Progress and challenges in using stable isotopes to trace plant carbon and water relations across scales

Stable isotope analysis is a powerful tool for assessing plant carbon and water relations and their impact on biogeochemical processes at different scales. Our processbased understanding of stable isotope signals, as well as technological developments, has progressed significantly, opening new frontiers in ecological and interdisciplinary research. This has promoted the broad utilisation of carbon, oxygen and hydrogen isotope applications to gain insight into plant carbon and water cycling and their interaction with the atmosphere and pedosphere. Here, we highlight specific areas of recent progress and new research challenges in plant carbon and water relations, using selected examples covering scales from the leaf to the regional scale. Further, we discuss strengths and limitations of recent technological Published by Copernicus Publications on behalf of the European Geosciences Union. 3084 C. Werner et al.: Progress and challenges in using stable isotopes developments and approaches and highlight new opportunities arising from unprecedented temporal and spatial resolution of stable isotope measurements.

1 Introduction dicated which survey the published literature and pioneering work; thereafter, we focus on selected examples from the Stable isotopes are a powerful tool for tracing biogeochemilast decade.Finally, we highlight strengths and limitations cal processes across spatio-temporal scales ) and present an berg, 2000).The stable isotope composition of plant mate-outlook (Sect. 4) on what we identify as main goals of the rial, animal tissues, sediments and trace gases can be used stable isotope research in carbon and water biogeochemistry.as indicators of ecological change (Dawson and Siegwolf, 2007).The assessment of the circulation of isotopes in the biosphere allows characterisation and quantification of bio-2 Isotope effects across temporal and spatial scales geochemical cycles as well as exploration of food webs (Fry, 2006).Stable isotope studies give insights into key reactions 2.1 Leaf-level processes of plant metabolism (Schmidt and Gleixner, 1998), can increase our understanding of water movement along the soil-2.1.1CO 2 and H 2 O exchange plant-atmosphere continuum (Dawson et al., 1998), and allows palaeoclimatic/-physiological reconstructions (Beerling Leaf CO 2 and H 2 O fluxes have unique and distinct isotope and Woodward, 1998).Moreover, the analysis of the iso-signals that carry useful physiological and biogeochemical topic composition of trace gases exchanged between ecosys-information.For example, environmental stresses, such as tems and the atmosphere gives insights in the underlying drought, cause systematic variation in carbon isotope disprocesses driving the source and sink strength of biomes crimination during photosynthesis ( 13 C, see Table 1), shedfor CO 2 , CH 4 and/or N 2 O (Flanagan et al., 2005).Stable ding light on different steps of CO 2 transfer from the atmocarbon, oxygen and hydrogen isotope composition of or-sphere to the chloroplasts (Evans et al., 1986;Farquhar et al., ganic matter and inorganic compounds such as CO 2 and 1989a, b;Brugnoli and Farquhar, 2000).On the other hand, H 2 O is altered during vegetation-soil-atmosphere exchange stomatal opening and, hence, transpiration, cause 18 O and processes, such as evapotranspiration, carbon assimilation deuterium ( 2 H) enrichment of water at the sites of evaporaand respiration.This leaves an isotopic imprint on soil, tion ( ev , see Table 2), which lead to the enrichment of the plant and atmospheric carbon and water pools and associtotal leaf water (Dongmann et al., 1974;Farquhar and Lloyd, ated fluxes.These isotopic fingerprints can then be used to 1993).The oxygen of leaf-dissolved CO 2 exchanges with the trace different processes involved in the transfer of carbon 18 O-enriched leaf water, entraining distinct 18 O discriminaand water across the plant-soil-atmosphere continuum.Par-tion ( 18 O) during photosynthetic CO 2 exchange (Farquhar ticularly the multiple-isotope approach, i.e. the simultane-et al., 1993).The knowledge of 13 C and 18 O together ous measurements of stable isotope composition of differcan then be used to assess the limitations to CO 2 transfer beent elements ( 2 H, 18  δ δ O and/or 13  δ C, for definition see Tatween the intercellular air space and the chloroplasts (Gillon bles 1 and 2), provides a unique way to investigate the in-and Yakir, 2000b).terrelation between water and carbon fluxes (Ehleringer et The theoretical understanding of the individual 13 C and  al., 1993; Griffiths, 1998; Flanagan et al., 2005; Yakir and  18 O fractionation phenomena (including transport/diffusion, Sternberg, 2000).The use of biological archives may en-transformations and exchange processes) in CO 2 and H 2 O able extrapolation of this information to longer time scales, is well established for systems in steady-state (Dongmann such as the Anthropocene.Methodological advances allow et al., 1974;Farquhar et al., 1982;Evans et al., 1986;Farisotopologue and compound-specific analyses at unprece-quhar et al., 1993).However, we are only at the beginning dented resolution, providing new insight into isotope frac-of gaining theoretical understanding for those in non-steadytionation processes in metabolic pathways and in biogeo-state.On-line isotope discrimination studies, i.e. instantachemical processes.Further, a more advanced mechanistic neous measurements of leaf/plant gas exchange and the assounderstanding of processes affecting the stable isotope comciated isotopic signals, during transient conditions and shortposition in various ecosystem compartments allows mod-Table 1. Introduction to terms and equations of carbon isotopes, photosynthetic discrimination and post-carboxylation fractionation.

R
The delta notation for carbon isotopes R standard Carbon has two stable isotopes, 12 C and 13 C, with natural abundances where in the case of carbon isotopes, y X is replaced by 13 C, and of 98.9 and 1.1 %, respectively.The relative abundance of 13 C in any R sample and R standard are the measured 13 C 12 / C ratios in the samsample is conventionally expressed in the δ notation (Eq. 1) which is ple and standard, respectively defined as the relative deviation of the isotope ratio R (R = 13 C 12  / C) of a sample relative to that of an international standard (and is often expressed in ‰).The international standard is the R of CO 2 from a fossil belemnite in the Pee Dee formation of South Carolina.Today, 13 C standards are obtained from the IAEA in Vienna and are referred to as V-PDB (Coplen, 1995(Coplen, , 2011)).
Carbon isotope discrimination 13 = 13   13   C (δ C a δ C p )/(1 (2) The change in relative abundance of 13 C between an educt and product where 13   − + δ C a and 13  δ C p are the 13  δ C values of the CO 2 in air and is called discrimination, often denoted with .In the case of CO 2 as the the plant, respectively.source and the plant material as the product of photo-and biosynthesis, carbon isotope discrimination is described in Eq. ( 2).Isotope discrimination during carbon assimilation has been modelled (3) by Farquhar et al. (1982Farquhar et al. ( , 1989a) ) for C 3 plants by Eq. (3).This equation where a b is the fractionation during diffusion in the boundary layer has been developed to describe leaf-level photosynthetic discrimination (2.9 ‰); a is fractionation during binary diffusion in air (4.4 ‰); e s during the light period, where e denotes the fractionation of mitochon-is discrimination during CO 2 dissolution (1.1 ‰ at 25 • C); a l is fracdrial respiration in the light, i.e. day respiration (Tcherkez et al., 2010) tionation during diffusion in the liquid phase (0.7 ‰); b is fractionaand * the compensation point in the absence of day respiration.tion during carboxylation in C 3 plants (≈ 29.5 ‰); p a is atmospheric When Eq. ( 3) is applied to analyse 13 C of bulk tissue as an integrative CO 2 partial pressure; p s is CO 2 partial pressure at the leaf surface; parameter for preceding photosynthetic discrimination during formap i is sub-stomatal CO 2 partial pressure; p c is CO 2 partial pressure tion of this material, e denotes the integrated respiratory discrimination at the site of carboxylation; e is fractionation during mitochondrial both during light and dark-respiration.However, additional factors such respiration; f is fractionation during photorespiration; k is carboxyas fractionation during carbon allocation, tissue turnover or carbon par-lation efficiency; and * is compensation point in the absence of titioning into different plant organs may affect the observed discrimina-mitochondrial respiration (R d ).tion.To date, we still lack a quantitative description of these processes (see Sects. 2.1.2 and 2.2).
Simplified model of C 3 photosynthetic isotope discrimination 13 p a Few empirical/experimental studies have used Eq. ( 3), partly due to lack where a is the fractionation during binary diffusion in air (4.4 ‰); of needed input data.Instead, a simplified version (Farquhar et al., 1982) b is fractionation during carboxylation in C 3 plants; and p a and p i has been used extensively (Eq. 4).This equation is valid on the condiare the atmospheric and sub-stomatal CO 2 partial pressures, respection that effects of boundary layer, internal conductance, photorespiratively.tion, day respiration and allocation are negligible.In strict terms, these conditions are met if boundary layer and internal conductance are infinitely high, photorespiration and respiration are infinitely low or nondiscriminating, and isotope discrimination during allocation and partitioning does not happen.To account for effects of the neglected terms in Eq. ( 4), the value of b is often slightly reduced (≈ 28 ‰) (Brugnoli and Farquhar, 2000).
Intrinsic water use efficiency In Eq. ( 4), 13   = s = = C is directly proportional to p i /p a , which is determined where A is photosynthetic assimilation, g s is stomatal conductance; by the relationship between photosynthetic assimilation (A) and stoma is the fractionation during binary diffusion in air (4.4 ‰); b is fracatal conductance ( .Therefore, 13  g s ) C is a measure of intrinsic water tionation during carboxylation in C 3 plants; and p a and p i are the atuse efficiency WUE i (or transpiration efficiency), the ratio of assimila-mospheric and sub-stomatal CO 2 partial pressures, respectively.The tion to transpiration, which can be estimated as WUE i / VPD (Farquhar factor 1.6 denotes the ratio of diffusivities of water vapour and CO 2 and Richards, 1984;Farquhar et al., 1989b). in air.O p and 2 δ δ H p is the oxygen and hydrogen isotopic composirichment of leaf water or plant organic matter is expressed as enrichtion, respectively, of leaf water or plant organic matter and 18  δ O sw and ment above source water (often assumed to be soil or xylem water) by 2 δ H sw are the respective isotope compositions of the source water.Eq. ( 6).

