Durham North Carolina In this pine forestsoil respiration has increased in response to elevated CO2(Andrews and Schlesinger 2001) In 1998 soil respirationincreased by about 25 or 220 grams (g) of carbon persquare meter (m2) per year in the elevated CO2 treatmentsThe heterotrophic component of this flux constituted an in-crease of 80 g carbon per m2 per year which corresponds wellwith the observation of Houmlgberg and Houmlgberg (2002)According to the isotopic results about 70 of the total in-crease in soil respiration in the plots with elevated CO2 orig-inated from carbon that was fixed since the beginning of theexperiment in September 1996 Although the absence of an


Articles

Figure 2 Schematic of ecosystem carbon transfer through various pools with examples of the carbon isotope ratio (d13C) ofcarbon pools from experiments on the effects of elevated carbon dioxide (CO2) Data from Hungate and colleagues (1997) andLin and colleagues (2001)

Leaf d13C (parts per mil)

Site Elevated CO2 Ambient CO2

Duke ndash395 plusmn 02 ndash283 plusmn 01

Corvallis ndash352 plusmn 03 ndash290 plusmn 02

Minnesota ndash351 plusmn 03 ndash272 plusmn 01

Wisconsin ndash389 plusmn 05 ndash273 plusmn 06

Photosynthesis

Litter

Soil organic matter

Leaves Tree rings Roots

Elevated CO2

Ambient CO2

June lsquo93

1992 1993 1994 1995Year

ndash26Jasper Ridge CA

June lsquo95June lsquo94

ndash30

ndash28

ndash40

ndash36

ndash32

ndash28

ndash24

Elevated CO2

Ambient CO2

Carbon movement through stocks

Corvallis OR

Soil

carb

on

d13 C

(par

ts p

er m

il)

Root

s lt

1 m

md1

3 C (p

arts

per

mil)

isotope tracer prevents the use of a comparable partitioningapproach in the control plots these results are a clear indicationthat much of the extra carbon fixed under elevated CO2 con-ditions may be transferred to short-lived carbon pools thatare quickly returned to the atmosphere

With the use of tracers in control plots experiments usingclosed-system mesocosms have found that elevated atmo-spheric CO2 stimulates decomposition of recently fixed carbon but suppresses decomposition of older soil organicmatter (Lin et al 2001) particularly in the presence of highnutrient availability (Calder et al 1992 Cheng and Johnson1998 Cardon et al 2001) In the shortgrass steppe open-topchamber (OTC) experiment in Colorado a change in the pro-portion of C3 and C4 plants which occurred before the ex-periment began caused currently growing plants and soil organic matter from previous vegetation to attain distinct iso-tope ratios This has allowed researchers to apply stable iso-topes to distinguish between root and rhizosphere respirationand heterotrophic decomposition in three treatments cham-bers with elevated CO2 chambers with ambient levels ofCO2 and nonchambered control plots In this experiment ele-vated CO2 resulted in doubled decomposition rates but nochange in root and rhizosphere respiration relative to ambi-ent conditions (Pendall et al 2003) In an OTC experimenton a California annual grassland Hungate and colleagues(1997) found increases in both rhizosphere and heterotrophicrespiration under conditions of elevated CO2 with the en-hancement of heterotrophic respiration originating primar-ily from the decomposition of senesced roots

These experiments illustrate both the importance of ex-amining changes in the whole-ecosystem carbon cycle in re-sponse to elevated CO2 and the utility of carbon isotopes toserve as valuable tracers Although plant biomass often in-creases under elevated CO2 isotopic meth-ods are providing a unique insight into changesin decomposition rates and the balance of au-totrophic and heterotrophic respiration Ingeneral isotopic evidence points toward greaterheterotrophic respiration rates under elevatedCO2 In many studies this increase has been attributable to greater decomposition of re-cently fixed labile carbon and to stimulationof microbial activity such that older carbonpools are also utilized These results have important implications for the long-term sequestration of carbon in the terrestrial biosphere

Quantifying the turnover rate of carbon in ecosystemsQuantifying the mean residence time of car-bon in various ecosystem pools is a principalgoal in several areas of carbon cycle researchparticularly with regard to belowground frac-tions of biomass and soil organic matter (Randerson et al 1999) In this regard isotope

analyses in experiments on elevated CO2 provide an oppor-tunity to advance our understanding of ecosystem functionThese experiments are analogs of labeling studies previouslyconducted in small greenhouses or closed chambers It isnow possible to understand the time scales of carbon trans-fer through belowground carbon pools in whole-ecosystemmanipulations using labeled CO2

In the Duke FACE experiment within one week of the ini-tiation of fumigation with depleted 13CO2 the soil CO2 effluxin experimental plots became isotopically depleted in 13Crelative to the plots with ambient levels of CO2 (Andrews etal 1999) providing unequivocal evidence of the rapid carbonfluxes between aboveground and belowground componentsThis 13C depletion reflects transport of recently fixed carbonbelow ground which caused a corresponding change in theisotope ratio of autotrophic respiration Within a year of theinitiation of fumigation root-free soil showed 13C depletionin the CO2 efflux which must have been derived from theheterotrophic component of respiration Yet after 32 days of incubation with no further supplemental carbon the 13C-depleted respiration signal returned to values similar to thoseof soil from ambient plots This indicated that the newlyfixed carbon had been transferred to labile soil organic mat-ter pools that were rapidly exhausted in the absence of con-tinuous inputs (figure 3) These and other results supportrecent studies of naturally occurring 13C that show correla-tions between d13C of soil-respired and ecosystem-respiredCO2 and environmental factors such as humidity and vaporpressure deficit (Ekblad and Houmlgberg 2001 Bowling et al2002) illustrating that a large portion of respired carbonwas recently fixed These isotopic studies allow further delineation of the carbon cycle and provide strong evidencefor rapid carbon cycling at the level of short-term photo-


Articles

Figure 3 The carbon isotope ratio (d13C) of soil carbon dioxide (CO2) flux during root-free incubation of soil from the Duke Forest FACE (free-air CO2enrichment) experiment Source Andrews and colleagues (1999)

Elevated CO2

Ambient CO2d13C

of C

O 2flu

x (p

arts

per

mil)

ndash27

ndash26

ndash25

ndash24

ndash23

10 20 300

ndash19

ndash20

ndash21

ndash22

Days of incubation

synthesis and respiration In addition analyses of naturallyabundant radiocarbon in a number of ecosystems have in-dicated that the carbon evolved in soil respiration has generallybeen fixed much more recently than the carbon in bulk soilorganic matter (Trumbore 2000) Soil CO2 flux appears to con-tain a large proportion of carbon from live roots and from de-composition of recently deposited labile organic matter

Precise determination of root turnover rates is difficult inthe field although it has been recognized that root turnoveris likely to be a major component of soil carbon inputs (Jack-son et al 1996) The 13C-depletion signature used to trace theadded carbon in FACE experiments can also be used to de-termine turnover rates of carbon in different pools includ-ing roots of different class sizes The mean residence time ofcarbon can be calculated from an exponential decay functionby monitoring the remaining d13C values and biomass of agiven root size class over the course of a long-term CO2fumigation experiment (table 2) Results from this isotopicapproach have indicated that the mean residence time ofcarbon in roots in the Duke FACE experiment varied from4 to 9 years depending on root size class and treatmentThese root ages are much greater than previously estimatedwith other techniques (Matamala and Schlesinger 2000) Yetthe estimated turnover rates based on 13C measurements areclose to root age estimates based on another isotopic ap-proach the natural abundance of 14C (Gaudinski et al 2001)Isotopic methods appear to suggest much longer fine-rootturnover times than traditional methods Luo (2003) suggeststhat the influence of older pretreatment carbon reserves onthe isotope ratio of fine roots could account for much of thisdiscrepancy although conventional methods of estimatingroot production and turnover (such as periodic soil samplingor the use of belowground cameras) are also subject to sig-nificant uncertainty (Lauenroth 2000) This issue merits fur-ther investigation because of the important implications forecosystem carbon balanceFor example at the Duke FACE siteif carbon has long mean residence times in live roots rootturnover cannot explain the observed increase in soil respi-ration under elevated CO2 Isotopic techniques thus suggestthat other rhizosphere processes such as microbial decom-position of root exudates increased mycorrhizal activity anddecomposition of very fine root tips must be a significantcomponent of the observed increase in respiration of new carbon under elevated CO2