Leaf water enrichment
The enrichment of the leaf water has been modelled with approaches ε + is the equilibrium fractionation between liquid water and water of increasing complexity (e.g.Cuntz et al., 2007).Steady-state isotopic vapour; ε k is the kinetic fractionation as vapour diffuses from leaf inenrichment of oxygen or hydrogen over source water at the site of evaptercellular spaces to the atmosphere (Farquhar et al., 1989a), v is the oration in the leaf ( e ) can be calculated by the Craig & Gordon model isotopic enrichment of water vapour relative to the source water taken (Craig and Gordon, 1965;Dongmann et al., 1974) by Eq. ( 7).In steady up by the plant, and e a /e i is the ratio of ambient to intercellular vapour state conditions (i.e.source water isotopic composition is equal to the pressures.one of transpired water), the isotopic enrichment of water vapour relative to the source water taken up by the plant ( v ) can be approximated by -ε + .This model was developed for open water surfaces and only applies to the water composition at the site of evaporation, and not the whole leaf (mean lamina mesophyll water).
Steady-state isotopic enrichment of leaf water ℘ C • D The steady-state isotopic enrichment of mean lamina mesophyll water where ℘ is the Peclet ´number • , E the leaf transpiration rate ( LsP ) can be described by correcting Eq. ( 7) for the so-called Peclet ´(mol m −2 s −1 ), L is the scaled effective path length (m) for water effect (Farquhar and Lloyd, 1993), as shown in Eq. ( 8).The Peclet ´effect movement from the xylem to the site of evaporation, C the molar conis the net effect of the convection of unenriched source water to the leaf centration of water (mol m −3 ), and D the tracer-diffusivity (m 2 s −1 ) of evaporative sites via the transpiration stream as opposed by the diffusion heavy water isotopologues (either H 18 O or 2 H 1 HO) in "normal" water. 2 of evaporatively enriched water away from the sites of evaporation.
L is a fitted parameter (using Eq. 8; Flanagan et al., 1993) as it cannot be measured directly.

+
Non-steady-state isotopic enrichment of leaf water Non-steady-state effects in lamina mesophyll water enrichment ( LnP ) where corresponding to + and ε α 1 ε, (α + and α k are ε , k have been added by Farquhar and Cernusak (2005) by Eq. ( 9).This = + respectively), V m is lamina leaf water molar concentration (mol m −2 ), equation has an analytical solution and can be calculated with the t is time (s), g "Solver" function in Excel.
t is the combined conductance of stomata and boundary layer for water vapour (mol m −2 s −1 ), and w i is the mole fraction of water vapour in the leaf intercellular air spaces (mol mol −1 ).The non-steady-state model of leaf water enrichment as given by Eq. ( 9) where r denotes the distance from the xylem to the evaporating site is a simplification of the advection-diffusion description of leaf wa-(m), v r is the advection speed of water in the mesophyll (m s −1 ), m ter enrichment ( LnAD ), as given by Cuntz et al. (2007) and Ogee ´et the volumetric water content of the mesophyll, and D r = m κ m D the al. (2007) in Eq. ( 10).
effective diffusivity of the water isotopologues (m 2 s −1 ), with κ m (< 1) Steady-state approaches often accurately describe leaf water isotopic the tortuosity factor of the water path through the mesophyll.The volenrichment (e.g.Welp et al., 2008), especially for longer times (weeks, umetric water content in the leaf mesophyll m is related to the water months or years) or spatial scales (ecosystem studies).If shorter times volume V m (per unit leaf area) and the mesophyll thickness r m through and spatial scales are considered (diel measurements or gradients across m = V m /(Cr m ) (Cuntz et al., 2007).a leaf), non steady-state approaches are more suitable, especially for modelling leaf water enrichment during the night (Cernusak et al., 2005).
Enrichment of organic matter Newly produced assimilates are assumed to obtain an imprint of the signature of the average bulk mesophyll leaf water at the time when they were produced.For oxygen, an equilibrium fractionation factor (ε wc ) results in carbonyl oxygen being ca.27 ‰ more enriched than water (Sternberg and DeNiro, 1983), which has been confirmed for cellulose (e.g.Yakir and DeNiro, 1990), leaf soluble organic matter (e.g.Barnard et al., 2007) and phloem sap sucrose (e.g.Cernusak et al., 2003b, Gessler et al., 2007a).
issues which need to be addressed to better understand the Ferrio et al., 2009;Kahmen et al., 2009;Ferrio et al., 2012).physiological information imprinted on plant material.
It is likely that in leaves, all water pools are involved during water transport (Yakir, 1998); however, the leaf water pools Progress and challenges might not be considered as perfectly mixed (e.g.Helliker and Ehleringer, 2000).Gan et al. (2002) compared differ-Mesophyll conductance ent leaf water evaporative enrichment models (i.e. the twopool model, the Peclet ´effect model and the string-of-lakes Mesophyll or leaf internal conductance (often referred to as model), which assume different water isotopic gradients and g m or g i ) has emerged as a significant (co-)limitation for different mixing of leaf water pools.The different models all CO 2 transport to the chloroplast, with large variation bedescribed large parts of the observed dynamics of leaf watween species and environmental scenarios of light, temperter enrichment but not all facets were captured by a single ature, drought and salinity (Warren and Adams, 2006;Flexas et al., 2008).On-line measurements of 13  model.The non-steady state Peclet ´model of Farquhar and C in conjunc-Cernusak (2005; see Table 2, Eq. 9) is the simplification of tion with gas exchange have been instrumental in detecting the diffusion-advection model of Cuntz et al. (2007;Eq. 10), these variations of g i .Variation in g i is related to developmenwhich assumes that the leaf represents a continuum of untal changes and morphological/structural features of leaves, enriched (source) and enriched (evaporative sites) water.The such as cell wall thickness, chloroplast arrangement, and leaf latter, more complex model is less sensitive to noise in the porosity (Flexas et al., 2008;Evans et al., 2009).Moreover, input data and gives smoother results.Cuntz et al. (2007), g i may be regulated via the expression of particular aquaporhowever, state that comparably good results could also be ins capable of transporting CO 2 across plasma membranes achieved with different well mixed metabolic pools of water.(cooporins) (Hanba et al., 2004;Flexas et al., 2007).Strong For epiphytic and non-vascular plants which lack permanent dynamic responses of g i to various environmental factors at access to soil water, it has been shown that a description of the scale of minutes to days have been reported (Flexas et al., water isotope dynamics requires consideration of distinct wa-2008; Bickford et al., 2009), and such variation has also been ter pools as well as water potentials (Helliker and Griffiths, observed at the canopy-scale (Schaufele ¨et al., 2011).So far, 2007;Hartard et al., 2009;Helliker, 2011).the metabolic basis of these short-term adjustments of g i is unknown.
Exchange of H ) 3 − proaches (Tcherkez et al., 2010) indicate that the isotopic (Badger and Price, 1994).This exchange underlies 18 O discomposition of day respiration is not the same as that of con-crimination during CO 2 exchange ( 18 O), and retroflux to currently fixed carbon dioxide.In part, the respiratory car-the atmosphere of CO 2 that has previously equilibrated with bon isotope fractionation during daytime (Tcherkez et al., leaf water, which has a strong effect on the 18 O content of at-2010) is related to fuelling of respiration by old carbon pools mospheric CO 2 (Farquhar et al., 1993).Also, this signal pro- (Nogues ´et al., 2004).This calls for further experimental vides a measure of photosynthetic activity of the terrestrial studies, a more detailed theoretical description of whole-leaf biosphere (Farquhar et al., 1993).At the leaf level, measure-13 C during daytime gas exchange (Wingate et al., 2007; ments of gas exchange, CA, 18 O and 13 C can help to par- Tcherkez et al., 2004), and consideration of this effect in car-tition mesophyll conductance into a cell wall and a chlorobon isotope-based estimations of g .plast component (Gillon and Yakir, 2000a).However, work i by Cousins et al. (2008) indicates that CA activity may not Water isotope enrichment in leaves be a good predictor for CO 2 -H 2 O isotopic exchange, endorsing the view that more work is needed to fully understand the Isotopic enrichment in leaf water is reasonably well undercontrol of 18 O and its physiological implications.stood (Craig and Gordon, 1965;Dongmann et al., 1974;Farquhar and Lloyd, 1993;Cuntz et al., 2007;Ogee ´et al., 2007), Ternary effects on CO 2 isotopes during gas exchange except for the parameter that characterises the effective path length for water movement from the xylem to the site of Ternary effects, i.e. effects of concurrent water vapor difevaporation (see Peclet ´effect, Table 2).This parameter is fusion on CO 2 diffusion through stomata, are taken into especially important for modelling leaf water enrichment in account for CO 2 exchange (von Caemmerer and Farquhar, non-steady state.Understanding how the effective length is 1981) but not for isotopes.Mesophyll conductance is a adjusted by environmental conditions requires knowledge of parameter greatly influenced by ternary effects.Farquhar how water transport inside the leaf is changing, for exam-and Cernusak (2012) recently showed that by applying the ple, with the leaf's water status (Barbour et al., 2004(Barbour et al., , 2007;; ternary correction, oxygen isotope composition of CO 2 in the C. Werner et al.: Progress and challenges in using stable isotopes chloroplast and mitochondria better match the oxygen iso-isotopic signatures of phloem sugars, with day sucrose being topic composition of water at the sites of evaporation.The 13 C-depleted, while night exported sucrose is 13 C-enriched ternary effect has been observed to be greatest when the leaf- (Tcherkez et al., 2004;Gessler et al., 2008).Analyses of to-air water vapor pressure deficit is large.Farquhar and Cer-sugar 13  δ C and its diurnal variations offer potential for imnusak ( 2012) also observed that a large impact of ternary cor-proved tracing of these changes in these metabolic activities.rections occurred when the difference in the isotopic composition of CO 2 between the leaf interior and the ambient Apparent respiratory fractionation air was large.The precision of current isotope fractionation During the last decade the knowledge of relevant apparmodels can be improved by applying the ternary correction ent fractionation in the respiratory pathways has signifiequations for isotope fractionation and isotope exchange durcantly advanced, demonstrating substantial variability in resing gas exchange measurements.
piratory fractionation among species, organs and functional groups, as well as with environmental conditions (see re-
The observed apparent respiratory fractionation and its vari-The carbon isotope signal, imprinted through photosynthetic 13 ability are mainly attributed to (i) non-homogeneous 13 C-C discrimination (the sum of terms one to four on the distribution within hexose molecules reported by Rossmann right hand side of Eq. 3, Table 1), can be altered by multiple et al. (1991) and modelled by Tcherkez et al. (2004), (ii) relprocesses in down-stream metabolic pathways (termed postative contributions of different pathways to respiration (recarboxylation fractionation), which will be reflected in difviewed by Ghashghaie et al., 2003), as well as (iii) enzymatic ferent carbon pools and respired CO 2 .Despite early evidence isotope effects during decarboxylation reactions (recently reby Park and Epstein (1961), carbon isotope fractionation durviewed by Tcherkez et al., 2011).ing dark respiration has long been considered negligible (Lin and Ehleringer, 1997).Systematic studies by Duranceau et Fragmentation fractionation al. (1999) and Ghashghaie et al. (2001) with a range of C 3 species again provided clear evidence for substantial and sys- The non-homogeneous intra-molecular distribution of 13 C in tematic variation in carbon isotope ratios of leaf dark respicarbohydrates, results in so-called "fragmentation fractionaration (see review by Ghashghaie et al., 2003).These aution" (Tcherkez et al., 2004), leaving its imprint on synthethors introduced the term "apparent respiratory fractionasized metabolites.If one of these products is decarboxylated, tion" to describe the manifested differences between carbon then respired CO 2 will carry an isotopic signature different isotope compositions of leaf dark-respired CO 2 and its pufrom the average sugar signature.New data indicate that the tative substrates (mainly carbohydrates), caused by multiple heterogeneous 13 C distribution in carbohydrates may vary processes in the respiratory pathways (see below).The work among species and with environmental conditions (Gilbert et of Ghashghaie and coworkers promoted a significant numal., 2011).Moreover, switches between substrates (Tcherkez ber of studies on post-carboxylation fractionations in downet al., 2003) during light-dark transition of leaves (i.e.light stream metabolic processes (Klumpp et al., 2005;Badeck enhanced dark respiration due to decarboxylation of lightet al., 2005;Cernusak et al., 2009;Tcherkez et al., 2011;accumulated malate, Barbour et al., 2007) and the oxidative Werner and Gessler, 2011).

Progress and challenges
The implication of these processes still needs to be explored.

Post-carboxylation fractionation
Temporal dynamics and apparent respiratory fractionation Already within the Calvin cycle, isotopic fractionation occurs mainly due to metabolic branching points and the use So far, a full quantitative understanding of apparent respiraof triose phosphates that can either be exported to the cy-tory fractionation has not yet been achieved.However, meatosol or continue to be used within the Calvin cycle.The surements with a high temporal resolution indicated remarktriose phosphates that are not exported are subject to cer-able diel dynamics in leaf respiratory 13  δ CO 2 , which diftain enzyme catalyzed reactions (aldolisation and transketoli-fered between functional plant groups (Priault et al., 2009;sation) which involve position-specific discrimination dur- Werner et al., 2009;Werner and Gessler, 2011).Feeding exing C-C bond making.As a result, the C-3 and C-4 po-periments with positionally labelled glucose or pyruvate can sitions within glucose are enriched in 13 C and thus a non-trace changes in carbon partitioning in the metabolic branchhomogeneous intra-molecular distribution of 13 C within car-ing points of the respiratory pathways (Tcherkez et al., 2004), bohydrates is established (Rossmann et al., 1991;Tcherkez which has been used at the organ level as evidence for an et al., 2004;Gilbert et al., 2009).Subsequently, photorespiraimportant contribution of PPP to root respiration (Batheltion and starch-sucrose partitioning result in diel changes in lier et al., 2009) as well as to identify differences between functional groups (Priault et al., 2009;Wegener et al., 2010).

Progress and challenges One challenge of labelling experiments is to find methods for channelling additional labelled metabolites into plant or-
The dual-isotope approach gans in vivo to shed further light on potential involvement of these metabolites and their metabolic pathways.Further-Combined analyses of the carbon and oxygen isotopic commore, the commitment of metabolites to alternative pathways position of bulk leaf biomass provide a means to distinat metabolic branching points needs to be quantified.This guish the separate effects of stomatal conductance and net is particularly relevant where metabolic channelling evokes photosynthesis on WUE i (Scheidegger et al., 2000).Prefercompartmentation in organelles with membranes, which are ably, however, carbon isotope discrimination 13 C b , and impermeable for intermediate products, as shown recently bulk biomass oxygen isotope discrimination, 18 O b , should for the Krebs cycle (Werner et al., 2011).
be used in such an approach to account for effects of differences in 13  δ C of assimilated CO 2 and variations of 18  δ O Intra-molecular site-specific isotope fractionation of source water.A distinction between stomatal and photosynthetic influences cannot be made from analysis of 13   , 2001;Guerrieri et al., 2009Guerrieri et al., , 2010;;Savard, 2010).the application of this new technology is to ensure sufficient Grams et al. (2007) extended the model to estimate stomsample quantities from metabolic pools of interest.Of speatal aperture directly for interpreting physiological changes.cial interest is that obtained data can be interpreted directly 18 O of bulk organic matter has also been used to deterin terms of isotope effects associated with specific enzymes.δ mine whether a change in WUE i results from the increase in atmospheric CO gwolf, 2007).Also, the En onmental effects 2 (Saurer and Sie vir effects of changes in vapor pressure deficit (VPD) resulting from increasing temperature or decreasing precipitation More studies of the sensitivity of respiratory 13  δ C to changes have been assessed along a Siberian transect (Sidorova et al., in environmental conditions and between organs are needed, 2009;Knorre et al., 2010).The dual-isotope approach has which will allow for a better understanding of temporal variproven a valuable concept for ecological applications.Howability in post-photosynthetic fractionation (Dubbert et al., ever, the interpretation of 13 C b in terms of WUE i under 2012) and could provide a basis for the use of respiratory natural changing environments is complex (e.g.Seibt et al.,13 δ CO 2 as an indicator of physiological activity (e.g.Bar-2008), as it provides a time-integrated record of photosynbour et al., 2011a, b).
thetic discrimination over the period that the carbon forming the leaf was fixed, which can be derived from multi-2.1.3Bulk leaf tissue 13 C and 18  δ δ O and water ple sources, e.g.fresh assimilates, carbon exported from mause efficiency ture leaves or even older storage pools.Moreover, different leaf carbon pools have different residence and turnover times The 13 C model by Farquhar et al. (1982;Eq. 3 in Table 1) (Nogues ´et al., 2004;Lehmeier et al., 2008Lehmeier et al., , 2010b)).Thus, predicts a linear relationship between 13 C and intrinsic wa-during leaf formation, growth and maintenance leaf bulk mater use efficiency (WUE i ; the ratio of net assimilation, A, terial integrates isotopic information from different time peto stomatal conductance, g s ), for conditions where meso-riods and sources, which is weighted by the amount of carbon phyll conductance is very high and (photo)respiratory 13 C incorporated from each source/period.Therefore, interpretadiscrimination is negligible (Eq.4).Empirical studies in con-tion of 13 C b in terms of WUE i under natural changing entrolled conditions confirmed this linear relationship between vironments requires several precautions, as described below. 13C, estimated from bulk biomass carbon isotope composition ( 13 C b ), and WUE i (Farquhar et al., 1982(Farquhar et al., , 1989b;; Interpretation of 13 C b and 18 O b in relation to leaf Ehleringer et al., 1993;Griffiths, 1998;Brugnoli and Fartraits quhar, 2000).In the following three decades, this linear (simplified) model of 13 C (Eq. 4) was used widely as an indica-A comparison of WUE i between different species based on tor of water use efficiency at the leaf, plant and ecosystem 13 C in bulk leaf material is nontrivial.Differences in not scale (e.g.Bonal et al., 2000;Lauteri et al., 2004; Saurer only leaf structural, anatomical, but also physiological traits et al., 2004;Ponton et al., 2006) in retrospective studies can modulate 13 C b (Ehleringer, 1993; Werner and Maguas, óf carbon-water relations based on biological archives (see 2010), as well as 18 O in lamina leaf water (Barbour and Sect.2.6), and in breeding crop varieties for improved yield Farquhar, 2003;Kahmen et al., 2008) and 18 O b .Differunder water-limited conditions (Condon et al., 2002).