Carbon and nitrogen interactionsEcosystem carbon and nitrogen cycles aretightly integrated and isotope-based approaches can help elucidate changes under elevated CO2 conditions Organicmaterial in the form of roots litter andwoody debris is decomposed by the soilmicrobial community and releases min-eral nitrogen available for plant uptake inthe form of ammonium and nitrate (nitrogen mineralization) In temperate

ecosystems the availability of nitrogen is often a limitingfactor for plant growth To sustain increased growth under el-evated CO2 in these ecosystems plants must either increasetheir nitrogen use efficiency (carbon fixed per unit of plantnitrogen) or increase their uptake of nitrogen Increased ni-trogen uptake may result from greater uptake of existingmineral nitrogen from increased rates of mineralization orfrom increased fixation by symbiotic or free-living organismsthat can utilize nitrogen from the atmosphere and make itavailable to plants

The number of possible alterations to the nitrogen cycle thatcould occur under elevated CO2 introduces uncertainty intointerpretations of experimental results It is essential to de-termine whether short-term observations of increased plantgrowth under elevated CO2 can be sustained in the future Thisquestion hinges in part on an understanding of carbon andnitrogen interactions To examine these interactions ecolo-gists have studied variations in naturally occurring levels ofnitrogen-15 (15N) and conducted experiments using artificiallyapplied 15N These studies present considerable opportuni-ties for understanding carbonndashnitrogen interactions in ex-periments with elevated CO2


Articles

Figure 4 The nitrogen isotope ratio (d15N) of leaves ofAmbrosia dumosa (Amdu) Krameria erecta (Krer)Pleuraphis rigida (Plri) and Larrea tridentata (Latr) at the Nevada Desert FACE (free-air carbon dioxide enrichment) facility September 1999 Modified fromBillings and colleagues (2002)

Table 2 Mean residence times of fine roots in the Duke Forest FACE (free-aircarbon dioxide enrichment) site estimated using carbon-13 and carbon-14isotope tracers

Mean residence times of fine rootsFine root class Ambient carbon dioxide Elevated carbon dioxide

lt 1 mm 4 to 5 years 42 years1 to 2 mm 6 to 7 years 59 yearsgt 2 mm 8 to 9 years 60 years

Note Incorporation of the isotope tracer into a given root size pool is measured as the quantity of roots that have the initial isotope ratios as a function of time fitted to an exponentialfunction (F[t] = endashkt mean residence time = 1k years)

d15N

(par

ts p

er m

il)

Elevated carbon dioxideAmbient carbon dioxide

Lessons from natural abundance of nitrogen-15 The nitro-gen isotope ratio (d15N) is a useful indicator of change inecosystem nitrogen cyclesThe d15N of plant-available nitrogenis a function of several variables including nitrogen inputs intothe ecosystem d15N of substrates used by soil microbial pop-ulations rates of microbial transformations gaseous lossesfrom the soil and changes in active plant rooting depthsChanges in any of these factors as a result of growth under ele-vated CO2 should alter the d15N of nitrogen available toplants and subsequently the d15N of the plants themselvesThereforea significant shift in plant d15N can be a conclusiveindicator of changes in the ecosystem nitrogen cycle longbefore small shifts in the overall nitrogen content of differ-ent ecosystem components can be detected

Recently a large shift in plant d15N under simulated con-ditions of elevated CO2 was observed at the Nevada DesertFACE facility Billings and colleagues (2002) documented a3permil increase in d15N of the dominant shrub (Larrea triden-tata) following exposure to elevated CO2 and seasonal shiftsof 2permil for shrubs grown under both ambient and elevatedCO2 conditions In September 1999 four shrub speciesshowed enrichment in d15N under conditions of elevatedCO2 relative to ambient conditions (figure 4) No significantdifferences were observed in root growth and water sourcesbetween the two treatments Moreover plant d15N was actu-ally greater than soil d15N suggesting enrichment of 15N in thepool of nitrogen available to plants as a result of mineraliza-tion This could be caused by an increase in gaseous nitrogenloss or by immobilization in microbial biomass because bothprocesses are accompanied by significant fractionation thatis preferential loss of 14N and enrichment of the remainingpool in 15N (Houmlgberg 1997) The shift in plant d15N andother supporting evidence in the Nevada study strongly suggest an increase in microbial activity under conditions of elevated CO2 associated with the greater inputs of organicsubstrate available for decomposition

The numerous transformations that occur in the nitrogencycle prohibit 15N at natural abundance levels from being usedas a tracer of nitrogen movement in ecosystems (Evans 2001Robinson 2001) One exception is that 15N discrimination isnot observed during plant uptake of inorganic nitrogen at typ-ical soil concentrations therefore whole-plant d15N will ac-curately trace that of the source if plants absorb ammoniumor nitrate (Mariotti et al 1982 Yoneyama and Kaneko 1989Evans et al 1996) A complication is that 15N fractionation mayoccur during mycorrhizal uptake and transfer of nitrogen tothe host but 15N discrimination will not be observed if mostof the nitrogen absorbed by the fungus is transferred to thehost (Hobbie et al 1999)

Lessons from tracer nitrogen-15 An alternative approach tothe use of natural abundance nitrogen isotopes is to study ni-trogen cycling with the application of an artificial 15N-enrichedtracer An important consideration is that the application ofan artificial tracer excludes the possibility of subsequent stud-ies of natural abundanceeven if the tracer is applied to a smallsubsection of the experimental plot Artificial 15N tracers arefar more enriched in 15N than naturally occurring organic material with 15N concentrations that are 1000 to 10000times higher It is difficult to avoid contamination underthese circumstances

One of the strengths of the 15N tracer technique is thepossibility of constructing a complete 15N budget by follow-ing the tracer into the various plant and soil pools This pro-vides an opportunity to address the question of whetherelevated CO2 leads to greater nitrogen use efficiency of plantsincreased available nitrogen or decreased nitrogen lossesfrom the plantndashsoil system Increases or decreases in nitrogenavailability could occur depending on whether elevated CO2stimulates or suppresses soil organic matter decompositionA number of studies have assessed how plant nitrogen use andtotal 15N losses have been altered under elevated CO2 condi-

tions Losses of applied 15Nwere found to be slightly lowerunder elevated CO2 in amicrocosm study when thegrass Danthonia richardsoniiwas grown under a high levelof applied mineral nitrogen(Lutze and Gifford 1998) AFACE experiment that madeuse of 15N-labeled fertilizer in-dicated that elevated CO2 con-centration had no significanteffect on total 15N losses (soiland plant) in swards of Loliumperenne (perennial ryegrass)and Trifolium repens (whiteclover) after 4 years of CO2enrichment (Hartwig et al2002) These studies suggestthat the total nitrogen in the


Articles

Figure 5 Isolines of the oxygen isotope ratio (d18O parts per mil) in precipitation (relative toVienna Standard Mean Ocean Water) in the United States Source Kendall and Coplen(2001)

d 18O (parts per mil)

gt ndash2ndash4 to ndash2ndash6 to ndash4

lt ndash18

ndash16 to ndash14ndash14 to ndash12ndash12 to ndash10

ndash10 to ndash8ndash8 to ndash6

ndash18 to ndash16

plantndashsoil system is not altered under elevated CO2 in theseecosystems Moreover in the Lo perenne system after 8 yearsof elevated and ambient CO2 concentrations 15N enrich-ments were similar in the various soil organic matter fractionsshowing that the soil organic matterndashnitrogen dynamics wereunaffected by prolonged exposure to elevated CO2 (vanGroenigen et al 2002)