C. Werner et al.:
Progress and challenges in using stable isotopes Kogami et al., 2001;Hanba et al., 2003) and thus 13 C b (see and Schleser, 2004).Moreover, during starch breakdown, Sect.2.1.1).Mesophyll conductance is generally neglected carbonyl oxygen atoms are exchanged with unenriched water when calculating WUE i from stable isotope discrimination in stems, causing these incorporated starch-derived sugars to (see Eq. 5).If there are varying influences of mesophyll con-be 18 O depleted as compared to sugars formed in transpiring ductance on 13 C among species, WUE i calculated from leaves (Gessler et al., 2007b).This "isotopic starch imprint" Eq. (5) will be not directly comparable.Leaf traits may also in the newly developed leaves is thought to be diluted duraffect the scaled effective path length for water movement ing the growing season by carbon turnover and the incorpofrom the xylem to the site of evaporation (Wang and Yakir, ration of new assimilates into bulk leaf organic matter (see 1995) and thus influence 18 O (cf.Eq. 8, able 2).The conof ulk leaf 13 b T e.g. the seasonal course b δ C of beech shown by ceptual model of Scheidegger (2000) does not account for Helle and Schleser, 2004).For interpreting the isotopic comsuch effects but strictly assumes oxygen isotope enrichment position of a deciduous leaf, it is thus important to consider to be only affected by the ratio of ambient to intracellular when the leaf was harvested during the season and that there water pressure (e a /e i ; cf.Eq. 7).Any other factor varying might be species-specific differences in the extent to which leaf water evaporative enrichment and thus 18 O b will thus the starch imprint or the influence of the assimilates incorpoconstrain the interpretation of the impact of stomatal conrated during the current growing season dominate the bulk ductance versus net photosynthesis on WUE i .Moreover, due isotope signal.In turn, coniferous needles can accumulate to differences in phenological phases and length of grow-large amounts of starch in spring, followed by mobilization ing period leaf 13 C b and 18 O b of co-occurring species towards the summer, and starch contents are generally low might provide an integrated signal over diverging environduring the winter (e.g.Ericsson, 1979).As a consequence, at mental conditions (e.g.Werner and Maguas, ´2010).Thus, least part of the 13 C b during the growing season is related species-specific differences in phenology, growth pattern and to the variation of starch content and isotopic composition leaf structures might constrain a direct comparison of bulk (Jaggi ¨et al., 2002), a fact that also needs to be taken into acleaf 13 C and 18 O between different species (Hanba et al., count when calculating WUE from 13   i C b and comparing it 2003; Warren and Adams, 2006).Moreover, ontogeny can among species.markedly alter the isotopic signature (Terwilliger et al., 2001; Thus, the complexity of processes influencing 13 C b may Bathellier et al., 2008;Salmon et al., 2011).Repetitive samconstrain its use in ecological field studies.Carbon pools pling and isotope analysis of tissues and compounds which with shorter turnover times and thus a better-defined origin are known to integrate shorter and more defined time peri-such as leaf soluble sugars (Brugnoli and Farquhar, 1988), ods such as phloem sugars (e.g.Keitel et al., 2003 , 1997).That material was to-sink transport, while tracing of carbon allocation with high derived from the photosynthetic activity of previous year temporal resolution in plants required the use of labelling exleaves, with different morpho-physiological characteristics periments (e.g.Hansen and Beck, 1990). in other environmental conditions, producing a previous year Within the last ten years it was, however, shown that the isotopic signal.Since starch can be 13 C-enriched by up to transport of newly assimilated carbon within the plant and 4 ‰ as compared to newly assimilated sugars (Gleixner et al., from the plant to the rhizosphere can also be followed by 1998), growing leaves supplied from storage pools are often natural abundance stable isotope techniques (e.g.Scartazza strongly 13 C enriched in spring (e.g.Terwilliger, 1997;Helle et al., 2004;Brandes et al., 2006;Wingate et al., 2010b).
Progress and challenges 13 C enrichment (Brandes et al., 2006;Wingate et al., 2010a) and no change in 13  δ C (Pate and Arthur, 1998; Gessler et al., Plant integrating information and phloem transport 2007a) to 13 C depletion (Rascher et al., 2010).The nature of these species-specific differences remains to be clarified and The 13  δ C of phloem organic matter is mincreasingly being might shed new light on mechanisms controlling assimilate used to derive information on carbon allocation, canopy intepartitioning in trees.grated WUE and canopy integrated mesophyll conductance in plants affected by environmental conditions (e.g.Keitel et Chemical composition of phloem sugars al., 2003;Gessler et al., 2004;Scartazza et al., 2004;Barbour et al., 2005;Rascher et al., 2010;Ubierna and Marshall, It is often assumed that only one major sugar, namely 2011; Dubbert et al., 2012).A dual-isotope approach ( 13δ C sucrose, is present in the phloem.However, besides suand 18  δ O see Sect.2.1.3)can also be successfully applied to crose, there are other transport carbohydrates -depending phloem sugars to distinguish whether net assimilation and/or on species and phloem loading mechanisms -such as myostomatal conductance is changing as a result of environmeninositol and raffinose family sugars (Karner et al., 2004).In tal conditions (Keitel et al., 2003, Cernusak et al., 2003, addition, it is still a matter of debate if hexoses are trans-2005; Brandes et al., 2006;Keitel et al., 2006).Even though ported in the phloem or not (van Bel and Hess, 2008; Liu the carbon and oxygen isotope composition of phloem oret al., 2012).Phloem sugar composition can vary with enviganic matter can, in principle, integrate leaf physiology over ronmental conditions, which could be one factor for changes the whole canopy and track transport of assimilates within in phloem 13  δ C (Merchant et al., 2010), independent of the the plant, it is now clear that several uncertainties constrain original leaf-borne isotope signal, since 13  δ C differs between the interpretation of phloem isotopic information.These are different carbohydrate molecules (Schmidt, 2003; Devaux et related to (i) the temporal integration of the isotope signal in al., 2009).Compound-specific analysis, provided by modern the phloem, (ii) potential changes of the isotope composition LC-and GC-IRMS techniques (Sect.3), will help to differof phloem sugars in basipetal direction, and (iii) the chemical entiate between changes in phloem 13  δ C that result from eicomposition of phloem transported organic matter.
ther changes in the chemical composition or changes in leaf level fractionation.In addition, comparable methods should Phloem sugars and temporal integration be used to characterise the compound-specific oxygen isotope composition of phloem organic matter.Short-term variations in the isotopic composition of leaf sug-Only recently, the natural abundance stable isotope inforars -induced by either an environmental signal or plant inmation in soil and ecosystem respired CO 2 cross-correlated ternal processes -might or might not be reflected in the isoto photosynthesis (or its proxies) has been used systematitopic composition of phloem organic matter.Twig phloem cally to characterise the speed of link between canopy and organic matter of trees (e.g.Gessler et al., 2007a) and the soil processes (see review by Kuzyakov and Gavrichkova, stem phloem of herbaceous species (e.g.Gessler et al., 2008) 18 2010;Wingate et al., 2010b).Even though such approaches can be applied to monitor diel variation of evaporative O 2 have significant potential, there is still debate about the physand H enrichment or carbon isotope fractionation.In the iological information conveyed by the isotope signal and of trunks of adult trees, however, the mixing of sugars of differthe processes involved (see review by Bruggemann ¨et al., ent metabolic origins can dampen the short-term variations 2011).and the isotope signatures provide time-integrated information on canopy processes instead (Keitel et al., 2006;Rascher The link between above-and belowground processes et al., 2010).

Change of the original isotope signal in
In their review, Kuzyakov and Gavrichkova (2010) postuphloem sugars lated that approaches which quantify time lags between proxies of photosynthetic activity and natural abundance 13  δ C The original isotope signal imprinted on sugars in the leaf in soil respiration are (besides other techniques) approprimay be altered during basipetal transport in the phloem ate to study the link between above-and belowground proof trees (e.g.Rascher et al., 2010).The transport of sugar cesses.Mencuccini and Holtta ¨( 2010) reviewed different apmolecules itself does not fractionate to a measurable extent.
proaches to assess the speed of link between assimilation However, carbon fixation by PEPc in the bark and oxygen and soil respiration and concluded, in contrast, "that isoatom exchange with stem water during metabolic processes topic approaches are not well suited to document whether in the stem tissue together with the continuous unloading changes in photosynthesis of tall trees can rapidly affect and loading of sugars from and to the phloem might con-soil respiration".These different opinions may be related tribute to the observed isotope patterns (Barnard et al., 2007; to uncertainties on the mechanisms involved, as described Gessler et al., 2009).The change in 13  δ C along the transport by Kayler et al. (2010a, b): on the one hand, there is evpath, however, varies strongly among species ranging from idence that pressure-concentration waves (  , 2004), which travel rapidly through the phloem Progress and challenges of plants (and not the supply of new assimilates transported via the phloem to the roots and the rhizosphere), are respon-Adding duel-isotope approaches to community ecology sible for the fast response of soil respiration to changes in 13 C and 18 O signals can trace biotic and abiotic interactions photosynthesis.On the other hand, the time-lag between the within the plant community and may contribute to identifyfixation of a carbon molecule during photosynthesis and its ing what shapes community-scale processes.However, indirespiration belowground may contain real and important invidual plants will not necessarily respond to environmental formation about plant physiology and carbon use as well as perturbations as "a community", but may respond accordthe degree to which plant and soil are coupled (Kayler et ing to species-specific traits and requirements and additional., 2010a).This information may be obtained by the assess-13 ally depend on the interactions with the surrounding enviment of δ C in respired CO 2 but also in respiratory subronment and other present species (e.g.Roscher et al., 2004; strates when points listed above are taken into account.To Gubsch et al., 2011).Competition and/or facilitation interunravel the importance of the different relevant processes, we actions between species, e.g. through depletion of a particneed novel, pertinent experiments which combine (a) conular resource, may also be a source of isotopic variation, tinuous measurements of the natural abundance stable isoas shown for plant-plant competition for above-and belowtope composition of soil respired CO 2 , as done by Wingate ground resources by combining 13  terations in nutrient, carbon and hydrological cycles after exindicates how long it takes until a molecule with a given isootic plant invasion, can also be traced through stable isotopes tope composition imprinted during photosynthesis is trans-(e.g.Rascher et al., 2011).ported from the leaves via the phloem to the roots where it is respired, (b) allows to detect the response time(s) of soil res-Tracing spatial interaction between species within plant piration towards changes in carbon assimilation, which might communities or might not be faster than the transport of a given molecule from the canopy to belowground.
Spatio-temporal variations in isotope ratios (i.e.isoscapes) contain a potential wealth of information regarding eco-2.3Community-scale processes logical processes (West et al., 2008;Bowen et al., 2009), which have, so far been applied at larger spatial scales Because different species living within the same habitat show (see Sect. 2.5).At the community scale, spatial heterogenemarked differences in the isotopic composition of their leaf ity ty ariation in 13  in resource availabili , differential resource utilisation tissues, characterising community-wide v δ C and/or 18  by neighbouring species and their interactions (competiδ O, can provide potentially powerful tools for investion and facilitation) occur in a spatially explicit dimension, tigating the physiological basis for niche partitioning among which may contain crucial information regarding community community members (Dawson et al., 2002).Good examples functioning.For example, hydraulic redistribution of water are utilization of different water sources and redistribution sources is a key process which can shape plant communi- (Caldwell et al., 1998;Ryel et al., 2003), which can in turn be ties (see review by Prieto et al., 2012).Recently it has been linked to community composition (Ehleringer et al., 1991), shown that downscaling isoscapes to the community level niche partitioning and spatial and temporal variations in plant allowed tracing the spatial impact of an invasive species distributions (e.g.Dawson et al., 2002;Snyder and Williams, on community functioning (Rascher et al., 2012), and may 2000; Stratton et al., 2000;Drake and Franks, 2003; Rose therefore open new possibilities in resolving the spatial comet al., 2003;Grams and Matyssek, 2010).As stated above ponent of within-community interactions.(Sect.2.1.3),it must be kept in mind when comparing different species that the isotopic composition of bulk leaf ma-Tracing functional groups/community composition terial might be influenced by multiple factors such as structural, anatomical and physiological traits but also phenology.
During the last decade, the functional group approach has There are only very few community-wide investigations on proved to be an efficient way to analyse plant functioning at this topic (see Smedley et al., 1991;Guehl et al., 2004;Kah-the community scale.Leaf bulk 13 C allows the distinction men and Buchmann, 2007).This scarcity may be in part reof broad plant functional types, differing in structural, phelated to difficulties in assigning cause and effect to observed nological and physiological leaf traits (Brooks et al., 1997; variation from either physical (e.g.water availability, light) Bonal et al., 2000;Werner and Maguas, ´2010).Functional or biological (e.g.resource competition) factors.
traits such as water or nutrient use strategies, carbon acquisition, growth behaviours, and phenological cycles contribute significantly to the observed variation in isotope composition (e.g. Warren and Adams, 2006;Gubsch et al., 2011;Salmon et al., 2011;Ramírez et al., 2012).However, the responsive-recent literature).The core of the variation behind patterns in ness of leaf 13 C as a functional tracer has to be verified 13 δ C of ecosystem respiration ( 13δ C R ) lies in photosynthetic for different communities and may differ with the predomidiscrimination, the magnitude of metabolic fluxes and sevnant environmental constraints for plant growth and survival eral post-carboxylation fractionation processes that differ be-(e.g.Caldeira et al., 2001).For example, in a tropical rainfor-tween autotrophic and heterotrophic organs (see Sect. 2.1.2est, 13 C was associated with differences in shade tolerance and references therein).How these components manifest into (Bonal et al., 2000;Guehl et al., 2004), whereas in an upland integrative measures such as ecosystem respiration is fundawater-limited grassland of Greece, a semi-arid Inner Mongo-mental to understanding ecosystem physiology and biogeolian steppe, and a Portuguese mediterranean macchia groupchemistry.It is clear that ecosystem responses to climate and ing according to 13 C was associated with species' com-land use change, or perturbations, such as drought or fire, petitive ability related to WUE i , nitrogen use efficiency, and are an integrative signal from a network of carbon pools structural adaptations to drought (Tsialtas et al., 2001; Gong and organisms linking legacy conditions to current observaet al., 2010; Werner and Maguas, ´2010).
tions (e.g.Buchmann et al., 1997aBuchmann et al., , b, 1998;;Ehleringer et al., 2000).Thus, to properly account for ecosystem trace gas ex-The role of water source partitioning on community change and partitioning by stable isotopes, a detailed knowlfunctioning edge of the physical and biological basis of the isotopic signals for each of the fluxes and their dynamics across spa-Several mixing models have been used to determine the con-tial and temporal scales in soil-biosphere-atmosphere intertribution of different water sources to plant and ecosystem actions is required.evapotranspiration: Linear mixing models can be applied if the differences of 18  δ O among the water sources and xylem Progress and challenges plant water are large enough; δD- 18  δ O plots can be used if the difference between water sources and xylem water Recent findings on component sources and fluxes is small (Ogle and Reynolds, 2004;Dawson and Simonin, 2011).The use of multiple source mass balance analyses can Previous ecosystem 13 C and 18 O isotope research primarily improve the capacity to quantitatively and objectively evalfocused on partitioning of soil and canopy sources, which uate complex patterns in stable isotope data for determining are now a mainstay of ecosystem isotopic investigations (e.g.possible contributions of different sources to total plant water Buchmann et al., 1998;Kaplan et al., 2002;Yakir and Sternuptake (see review by Hu et al., 2009).Furthermore, combin-berg, 2000 and literature therein).The inherent complexity ing water source partitioning with indicators of species func-behind ecosystem respiration lies behind the many contributtional responses (e.g.changes in leaf water potential and car-ing sources.Nowadays, studies of these components have bon isotope discrimination) lent insight regarding the degree expanded to include stem CO 2 flux, mycorrhizal and microof plasticity among individual members of a given plant com-bial contributions (Esperschutz ¨et al., 2009), litter decompomunity (Maguas ´et al., 2011).However, there is increasing sition (Bird et al., 2008;Rubino et al., 2010), dissolved orawareness that the utilisation of simple linear mixing mod-ganic carbon (Sanderman and Amundson, 2008; Muller ¨et els to infer plant water uptake by comparing D and 18  δ δ O of al., 2009), erosion (Schaub andAlewell, 2009), soil organic xylem or root crown, on the one hand, and soil water, on the matter dynamics (Klumpp et al., 2007;Kayler et al., 2011) other hand, does not adequately reflect the high heterogeneity and CO 2 storage in soil air and solution (Gamnitzer et al., of water sources that may be available for a plant.Given the 2011).Labelling has also played a central role in achieving importance of resource variability at the community level, a higher level of certainty in observing single source temthe utilisation of more complex mixing models (for example, poral patterns (Ubierna et al., 2009;Powers and Marshall, by Phillips, 2001;Phillips andGregg, 2001, 2003;Parnell et 2011).Similarly, the water oxygen and hydrogen isotope al., 2010) as well as Bayesian models (Ogle et al., 2004) may composition has been used as natural or artificial tracer of be fruitful.
the ecosystem and component water fluxes and to partition evaporation and transpiration (e.g.Yakir and Wang, 1996;2.4Use of stable isotopes to disentangle ecosystem Yepez et al., 2005Yepez et al., , 2007;;Williams et al., 2004;Lai et al., exchange processes 2006;Rothfuss et al., 2010;Wang et al., 2010;Kim and Lee, 2011) to assess ecosystem water use efficiency (WUE) (Pon-At the ecosystem scale, stable isotopes can provide insight ton et al., 2006) and, e.g. the effects of hydraulic redistribuinto the complex interaction between vegetation, soil and attion by roots and mycorrhiza (e.g.Ludwig et al., 2004; Kurzmosphere exchange of carbon and water fluxes, including Besson et al., 2006;see Sect. 2.3.2).These detailed studies their responses and feed-backs to environmental drivers (e.g. are important because inferences can be drawn concerning Flanagan and Ehleringer, 1998;Dawson et al., 2002;Yakir carbon and water dynamics at larger time scales (e.g.eroand Sternberg, 2000;Yakir, 2003, Hemming et al., 2005; sion, soil organic matter transformations), and spatial variplease see Bowling et al., 2008 for review of pioneer and ability across the ecosystem can be better described.The advantage of these studies is two-fold: (1) underlying con-transport times of recent assimilates (Epron et al., 2011) nections between ecosystem carbon pools and fluxes and the can potentially delay the photosynthetic response signal in influence of changes in environmental drivers can be char-13 δ C R , or abiotic phenomena (following section) can obscure acterised, and ( 2) results can be used in models designed to component iso-fluxes (e.g.Ekblad et al., 2005;Knohl et al., partition ecosystem respiration. 2005).In these cases, deciphering the drivers behind 13  δ C R may become increasingly difficult.

Canopy labelling
Abiotic influences Whole ecosystem dynamics studied in situ using isotopes, at first pioneered through girdling (Hogber ¨g et al., 2001), have Analyses of 13  δ C R may lead to the identification of drivers increased in number through whole canopy tracer applica-and mechanisms underlying the dynamics of ecosystem tion.Advances in our understanding of ecosystem processes metabolism; yet, other abiotic processes that are also affected through canopy labelling include assessing photosynthetic-by biological drivers (e.g.temperature) may amplify, dampen soil-respiration coupling strength (Steinmann et al., 2004; or time-lag responses in 13  δ C R , obfuscating the signal of bi-Hogber ¨g et al., 2008;Bahn et al., 2009;Gamnitzer et ological respiration (Bruggemann ¨et al., 2011). Soil respiraal., 2009, 2011) 2006;Kayler et al., 2008Kayler et al., , 2010a;;Nickerson and Risk, 2009; and the turnover times of seasonally dynamic carbon pools Ohlsson, 2009;Gamnitzer et al., 2011).Likewise, the oxy- (Epron et al., 2012), studies that were previously limited to gen isotope composition of soil respired CO ( 182 δ O S ) not laboratory studies (e.g.Schnyder et al., 2003;Lehmeier et al., only carries the isotopic signature of the soil water it inter-2008, 2010a, b) or inferred from annual changes in biomass acted with, but also is influenced by the carbonic anhydrase measured in the field.Recent findings have illustrated the in soil microorganisms that accelerate isotopic equilibration complexity of dynamic processes that interact at the ecosys-between CO 2 and soil water (Wingate et al., 2009(Wingate et al., , 2010b)).tem scale.This calls into question the applicability of simple Despite their potential to propagate uncertainties in isotopic two-end member mixing models in complex systems with information through the soil-canopy continuum, such promultiple sources (Kayler et al., 2010a) and poses a significesses merit inclusion in isotope ecosystem models, enhanccant challenge for ecosystem studies, as outlined below.
ing the interpretation of patterns and drivers of 13 δ C R .