Nitrogen fixation plays a key role in the overall nitrogen cycle and the total input in terrestrial ecosystems is esti-mated at around 100 teragrams per year (Mosier 2001) Es-timates for molecular nitrogen (N2)ndashfixing activity in thefield are based on the dilution of 15N in N2-fixing legumescompared with non-N2-fixing reference plants Using the15N dilution approach the percentage of nitrogen derived fromN2 fixation by T repens grown under elevated CO2 (FACE)conditions increased by 8 over a 3-year period (Zanetti etal 1996) Similar findings were reported for free-living N2-fixing bacteria in stands of a C3 sedge and a C4 grass grownunder elevated CO2 (Dakora and Drake 2000) It will be ofinterest to determine whether the increase in N2 fixation un-der elevated CO2 conditions will be sustained Once the feed-back mechanisms between the demand of nitrogen by theplant available soil nitrogen and N2-fixing activity havereached a new equilibrium under elevated CO2 the increasein N2 fixation may subside

Integrating altered water balanceIn water-limited ecosystems water availability has been shownto be the critical variable controlling plant growth responsesunder elevated CO2 (Mooney et al 1999) Elevated CO2 canimprove plant water relations increase soil water availabilityand indirectly alter organic matter decomposition nutrientcycling and carbon storage In addition alterations to evapo-transpiration and biospherendashatmosphere exchange ofwater vapor may have important implications for climate andthe hydrologic cycle Stable isotopes of carbon oxygen andhydrogen record the influence of environmental conditionsand plant water sources on plant function Isotopic water balance studies take advantage of the large environmental gradients in the isotopic composition of precipitation and soil water Such gradients provide natural tracers that avoidthe need for artificial labeling (figure 5) Unlike short-termmeasurements of leaf gas exchange and temperature iso-topic methods may be used to study long-term changes inplant energy and water balance

Lessons from carbon isotopes The carbon isotope ratio ofplant organic matter (d13CP) is an integrated measure ofphotosynthetic discrimination (D13C) during CO2 fixationThe parameter D13C takes into account variations in d13C of source air (d13Csource box 1) in the formation of plant material It is particularly important to take into accountd13Csource in studies of elevated CO2 in which d13Csource is often significantly different from ambient levelsWhen effectsof d13Csource are removed D13CP in C3 plants is a function ofthe ratio of the partial pressures of CO2 inside and outside the

leaf (cica box 2) This ratio reflects the balance between CO2diffusion into the leaf as regulated by stomatal conductanceand CO2 removal as regulated by the photosynthetic rateTherefore the cica ratio can be considered an integrated setpoint of plant gas exchange (Ehleringer and Cerling 1995)By applying estimates of d13Csource and measurements ofd13CP to estimating cica integrated over time in studies of ele-vated CO2 we can assess the effects of elevated CO2 on thisset point that is how stomatal regulation of photosyntheticgas exchange has responded to a higher-CO2 environment

Short-term instantaneous leaf-level estimates of cica canbe obtained with traditional gas exchange measurementsWhile several studies have reported no short-term changes ininstantaneous cica under elevated CO2 (see Morison 1985Drake et al 1997) longer-term measurements using d13CP toestimate assimilation-weighted cica values have suggestedotherwise One of the authors (G L) estimated cica valuesfrom plants grown under elevated CO2 in a variety of ecosys-tem facilities including chambers micro- and mesocosmsFACE experiments and CO2 springs and concluded thatabout two-thirds of studies examined showed significant in-creases in the cica ratio under elevated CO2 while the restshowed no change in cica This is a key observation becauselong-term increases in cica are associated with increasedwater-use efficiency defined as the ratio of carbon fixed to water lost (Farquhar et al 1989) These 13C measurements willnot only help resolve uncertainty in physiological changes inresponse to rising CO2 concentrations over the long term butalso improve parameterization of ecological models forwhich cica is often a critical variable (see Katul et al 2000)

Lessons from oxygen and hydrogen isotopes The isotopiccomposition of plant stem water reflects the isotopic com-position of the plantrsquos water source because with the exceptionof several halophytes (Lin and Sternberg 1993) there is nofractionation against the heavy hydrogen isotope deuteriumand oxygen (d18O) isotopes during plant water uptake (Whiteet al 1985 Dawson and Ehleringer 1991 Lin and Sternberg1993) However because of evaporative effects leaf waterbecomes isotopically enriched in 18O by 15permil to 30permil and indeuterium by 40permil to 60permil relative to stem water (box 3) Inthe leaf oxygen in CO2 readily exchanges with oxygen in liq-uid water a rapid reaction in the presence of the plant enzymecarbonic anhydrase Because plant leaves contain much morewater than CO2 the d18O of CO2 takes on the d18O value ofleaf water Farquhar and colleagues (1993) and Farquharand Lloyd (1993) have shown how measurements of thed18O values of leaf water and CO2 can be used to estimate grossphotosynthetic rates In addition organic matter fixed withinleaves (ie sugars and cellulose) records the enriched d18O val-ues of leaf water Because 18O and deuterium have similar ap-plications and are subject to similar mechanisms ofevaporative enrichmentwe will restrict our discussion belowto 18O for convenience

Models predicting the isotopic composition of leafwater and plant cellulose are shown in box 3 These


Articles

equations can be used to interpret the effects of elevated CO2on d18O of bulk leaf water which may occur as a result ofseveral possible changes in plant function and canopy struc-ture Leaf water of plants grown under elevated CO2 can beexpected to become enriched in d18O if transpiration is re-duced (box 3 eq 2) and even more so if this reduction causesan increase in leaf temperature and therefore an increase inthe leaf saturation vapor pressure (box 3 eq 1) Converselyreduced enrichment is expected if transpiration increases un-der elevated CO2 because of increased plant size and root-ing volume (box 3 eq 2)

Preliminary results from a Swiss forest experiment showedthat leaves of deciduous trees exposed to elevated CO2 wereenriched in 18O relative to those grown under ambient CO2However the humidity leaf temperature and air temperaturein the canopy under elevated CO2 did not differ significantlyfrom ambient conditions Thus it is likely that reduced tran-spiration under elevated CO2 caused the enrichment in H2

18O(see box 3 eq 2) Cooper and Norby (1994) found species-specific effects on H2

18O enrichment Both Quercus alba andLiriodendron tulipifera showed reductions in transpiration un-der elevated CO2 but only Li tulipifera showed 18O-enrichedleaf water and leaf cellulose (figure 6) These results demon-strate that subtle changes in water and energy balancewhich

may be difficult to detect with traditional meteorologicaland physiological measurements may occur in plants grownunder elevated CO2 Isotopic approaches allow detection ofthe integrated effects over time of changes in the isotopic com-position of leaf water and organic components In additionthese studies have consequences for the interpretation ofchanges in isotope ratios in long-term data sets such as treering data that encompass historical changes in atmosphericCO2 concentrations

A mechanistic assessment of the effects of elevated CO2 onwater fluxes is facilitated by the use of stable isotopes in soilwater Soil water near the surface undergoes evaporation andbecomes 18O enriched whereas water from deeper layers is un-changed isotopically from precipitation inputs Since plantsdo not fractionate soil water during uptake and water vaporleaving the leaf is in isotopic equilibrium with stem waterwater evaporating from soil and water transpired by plants often have isotopically distinct compositions This differenceaffords a means of comparing the magnitudes of the soiland plant components of evapotranspiration

This approach was applied at the Colorado OTC exper-iment which had as its subject a perennial grassland In thisstudy the oxygen isotope ratio of soil CO2 was used as aproxy for d18O of soil water Measurements of d18O in soil