Heterogeneous flux sources Flux partitioning
Ecosystem respiration is a complex mixture of isofluxes from Conventional partitioning methods based on eddy covariance a range of biotic and abiotic sources that span the soil to vegmethods typically require several days or weeks of data to etation canopy continuum (see Badeck et al., 2005 and Bowl-cover key phenological periods in order to obtain robust reing et al., 2008 and literature therein).These sources con-gression parameters (e.g.Reichstein et al., 2005), neglecting tribute with distinct isotopic signatures at time scales from ecosystem responses at shorter time scales.These are, for daily (Bowling et al., 2003;Mortazavi et al., 2006;Werner et example, "switches" of ecosystem states (Baldocchi et al., al., 2006;Kodama et al., 2008;Unger et al., 2010a;Wingate 2006;Lee et al., 2007) or the pulse-like response of soil reset al., 2010a) to seasonal cycles (Griffis et al., 2004;McDow-piration to strong rain events, occurring at time scales from ell et al., 2004;Ponton et al., 2006;Alstad et al., 2007;Scha-minutes to hours (e.g. Xu et al., 2004;Unger et al., 2010bUnger et al., , effer et al., 2008)).These are based partly on phenology and 2012).It would be helpful if the partitioning scheme could disturbance regimes and all exhibit different effects on com-resolve episodic responses of this kind, because it is the tranponent fluxes.The challenge to advance our understanding sient, non-equilibrium responses that provide a rigorous test of 13  δ C R lies in identifying and quantifying these fluxes and of model performance and validity.Assimilating continuisotopic signatures of important ecosystem components (e.g.ous measurements of CO 2 and H 2 O fluxes and their isotopic Unger et al., 2010a;Epron et al., 2011;Barbour et al., 2011a).
compositions (e.g. 13  δ C, 18  δ O, 2 δ H) into process-based mod-This is especially important to test hypotheses about tempo-els should therefore provide a better-constrained solution.ral 13 C patterns, for e if 13 δ R xample, δ C R dynamics are heav-Similarly, assimilating chamber-based flux measurements of ily influenced by a sole component flux, resulting in a poorly these isotopic fluxes should help to explain and constrain mixed ecosystem source signal.Similarly, species-specific our model predictions during metabolic switches, especially when photosynthetic products may become limiting such as extent to which these moisture sources and climate phases during drought (Unger et al., 2010a), rainy periods (Wingate are recorded and how plant physiology alters the source waet al., 2010a) or when post-photosynthetic fractionation proter signal in the long-term growth record of trees is one of cesses dominate the isofluxes, e.g. at dawn (Barbour et al., the great challenges today.2011a).
Isotope tracers in back trajectory analysis 2.5 Regional scale isotope variation in precipitation and Back trajectory analysis of weather and thus precipitation linkages to carbon cycling (Draxler and Hess, 2004;Sjostrom and Welker, 2009) is a modelling tool that has been used extensively to quantify High frequency, spatially dense precipitation isoscapes (i.e. long-distance transport of pollutants, and more recently for spatial distribution maps of isotope records) over long time studies of isotopic characteristics of precipitation (Burnett et periods are continuing to assist our understanding of plant al., 2004).Combining this tool with isotopic measurements water relations, water sources and the extent to which they of continental precipitation and water vapour (e.g.networks are driven by seasonally varying water sources and how these such as MIBA and GNIP) and carbon and water fluxes (e.g.sources vary at the regional, inter-annual to inter-decadal networks such as Fluxnet) may be means by which almost scales (Rozanski et al., 1993;Welker, 2000;Vachon et al., real-time linkages between climate phases, moisture sources, 2007).Knowledge on the spatial distribution of the isotope plant water relations, carbon exchange and continental carsignatures of the source water taken up by plants is also prebon cycling may be possible.requisite to disentangle the climatic and physiological information laid down as 18 O signal in plant organic matter and 2.6 Isotopic archives and relevant aspects of isotopic archives (Augusti and Schleucher, 2007) on larger spatio-temporal integration scales.At the regional scale, we now fully appreciate that seasonally snow covered systems provide meltwater to soils Over the past decades, the use of stable isotope ratios in a and river systems that reflect the highly depleted values of wide range of materials -from tree, sediment and ice cores winter precipitation (Dutton et al., 2005;Vachon et al., 2010), to corals, hair, cactus spines, the balleen of whales and fish and that this snow meltwater allows high rates of stomatal odoliths -has provided some of the most important and novel conductance and high rates of carbon fixation (Alstad et al., insights into the patterns of past environmental changes and 1999).The duration and extent to which snowmelt and sumorganismal response to these changes of almost any type of mer precipitation sources are available to the vegetation may recorder (Dawson and Siegwolf, 2007).Such archives not be critical to supporting higher plant water use, thus affectonly provide a way to look back in time but more recent exing stomatal conductance as well as carbon fixation and gross amples show that one can also assign causes to responses ecosystem production.The complexity of seasonal patterns to environmental changes on a mechanistic basis (e.g.Ogee óf snow meltwater availability at the regional scale is reet al., 2009).Stable isotope analysis of biological or abiotic flected in the vegetation at higher latitudes where Arctic and archives can thus provide excellent tracers for spatial-and North Atlantic Oscillation phase changes are recorded in the temporal-integration over different scales.Here we discuss carbon and oxygen isotope composition (Welker et al., 2005).The advantages of tree rings are that they (i) can be reliably vary with climate phases (i.e.climate oscillations and modes, dated with a high temporal and spatial resolution; (ii) consuch as El Nino) ˜as it affects vegetation carbon fixation is tain several proxies (stable C, H, O and N isotopes, tree ring unknown (Holmgren et al., 2001;Birks and Edwards, 2009; width and tree ring density) in the same matrix (tree ring Sjostrom and Welker, 2009).We continue to recognize that wood/cellulose), which was formed at the same time, locatree rings may be recorders of the general isotopic history tion, and environmental conditions; and (iii) mostly the incluof source water (Briffa, 2000;Csank et al., 2011) regardsion of a limited number of trees and species may provide a less of geologic time period.However, understanding the strong signal.However, single tree ring chronologies provide www.biogeosciences.net/9/3083/2012/Biogeosciences, 9, 3083-3111, 2012 only limited spatial and community integration and report Micro-scale environmental record only local signals (ca. 10 −1 to 10 2 m).Signal strength is further reduced by species-specific responses to environmental A particular case of small-scale environmental records are impacts.Furthermore, the tree response is strongly affected carbon and oxygen isotope ratios of non-vascular plants by ontogeny (e.g. Monserud and Marshall, 2001) and site (NVP).Cyanobacteria, algae, lichens, and bryophytes intespecific properties such as competition, soil type, water and grate local changes of CO 2 (e.g.Maguas ´and Brugnoli, 1996) nutrient availability, resulting in a considerable variability of and water over long time periods due to their passive exthe signal expression, even within the same species (Saurer change with environmental conditions, low growth rates (ca. et al., 1997).Thus, constructions of ecosystem chronologies 0.02-30 mm a −1 ) and long life spans (hundreds to thousands depend on the combination of several tree ring records from of years).Therefore, NVP can be used, for example, for trees of different locations within the same site.This requires geochronologic aging (e.g.lichenometry), particularly dating additional information, such as knowledge of past species deposited surfaces over the past 500 years with an accuracy of 10 % error (Armstrong 2004).The 13 dynamics.
δ C of NVP archives environmental impacts over the whole life span in bulk or-Isotopic archives of herbaceous vegetation ganic material, and over several years if a chronosequence is sampled from the thallus margins or young shoots.Short-The life span of herbaceous vegetation is much shorter than term and online records can be obtained from analysing that of trees.However, isotopic reconstructions of climate respired CO 2 and extracted bulk water.2006).Additionally, epiphytic plants function as atmospheric general, herbarium specimens have been sampled at different water traps (Helliker and Griffiths, 2007;Helliker 2011).Belocations, so that long-term isotopic records from these incause epiphytic NVP are commonly in equilibrium with wavolve a spatially disperse representation of a species' changter vapour, it is inferred that 18  δ O of bulk water and organic ing isotopic composition.Because of site differences, such material might serve as a short and long-term recorder for atisotopic records display relatively high variation.Rare opmospheric vapour, respectively (Helliker and Griffiths, 2007; portunities for community-scale isotopic reconstructions are Hartard et al., 2008Hartard et al., , 2009)).In the same line, peat mosses presented by long-term (agro-) ecological experiments with serve as proxies for palaeoenvironmental changes (Loader et crops and grassland where biomass samples have been stored al., 2007;Lamentowicz et al., 2008;Moschen et al., 2009;in dedicated archives (Zhao et al., 2001;Kohler ¨et al., 2010).Loisel et al., 2010).However, approaches that use oxygen isotopes as long-term recorder of environmental conditions Grazer tissues as isotopic archives need to account for the contributions of the different water signals from rain, dew and vapour, as well as physiological For grassland, an analogy to tree rings is given by the yearly offsets which add considerable uncertainties (Moschen et al., rings (annuli) of horns (or hoofs) of obligate grazers (Bar-2009(Bar- ). bosa et al., 2009(Bar- , 2010)).These can yield carbon isotopic records over many years, which reflect that of grassland vegetation (Schnyder et al., 2006).The spatial integrations of 3 New technical and methodological developments in tree and horn ring isotope compositions are quite contrasting: stable isotope research local and stationary for the tree, and vast and cyclic for horns, reflecting visits of the different parts of the year-round graz- The past decade has seen tremendous progress in the develing grounds.Still, the use of grazer tissue for reconstructions opment of new techniques that complement or rival tradiof grassland isotopic chronologies usually rests on a number tional Isotope Ratio Mass Spectrometry (IRMS) for the deof assumptions, e.g.concerning the selectivity of grazing, the termination of stable isotope abundances.This has lead to constancy of the relationships between isotopic composition new dimensions in measurement speed, number of quantifiof grazer tissues, and contributions of diet components of dif-able isotopologues and sensitivity, increased the repeatabilferential digestibility (Wittmer et al., 2010).Such assump-ity, precision and sample turn-over considerably, and offered tions can be and should be validated (Wittmer et al., 2010).
new opportunities for in situ observations at natural abun-A significant advantage of keratinous tissue (horn, hair/wool dance and in tracer experiments.Most important for carbon and hoofs) is given by its homogenous chemical composi-and water cycle research was the introduction of instruments tion, which reduces variation associated with metabolic iso-using light absorption properties of small molecules for detope fractionation that can be significant in chemically het-termination of stable isotope abundances, as well as the introerogeneous tissue.
duction of innovative techniques for compound-specific sample extraction.

Laser absorption spectroscopy (LAS) Nuclear Magnetic Resonance (NMR)
The development of absorption spectroscopy instrumenta-At the advent of development of new techniques for nution (LAS) provided new dimensions of measurement speed clear magnetic resonance spectroscopy (NMR), new options and number of quantifiable isotopologues offering data richarise for studies of, e.g.starch-sugar partitioning and comness that had never been possible to achieve in fieldplementary information on (photo-)respiration by analyses deployable instrumentation (e.g.Bowling et al., 2003, Kam-of non-homogeneous distribution of 13 C within carbohydrate  mer et al., 2011, Sturm et al., 2012; but see Schnyder et molecules (e.g. Gilbert et al., 2009, 2011). Analogously, opal., 2004).The laser absorption spectroscopy is based on tions to distinguish between different water pools within the analysis of absorption of light in selected wavelengths in plant arise from new techniques for 18 O positional analyses the near and mid-infrared to determine the mole fractions by novel derivatiation approaches (Sternberg et al., 2006). of individual isotopologues (Kerstel, 2004;Kerstel and Gianfrani, 2008;Fried and Richter, 2006).Optical measure-Nano-scale secondary ion mass spectrometers ment methods based on Fourier Transform Infrared Spec-(NanoSIMS) troscopy (FTIR), Cavity Ringdown Spectroscopy (CRDS), Linking isotopic analysis with high resolution microscopy Integrated Cavity Output Spectroscopy (ICOS) and Tunable has provided significant progress of spatially resolved infor-Diode Laser Absorption Spectroscopy (TDLAS) now apmation on the molecular and isotopic compositions of (biproach levels of detection of small-molecule isotopologues ological) materials.New Nano-scale Secondary Ion Mass comparable to laboratory-based isotope ratio mass spectrom-Spectrometers (NanoSIMS) represent a significant improveeters (IRMS).Measurement by absorption spectroscopy is ment in sensitivity and spatial resolution (down to 50 nm).non-destructive and can therefore be repeated to increase In a destructive manner, NanoSIMS analysis involves conmeasurement precision (Werle et al., 2004).Furthermore, tinuous bombardment of the sample surface with an ion LAS enables a high temporal resolution of accurate isotope beam and subsequent analysis of the released secondary ions ratios, an ideal property for the visualisation of processes and according to their mass-to-charge ratios (Herrmann et al., temporal variability (e.g.Bowling et al., 2003;Lee et al., 2007).Although adequate sample preparation remains chal-2005; Tuzson, 2008).Further, new multi-species instruments lenging, imaging mass spectrometry via NanoSIMS reprethat are becoming available enable so-called "clumped isosents a promising avenue for mapping the spatial organisatope" measurements (Eiler, 2007), wherein the occurrence tion, metabolic pathways and resource fluxes within cells, of two heavy isotopes in the same molecule can serve as a plants and at the root-fungus-soil interface, in particular in unique stable isotope tracer itself.