Articles

Although there is no fractionation in plant water uptake or transport (White et al 1985) leaf water is isotopically enrichedrelative to the plant water source because of evaporation To quantify this effect plant leaves can generally be treated asevaporative pools of water However unlike large water reservoirs leaves have a high ratio of evaporative flux to water vol-ume and have nearly equal amounts of water entering from the xylem and leaving from evaporation Under these steady-state conditions we can predict the isotopic composition of leaf water at the site of evaporation with a simplified version ofthe model given by Craig and Gordon (1965)

(1) dLs = ds + eeq + ek + (da ndash ds ndash ek) acute eaei

where d stands for d18O the subscript a is ambient water vapor eeq is the equilibrium fractionation which is temperaturedependent (Majoube 1971) ek is the kinetic fractionation of water which depends on the molecular diffusion of 18O in airand the aerodynamic nature of the boundary conditions (Merlivat 1978) and eaei is the ratio of the partial pressure ofwater vapor outside and inside the leaf (Farquhar and Lloyd 1993 Yakir 1998) The 18O signal from leaf water at the site ofevaporation is transferred to organic matter (mostly cellulose) via biosynthesis from sucrose and starch with an averagedoxygen fractionation value of 27permil (DeNiro and Epstein 1979 Sternberg et al 1989 Yakir 1992)

Although equation 1 (above) is useful for predicting the isotopic composition of water at the site of evaporation it is ofteninadequate to predict the isotope ratio of bulk leaf water Not all parts of the leaf are equally exposed to evaporation andsignificant spatial heterogeneity within leaves has been found (Yakir et al 1989 Luo and Sternberg 1992 Yakir et al 1994Helliker and Ehleringer 2000) Some of this heterogeneity may be caused by patchiness (Terashima et al 1988) leaf com-partmentalization (Yakir et al 1993) and incomplete mixing away from the site of evaporation at the cell walls (Farquharand Lloyd 1993) To incorporate the effect of the decay of the 18O enrichment signal between the surface of evaporation andthe isotopic signature of the source water entering the leaf Farquhar and Lloyd (1993) proposed

(2) dLW = ds + (dLs ndash ds) acute (1 ndash endashP) P

as modified by Yakir (1998) where dLW is the d18O bulk leaf water dLs in equation 1 and P is the Peacuteclet number P is relatedto the transpiration rate (moles per square meter per second) and the effective maximal mixing path length as well as physi-cal constants An alternative approach to estimating transpiration rate to determine dLW has been proposed by Roden andEhleringer (1999) who apply best-fit empirical coefficients to leaf water mixing

Box 3 Models of fractionation in plant water isotopes

CO2 soil water and precipitation as well as the amount ofprecipitation were used in a mass balance model to assessevaporation and transpiration (Ferretti et al 2003) Soilwater d18O in chambers with ambient CO2 was more enriched than in chambers with elevated CO2 The mass balance showed that this enrichment was in part a conse-quence of year-to-year variability in the effects of elevatedCO2 on ecosystem water balance During an unusually drygrowing season transpiration in chambers with elevated CO2was higher than in those with ambient CO2 reflecting thelarger plant biomass under elevated CO2 conditions Inyears of average precipitation however transpiration wassimilar under both treatments Despite these differencestranspiration use efficiency defined as the amount of bio-mass fixed per unit of transpiration was greater in cham-bers with elevated CO2 in dry and normal years Theseresults illustrate a unique application of stable isotope trac-ers to understanding changes in ecosystem water use underelevated CO2 Such applications could be particularly valu-able in light of the difficulties in distinguishing the com-ponents of evapotranspiration with nonisotopic methods

Trophic- level interactionsChanges in plant tissue composition and chemistry inducedby elevated CO2 have important implications not only for car-bon and nutrient cycles but also for interactions betweenplants and communities of organisms at higher trophic levels Although there have been few studies of the effects ofelevated CO2 on fauna effects on leaf carbon-to-nitrogen(CN) ratios and on the levels and amounts of the secondarymetabolic compounds that deter herbivory have been shownto change leaf consumption by herbivorous insects and con-sequently the insectsrsquo growth rate (Lindroth 1996a 1996b)There is considerably less information available on the effectsof elevated CO2 on the chemistry of fine roots and root ex-udates but such changes have been hypothesized to have astrong influence on soil microbial community dynamics andsubsequently on the processes of nitrogen mineralizationand immobilization that control the availability of nitrogento plants (Zak et al 2000)

The isotopic composition of animals reflects that of theirfood source when tissue-dependent metabolic fractionationeffects have been accounted for (DeNiro and Epstein 1978Tiezen et al 1983 Focken and Becker 1998) Thereforestable carbon isotopes have been used extensively in studiesof terrestrial aquatic and marine food webs (Rounick andWinterbourn 1986) Carbon dioxide fumigation with a highly13C-depleted source creates a unique isotopic signature in plantmaterial which can be traced to higher trophic levelsThis pro-vides a unique opportunity to evaluate changes in diet in ahigh-CO2 environment

In a novel application of this idea Cotrufo and Drake(2000) collected litter from several FACE experiments withcontrasting d13C values of approximately ndash27permil in plotswith ambient levels of CO2 and ndash40permil in plots with elevatedlevels of CO2 Litter from plots with elevated CO2 litter from

plots with ambient CO2 and a 50ndash50 mixture of the two werethen inoculated with isopods and incubated The CO2 respiredfrom each incubation was collected and analyzed for 13CThe results showed that CO2 respired from the 50ndash50 mix-ture had a carbon isotope ratio of 34permil an amount inter-mediate between the ratios in CO2 evolved from litter fromplots with ambient and with elevated CO2 The results indi-cate that the detritivores showed no preference for litter fromplots with ambient CO2 over litter from plots with elevatedCO2 This finding contrasts with hypotheses based on thecommonly observed increase in leaf CN ratios under elevatedCO2 These hypotheses propose that decomposition shouldoccur more slowly in litter with a higher CN ratio reducingthe rate of litter decomposition under elevated CO2 (Strainand Bazzaz 1983) However the results were consistent witha recent meta-analysis that found no overall CO2 effect on lit-ter decomposition rates from the range of results reported todate (Norby et al 2001) despite the differences in live leaf CNratios and the increased levels of secondary compoundsboth of which affect herbivory in plants grown under elevatedCO2 Similar approaches may be used to understand herbi-vore preferences for plant material grown under elevatedCO2 and ambient conditions


Articles

Figure 6 Oxygen isotope ratios of Liriodendron tulip-ifera and its source water (relative to Vienna StandardMean Ocean Water) grown under elevated (+300 partsper mil) and ambient concentrations of carbon dioxideData from Cooper and Norby (1994)

ndash87

ndash87

+8

+10

+275

+255

In addition to the CO2 fumigation tracer other labelingtechniques have been used to examine the flow of carbon intothe soil microbial community under elevated CO2 One suchtechnique is pulse labeling with 14C In this technique plantsare exposed to a brief pulse of CO2 containing elevated 14CIn an experiment on the effects of elevated CO2 on crops pulselabeling showed that high CO2 caused reduced allocation ofcarbon into the microbial carbon pool probably the result ofa change in the types and amounts of carbon compoundsavailable to microbes under elevated CO2 (Patterson et al1996) It is also likely that alterations in the amount andcomposition of carbon deposited to the soil can affect mi-crobial community composition as well as biomass but thishas been difficult to assess Radajewski and colleagues (2000)reported that taxonomic groups with distinct metabolicprocesses could be distinguished by supplying a 13C-enrichedsubstrate to the soil Microbes that utilize the substrate in-corporate the isotope into their DNA and density-gradientcentrifugation can then be employed to separate DNA con-taining the added 13C isotope from DNA without the addedisotope Sequence analysis of the DNA thus separated can iden-tify the phylogenetic group that metabolizes a particular sub-strate This technique holds promise for evaluating shifts inmicrobial community composition in response to environ-mental perturbations such as elevated CO2 Thus a variety ofisotopic tracers can be used to evaluate hypothesized shifts inplantndashanimal interactions and microbial food webs resultingfrom changes in plant chemistry

Conclusions Carbon nitrogen oxygen and hydrogen isotope ratios at nat-ural abundance levels provide valuable tracers for integratedanalyses of terrestrial ecosystems in experiments with elevatedlevels of CO2 The additional CO2 supplied in these experi-ments provides a useful label that can be used to trace car-bon flows and turnover rates in different pools both underelevated CO2 treatment conditions and after supplementalCO2 is withdrawn In experiments in which isotope ratio massspectrometry is not available dry biomass and soil samplescan be archived for future isotopic analysis Frozen leaf andsoil samples or if possible water extracted and stored insealed vials can provide valuable information about the effect of elevated atmospheric CO2 on leaf water budgets andpossibly on leaf energy balance Linking these stable isotopeanalyses with traditional methods in elevated-CO2 ecosystemstudies provides a unique opportunity for tracing short-term and long-term changes in the carbon cycle and their impacts on key ecosystem biogeochemical cycles as well as trophic-level interactions

AcknowledgmentsThis article is a product of a workshop funded by the Biologicaland Environmental Research Program (BER) US Departmentof Energy (DOE) Research reported in this manuscript wasfunded by US DOE BER through the Great Plains RegionalCenter of the National Institute for Global Environmental

Change under Cooperative Agreement No DE-FC03-90ER61010 and by US National Science Foundation grants98-14358 02-12819 and DEB-0120169 This work con-tributes to the Global Change and Terrestrial Ecosystems(GCTE) core project of the International GeospherendashBiosphere Programme GCTE Focus 1 is supported by the National Aeronautics and Space Administrationrsquos Earth Science Enterprise

References citedAcklerly DD Bazzaz FA 1995 Plant growth and reproduction along CO2

gradients Non-linear responses and implications for community changeGlobal Change Biology 1 199ndash207

Allison GB Barnes CJ Hughes MW 1983 The distribution of deuterium and18O in dry soils 2 Experimental Journal of Hydrology 64 377ndash397

Andrews JA Schlesinger WH 2001 Soil CO2 dynamics acidification andchemical weathering in a temperate forest with experimental CO2enrichment Global Biogeochemical Cycles 15 149ndash162

Andrews JA Harrison KG Matamala R Schlesinger WH 1999 Separationof root respiration from total soil respiration using carbon-13 labelingduring free-air carbon dioxide enrichment (FACE) Soil Science Society of America Journal 63 1429ndash1435

Billings SA Schaeffer SM Zitzer SF Charlet TH Smith SD Evans RD 2002Alterations of nitrogen dynamics under elevated carbon dioxide in anintact Mojave Desert ecosystem Evidence from nitrogen-15 naturalabundance Oecologia 131 463ndash467

Bowling DR McDowell NG Bond BJ Law BE Ehleringer JR 2002 13Ccontent of ecosystem respiration is linked to precipitation and vapor pressure deficit Oecologia 131 113ndash124

Calder IR Swaminath MH Kariyappa GS Srinivasalu NV Srinivasa MurtyKV Mumtaz J 1992 Deuterium tracing for the estimation of transpirationfrom trees Part 3 Measurements of transpiration from Eucalyptus plan-tations India Journal of Hydrology 130 37ndash48

Cardon ZG Hungate BA Cambardella CA Chapin FS Field CB HollandEA Mooney HA 2001 Contrasting effects of elevated CO2 on old andnew soil carbon pools Soil Biology and Biochemistry 33 365ndash373

Cheng W Johnson DW 1998 Elevated CO2 rhizosphere processes and soilorganic matter decomposition Plant and Soil 202 167ndash174

Cooper LW Norby RJ 1994 Atmospheric CO2 enrichment can increase the18O content of leaf water and cellulose Paleoclimatic and ecophysiologicalimplications Climate Research 4 1ndash11

Cotrufo MF Drake BG2000 Effects of elevated CO2 on leaf litter palatabilityA stable isotope approach Eos Transactions 81 F281

Craig H Gordon LI 1965 Deuterium and oxygen 18 variations in the oceanand marine atmospherePages 9ndash130 in Tongiorigi E ed Stable Isotopesin Oceanographic Studies and Paleotemperatures Pisa (Italy) Con-siglio Nazionale Delle Ricerche Laboratorio di Geologia Nucleare

Dakora RD Drake BG 2000 Elevated CO2 stimulates associative N2 fixationin a C3 plant of the Chesapeake Bay wetland Plant Cell and Environ-ment 23 943ndash953

Dawson TE Ehleringer JR 1991 Streamside trees that do not use stream water Nature 350 335ndash337

DeLucia EH et al 1999 Net primary production of a forest ecosystem withexperimental CO2 enrichment Science 284 1177ndash1179

DeNiro MJ Epstein S 1978 Influence of diet on the distribution of carbonisotopes in animals Geochemica et Cosmochimica Acta 42 492ndash506

mdashmdashmdash 1979 Relationship between the oxygen isotope ratios of terrestrialplant cellulose carbon dioxide and water Science 204 51ndash53

Dorr H Munnich KO 1986 Annual variations of the 14C content of soil CO2Radiocarbon 28 338ndash345

Drake BG Gonzalez-Meler MA Long SP 1997 More efficient plants Aconsequence of rising atmospheric CO2 Annual Review of Plant Phys-iology 48 608ndash637


Articles

Edwards NT Norby RJ 1999 Belowground respiratory responses of sugarmaple and red maple saplings to atmospheric CO2 enrichment and elevated air temperature Plant and Soil 206 85ndash97

Ehleringer JR Cerling TE 1995 Atmospheric CO2 concentrations and theratio of intercellular to ambient CO2 in plants Tree Physiology 15105ndash111

Ekblad A Houmlgberg P 2001 Natural abundance of 13C in CO2 respired fromforest soils reveals speed of link between tree photosynthesis and root res-piration Oecologia 127 305ndash308

Evans RD 2001 Physiological mechanisms influencing plant nitrogen iso-tope composition Trends in Plant Science 6 121ndash126

Evans RD Bloom AJ Sukrapanna SS Ehleringer JR 1996 Nitrogen isotopecomposition of tomato (Lycopersicon esculentum Mill cv T-5) grown under ammonium or nitrate nutrition PlantCell and Environment 191317ndash1323

Farquhar GD 1983 On the nature of carbon isotope discrimination in C4species Australian Journal of Plant Physiology 10 205ndash226

Farquhar GD Lloyd J 1993 Carbon and oxygen isotope effects in the exchangeof carbon dioxide between plants and the atmosphere Pages 47ndash70 inEhleringer JR Hall AE Farquhar GD eds Stable Isotopes and Plant CarbonndashWater Relations New York Academic Press

Farquhar GD OrsquoLeary MH Berry JA 1982 On the relationship between carbon isotope discrimination and the intercellular carbon dioxide con-centration in leaves Australian Journal of Plant Physiology 9 121ndash137

Farquhar GD Ehleringer JR Hubick KT 1989 Carbon isotope discrimina-tion and photosynthesis Annual Review of Plant Physiology and PlantMolecular Biology 40 503ndash537

Farquhar GD Lloyd J Taylor JA Flanagan LB Syvertsen JP Hubick KT WongSC Ehleringer JR 1993 Vegetation effects on the isotope compositionof oxygen in atmospheric CO2 Nature 363 439ndash443

Ferretti DF Pendall E Morgan JA Nelson JA Lecain DR Mosier AR 2003Partitioning evapotranspiration fluxes from a Colorado grassland usingstable isotopes Seasonal variations and ecosystem implications of ele-vated atmospheric CO2 Plant and Soil Forthcoming

Focken U Becker K 1998 Metabolic fractionation of stable carbon isotopesImplications of different proximate compositions for studies of theaquatic food webs using d13C data Oecologia 115 337ndash343