Compound Specific Isotope Analysis (CSIA)
Progress and challenges The IRMS has experienced technological and methodological development, particularly Compound Specific Iso- The pool of various new and improved techniques currently tope Analysis (CSIA), which includes IRMS coupled to available for application of stable isotopes in environmen-Gas Chromatography-Combustion (GC-C-IRMS; Maier-tal, physiological and ecological research is large.However, Augenstein, 1999) or Liquid Chromatography (LC-IRMS; from the user's perspective, particularly for laser absorp- Godin et al., 2007).This facilitates the analysis of different tion spectroscopy, some general issues, as described below, compounds such as structural and labile carbohydrates ex-should be resolved.tracted from plant organs, leaf wax alkanes, phloem sap and Instrument accuracy and calibration soil fractions (see review by Sachse et al., 2012).For compound specific isotope analysis, the extraction method is cru-Calibration biases for water vapour isotope laser spectromcial and might strongly affect the results obtained (Richter eters can result, for instance, from evaporation efficiency of et al., 2009).Moreover, the need for derivatization of pothe reference water, instrument nonlinearity and impurity of lar metabolites for GC-MS (gas chromatography-mass specthe carrier gas.Calibration of water vapour analysers is done, trometry) analysis and thus the introduction of additional carfor example, using a liquid water injector ("dripper") into a bon and oxygen via the derivatization agent to the analyte flow of dry air (Lee et al., 2005;Wen et al., 2008; Baker and complicates the measurement of natural abundance stable Griffis, 2010;Griffis et al., 2010;Sturm and Knohl, 2010).In isotope composition of these compound classes (e.g.Gross addition, a heated vaporisation system is used wherein the and Glaser, 2004).This problem does not occur with LCliquid standard is completely vaporized without fractiona-IRMS systems, which are currently, however, restricted to tion.Nevertheless, any concentration dependence in the analcarbon isotope analyses.
yser itself can bias the overall calibration of the instrument, especially when measuring ambient water vapour of widely varying mixing ratios (Lee et al., 2005;Wen et al., 2008;Schmidt et al., 2010;Sturm and Knohl, 2010).For CO 2 , cal-Multi-isotopologue instruments ibration against two or more mixtures of CO 2 and dry air, which are tied to international reference standards, are criti-New instruments are becoming available and enable socal.Impurities in the water sample to be analysed can cause called "clumped isotope" measurements (Eiler, 2007), a spectral interference with organic contaminants and have wherein the occurrence of two heavy isotopes in the same been observed in analysis of liquid samples extracted from molecule can serve as a unique stable isotope tracer itself.biological sources, e.g.leaf water (West et al., 2010;Schultz Currently its applicability as a paleo-thermometer is being et al., 2011).
tested.The basic assumption is based on the observation that the heavier molecules and atoms are not randomly distributed High instrument precision at short detection intervals within the same matrix, but rather form a clumped aggregate of substrates with the heavier isotope.For the distinc-Free of sample preparation and processing, new optical techtion of such clumped isotopes highly sensitive instruments niques can achieve much faster detection than IRMS.Inare needed.situ measurements of CO 2 and H 2 O isotope ratios in ambient air, especially if made on a long-term basis and calibrated precisely, can provide a powerful tool for atmospheric 4 Outlook inverse analysis of the terrestrial carbon sink and tracking of water transport in the atmosphere.However, to measure New research opportunities at all scales of isotope biogeothe source/sink signature properly near the land surface, chemistry of carbon and water are arising from deepened one should interface the isotopic analyser with plant (Barprocess-based understanding and improved analysis tools, bour et al., 2007;Barthel et al., 2011) and soil chambers together with the development of mechanistic models.Espe-(e.g.Wingate et al., 2010a, b) and deploy it in the gradientcially combinations of multiple isotope and non-isotope varidiffusion mode either over the vegetation (Griffis et al., 2004) ables have the potential to stimulate our understanding across or over the soil surface inside the canopy (Santos et al., a wide range of scales, including leaves, plants, mesocosms, 2010), or combine it with a sonic anemometer for direct eddy natural ecosystems, and the atmosphere.The scale-spanning covariance measurement of isotopic fluxes (Lee et al., 2005; assessment of carbon and water fluxes is, on the one hand, a Griffis et al., 2008Griffis et al., , 2010) ) or landscape scale measurements great opportunity offered by stable isotope approaches.On in high elevation or airborne conditions (e.g.Tuszon et al., the other hand, deeper insights into the multitude of pro-2010).In all these configurations, suitable interfaces between cesses affecting carbon and oxygen isotope discrimination the analyser and the sample and calibration periphery are during photosynthesis and transpiration, as well as during useful.The system as a whole must be robust and designed downstream metabolic processes, are challenging a generaland tuned for minimal interference, memory effects or signal isation of the information contained in the isotopic signature drifts.Fast detection is particularly critical for eddy covariand a transfer to higher temporal and/or spatial scale.One ance applications, which require an instrument response to example is the knowledge that plant phenology or growth be faster than 10 Hz and relies on continuous-flow sampling.patterns (Sect.2.1.3)might complicate the comparison of However, fast detection is also desired for chamber based the isotopic composition of bulk material between species.measurements in studies of short-term events, such as wa-However, we can apply appropriate techniques such as the ter vapour and CO 2 flux pulses after rain (Santos et al., 2010; assessment of organic matter pools with a well defined turn- Unger et al., 2012).Maximizing precision at short integraover time and chemical composition to avoid misinterpretation times and maintaining accuracy for long periods should tion.Moreover, experimental designs focussing on changes be a high priority in future instrument development.
in environmental conditions or species interactions and on the effect of such changes on the isotopic composition can Instrument and infrastructure cost often overcome the problem.While the isotopic composition might not be directly comparable between species, the direc-High instrument and maintenance costs limit the broad adoption and magnitude of change can give quantitative information of new technologies in field research.It is highly desirtion on physiological reactions within and between species, able that the costs are brought down to a level comparable communities and ecosystems.to that of a broadband infrared gas analyser, which is now At the leaf-level (see Sect. 2.1), combined analyses of difan indispensible tool for ecosystem carbon and water flux ferent isotopes might lead to a better understanding of mesmonitoring worldwide (Baldocchi et al., 2001).We envision ophyll conductance and related components, including difthe development of a network with real-time observations of fusion through intercellular airspaces and transport through isotopic fluxes of CO 2 and H 2 O to help diagnose changes barriers in cells such as the cell wall, membranes, or stroma.in biospheric processes.This can become a realistic goal if It might also help to assess the possible role of cooporins instrumentation costs are lower.
(membrane proteins acting as pores for CO 2 ) in facilitating and controlling transport of CO 2 .Combined measurements of the isotopologues of CO 2 and H 2 O will further allow It is, therefore, not surprising that the interpretation quantifying the extent of equilibration between dissolved of ecosystem scale fractionation remains challenging (see CO 2 and leaf water, and thus can provide a non-invasive re-Sect.2.4).We expect that significant steps for resolving this construction of leaf water dynamics.These are critical as-complexity will include similar approaches as advocated for pects for validation and further development of carbon and leaf-and plant-level studies: (i) joint flux measurements of water isotope approaches and models.Information on differ-the different isotopologues of CO 2 and H 2 O in natural sysent species and ecotypes will in turn enhance our understand-tems -which will enable a better distinction of the CO 2 and ing of the different morpho-physiological factors controlling H 2 O flux components and pools, (ii) tracing metabolite and carbon and water fluxes and, hence, water use efficiency of intramolecular labelling patterns between and within system leaves.
components in artificial setups as well as in field labelling ex-Although the last ten years have seen a large increase in periments -shedding light on allocation, turnover of differknowledge of post-carboxylation fractionation phenomena ent carbon pools as well as plant-soil-atmosphere interaction, (see Sect. 2.1.2),we expect no slowdown in the develop-and (iii) hypothesis-testing mesocosm-scale experimentsment of this field.In part, empirical progress will be faciltesting our system-scale understanding.Insights from these itated by improvements in NMR technology as well as in approaches may then help to improve and test stable-isotopederivatisation techniques (see Sect. 3) which permit mea-enabled models of carbon and water fluxes at the ecosystem surements of natural intramolecular isotope distribution pat-scale.terns in intermediates of primary and secondary metabolism, Regional-scale studies (see Sect. 2.5) of the water isoand respiratory substrates.Dynamic labelling experiments tope cycle are becoming more important to our understandwith 13 C-enriched or depleted CO 2 or with (intra-molecular) ing of synoptic climates, ecosystem processes, the role of position-labelled metabolites will permit better assessment abiotic processes (e.g.temperature of condensation), moisof metabolic networks and turnover times of different car-ture sources, and storm tracks on the ecohydrology of enbon pools.Such work will also enhance our understanding tire landscapes and continents.However, isotope fractionfor the metabolic causes of variations in post-carboxylation ations are quite uncertain on global and continental scales fractionation.Temporal dynamics of apparent fractionation and it is therefore important to identify robust features that during dark respiration may vary, depending on the identity can be constrained by large-scale isotope observations.C 4of the different metabolic intermediates, their synthesis path-plant distribution is one such feature that might become well ways and metabolic functions as well as on the demand for constrained by 13 C isotopes.But isotope studies will benefit substrates in the respiratory pathways.
greatly from the combination with other non-isotope tracers Investigations of natural intra-molecular 13 C and 18 O dis-also on landscape, regional and continental scales.It might tribution patterns might also be key to quantify isotope fracbe the tapping into the above-mentioned multitude of infortionation phenomena during loading, phloem transport and mation that will advance the usage of isotope signals on the unloading of different organic compounds (see Sect. 2.2).
global scale.These include assessments of isotopic exchange reactions In conclusion, we are in the midst of a rapid growth in along the path from leaves to sites of assimilate use, and process-based understanding of the behaviour of carbon and fractionation or isotopic exchange during biosynthetic pro-oxygen stable isotopes in organisms and in the environment.cesses such as cellulose synthesis.Such approaches may On the one hand, we are increasingly recognising the comelucidate the mechanisms underlying spatio-temporal variplexity of 13 C and 18 O fractionation processes and their spaation of 13 C 18 δ and δ O during transfer from the chloro-tial and temporal variation.On the other hand, new technoloplast to heterotrophic tissues, the rhizosphere/soil and atmo-gies (see Sect. 3) can deliver high resolution records of shortsphere.The mechanistic understanding, on the other hand, and long-term variability in isotope signatures, overcoming will strengthen climatological and physiological interpreta-the constraints of earlier laborious procedures.New analytition of tree ring cellulose and similar isotope archives such cal tools and process-based understanding will allow further as grass, sediments, hair, horn, or tooth enamel of herbivores development of isotope-enabled biogeochemical models for (see Sect. 2.6).We are of the strong opinion that a deeper investigations of the complex interplay of soil, plant, ecosysknowledge of fractionation during photosynthesis, transport tem and atmosphere processes in the carbon and water cyand post-carboxylation metabolism is an important basis for cles.understanding ecosystem-scale isotope discrimination and for linking the carbon balance with water relations at different scales.Whereas the mechanistic understanding of photosynthetic carbon isotope fractionation and evaporative 18 O enrichment of water in leaves is relatively advanced, equivalent understanding of fractionation phenomena in the downstream metabolism -as expressed in quantitative models -is still in its infancy.
Photosynthetic carbon isotope discrimination C b alone.The conceptual model of Scheidegger et al. (2000) was Recent results of Gilbert et al. (2011) demonstrated that acsuccessfully applied in the field (Keitel et al., 2003, Sullivan cess to site-specific isotope fractionation is now possible us-13 and Welker, 2007) and further adapted for air pollution studing C NMR (see Sect. 3) to directly determine intramolec-13 ies evaluating the effect of NO x on plant metabolism (Siegular C distributions at natural abundance.The challenge in wolf et al. 18progress and challenges of a few selected examples of bio-Our emerging understanding of the temporal patterns of δ O logical archives.and δD during swings in the major climate oscillations provides a modern basis for calibrating storm track, climate os-Progress and challenges cillations and the source water of vegetation in conjunction with carbon fixation rates and variability.Isotopic archives in treesProgress and challengesTree rings enable retrospective analyses of intra-and interannual variation of carbon and oxygen isotope composi-Tracing climate phase variation tion and the related ecophysiological drivers over many centuries(Sidorova et al., 2009;Nock et al., 2010;  Knorre et Understanding the role of moisture sourced from multiple al., 2010; Andreu-Hayles et al., 2011; Penuelas ˜et al., 2011).regions (i.e.different storm tracks), and how those sources

Table 2 .
Introduction to terms and equations of oxygen and hydrogen isotopes and evaporative enrichment. 2 H (or D; deuwhere y X is replaced by18O or 2H in the case of oxygen and hydroterium), oxygen possesses the isotopes 16 O, 17 O and18O.Since the natgen isotopes and R sample and R standard are the measured18 O 16 RThe delta notation for oxygen and hydrogen isotopes y = R Whereas hydrogen has two stable isotopes, 1 standard H and

.2 13 C and 18 O isotopes to trace plant integrated the
intra-annual analysis of tree ring, whole wood or cellu-

processes and plant-soil coupling lose
(Helle and Schleser, 2004dy periods during the growing season when the isotopic signature in this archive is directly coupled to leaf physiology(Helle and Schleser, 2004; Offer-With their pioneering work on the phloem carbon isotopic mann et al., 2011).composition of grasses (Yoneyama et al., 1997) and trees (Pate and Arthur, 1998), two groups paved the way to getting Interpretation

3111, 2012 C. Werner et al.: Progress and challenges in using stable isotopes Holbrook
Thompson and www.biogeosciences.net/9/3083/2012/Biogeosciences, 9, 3083- , carbon allocation patterns (Kuptz et al., tion, which can represent 20 to 70 % of total ecosystem res-2011; Epron et al., 2011), and shading impacts (Warren piration, is an integrative signal driven by many abiotic and et al., 2012), to name a few.Quantitative methods of biological processes.Recent studies have shown that factors canopy labelling in connection with on-line tracer mea-such as diffusivity of soil CO 2 , dissolution of CO 2 from bisurement techniques (Sect.3) and modelling of the tracer carbonates, and advection of soil gas may be responsible for distribution data (e.g. by compartmental analysis), holds strong 13 C-isotopic disequilibria between the CO 2 efflux at the promise of testing hypotheses of ecosystem physiology, the soil surface and concurrent soil respiration (Crow et al., aboveground-belowground response to a changing climate, 13 δ C of NVP can be change in herbaceous vegetation (crops and grassland) are used to trace environmental CO 2 gradients (Flanagan et al., possible if plants were sampled and preserved during the 1999; Lakatos et al., 2007; Meyer et al., 2008), whereas fossil epoch.Such archives are relatively rare and are mainly reprebryophytes record ancient CO 2 levels (Fletcher et al., 2005, sented by herbaria (e.g.Penuelas and Azcon-Bieto, ´1992).In