Gaudinski JB Trumbore SE Davidson EA Cook AC Markewitz D RichterDD 2001 The age of fine-root carbon in three forests of the easternUnited States measured by radiocarbon Oecologia 129 420ndash429

Hartwig UA Luumlscher A Nosberger J van Kessel C 2002 Nitrogen-15 budget in model ecosystems of white clover and in perennial ryegrass exposed for four years at elevated atmospheric CO2 Plant Cell and Environment 24 957ndash965

Helliker BR Ehleringer JR 2000 Establishing a grassland signature in veins18O in the leaf water of C3 and C4 grasses Proceedings of the NationalAcademy of Sciences 97 7894ndash7898

Hobbie EA Macko SA Shugart HH 1999 Insights into nitrogen and carbon dynamics of ectomycorrhizal and saprotrophic fungi from isotopic evidence Oecologia 118 353ndash360

Houmlgberg P 1997 15N natural abundance in soilndashplant systems New Phytologist 137 179ndash203

Houmlgberg P Ekblad A 1996 Substrate-induced respiration measured in situin a C3-plant ecosystem using additions of C4-sucrose Soil Biology andBiochemistry 28 1131ndash1138

Houmlgberg P Houmlgberg M 2002 Extramatrical ectomycorrhizal myceliumcontributes one-third of microbial biomass and produces together withassociated roots half the dissolved organic carbon in a forest soil NewPhytologist 154 791ndash795

Houmlgberg P Nordgren A Buchmann N Taylor AFS Ekblad A Houmlgberg MNyberg G Ottosson-Loumlfvenius M Read DJ 2001 Large-scale forestgirdling shows that current photosynthesis drives soil respirationNature 411 789ndash792

Hungate BA Holland EA Jackson RB Chapin FS Mooney HA Field CB1997 The fate of carbon in grasslands under carbon dioxide enrichmentNature 388 576ndash579

Ineson P Cotrufo MF Bol R Harkness DD Blum H 1996 Quantificationof soil carbon inputs under elevated CO2 C3 plants in a C4 soil Plant andSoil 187 345ndash350

[IPCC] Intergovernmental Panel on Climate Change 2001 Climate Change2001 The Scientific Basis Cambridge (United Kingdom) Cambridge University Press

Jackson RB Canadell J Ehleringer JR Mooney HA Sala OE Schulze E-D1996 A global analysis of root distributions for terrestrial biomesOecologia 108 389ndash411

Jastrow JD Boutton TW Miller RM 1996 Carbon dynamics of aggregate-associated organic matter estimated by 13C natural abundance Soil Science Society of America Journal 60 801ndash807

Jenkinson DS Harkness DDVance EDAdam DE 1992 Calculating net pri-mary production and annual input of organic matter to soil from theamount and radiogenic content of soil organic matter Soil Biology andBiochemistry 24 295ndash305

Katul G Ellsworth DS Lai C 2000 Modelling assimilation and intercellularCO2 from measured conductance A synthesis of approaches PlantCell and Environment 23 1313ndash1328

Keeling CD Whorf TP 2002 Atmospheric CO2 records from sites in the SIOair sampling network In Trends A Compendium of Data on GlobalChange Oak Ridge (TN) Carbon Dioxide Information Analysis Center Oak Ridge National Laboratory (29 July 2003 httpcdiacesdornlgovtrendsco2sio-keelhtm)

Kendall C Coplen TB2001 Distribution of oxygen-18 and deuterium in riverwaters across the United States Hydrological Processes 15 1363ndash1393

Koumlrner C 2000 Biosphere responses to CO2 enrichment Ecological Appli-cations 10 1590ndash1619

Lauenroth WK 2000 Methods of estimating belowground new primary production Pages 58ndash71 in Sala OE Jackson RB Mooney HAHowarthRW eds Methods in Ecosystem Science New York Springer

Leavitt SW et al 1994 Carbon isotope dynamics of CO2-enriched FACE cotton and soils Agricultural and Forest Meteorology 70 87ndash101

Leavitt SW Paul EA Galadima A Nakayama FS Danzer SR Johnson HKimball BA 1996 Carbon isotopes and carbon turnover in cotton andwheat FACE experiments Plant and Soil 187 147ndash155

Leavitt SW Pendall E Paul EAKimball BA Pinter PJ Johnson HB MatthiasADWall GW LaMorte RL 2001 Stable-carbon isotopes and soil organiccarbon in wheat under CO2 enrichment New Phytologist 150 305ndash314

Lin G Sternberg LL 1993 Hydrogen isotopic fractionation by plant rootsduring water uptake in coastal wetland plantsPages 497ndash510 in EhleringerJR Hall AE Farquhar GD eds Stable Isotopes and Plant CarbonndashWater Relations San Diego Academic Press

Lin G Ehleringer JR Rygiewicz PT Johnson MG Tingey DT 1999 ElevatedCO2 and temperature impacts on different components of soil CO2efflux in Douglas-fir terracosms Global Change Biology 5 157ndash168

Lin G Rygiewicz PT Ehleringer JR Johnson MG Tingey DT 2001 Time-dependent responses of soil CO2 efflux components to elevated atmos-pheric CO2 and temperature in experimental forest mesocosms Plantand Soil 229 259ndash270

Lindroth RL 1996a Consequences of elevated atmospheric CO2 for forestinsects Pages 347ndash362 in Koumlrner C Bazzaz FA eds Carbon DioxidePopulations and Communities San Diego Academic Press

mdashmdashmdash 1996b CO2-mediated changes in tree chemistry and treendashLepidoptera interactions Pages 105ndash120 in Koch GW Mooney HAeds Carbon Dioxide and Terrestrial Ecosystems San Diego AcademicPress

Luo Y 2003 Uncertainties in interpretation of isotope signals for estimationof fine root longevity Theoretical considerations Global Change Biol-ogy 9 1ndash12

Luo Y-H Sternberg L 1992 Spatial DH heterogeneity of leaf water PlantPhysiology 99 348ndash350

Lutze J Gifford RM 1998 Acquisition and allocation of carbon and nitro-gen by Danthonia richardsonii in response to restricted nitrogen supplyand CO2 enrichment Plant Cell and Environment 21 1133ndash1141


Articles

Majoube M 1971 Fractionnement en oxygene 18 et en deuterium entre lrsquoeauet savapeur Journal de Chimie Physique et de Physico-Chemie Biologique 58 1423ndash1436

Mariotti A Mariotti F Champigny M-L Amarger N Moyse A 1982 Nitro-gen isotope fractionation associated with nitrate reductase activity anduptake of NO3

ndash by pearl millet Plant Physiology 69 880ndash884Matamala R Schlesinger WH 2000 Effects of elevated atmospheric CO2 on

fine root production and activity in an intact temperate forest ecosystemGlobal Change Biology 6 967ndash979

Merlivat L 1978 Molecular diffusivities of H216O HD16O and H2

18O ingases Journal of Chemical Physics 69 2864ndash2871

Mooney HA Canadell J Chapin FS Ehleringer JR Koumlrner Ch McMurtrieRE Parton WJ Pitelka LF Schulze E-D 1999 Ecosystem physiology responses to global change Pages 141ndash189 in Walker BH Steffen WLCanadell J Ingram JSI eds The Terrestrial Biosphere and Global ChangeImplications for Natural and Managed Ecosystems Cambridge (UnitedKingdom) Cambridge University Press

Morison J 1985 Sensitivity of stomata and water use efficiency to highCO2 Plant Cell and Environment 8 467ndash474

mdashmdashmdash 1998 Stomatal response to increased CO2 concentration Journalof Experimental Botany 49 443ndash452

Mosier AR 2001 Exchange of gaseous nitrogen compounds between ter-restrial systems and the atmospherePages 291ndash309 in Follet RF HatfieldJL eds Nitrogen in the Environment Sources Problems and Manage-ment Amsterdam Elsevier

Nitschelm J Luscher AHartwig UA van Kessel C 1997 Using stable isotopesto determine soil carbon input differences under ambient and elevatedatmospheric CO2 conditions Global Change Biology 3 411ndash416

Norby RJ Cotrufo MF Ineson P OrsquoNeill EGCanadell JG 2001 Elevated CO2litter chemistry and decompositionA synthesis Oecologia 127 153ndash165

OrsquoLeary MH 1981 Carbon isotope fractionation in plants Phytochem-istry 20 553ndash567

Owensby CE Cochran RC Auen LM 1996 Effects of elevated carbon diox-ide on forage quality for ruminants Pages 363ndash374 in Koumlrner C BazzazFA eds Carbon Dioxide Populations and Communities San DiegoAcademic Press

Patterson E Rattray EASKillhan K 1996 Effect of elevated atmospheric CO2concentration on C-partitioning and rhizosphere C-flow for 3 plantspecies Soil Biology and Biochemistry 28 195ndash201

Pendall E et al 2001 Elevated CO2 stimulates soil respiration in a FACE wheatfield Basic and Applied Ecology 2 193ndash201

mdashmdashmdash 2003 Elevated atmospheric CO2 effects and soil water feedbacks onsoil respiration components in a Colorado grassland Global Bio-geochemical Cycles 17 (2) 15-1ndash15-13

Pepin S Koumlrner C 2002 Web-FACE A new canopy free-air CO2 enrichmentsystem for tall trees in mature forests Oecologia 133 1ndash9

Radajewski S Ineson P Parekh NR Murrell JC 2000 Stable-isotope probing as a tool in microbial ecology Nature 403 646ndash649

Randerson JT Thompson MV Field CB 1999 Linking 13C-based estimatesof land and ocean sinks with predictions of carbon storage from CO2fertilization of the plant growth Tellus 51B 668ndash678

Robinson D 2001 d15N as an integrator of the nitrogen cycle Trends in Ecology and Evolution 16 153ndash162

Rochette P Flanagan LB 1997 Quantifying rhizosphere respiration in acorn crop under field conditions Soil Science Society of America Journal 61 466ndash474

Roden JS Ehleringer JR 1999 Observations of hydrogen and oxygen isotopesin leaf water confirm the CraigndashGordon model under wide-ranging environmental conditions Plant Physiology 120 1165ndash1173

Rounick JS Winterbourn MJ 1986 Stable carbon isotopes and carbon flowin ecosystems BioScience 36 171ndash177

Schimel D et al 2000 Contribution of increasing CO2 and climate to carbon storage by ecosystems in the United States Science 287 2004ndash2006

Schlesinger WH Lichter J 2001 Limited carbon storage in soil and litter ofexperimental forest plots under increased atmospheric CO2 Nature411 466ndash469

Sternberg LL Mulkey SS Wright SJ 1989 Oxygen isotope ratio stratificationin a tropical moist forest Oecolgia 81 51ndash56

Strain BR Bazzaz FA 1983 Terrestrial plant communities Pages 177ndash222 inLemon ER ed CO2 and Plants Boulder (CO) Westview

Stuiver MPolach H 1977 Reporting of 14C dataRadiocarbon 19 355ndash363Terashima I Wong S-C Osmond CB Farquhar GD 1988 Characterisation

of non-uniform photosynthesis induced by abscisic acid in leaves having different mesophyll anatomies Plant and Cell Physiology 29385ndash394

Tiezen LL Boutton TW Tesdahl KG Slade NA 1983 Fractionation andturnover of stable carbon isotopes in animal tissues Implications for d13Canalysis of diet Oecologia 57 32ndash37

Torbert HA Rogers HH Prior SA Schlesinger WH Runion GB 1997Effects of elevated atmospheric CO2 in agro-ecosystems on soil carbonstorage Global Change Biology 513ndash521

Trumbore SE 1993 Comparison of carbon dynamics in two soils using measurements of radiocarbon in pre- and post-bomb soils Global Bio-geochemical Cycles 7 275ndash290

mdashmdashmdash 2000 Age of soil organic matter and soil respiration Radiocarbonconstraints on belowground C dynamics Ecological Applications 10399ndash411

van Groenigen KJ Harris D Horwath WR Hartwig UA van Kessel C 2002Linking sequestration of 13C and 15N in aggregates in a pasture soil fol-lowing 8 years of elevated atmospheric CO2 Global Change Biology 81094ndash1108

van Kessel C Horwath WR Hartwig UA Harris D Luscher A 2000 Net soilcarbon input under ambient and elevated CO2 concentrations Isotopicevidence after 4 years Global Change Biology 6 435ndash444

White JWC Cook ER Lawrence JR Broecker WS 1985 The DH ratios ofsap in trees Implications for water sources and tree ring DH ratiosGeochimica et Cosmochimica Acta 49 237ndash246

Yakir D 1992 Variations in the natural abundance of oxygen-18 and deuterium in plant carbohydrates Plant Cell and Environment 151005ndash1020

mdashmdashmdash 1998 Oxygen-18 of leaf water A crossroad for plant-associated iso-topic signals Pages 147ndash168 in Griffiths H ed Stable Isotopes Integra-tions of Biological Ecological and Geochemical Processes Oxford(United Kingdom) Bios

Yakir D DeNiro MJ Rundel PW 1989 Isotopic inhomogeneity of leafwater Evidence and implications for the use of isotopic signals transducedby plants Geochimica et Cosmochimica Acta 53 2769ndash2773

Yakir D Berry JA Giles L Osmond CB 1993 The 18O of water in the meta-bolic compartment of transpiring leaves Pages 529ndash540 in Ehleringer JRHall AE Farquhar GD eds Stable Isotopes and Plant CarbonndashWater Relations San Diego Academic Press

mdashmdashmdash 1994 Isotopic heterogeneity of water in transpiring leaves Identi-fication of the component that controls the 18O of atmospheric O2 andCO2 Plant Cell and Environment 17 73ndash80

Yoneyama T Kaneko A 1989 Variations in the natural abundance of 15N innitrogenous fractions of komatsuna plants supplied with nitrate PlantCell Physiology 30 957ndash962

Zak DR Pregitzer KS King JS Holmes WE 2000 Elevated atmospheric CO2fine roots and the response of soil microorganisms A review and hypothesis New Phytologist 147 201ndash222

Zanetti S Hartwig UA Luumlscher A Hebeisen T Frehner M Fischer BUHendrey GR Blum H Nosberger J 1996 Stimulation of symbiotic N2fixation in Trifolium repens L under elevated atmosphere CO2 in agrassland ecosystem Plant Physiology 112 575ndash583


Articles

The need to develop methods to quantify ecosystem re-sponses to elevated CO2 in relatively small-scale and short-term experiments presents both challenges and opportunitiesfor ecologists Measurements of the abundance of various car-bon hydrogen oxygen and nitrogen isotopes in sampleshave long been used in various subdisciplines of ecology asunique tools that integrate physical and physiological processesover space and time (table 1 box 1) New developments inmass spectrometry allow stable isotope analyses to be con-ducted using very small samples of biomass soil organicmatter or water These advances have improved our abilityto make inferences about plant physiology biogeochemistryand ecosystem function This is particularly valuable in stud-ies of elevated CO2 in which destructive measurements mustbe limited to maintain the integrity of experimental plots forlong-term study Many studies of elevated CO2 effects are intrinsically isotopic tracer experiments because the ratios ofcarbon-13 (13C) to carbon-12 (12C) and of carbon-14 (14C)to 12C in the added CO2 are distinct from the ratios in at-mospheric CO2 Plants grown in conditions of elevated CO2that contain such isotopic tracers will incorporate the tracerin carbon-containing compounds The two labels can then beused to trace the fate of carbon newly incorporated into theecosystem under high CO2 conditions and to calculate the residence time of carbon in various pools

Here we discuss current and potential research that com-bines isotope ecology and experiments with elevated CO2 toimprove our understanding of interactions between eco-system function and global change

The carbon-13 labelThe mean d13C (see box 1) of atmospheric CO2 is currentlyabout ndash8 parts per mil (permil) (Keeling and Whorf 2002) In con-trast commercial CO2 originating from geologically ancientcarbon (eg carbon from petroleum byproducts or wells) isoften relatively depleted in 13C with d13C values typicallyranging from ndash27permil to ndash45permil In most experiments with elevated CO2 the increase in CO2 concentration is accom-plished by mixing commercial CO2 with ambient air As a result short-term fluctuations in canopy CO2 concentration

above and below the targeted concentration may occur andthe concentration may vary from one place to anotherAssessing actual exposure of plants to CO2 therefore be-comes a significant challengeA further complication is thatd13C of the commercial CO2 may vary if the source changesover time Periodic sampling of the added CO2 can track itsisotopic composition and subsequent modeling of CO2circulation in CO2 plots can provide estimates of the averageexposure of experimental plants to CO2 and 13CO2 how-ever the best way of measuring plantsrsquo CO2 exposure over timemay be to measure of isotopic ratios of carbon in C4 plants(box 2 Pepin and Koumlrner 2002) Such plants utilize a photo-synthetic pathway in which the initial fixation of carbon occurs through the action of the enzyme phosphoenolpyru-vate (PEP) carboxylase This enzyme has a much lower iso-topic fractionation factor than rubisco (ribulose bisphosphatecarboxylaseoxygenase) the initial carbon-fixing enzyme forplants utilizing the more common C3 photosynthetic pathway (OrsquoLeary 1981 Farquhar et al 1989) that is PEP carboxylase discriminates less against 13C when it fixes carbon The isotopic composition of C4 plants thereforechanges less in response to factors that influence stomatalopening (such as light temperature and humidity) than thatof C3 plants and its variation tracks the isotopic content ofatmospheric CO2

This property of C4 photosynthesis is being used in a Swissexperiment in which a mature temperate forest is fumigatedwith CO2 from tubing distributed throughout the canopy C4plants placed in pots throughout the canopy are periodicallysampled to calculate the effective atmospheric CO2 concen-tration from d13C of C4 plant material (box 2 Pepin andKoumlrner 2002) This method agrees well with estimates ofaverage CO2 level derived from measuring the CO2 concen-tration in sampled air (figure 1) and greatly reduces the needfor air turbulence modeling or rapid gas sampling deviceswhile still providing spatial information about the CO2 gra-dients that may occur in these experiments Once the 13CO2exposure of experimental plants has been characterized the13C label can then be used to follow the fate of experimentallyadded CO2 in the ecosystem

The carbon- 14 labelIf commercial CO2 is derived from fossil sources it containsno measurable 14C because it has effectively all decayed to nitrogen-14 (14N) during its long isolation from the naturalsource of 14C the action of cosmic rays in the atmosphere andon Earthrsquos surface (The half-life of 14C is 5730 years) The ab-sence of 14C in fossil CO2 provides a useful isotopic tracer thatmay show less temporal variation than the 13C isotope ratioWhen the 14C-free commercial CO2 is added to ambient air14C concentrations in the air are diluted This signal is presentin the plots with elevated CO2 in many completed and ongoingFACE experiments but to date only one has attempted to takeadvantage of itAt the FACE experiment in Maricopa Arizonathe 14C-depleted CO2 added to experimental plots resulted incotton plants that contained only 70 of the 14C activity of


Articles

Table 1 Isotopes of common elements and their naturalabundanceElement Isotope Abundance ()

Hydrogen 1H 999852H 0015

Carbon 12C 988913C 11114C lt 0001

Nitrogen 14N 996315N 037

Oxygen 16O 997617O 00418O 020

Note All the isotopes except 14C are stable





Articles













CO2

conc

entra

tions

(mic

rom

oles

per

mol

)










Articles














Articles








Photosynthesis

Litter

Soil organic matter


Elevated CO2

Ambient CO2

June lsquo93

1992 1993 1994 1995Year



ndash30

ndash28

ndash40

ndash36

ndash32

ndash28

ndash24

Elevated CO2

Ambient CO2


Corvallis OR

Soil

carb

on

d13 C

(par

ts p

er m

il)

Root

s lt

1 m

md1

3 C (p

arts

per

mil)








Articles


Elevated CO2

Ambient CO2d13C

of C

O 2flu

x (p

arts

per

mil)

ndash27

ndash26

ndash25

ndash24

ndash23

10 20 300

ndash19

ndash20

ndash21

ndash22

Days of incubation







Articles






d15N

(par

ts p

er m

il)









Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles





Articles













CO2

conc

entra

tions

(mic

rom

oles

per

mol

)










Articles














Articles








Photosynthesis

Litter

Soil organic matter


Elevated CO2

Ambient CO2

June lsquo93

1992 1993 1994 1995Year



ndash30

ndash28

ndash40

ndash36

ndash32

ndash28

ndash24

Elevated CO2

Ambient CO2


Corvallis OR

Soil

carb

on

d13 C

(par

ts p

er m

il)

Root

s lt

1 m

md1

3 C (p

arts

per

mil)








Articles


Elevated CO2

Ambient CO2d13C

of C

O 2flu

x (p

arts

per

mil)

ndash27

ndash26

ndash25

ndash24

ndash23

10 20 300

ndash19

ndash20

ndash21

ndash22

Days of incubation







Articles






d15N

(par

ts p

er m

il)









Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles










Articles














Articles








Photosynthesis

Litter

Soil organic matter


Elevated CO2

Ambient CO2

June lsquo93

1992 1993 1994 1995Year



ndash30

ndash28

ndash40

ndash36

ndash32

ndash28

ndash24

Elevated CO2

Ambient CO2


Corvallis OR

Soil

carb

on

d13 C

(par

ts p

er m

il)

Root

s lt

1 m

md1

3 C (p

arts

per

mil)








Articles


Elevated CO2

Ambient CO2d13C

of C

O 2flu

x (p

arts

per

mil)

ndash27

ndash26

ndash25

ndash24

ndash23

10 20 300

ndash19

ndash20

ndash21

ndash22

Days of incubation







Articles






d15N

(par

ts p

er m

il)









Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles






Articles








Photosynthesis

Litter

Soil organic matter


Elevated CO2

Ambient CO2

June lsquo93

1992 1993 1994 1995Year



ndash30

ndash28

ndash40

ndash36

ndash32

ndash28

ndash24

Elevated CO2

Ambient CO2


Corvallis OR

Soil

carb

on

d13 C

(par

ts p

er m

il)

Root

s lt

1 m

md1

3 C (p

arts

per

mil)








Articles


Elevated CO2

Ambient CO2d13C

of C

O 2flu

x (p

arts

per

mil)

ndash27

ndash26

ndash25

ndash24

ndash23

10 20 300

ndash19

ndash20

ndash21

ndash22

Days of incubation







Articles






d15N

(par

ts p

er m

il)









Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles








Articles


Elevated CO2

Ambient CO2d13C

of C

O 2flu

x (p

arts

per

mil)

ndash27

ndash26

ndash25

ndash24

ndash23

10 20 300

ndash19

ndash20

ndash21

ndash22

Days of incubation







Articles






d15N

(par

ts p

er m

il)









Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles







Articles






d15N

(par

ts p

er m

il)









Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles








Articles




lt ndash18



ndash18 to ndash16










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles










Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles









Articles














Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles







Articles


ndash87

ndash87

+8

+10

+275

+255


























Articles












































Articles















































Articles


























Articles












































Articles















































Articles












































Articles















































Articles















































Articles

Tracing Changes in Ecosystem Function under …basin.yolasite.com/resources/Pataki et al. 2003...

Documents

Transcript of Tracing Changes in Ecosystem Function under …basin.yolasite.com/resources/Pataki et al. 2003...